MyArxiv
Robotics 41
☆ A Passive Elastic-Folding Mechanism for Stackable Airdrop Sensors ICRA 2026
Air-dispersed sensor networks deployed from aerial robotic systems (e.g., UAVs) provide a low-cost approach to wide-area environmental monitoring. However, existing methods often rely on active actuators for mid-air shape or trajectory control, increasing both power consumption and system cost. Here, we introduce a passive elastic-folding hinge mechanism that transforms sensors from a flat, stackable form into a three-dimensional structure upon release. Hinges are fabricated by laminating commercial sheet materials with rigid printed circuit boards (PCBs) and programming fold angles through a single oven-heating step, enabling scalable production without specialized equipment. Our geometric model links laminate geometry, hinge mechanics, and resulting fold angle, providing a predictive design methodology for target configurations. Laboratory tests confirmed fold angles between 10 degrees and 100 degrees, with a standard deviation of 4 degrees and high repeatability. Field trials further demonstrated reliable data collection and LoRa transmission during dispersion, while the Horizontal Wind Model (HWM)-based trajectory simulations indicated strong potential for wide-area sensing exceeding 10 km.
comment: 8 pages, 8 figures, The 2026 IEEE International Conference on Robotics and Automation (ICRA 2026)
☆ V-Dreamer: Automating Robotic Simulation and Trajectory Synthesis via Video Generation Priors
Training generalist robots demands large-scale, diverse manipulation data, yet real-world collection is prohibitively expensive, and existing simulators are often constrained by fixed asset libraries and manual heuristics. To bridge this gap, we present V-Dreamer, a fully automated framework that generates open-vocabulary, simulation-ready manipulation environments and executable expert trajectories directly from natural language instructions. V-Dreamer employs a novel generative pipeline that constructs physically grounded 3D scenes using large language models and 3D generative models, validated by geometric constraints to ensure stable, collision-free layouts. Crucially, for behavior synthesis, we leverage video generation models as rich motion priors. These visual predictions are then mapped into executable robot trajectories via a robust Sim-to-Gen visual-kinematic alignment module utilizing CoTracker3 and VGGT. This pipeline supports high visual diversity and physical fidelity without manual intervention. To evaluate the generated data, we train imitation learning policies on synthesized trajectories encompassing diverse object and environment variations. Extensive evaluations on tabletop manipulation tasks using the Piper robotic arm demonstrate that our policies robustly generalize to unseen objects in simulation and achieve effective sim-to-real transfer, successfully manipulating novel real-world objects.
comment: 8 pages, 6 figures
☆ "You've got a friend in me": Co-Designing a Peer Social Robot for Young Newcomers' Language and Cultural Learning
Community literacy programs supporting young newcomer children in Canada face limited staffing and scarce one-to-one time, which constrains personalized English and cultural learning support. This paper reports on a co-design study with United for Literacy tutors that informed Maple, a table-top, peer-like Socially Assistive Robot (SAR) designed as a practice partner within tutor-mediated sessions. From shadowing and co-design interviews, we derived newcomer-specific requirements and added them in an integrated prototype that uses short story-based activities, multi-modal scaffolding (speech, facial feedback, gesture), and embedded quizzes that support attention while producing tutor-actionable formative signals. We contribute system design implications for tutor-in-the-loop SARs supporting language socialization in community settings and outline directions for child-centered evaluation in authentic programs.
☆ ViTac-Tracing: Visual-Tactile Imitation Learning of Deformable Object Tracing ICRA2026
Deformable objects often appear in unstructured configurations. Tracing deformable objects helps bringing them into extended states and facilitating the downstream manipulation tasks. Due to the requirements for object-specific modeling or sim-to-real transfer, existing tracing methods either lack generalizability across different categories of deformable objects or struggle to complete tasks reliably in the real world. To address this, we propose a novel visual-tactile imitation learning method to achieve one-dimensional (1D) and two-dimensional (2D) deformable object tracing with a unified model. Our method is designed from both local and global perspectives based on visual and tactile sensing. Locally, we introduce a weighted loss that emphasizes actions maintaining contact near the center of the tactile image, improving fine-grained adjustment. Globally, we propose a tracing task loss that helps the policy to regulate task progression. On the hardware side, to compensate for the limited features extracted from visual information, we integrate tactile sensing into a low-cost teleoperation system considering both the teleoperator and the robot. Extensive ablation and comparative experiments on diverse 1D and 2D deformable objects demonstrate the effectiveness of our approach, achieving an average success rate of 80% on seen objects and 65% on unseen objects.
comment: The paper has been accepted by ICRA2026
☆ Empathetic Motion Generation for Humanoid Educational Robots via Reasoning-Guided Vision--Language--Motion Diffusion Architecture
This article suggests a reasoning-guided vision-language-motion diffusion framework (RG-VLMD) for generating instruction-aware co-speech gestures for humanoid robots in educational scenarios. The system integrates multi-modal affective estimation, pedagogical reasoning, and teaching-act-conditioned motion synthesis to enable adaptive and semantically consistent robot behavior. A gated mixture-of-experts model predicts Valence/Arousal from input text, visual, and acoustic features, which then mapped to discrete teaching-act categories through an affect-driven policy.These signals condition a diffusion-based motion generator using clip-level intent and frame-level instructional schedules via additive latent restriction with auxiliary action-group supervision. Compared to a baseline diffusion model, our proposed method produces more structured and distinctive motion patterns, as verified by motion statics and pairwise distance analysis. Generated motion sequences remain physically plausible and can be retargeted to a NAO robot for real-time execution. The results reveal that reasoning-guided instructional conditioning improves gesture controllability and pedagogical expressiveness in educational human-robot interaction.
☆ ROFT-VINS: Robust Feature Tracking-based Visual-Inertial State Estimation for Harsh Environment
SLAM (Simultaneous Localization and Mapping) and Odometry are important systems for estimating the position of mobile devices, such as robots and cars, utilizing one or more sensors. Particularly in camera-based SLAM or Odometry, effectively tracking visual features is important as it significantly impacts system performance. In this paper, we propose a method that leverages deep learning to robustly track visual features in monocular camera images. This method operates reliably even in textureless environments and situations with rapid lighting changes. Additionally, we evaluate the performance of our proposed method by integrating it into VINS-Fusion (Monocular-Inertial), a commonly used Visual-Inertial Odometry (VIO) system.
comment: 6 pages, published ICCAS 2024
☆ CSSDF-Net: Safe Motion Planning Based on Neural Implicit Representations of Configuration Space Distance Field
High-dimensional manipulator operation in unstructured environments requires a differentiable, scene-agnostic distance query mechanism to guide safe motion generation. Existing geometric collision checkers are typically non-differentiable, while workspace-based implicit distance models are hindered by the highly nonlinear workspace--configuration mapping and often suffer from poor convergence; moreover, self-collision and environment collision are commonly handled as separate constraints. We propose Configuration-Space Signed Distance Field-Net (CSSDF-Net), which learns a continuous signed distance field directly in configuration space to provide joint-space distance and gradient queries under a unified geometric notion of safety. To enable zero-shot generalization without environment-specific retraining, we introduce a spatial-hashing-based data generation pipeline that encodes robot-centric geometric priors and supports efficient retrieval of risk configurations for arbitrary obstacle point sets. The learned distance field is integrated into safety-constrained trajectory optimization and receding-horizon MPC, enabling both offline planning and online reactive avoidance. Experiments on a planar arm and a 7-DoF manipulator demonstrate stable gradients, effective collision avoidance in static and dynamic scenes, and practical inference latency for large-scale point-cloud queries, supporting deployment in previously unseen environments.
☆ REST: Receding Horizon Explorative Steiner Tree for Zero-Shot Object-Goal Navigation
Zero-shot object-goal navigation (ZSON) requires navigating unknown environments to find a target object without task-specific training. Prior hierarchical training-free solutions invest in scene understanding (\textit{belief}) and high-level decision-making (\textit{policy}), yet overlook the design of \textit{option}, i.e., a subgoal candidate proposed from evolving belief and presented to policy for selection. In practice, options are reduced to isolated waypoints scored independently: single destinations hide the value gathered along the journey; an unstructured collection obscures the relationships among candidates. Our insight is that the option space should be a \textit{tree of paths}. Full paths expose en-route information gain that destination-only scoring systematically neglects; a tree of shared segments enables coarse-to-fine LLM reasoning that dismisses or pursues entire branches before examining individual leaves, compressing the combinatorial path space into an efficient hierarchy. We instantiate this insight in \textbf{REST} (Receding Horizon Explorative Steiner Tree), a training-free framework that (1) builds an explicit open-vocabulary 3D map from online RGB-D streams; (2) grows an agent-centric tree of safe and informative paths as the option space via sampling-based planning; and (3) textualizes each branch into a spatial narrative and selects the next-best path through chain-of-thought LLM reasoning. Across the Gibson, HM3D, and HSSD benchmarks, REST consistently ranks among the top methods in success rate while achieving the best or second-best path efficiency, demonstrating a favorable efficiency-success balance.
☆ Benchmarking Visual Feature Representations for LiDAR-Inertial-Visual Odometry Under Challenging Conditions
Accurate localization in autonomous driving is critical for successful missions including environmental mapping and survivor searches. In visually challenging environments, including low-light conditions, overexposure, illumination changes, and high parallax, the performance of conventional visual odometry methods significantly degrade undermining robust robotic navigation. Researchers have recently proposed LiDAR-inertial-visual odometry (LIVO) frameworks, that integrate LiDAR, IMU, and camera sensors, to address these challenges. This paper extends the FAST-LIVO2-based framework by introducing a hybrid approach that integrates direct photometric methods with descriptor-based feature matching. For the descriptor-based feature matching, this work proposes pairs of ORB with the Hamming distance, SuperPoint with SuperGlue, SuperPoint with LightGlue, and XFeat with the mutual nearest neighbor. The proposed configurations are benchmarked by accuracy, computational cost, and feature tracking stability, enabling a quantitative comparison of the adaptability and applicability of visual descriptors. The experimental results reveal that the proposed hybrid approach outperforms the conventional sparse-direct method. Although the sparse-direct method often fails to converge in regions where photometric inconsistency arises due to illumination changes, the proposed approach still maintains robust performance under the same conditions. Furthermore, the hybrid approach with learning-based descriptors enables robust and reliable visual state estimation across challenging environments.
comment: 14 pages, Publised IEEE Access2026
☆ TiBCLaG: A Trigger-induced Bistable Compliant Laparoscopic Grasper
Industrial laparoscopic graspers use multi-link rigid mechanisms manufactured to tight tolerances, resulting in high manufacturing and assembly costs. This work presents the design and proof-of-concept validation of a monolithic, fully compliant, bistable, laparoscopic grasper that eliminates the need for multiple rigid links, thereby reducing part count. The device integrates a compliant trigger and a compliant gripper end-effector, coupled via a control push-rod, to achieve stable grasping without continuous user input. The trigger mechanism is synthesized using a Two-Element Beam Constraint Model as a design framework to control the deformation and stiffness of V-beam-like elements. This technique enables elastic energy storage while preventing snap-through instability. The end-effector is designed as a compliant gripper to achieve adaptive grasping through elastic deformation. Jaws' opening-and-closing performance is demonstrated using nonlinear finite element analysis. The laparoscopic design presented here is fabricated using fused deposition 3D printing. The fabricated prototype demonstrates reliable bistable actuation, confirming the feasibility of such compliant laparoscopic grasper architectures.
comment: 17 pages, 13 figures
☆ Inductance-Based Force Self-Sensing in Fiber-Reinforced Pneumatic Twisted-and-Coiled Actuators
Fiber-reinforced pneumatic twisted-and-coiled actuators (FR-PTCAs) offer high power density and compliance but their strong hysteresis and lack of intrinsic proprioception limit effective closed-loop control. This paper presents a self-sensing FR-PTCA integrated with a conductive nickel wire that enables intrinsic force estimation and indirect displacement inference via inductance feedback. Experimental characterization reveals that the inductance of the actuator exhibits a deterministic, low-hysteresis inductance-force relationship at constant pressures, in contrast to the strongly hysteretic inductance-length behavior. Leveraging this property, this paper develops a parametric self-sensing model and a nonlinear hybrid observer that integrates an Extended Kalman Filter (EKF) with constrained optimization to resolve the ambiguity in the inductance-force mapping and estimate actuator states. Experimental results demonstrate that the proposed approach achieves force estimation accuracy comparable to that of external load cells and maintains robust performance under varying load conditions.
☆ HEP Statistical Inference for UAV Fault Detection: CLs, LRT, and SBI Applied to Blade Damage
This paper transfers three statistical methods from particle physics to multirotor propeller fault detection: the likelihood ratio test (LRT) for binary detection, the CLs modified frequentist method for false alarm rate control, and sequential neural posterior estimation (SNPE) for quantitative fault characterization. Operating on spectral features tied to rotor harmonic physics, the system returns three outputs: binary detection, controlled false alarm rates, and calibrated posteriors over fault severity and motor location. On UAV-FD, a hexarotor dataset of 18 real flights with 5% and 10% blade damage, leave-one-flight-out cross-validation gives AUC 0.862 +/- 0.007 (95% CI: 0.849--0.876), outperforming CUSUM (0.708 +/- 0.010), autoencoder (0.753 +/- 0.009), and LSTM autoencoder (0.551). At 5% false alarm rate the system detects 93% of significant and 81% of subtle blade damage. On PADRE, a quadrotor platform, AUC reaches 0.986 after refitting only the generative models. SNPE gives a full posterior over fault severity (90% credible interval coverage 92--100%, MAE 0.012), so the output includes uncertainty rather than just a point estimate or fault flag. Per-flight sequential detection achieves 100% fault detection with 94% overall accuracy.
comment: 12 Pages, 8 Figures
☆ Scaling Sim-to-Real Reinforcement Learning for Robot VLAs with Generative 3D Worlds
The strong performance of large vision-language models (VLMs) trained with reinforcement learning (RL) has motivated similar approaches for fine-tuning vision-language-action (VLA) models in robotics. Many recent works fine-tune VLAs directly in the real world to avoid addressing the sim-to-real gap. While real-world RL circumvents sim-to-real issues, it inherently limits the generality of the resulting VLA, as scaling scene and object diversity in the physical world is prohibitively difficult. This leads to the paradoxical outcome of transforming a broadly pretrained model into an overfitted, scene-specific policy. Training in simulation can instead provide access to diverse scenes, but designing those scenes is also costly. In this work, we show that VLAs can be RL fine-tuned without sacrificing generality and with reduced labor by leveraging 3D world generative models. Using these models together with a language-driven scene designer, we generate hundreds of diverse interactive scenes containing unique objects and backgrounds, enabling scalable and highly parallel policy learning. Starting from a pretrained imitation baseline, our approach increases simulation success from 9.7% to 79.8% while achieving a 1.25$\times$ speedup in task completion time. We further demonstrate successful sim-to-real transfer enabled by the quality of the generated digital twins together with domain randomization, improving real-world success from 21.7% to 75% and achieving a 1.13$\times$ speedup. Finally, we further highlight the benefits of leveraging the effectively unlimited data from 3D world generative models through an ablation study showing that increasing scene diversity directly improves zero-shot generalization.
☆ Robotic Agentic Platform for Intelligent Electric Vehicle Disassembly
Electric vehicles (EV) create an urgent need for scalable battery recycling, yet disassembly of EV battery packs remains largely manual due to high design variability. We present our Robotic Agentic Platform for Intelligent Disassembly (RAPID), designed to investigate perception-driven manipulation, flexible automation, and AI-assisted robot programming in realistic recycling scenarios. The system integrates a gantry-mounted industrial manipulator, RGB-D perception, and an automated nut-running tool for fastener removal on a full-scale EV battery pack. An open-vocabulary object detection pipeline achieves 0.9757 mAP50, enabling reliable identification of screws, nuts, busbars, and other components. We experimentally evaluate (n=204) three one-shot fastener removal strategies: taught-in poses (97% success rate, 24 min duration), one-shot vision execution (57%, 29 min), and visual servoing (83%, 36 min), comparing success rate and disassembly time for the battery's top cover fasteners. To support flexible interaction, we introduce agentic AI specifications for robotic disassembly tasks, allowing LLM agents to translate high-level instructions into robot actions through structured tool interfaces and ROS services. We evaluate SmolAgents with GPT-4o-mini and Qwen 3.5 9B/4B on edge hardware. Tool-based interfaces achieve 100% task completion, while automatic ROS service discovery shows 43.3% failure rates, highlighting the need for structured robot APIs for reliable LLM-driven control. This open-source platform enables systematic investigation of human-robot collaboration, agentic robot programming, and increasingly autonomous disassembly workflows, providing a practical foundation for research toward scalable robotic battery recycling.
☆ Computationally Efficient Density-Driven Optimal Control via Analytical KKT Reduction and Contractive MPC
Efficient coordination for collective spatial distribution is a fundamental challenge in multi-agent systems. Prior research on Density-Driven Optimal Control (D2OC) established a framework to match agent trajectories to a desired spatial distribution. However, implementing this as a predictive controller requires solving a large-scale Karush-Kuhn-Tucker (KKT) system, whose computational complexity grows cubically with the prediction horizon. To resolve this, we propose an analytical structural reduction that transforms the T-horizon KKT system into a condensed quadratic program (QP). This formulation achieves O(T) linear scalability, significantly reducing the online computational burden compared to conventional O(T^3) approaches. Furthermore, to ensure rigorous convergence in dynamic environments, we incorporate a contractive Lyapunov constraint and prove the Input-to-State Stability (ISS) of the closed-loop system against reference propagation drift. Numerical simulations verify that the proposed method facilitates rapid density coverage with substantial computational speed-up, enabling long-horizon predictive control for large-scale multi-agent swarms.
☆ MemoAct: Atkinson-Shiffrin-Inspired Memory-Augmented Visuomotor Policy for Robotic Manipulation
Memory-augmented robotic policies are essential in handling memory-dependent tasks. However, existing approaches typically rely on simple observation window extensions, struggling to simultaneously achieve precise task state tracking and robust long-horizon retention. To overcome these challenges, inspired by the Atkinson-Shiffrin memory model, we propose MemoAct, a hierarchical memory-based policy that leverages distinct memory tiers to tackle specific bottlenecks. Specifically, lossless short-term memory ensures precise task state tracking, while compressed long-term memory enables robust long-horizon retention. To enrich the evaluation landscape, we construct MemoryRTBench based on RoboTwin 2.0, specifically tailored to assess policy capabilities in task state tracking and long-horizon retention. Extensive experiments across simulated and real-world scenarios demonstrate that MemoAct achieves superior performance compared to both existing Markovian baselines and history-aware policies. The project page is \href{https://tlf-tlf.github.io/MemoActPage/}{available}.
☆ Fundamental Limits for Sensor-Based Control via the Gibbs Variational Principle
Fundamental limits on the performance of feedback controllers are essential for benchmarking algorithms, guiding sensor selection, and certifying task feasibility -- yet few general-purpose tools exist for computing them. Existing information-theoretic approaches overestimate the information a sensor must provide by evaluating it against the uncontrolled system, producing bounds that degrade precisely when feedback is most valuable. We derive a lower bound on the minimum expected cost of any causal feedback controller under partial observations by applying the Gibbs variational principle to the joint path measure over states and observations. The bound applies to nonlinear, nonholonomic, and hybrid dynamics with unbounded costs and admits a self-consistent refinement: any good controller concentrates the state, which limits the information the sensor can extract, which tightens the bound. The resulting fixed-point equation has a unique solution computable by bisection, and we provide conditions under which the free energy minimization is provably convex, yielding a certifiably correct numerical bound. On a nonlinear Dubins car tracking problem, the self-consistent bound captures most of the optimal cost across sensor noise levels, while the open-loop variant is vacuous at low noise.
comment: 6 pages, 1 figure
☆ Efficient and Versatile Quadrupedal Skating: Optimal Co-design via Reinforcement Learning and Bayesian Optimization
In this paper, we present a hardware-control co-design approach that enables efficient and versatile roller skating on quadrupedal robots equipped with passive wheels. Passive-wheel skating reduces leg inertia and improves energy efficiency, particularly at high speeds. However, the absence of direct wheel actuation tightly couples mechanical design and control. To unlock the full potential of this modality, we formulate a bilevel optimization framework: an upper-level Bayesian Optimization searches the mechanical design space, while a lower-level Reinforcement Learning trains a motor control policy for each candidate design. The resulting design-policy pairs not only outperform human-engineered baselines, but also exhibit versatile behaviors such as hockey stop (rapid braking by turning sideways to maximize friction) and self-aligning motion (automatic reorientation to improve energy efficiency in the direction of travel), offering the first system-level study of dynamic skating motion on quadrupedal robots.
☆ Graph-of-Constraints Model Predictive Control for Reactive Multi-agent Task and Motion Planning ICRA 2026
Sequences of interdependent geometric constraints are central to many multi-agent Task and Motion Planning (TAMP) problems. However, existing methods for handling such constraint sequences struggle with partially ordered tasks and dynamic agent assignments. They typically assume static assignments and cannot adapt when disturbances alter task allocations. To overcome these limitations, we introduce Graph-of-Constraints Model Predictive Control (GoC-MPC), a generalized sequence-of-constraints framework integrated with MPC. GoC-MPC naturally supports partially ordered tasks, dynamic agent coordination, and disturbance recovery. By defining constraints over tracked 3D keypoints, our method robustly solves diverse multi-agent manipulation tasks-coordinating agents and adapting online from visual observations alone, without relying on training data or environment models. Experiments demonstrate that GoC-MPC achieves higher success rates, significantly faster TAMP computation, and shorter overall paths compared to recent baselines, establishing it as an efficient and robust solution for multi-agent manipulation under real-world disturbances. Our supplementary video and code can be found at https://sites.google.com/view/goc-mpc/home .
comment: 8 main content pages, 4 main content figures, camera ready version submitted to IEEE International Conference on Robotics and Automation (ICRA 2026)
☆ RE-SAC: Disentangling aleatoric and epistemic risks in bus fleet control: A stable and robust ensemble DRL approach
Bus holding control is challenging due to stochastic traffic and passenger demand. While deep reinforcement learning (DRL) shows promise, standard actor-critic algorithms suffer from Q-value instability in volatile environments. A key source of this instability is the conflation of two distinct uncertainties: aleatoric uncertainty (irreducible noise) and epistemic uncertainty (data insufficiency). Treating these as a single risk leads to value underestimation in noisy states, causing catastrophic policy collapse. We propose a robust ensemble soft actor-critic (RE-SAC) framework to explicitly disentangle these uncertainties. RE-SAC applies Integral Probability Metric (IPM)-based weight regularization to the critic network to hedge against aleatoric risk, providing a smooth analytical lower bound for the robust Bellman operator without expensive inner-loop perturbations. To address epistemic risk, a diversified Q-ensemble penalizes overconfident value estimates in sparsely covered regions. This dual mechanism prevents the ensemble variance from misidentifying noise as a data gap, a failure mode identified in our ablation study. Experiments in a realistic bidirectional bus corridor simulation demonstrate that RE-SAC achieves the highest cumulative reward (approx. -0.4e6) compared to vanilla SAC (-0.55e6). Mahalanobis rareness analysis confirms that RE-SAC reduces Oracle Q-value estimation error by up to 62% in rare out-of-distribution states (MAE of 1647 vs. 4343), demonstrating superior robustness under high traffic variability.
☆ Contact Status Recognition and Slip Detection with a Bio-inspired Tactile Hand
Stable and reliable grasp is critical to robotic manipulations especially for fragile and glazed objects, where the grasp force requires precise control as too large force possibly damages the objects while small force leads to slip and fall-off. Although it is assumed the objects to manipulate is grasped firmly in advance, slip detection and timely prevention are necessary for a robot in unstructured and universal environments. In this work, we addressed this issue by utilizing multimodal tactile feedback from a five-fingered bio-inspired hand. Motivated by human hands, the tactile sensing elements were distributed and embedded into the soft skin of robotic hand, forming 24 tactile channels in total. Different from the threshold method that was widely employed in most existing works, we converted the slip detection problem to contact status recognition in combination with binning technique first and then detected the slip onset time according to the recognition results. After the 24-channel tactile signals passed through discrete wavelet transform, 17 features were extracted from different time and frequency bands. With the optimal 120 features employed for status recognition, the test accuracy reached 96.39% across three different sliding speeds and six kinds of materials. When applied to four new unseen materials, a high accuracy of 91.95% was still achieved, which further validated the generalization of our proposed method. Finally, the performance of slip detection is verified based on the trained model of contact status recognition.
comment: 7 pages, 9 figures
♻ ☆ Multimodal Fused Learning for Solving the Generalized Traveling Salesman Problem in Robotic Task Planning
Effective and efficient task planning is essential for mobile robots, especially in applications like warehouse retrieval and environmental monitoring. These tasks often involve selecting one location from each of several target clusters, forming a Generalized Traveling Salesman Problem (GTSP) that remains challenging to solve both accurately and efficiently. To address this, we propose a Multimodal Fused Learning (MMFL) framework that leverages both graph and image-based representations to capture complementary aspects of the problem, and learns a policy capable of generating high-quality task planning schemes in real time. Specifically, we first introduce a coordinate-based image builder that transforms GTSP instances into spatially informative representations. We then design an adaptive resolution scaling strategy to enhance adaptability across different problem scales, and develop a multimodal fusion module with dedicated bottlenecks that enables effective integration of geometric and spatial features. Extensive experiments show that our MMFL approach significantly outperforms state-of-the-art methods across various GTSP instances while maintaining the computational efficiency required for real-time robotic applications. Physical robot tests further validate its practical effectiveness in real-world scenarios.
comment: 14 pages, 6 figures, under review
♻ ☆ From Vocal Instructions to Household Tasks: The Inria TIAGo++ in the euROBIN Service Robots Coopetition
This paper describes the Inria team's integrated robotics system used in the 1st euROBIN \textit{coopetition}, during which service robots performed voice-activated household tasks in a kitchen setting. The team developed a modified TIAGo++ platform that leverages a whole-body control stack for autonomous and teleoperated modes, and an LLM-based pipeline for instruction understanding and task planning. The key contributions (opens-sourced) are the integration of these components and the design of custom teleoperation devices, addressing practical challenges in the deployment of service robots.
♻ ☆ TrajBooster: Boosting Humanoid Whole-Body Manipulation via Trajectory-Centric Learning
Recent Vision-Language-Action models show potential to generalize across embodiments but struggle to quickly align with a new robot's action space when high-quality demonstrations are scarce, especially for bipedal humanoids. We present TrajBooster, a cross-embodiment framework that leverages abundant wheeled-humanoid data to boost bipedal VLA. Our key idea is to use end-effector trajectories as a morphology-agnostic interface. TrajBooster (i) extracts 6D dual-arm end-effector trajectories from real-world wheeled humanoids, (ii) retargets them in simulation to Unitree G1 with a whole-body controller trained via a heuristic-enhanced harmonized online DAgger to lift low-dimensional trajectory references into feasible high-dimensional whole-body actions, and (iii) forms heterogeneous triplets that couple source vision/language with target humanoid-compatible actions to post-pre-train a VLA, followed by only 10 minutes of teleoperation data collection on the target humanoid domain. Deployed on Unitree G1, our policy achieves beyond-tabletop household tasks, enabling squatting, cross-height manipulation, and coordinated whole-body motion with markedly improved robustness and generalization. Results show that TrajBooster allows existing wheeled-humanoid data to efficiently strengthen bipedal humanoid VLA performance, reducing reliance on costly same-embodiment data while enhancing action space understanding and zero-shot skill transfer capabilities. For more details, For more details, please refer to our \href{https://jiachengliu3.github.io/TrajBooster/}.
♻ ☆ Accelerated Multi-Modal Motion Planning Using Context-Conditioned Diffusion Models ICRA 2026
Classical methods in robot motion planning, such as sampling-based and optimization-based methods, often struggle with scalability towards higher-dimensional state spaces and complex environments. Diffusion models, known for their capability to learn complex, high-dimensional and multi-modal data distributions, provide a promising alternative when applied to motion planning problems and have already shown interesting results. However, most of the current approaches train their model for a single environment, limiting their generalization to environments not seen during training. The techniques that do train a model for multiple environments rely on a specific camera to provide the model with the necessary environmental information and therefore always require that sensor. To effectively adapt to diverse scenarios without the need for retraining, this research proposes Context-Aware Motion Planning Diffusion (CAMPD). CAMPD leverages a classifier-free denoising probabilistic diffusion model, conditioned on sensor-agnostic contextual information. An attention mechanism, integrated in the well-known U-Net architecture, conditions the model on an arbitrary number of contextual parameters. CAMPD is evaluated on a 7-DoF robot manipulator and benchmarked against state-of-the-art approaches on real-world tasks, showing its ability to generalize to unseen environments and generate high-quality, multi-modal trajectories, at a fraction of the time required by existing methods.
comment: Accepted for publication at the 2026 IEEE International Conference on Robotics & Automation (ICRA 2026)
♻ ☆ RoboForge: Physically Optimized Text-guided Whole-Body Locomotion for Humanoids
While generative models have become effective at producing human-like motions from text, transferring these motions to humanoid robots for physical execution remains challenging. Existing pipelines are often limited by retargeting, where kinematic quality is undermined by physical infeasibility, contact-transition errors, and the high cost of real-world dynamical data. We present a unified latent-driven framework that bridges natural language and whole-body humanoid locomotion through a retarget-free, physics-optimized pipeline. Rather than treating generation and control as separate stages, our key insight is to couple them bidirectionally under physical constraints.We introduce a Physical Plausibility Optimization (PP-Opt) module as the coupling interface. In the forward direction, PP-Opt refines a teacher-student distillation policy with a plausibility-centric reward to suppress artifacts such as floating, skating, and penetration. In the backward direction, it converts reward-optimized simulation rollouts into high-quality explicit motion data, which is used to fine-tune the motion generator toward a more physically plausible latent distribution. This bidirectional design forms a self-improving cycle: the generator learns a physically grounded latent space, while the controller learns to execute latent-conditioned behaviors with dynamical integrity.Extensive experiments on the Unitree G1 humanoid show that our bidirectional optimization improves tracking accuracy and success rates. Across IsaacLab and MuJoCo, the implicit latent-driven pipeline consistently outperforms conventional explicit retargeting baselines in both precision and stability. By coupling diffusion-based motion generation with physical plausibility optimization, our framework provides a practical path toward deployable text-guided humanoid intelligence.
comment: 10 pages, 5 figures
♻ ☆ TwinRL-VLA: Digital Twin-Driven Reinforcement Learning for Real-World Robotic Manipulation
Despite strong generalization capabilities, Vision-Language-Action (VLA) models remain constrained by the high cost of expert demonstrations and insufficient real-world interaction. While online reinforcement learning (RL) has shown promise in improving general foundation models, applying RL to VLA manipulation in real-world settings is still hindered by low exploration efficiency and a restricted exploration space. Through systematic real-world experiments, we observe that the effective exploration space of online RL is closely tied to the data distribution of supervised fine-tuning (SFT). Motivated by this observation, we propose TwinRL, a digital twin-real-world collaborative RL framework designed to scale and guide exploration for VLA models. First, a high-fidelity digital twin is efficiently reconstructed from smartphone-captured scenes, enabling realistic bidirectional transfer between real and simulated environments. During the SFT warm-up stage, we introduce an exploration space expansion strategy using digital twins to broaden the support of the data trajectory distribution. Building on this enhanced initialization, we propose a sim-to-real guided exploration strategy to further accelerate online RL. Specifically, TwinRL performs efficient and parallel online RL in the digital twin prior to deployment, effectively bridging the gap between offline and online training stages. Subsequently, we exploit efficient digital twin sampling to identify failure-prone yet informative configurations, which are used to guide targeted human-in-the-loop rollouts on the real robot. In our experiments, TwinRL approaches 100% success in both in-distribution regions covered by real-world demonstrations and out-of-distribution regions, delivering at least a 30% speedup over prior real-world RL methods and requiring only about 20 minutes on average across four tasks.
♻ ☆ FoldNet: Learning Generalizable Closed-Loop Policy for Garment Folding via Keypoint-Driven Asset and Demonstration Synthesis
Due to the deformability of garments, generating a large amount of high-quality data for robotic garment manipulation tasks is highly challenging. In this paper, we present a synthetic garment dataset that can be used for robotic garment folding. We begin by constructing geometric garment templates based on keypoints and applying generative models to generate realistic texture patterns. Leveraging these keypoint annotations, we generate folding demonstrations in simulation and train folding policies via closed-loop imitation learning. To improve robustness, we propose KG-DAgger, which uses a keypoint-based strategy to generate demonstration data for recovering from failures. KG-DAgger significantly improves the model performance, boosting the real-world success rate by 25\%. After training with 15K trajectories (about 2M image-action pairs), the model achieves a 75\% success rate in the real world. Experiments in both simulation and real-world settings validate the effectiveness of our proposed framework.
comment: Project: https://pku-epic.github.io/FoldNet/
♻ ☆ Manual2Skill++: Connector-Aware General Robotic Assembly from Instruction Manuals via Vision-Language Models
Assembly hinges on reliably forming connections between parts; yet most robotic approaches plan assembly sequences and part poses while treating connectors as an afterthought. Connections represent the foundational physical constraints of assembly execution; while task planning sequences operations, the precise establishment of these constraints ultimately determines assembly success. In this paper, we treat connections as explicit, primary entities in assembly representation, directly encoding connector types, specifications, and locations for every assembly step. Drawing inspiration from how humans learn assembly tasks through step-by-step instruction manuals, we present Manual2Skill++, a vision-language framework that automatically extracts structured connection information from assembly manuals. We encode assembly tasks as hierarchical graphs where nodes represent parts and sub-assemblies, and edges explicitly model connection relationships between components. A large-scale vision-language model parses symbolic diagrams and annotations in manuals to instantiate these graphs, leveraging the rich connection knowledge embedded in human-designed instructions. We curate a dataset containing over 20 assembly tasks with diverse connector types to validate our representation extraction approach, and evaluate the complete task understanding-to-execution pipeline across four complex assembly scenarios in simulation, spanning furniture, toys, and manufacturing components with real-world correspondence. More detailed information can be found at https://nus-lins-lab.github.io/Manual2SkillPP/
♻ ☆ AdaptPNP: Integrating Prehensile and Non-Prehensile Skills for Adaptive Robotic Manipulation
Non-prehensile (NP) manipulation, in which robots alter object states without forming stable grasps (for example, pushing, poking, or sliding), significantly broadens robotic manipulation capabilities when grasping is infeasible or insufficient. However, enabling a unified framework that generalizes across different tasks, objects, and environments while seamlessly integrating non-prehensile and prehensile (P) actions remains challenging: robots must determine when to invoke NP skills, select the appropriate primitive for each context, and compose P and NP strategies into robust, multi-step plans. We introduce ApaptPNP, a vision-language model (VLM)-empowered task and motion planning framework that systematically selects and combines P and NP skills to accomplish diverse manipulation objectives. Our approach leverages a VLM to interpret visual scene observations and textual task descriptions, generating a high-level plan skeleton that prescribes the sequence and coordination of P and NP actions. A digital-twin based object-centric intermediate layer predicts desired object poses, enabling proactive mental rehearsal of manipulation sequences. Finally, a control module synthesizes low-level robot commands, with continuous execution feedback enabling online task plan refinement and adaptive replanning through the VLM. We evaluate ApaptPNP across representative P&NP hybrid manipulation tasks in both simulation and real-world environments. These results underscore the potential of hybrid P&NP manipulation as a crucial step toward general-purpose, human-level robotic manipulation capabilities. Project Website: https://adaptpnp.github.io/
♻ ☆ U-ARM : Ultra low-cost general teleoperation interface for robot manipulation
We propose U-Arm, a low-cost and rapidly adaptable leader-follower teleoperation framework designed to interface with most of commercially available robotic arms. Our system supports teleoperation through three structurally distinct 3D-printed leader arms that share consistent control logic, enabling seamless compatibility with diverse commercial robot configurations. Compared with previous open-source leader-follower interfaces, we further optimized both the mechanical design and servo selection, achieving a bill of materials (BOM) cost of only \$50.5 for the 6-DoF leader arm and \$56.8 for the 7-DoF version. To enhance usability, we mitigate the common challenge in controlling redundant degrees of freedom by %engineering methods mechanical and control optimizations. Experimental results demonstrate that U-Arm achieves 39\% higher data collection efficiency and comparable task success rates across multiple manipulation scenarios compared with Joycon, another low-cost teleoperation interface. We have open-sourced all CAD models of three configs and also provided simulation support for validating teleoperation workflows. We also open-sourced real-world manipulation data collected with U-Arm. The project website is https://github.com/MINT-SJTU/LeRobot-Anything-U-Arm.
♻ ☆ Aegis: Automated Error Generation and Attribution for Multi-Agent Systems
Large language model based multi-agent systems (MAS) have unlocked significant advancements in tackling complex problems, but their increasing capability introduces a structural fragility that makes them difficult to debug. A key obstacle to improving their reliability is the severe scarcity of large-scale, diverse datasets for error attribution, as existing resources rely on costly and unscalable manual annotation. To address this bottleneck, we introduce Aegis, a novel framework for Automated error generation and attribution for multi-agent systems. Aegis constructs a large dataset of 9,533 trajectories with annotated faulty agents and error modes, covering diverse MAS architectures and task domains. This is achieved using a LLM-based manipulator that can adaptively inject context-aware errors into successful execution trajectories. Leveraging fine-grained labels and the structured arrangement of positive-negative sample pairs, Aegis supports three different learning paradigms: Supervised Fine-Tuning, Reinforcement Learning, and Contrastive Learning. We develop learning methods for each paradigm. Comprehensive experiments show that trained models consistently achieve substantial improvements in error attribution. Notably, several of our fine-tuned LLMs demonstrate performance competitive with or superior to proprietary models an order of magnitude larger, validating our automated data generation framework as a crucial resource for developing more robust and interpretable multi-agent systems. Our project website is available at https://kfq20.github.io/Aegis-Website/.
♻ ☆ RhoMorph: Rhombus-shaped Deformable Modular Robots for Stable, Medium-Independent Reconfiguration Motion
In this paper, we present RhoMorph, a novel deformable planar lattice modular self-reconfigurable robot (MSRR) with a rhombus shaped module. Each module consists of a parallelogram skeleton with a single centrally mounted actuator that enables folding and unfolding along its diagonal. The core design philosophy is to achieve essential MSRR functionalities such as morphing, docking, and locomotion with minimal control complexity. This enables a continuous and stable reconfiguration process that is independent of the surrounding medium, allowing the system to reliably form various configurations in diverse environments. To leverage the unique kinematics of RhoMorph, we introduce morphpivoting, a novel motion primitive for reconfiguration that differs from advanced MSRR systems, and propose a strategy for its continuous execution. Finally, a series of physical experiments validate the module's stable reconfiguration ability, as well as its positional and docking accuracy.
♻ ☆ Whole-Body Safe Control of Robotic Systems with Koopman Neural Dynamics
Controlling robots with strongly nonlinear, high-dimensional dynamics remains challenging, as direct nonlinear optimization with safety constraints is often intractable in real time. The Koopman operator offers a way to represent nonlinear systems linearly in a lifted space, enabling the use of efficient linear control. We propose a data-driven framework that learns a Koopman embedding and operator from data, and integrates the resulting linear model with the Safe Set Algorithm (SSA). This allows the tracking and safety constraints to be solved in a single quadratic program (QP), ensuring feasibility and optimality without a separate safety filter. We validate the method on a Kinova Gen3 manipulator and a Go2 quadruped, showing accurate tracking and obstacle avoidance.
♻ ☆ From Optimizable to Interactable: Mixed Digital Twin-Empowered Testing of Vehicle-Infrastructure Cooperation Systems
Sufficient testing under corner cases is critical for the long-term operation of vehicle-infrastructure cooperation systems (VICS). However, existing corner-case generation methods are primarily AI-driven, and VICS testing under corner cases is typically limited to simulation. In this paper, we introduce an L5 ''Interactable'' level to the VICS digital twin (VICS-DT) taxonomy, extending beyond the conventional L4 ''Optimizable'' level. We further propose an L5-level VICS testing framework, IMPACT (Interactive Mixed-digital-twin Paradigm for Advanced Cooperative vehicle-infrastructure Testing). By enabling direct human interactions with VICS entities, IMPACT incorporates highly uncertain and unpredictable human behaviors into the testing loop, naturally generating high-quality corner cases that complement AI-based methods. Furthermore, the mixedDT-enabled ''Physical-Virtual Action Interaction'' facilitates safe VICS testing under corner cases, incorporating real-world environments and entities rather than purely in simulation. Finally, we implement IMPACT on the I-VIT (Interactive Vehicle-Infrastructure Testbed), and experiments demonstrate its effectiveness. The experimental videos are available at our project website: https://dongjh20.github.io/IMPACT.
♻ ☆ Fast Confidence-Aware Human Prediction via Hardware-accelerated Bayesian Inference for Safe Robot Navigation
As robots increasingly integrate into everyday environments, ensuring their safe navigation around humans becomes imperative. Efficient and safe motion planning requires robots to account for human behavior, particularly in constrained spaces such as grocery stores or care homes, where interactions with multiple individuals are common. Prior research has employed Bayesian frameworks to model human rationality based on navigational intent, enabling the prediction of probabilistic trajectories for planning purposes. In this work, we present a simple yet novel approach for confidence-aware prediction that treats future predictions as particles. This framework is highly parallelized and accelerated on an graphics processing unit (GPU). As a result, this enables longer-term predictions at a frequency of 125 Hz and can be easily extended for multi-human predictions. Compared to existing methods, our implementation supports finer prediction time steps, yielding more granular trajectory forecasts. This enhanced resolution allows motion planners to respond effectively to subtle changes in human behavior. We validate our approach through real-world experiments, demonstrating a robot safely navigating among multiple humans with diverse navigational goals. Our results highlight the methods potential for robust and efficient human-robot coexistence in dynamic environments.
comment: Update the paper
♻ ☆ Embodied Foundation Models at the Edge: A Survey of Deployment Constraints and Mitigation Strategies
Deploying foundation models in embodied edge systems is fundamentally a systems problem, not just a problem of model compression. Real-time control must operate within strict size, weight, and power constraints, where memory traffic, compute latency, timing variability, and safety margins interact directly. The Deployment Gauntlet organizes these constraints into eight coupled barriers that determine whether embodied foundation models can run reliably in practice. Across representative edge workloads, autoregressive Vision-Language-Action policies are constrained primarily by memory bandwidth, whereas diffusion-based controllers are limited more by compute latency and sustained execution cost. Reliable deployment therefore depends on system-level co-design across memory, scheduling, communication, and model architecture, including decompositions that separate fast control from slower semantic reasoning.
♻ ☆ Agentic Vehicles for Human-Centered Mobility: Definition, Prospects, and System Implications
Autonomy, from the Greek autos (self) and nomos (law), refers to the capacity to operate according to internal rules without external control. Autonomous vehicles (AuVs) are therefore understood as systems that perceive their environment and execute pre-programmed tasks independently of external input, consistent with the SAE levels of automated driving. Yet recent research and real-world deployments have begun to showcase vehicles that exhibit behaviors outside the scope of this definition. These include natural language interaction with humans, goal adaptation, contextual reasoning, external tool use, and the handling of unforeseen ethical dilemmas, enabled in part by multimodal large language models (LLMs). These developments highlight not only a gap between technical autonomy and the broader cognitive and social capacities required for human-centered mobility, but also the emergence of a form of vehicle intelligence that currently lacks a clear designation. To address this gap, the paper introduces the concept of agentic vehicles (AgVs): vehicles that exhibit agency, the capacity for goal-driven reasoning, strategic adaptation, self-reflection, and purposeful engagement with complex environments. We conclude by outlining key challenges in the development and governance of AgVs and their potential role in shaping future agentic transportation systems that align with user and societal needs.
♻ ☆ Path Integral Particle Filtering for Hybrid Systems via Saltation Matrices
We present an optimal-control-based particle filtering method for state estimation in hybrid systems that undergo intermittent contact with their environments. We follow the path integral filtering framework that exploits the duality between the smoothing problem and optimal control. We leverage saltation matrices to map out the uncertainty propagation during contact events for hybrid systems. The resulting path integral optimal control problem allows for a state estimation algorithm robust to outlier effects, flexible to non-Gaussian noise distributions, that also handles the challenging contact dynamics in hybrid systems. This work offers a computationally efficient and reliable estimation algorithm for hybrid systems with stochastic dynamics. We also present extensive experimental results demonstrating that our approach consistently outperforms strong baselines across multiple settings.
♻ ☆ HaltNav: Reactive Visual Halting over Lightweight Topological Priors for Robust Vision-Language Navigation
Vision-and-Language Navigation (VLN) is shifting from rigid, step-by-step instruction following toward open-vocabulary, goal-oriented autonomy. Achieving this transition without exhaustive routing prompts requires agents to leverage structural priors. While prior work often assumes computationally heavy 2D/3D metric maps, we instead exploit a lightweight, text-based osmAG (OpenStreetMap Area Graph), a floorplan-level topological representation that is easy to obtain and maintain. However, global planning over a prior map alone is brittle in real-world deployments, where local connectivity can change (e.g., closed doors or crowded passages), leading to execution-time failures. To address this gap, we propose a hierarchical navigation framework HaltNav that couples the robust global planning of osmAG with the local exploration and instruction-grounding capability of VLN. Our approach features an MLLM-based brain module, which is capable of high-level task grounding and obstruction awareness. Conditioned on osmAG, the brain converts the global route into a sequence of localized execution snippets, providing the VLN executor with prior-grounded, goal-centric sub-instructions. Meanwhile, it detects local anomalies via a mechanism we term Reactive Visual Halting (RVH), which interrupts the local control loop, updates osmAG by invalidating the corresponding topology, and triggers replanning to orchestrate a viable detour. To train this halting capability efficiently, we introduce a data synthesis pipeline that leverages generative models to inject realistic obstacles into otherwise navigable scenes, substantially enriching hard negative samples. Extensive experiments demonstrate that our hierarchical framework outperforms several baseline methods without tedious language instructions, and significantly improves robustness for long-horizon vision-language navigation under environmental changes.
♻ ☆ AI-driven Dispensing of Coral Reseeding Devices for Broad-scale Restoration of the Great Barrier Reef
Coral reefs are on the brink of collapse, with climate change, ocean acidification, and pollution leading to a projected 70-90% loss of coral species within the next decade. Reef restoration is crucial, but its success hinges on introducing automation to upscale efforts. In this work, we present a highly configurable AI pipeline for the real-time deployment of coral reseeding devices. The pipeline consists of three core components: (i) the image labeling scheme, designed to address data availability and reduce the cost of expert labeling; (ii) the classifier which performs automated analysis of underwater imagery, at the image or patch-level, while also enabling quantitative coral coverage estimation; and (iii) the decision-making module that determines whether deployment should occur based on the classifier's analysis. By reducing reliance on manual experts, our proposed pipeline increases operational range and efficiency of reef restoration. We validate the proposed pipeline at five sites across the Great Barrier Reef, benchmarking its performance against annotations from expert marine scientists. The pipeline achieves 77.8% deployment accuracy, 89.1% accuracy for sub-image patch classification, and real-time model inference at 5.5 frames per second on a Jetson Orin. To address the limited availability of labeled data in this domain and encourage further research, we publicly release a comprehensive, annotated dataset of substrate imagery from the surveyed sites.
comment: 8 pages, 5 figures
Robotics 88
☆ KineVLA: Towards Kinematics-Aware Vision-Language-Action Models with Bi-Level Action Decomposition
In this paper, we introduce a novel kinematics-rich vision-language-action (VLA) task, in which language commands densely encode diverse kinematic attributes (such as direction, trajectory, orientation, and relative displacement) from initiation through completion, at key moments, unlike existing action instructions that capture kinematics only coarsely or partially, thereby supporting fine-grained and personalized manipulation. In this setting, where task goals remain invariant while execution trajectories must adapt to instruction-level kinematic specifications. To address this challenge, we propose KineVLA, a vision-language-action framework that explicitly decouples goal-level invariance from kinematics-level variability through a bi-level action representation and bi-level reasoning tokens to serve as explicit, supervised intermediate variables that align language and action. To support this task, we construct the kinematics-aware VLA datasets spanning both simulation and real-world robotic platforms, featuring instruction-level kinematic variations and bi-level annotations. Extensive experiments on LIBERO and a Realman-75 robot demonstrate that KineVLA consistently outperforms strong VLA baselines on kinematics-sensitive benchmarks, achieving more precise, controllable, and generalizable manipulation behaviors.
☆ Interpreting Context-Aware Human Preferences for Multi-Objective Robot Navigation
Robots operating in human-shared environments must not only achieve task-level navigation objectives such as safety and efficiency, but also adapt their behavior to human preferences. However, as human preferences are typically expressed in natural language and depend on environmental context, it is difficult to directly integrate them into low-level robot control policies. In this work, we present a pipeline that enables robots to understand and apply context-dependent navigation preferences by combining foundational models with a Multi-Objective Reinforcement Learning (MORL) navigation policy. Thus, our approach integrates high-level semantic reasoning with low-level motion control. A Vision-Language Model (VLM) extracts structured environmental context from onboard visual observations, while Large Language Models (LLM) convert natural language user feedback into interpretable, context-dependent behavioral rules stored in a persistent but updatable rule memory. A preference translation module then maps contextual information and stored rules into numerical preference vectors that parameterize a pretrained MORL policy for real-time navigation adaptation. We evaluate the proposed framework through quantitative component-level evaluations, a user study, and real-world robot deployments in various indoor environments. Our results demonstrate that the system reliably captures user intent, generates consistent preference vectors, and enables controllable behavior adaptation across diverse contexts. Overall, the proposed pipeline improves the adaptability, transparency, and usability of robots operating in shared human environments, while maintaining safe and responsive real-time control.
☆ From Optimizable to Interactable: Mixed Digital Twin-Empowered Testing of Vehicle-Infrastructure Cooperation Systems
Sufficient testing under corner cases is critical for the long-term operation of vehicle-infrastructure cooperation systems (VICS). However, existing corner-case generation methods are primarily AI-driven, and VICS testing under corner cases is typically limited to simulation. In this paper, we introduce an L5 ''Interactable'' level to the VICS digital twin (VICS-DT) taxonomy, extending beyond the conventional L4 ''Optimizable'' level. We further propose an L5-level VICS testing framework, IMPACT (Interactive Mixed-digital-twin Paradigm for Advanced Cooperative vehicle-infrastructure Testing). By enabling direct human interactions with VICS entities, IMPACT incorporates highly uncertain and unpredictable human behaviors into the testing loop, naturally generating high-quality corner cases that complement AI-based methods. Furthermore, the mixedDT-enabled ''Physical-Virtual Action Interaction'' facilitates safe VICS testing under corner cases, incorporating real-world environments and entities rather than purely in simulation. Finally, we implement IMPACT on the I-VIT (Interactive Vehicle-Infrastructure Testbed), and experiments demonstrate its effectiveness. The experimental videos are available at our project website: https://dongjh20.github.io/IMPACT.
☆ Bringing Network Coding into Multi-Robot Systems: Interplay Study for Autonomous Systems over Wireless Communications
Communication is a core enabler for multi-robot systems (MRS), providing the mechanism through which robots exchange state information, coordinate actions, and satisfy safety constraints. While many MRS autonomy algorithms assume reliable and timely message delivery, realistic wireless channels introduce delay, erasures, and ordering stalls that can degrade performance and compromise safety-critical decisions of the robot task. In this paper, we investigate how transport-layer reliability mechanisms that mitigate communication losses and delays shape the autonomy-communication loop. We show that conventional non-coded retransmission-based protocols introduce long delays that are misaligned with the timeliness requirements of MRS applications, and may render the received data irrelevant. As an alternative, we advocate for adaptive and causal network coding, which proactively injects coded redundancy to achieve the desired delay and throughput that enable relevant data delivery to the robotic task. Specifically, this method adapts to channel conditions between robots and causally tunes the communication rates via efficient algorithms. We present two case studies: cooperative localization under delayed and lossy inter-robot communication, and a safety-critical overtaking maneuver where timely vehicle-to-vehicle message availability determines whether an ego vehicle can abort to avoid a crash. Our results demonstrate that coding-based communication significantly reduces in-order delivery stalls, preserves estimation consistency under delay, and improves deadline reliability relative to retransmission-based transport. Overall, the study highlights the need to jointly design autonomy algorithms and communication mechanisms, and positions network coding as a principled tool for dependable multi-robot operation over wireless networks.
☆ P$^{3}$Nav: End-to-End Perception, Prediction and Planning for Vision-and-Language Navigation
In Vision-and-Language Navigation (VLN), an agent is required to plan a path to the target specified by the language instruction, using its visual observations. Consequently, prevailing VLN methods primarily focus on building powerful planners through visual-textual alignment. However, these approaches often bypass the imperative of comprehensive scene understanding prior to planning, leaving the agent with insufficient perception or prediction capabilities. Thus, we propose P$^{3}$Nav, a novel end-to-end framework integrating perception, prediction, and planning in a unified pipeline to strengthen the VLN agent's scene understanding and boost navigation success. Specifically, P$^{3}$Nav augments perception by extracting complementary cues from object-level and map-level perspectives. Subsequently, our P$^{3}$Nav predicts waypoints to model the agent's potential future states, endowing the agent with intrinsic awareness of candidate positions during navigation. Conditioned on these future waypoints, P$^{3}$Nav further forecasts semantic map cues, enabling proactive planning and reducing the strict reliance on purely historical context. Integrating these perceptual and predictive cues, a holistic planning module finally carries out the VLN tasks. Extensive experiments demonstrate that our P$^{3}$Nav achieves new state-of-the-art performance on the REVERIE, R2R-CE, and RxR-CE benchmarks.
☆ FloorPlan-VLN: A New Paradigm for Floor Plan Guided Vision-Language Navigation
Existing Vision-Language Navigation (VLN) task requires agents to follow verbose instructions, ignoring some potentially useful global spatial priors, limiting their capability to reason about spatial structures. Although human-readable spatial schematics (e.g., floor plans) are ubiquitous in real-world buildings, current agents lack the cognitive ability to comprehend and utilize them. To bridge this gap, we introduce \textbf{FloorPlan-VLN}, a new paradigm that leverages structured semantic floor plans as global spatial priors to enable navigation with only concise instructions. We first construct the FloorPlan-VLN dataset, which comprises over 10k episodes across 72 scenes. It pairs more than 100 semantically annotated floor plans with Matterport3D-based navigation trajectories and concise instructions that omit step-by-step guidance. Then, we propose a simple yet effective method \textbf{FP-Nav} that uses a dual-view, spatio-temporally aligned video sequence, and auxiliary reasoning tasks to align observations, floor plans, and instructions. When evaluated under this new benchmark, our method significantly outperforms adapted state-of-the-art VLN baselines, achieving more than a 60\% relative improvement in navigation success rate. Furthermore, comprehensive noise modeling and real-world deployments demonstrate the feasibility and robustness of FP-Nav to actuation drift and floor plan distortions. These results validate the effectiveness of floor plan guided navigation and highlight FloorPlan-VLN as a promising step toward more spatially intelligent navigation.
☆ SafeLand: Safe Autonomous Landing in Unknown Environments with Bayesian Semantic Mapping
Autonomous landing of uncrewed aerial vehicles (UAVs) in unknown, dynamic environments poses significant safety challenges, particularly near people and infrastructure, as UAVs transition to routine urban and rural operations. Existing methods often rely on prior maps, heavy sensors like LiDAR, static markers, or fail to handle non-cooperative dynamic obstacles like humans, limiting generalization and real-time performance. To address these challenges, we introduce SafeLand, a lean, vision-based system for safe autonomous landing (SAL) that requires no prior information and operates only with a camera and a lightweight height sensor. Our approach constructs an online semantic ground map via deep learning-based semantic segmentation, optimized for embedded deployment and trained on a consolidation of seven curated public aerial datasets (achieving 70.22% mIoU across 20 classes), which is further refined through Bayesian probabilistic filtering with temporal semantic decay to robustly identify metric-scale landing spots. A behavior tree then governs adaptive landing, iteratively validates the spot, and reacts in real time to dynamic obstacles by pausing, climbing, or rerouting to alternative spots, maximizing human safety. We extensively evaluate our method in 200 simulations and 60 end-to-end field tests across industrial, urban, and rural environments at altitudes up to 100m, demonstrating zero false negatives for human detection. Compared to the state of the art, SafeLand achieves sub-second response latency, substantially lower than previous methods, while maintaining a superior success rate of 95%. To facilitate further research in aerial robotics, we release SafeLand's segmentation model as a plug-and-play ROS package, available at https://github.com/markus-42/SafeLand.
☆ Physics-informed Deep Mixture-of-Koopmans Vehicle Dynamics Model with Dual-branch Encoder for Distributed Electric-drive Trucks
Advanced autonomous driving systems require accurate vehicle dynamics modeling. However, identifying a precise dynamics model remains challenging due to strong nonlinearities and the coupled longitudinal and lateral dynamic characteristics. Previous research has employed physics-based analytical models or neural networks to construct vehicle dynamics representations. Nevertheless, these approaches often struggle to simultaneously achieve satisfactory performance in terms of system identification efficiency, modeling accuracy, and compatibility with linear control strategies. In this paper, we propose a fully data-driven dynamics modeling method tailored for complex distributed electric-drive trucks (DETs), leveraging Koopman operator theory to represent highly nonlinear dynamics in a lifted linear embedding space. To achieve high-precision modeling, we first propose a novel dual-branch encoder which encodes dynamic states and provides a powerful basis for the proposed Koopman-based methods entitled KODE. A physics-informed supervision mechanism, grounded in the geometric consistency of temporal vehicle motion, is incorporated into the training process to facilitate effective learning of both the encoder and the Koopman operator. Furthermore, to accommodate the diverse driving patterns of DETs, we extend the vanilla Koopman operator to a mixture-of-Koopman operator framework, enhancing modeling capability. Simulations conducted in a high-fidelity TruckSim environment and real-world experiments demonstrate that the proposed approach achieves state-of-the-art performance in long-term dynamics state estimation.
comment: 13 pages, 8 tables, 7 figures
☆ OmniVLN: Omnidirectional 3D Perception and Token-Efficient LLM Reasoning for Visual-Language Navigation across Air and Ground Platforms
Language-guided embodied navigation requires an agent to interpret object-referential instructions, search across multiple rooms, localize the referenced target, and execute reliable motion toward it. Existing systems remain limited in real indoor environments because narrow field-of-view sensing exposes only a partial local scene at each step, often forcing repeated rotations, delaying target discovery, and producing fragmented spatial understanding; meanwhile, directly prompting LLMs with dense 3D maps or exhaustive object lists quickly exceeds the context budget. We present OmniVLN, a zero-shot visual-language navigation framework that couples omnidirectional 3D perception with token-efficient hierarchical reasoning for both aerial and ground robots. OmniVLN fuses a rotating LiDAR and panoramic vision into a hardware-agnostic mapping stack, incrementally constructs a five-layer Dynamic Scene Graph (DSG) from mesh geometry to room- and building-level structure, and stabilizes high-level topology through persistent-homology-based room partitioning and hybrid geometric/VLM relation verification. For navigation, the global DSG is transformed into an agent-centric 3D octant representation with multi-resolution spatial attention prompting, enabling the LLM to progressively filter candidate rooms, infer egocentric orientation, localize target objects, and emit executable navigation primitives while preserving fine local detail and compact long-range memory. Experiments show that the proposed hierarchical interface improves spatial referring accuracy from 77.27\% to 93.18\%, reduces cumulative prompt tokens by up to 61.7\% in cluttered multi-room settings, and improves navigation success by up to 11.68\% over a flat-list baseline. We will release the code and an omnidirectional multimodal dataset to support reproducible research.
☆ DexEXO: A Wearability-First Dexterous Exoskeleton for Operator-Agnostic Demonstration and Learning
Scaling dexterous robot learning is constrained by the difficulty of collecting high-quality demonstrations across diverse operators. Existing wearable interfaces often trade comfort and cross-user adaptability for kinematic fidelity, while embodiment mismatch between demonstration and deployment requires visual post-processing before policy training. We present DexEXO, a wearability-first hand exoskeleton that aligns visual appearance, contact geometry, and kinematics at the hardware level. DexEXO features a pose-tolerant thumb mechanism and a slider-based finger interface analytically modeled to support hand lengths from 140~mm to 217~mm, reducing operator-specific fitting and enabling scalable cross-operator data collection. A passive hand visually matches the deployed robot, allowing direct policy training from raw wrist-mounted RGB observations. User studies demonstrate improved comfort and usability compared to prior wearable systems. Using visually aligned observations alone, we train diffusion policies that achieve competitive performance while substantially simplifying the end-to-end pipeline. These results show that prioritizing wearability and hardware-level embodiment alignment reduces both human and algorithmic bottlenecks without sacrificing task performance. Project Page: https://dexexo-research.github.io/
comment: https://dexexo-research.github.io/
☆ Physics-informed offline reinforcement learning eliminates catastrophic fuel waste in maritime routing
International shipping produces approximately 3% of global greenhouse gas emissions, yet voyage routing remains dominated by heuristic methods. We present PIER (Physics-Informed, Energy-efficient, Risk-aware routing), an offline reinforcement learning framework that learns fuel-efficient, safety-aware routing policies from physics-calibrated environments grounded in historical vessel tracking data and ocean reanalysis products, requiring no online simulator. Validated on one full year (2023) of AIS data across seven Gulf of Mexico routes (840 episodes per method), PIER reduces mean CO2 emissions by 10% relative to great-circle routing. However, PIER's primary contribution is eliminating catastrophic fuel waste: great-circle routing incurs extreme fuel consumption (>1.5x median) in 4.8% of voyages; PIER reduces this to 0.5%, a 9-fold reduction. Per-voyage fuel variance is 3.5x lower (p<0.001), with bootstrap 95% CI for mean savings [2.9%, 15.7%]. Partial validation against observed AIS vessel behavior confirms consistency with the fastest real transits while exhibiting 23.1x lower variance. Crucially, PIER is forecast-independent: unlike A* path optimization whose wave protection degrades 4.5x under realistic forecast uncertainty, PIER maintains constant performance using only local observations. The framework combines physics-informed state construction, demonstration-augmented offline data, and a decoupled post-hoc safety shield, an architecture that transfers to wildfire evacuation, aircraft trajectory optimization, and autonomous navigation in unmapped terrain.
☆ ReSteer: Quantifying and Refining the Steerability of Multitask Robot Policies
Despite strong multi-task pretraining, existing policies often exhibit poor task steerability. For example, a robot may fail to respond to a new instruction ``put the bowl in the sink" when moving towards the oven, executing ``close the oven", even though it can complete both tasks when executed separately. We propose ReSteer, a framework to quantify and improve task steerability in multitask robot policies. We conduct an exhaustive evaluation of state-of-the-art policies, revealing a common lack of steerability. We find that steerability is associated with limited overlap among training task trajectory distributions, and introduce a proxy metric to measure this overlap from policy behavior. Building on this insight, ReSteer improves steerability via three components: (i) a steerability estimator that identifies low-steerability states without full-rollout evaluation, (ii) a steerable data generator that synthesizes motion segments from these states, and (iii) a self-refinement pipeline that improves policy steerability using the generated data. In simulation on LIBERO, ReSteer improves steerability by 11\% over 18k rollouts. In real-world experiments, we show that improved steerability is critical for interactive use, enabling users to instruct robots to perform any task at any time. We hope this work motivates further study on quantifying steerability and data collection strategies for large robot policies.
comment: Project website: https://resteer-vla.github.io/
☆ Neural Radiance Maps for Extraterrestrial Navigation and Path Planning
Autonomous vehicles such as the Mars rovers currently lead the vanguard of surface exploration on extraterrestrial planets and moons. In order to accelerate the pace of exploration and science objectives, it is critical to plan safe and efficient paths for these vehicles. However, current rover autonomy is limited by a lack of global maps which can be easily constructed and stored for onboard re-planning. Recently, Neural Radiance Fields (NeRFs) have been introduced as a detailed 3D scene representation which can be trained from sparse 2D images and efficiently stored. We propose to use NeRFs to construct maps for online use in autonomous navigation, and present a planning framework which leverages the NeRF map to integrate local and global information. Our approach interpolates local cost observations across global regions using kernel ridge regression over terrain features extracted from the NeRF map, allowing the rover to re-route itself around untraversable areas discovered during online operation. We validate our approach in high-fidelity simulation and demonstrate lower cost and higher percentage success rate path planning compared to various baselines.
comment: Published in the Proceedings of the ION GNSS+ 2023 Conference
☆ Full Stack Navigation, Mapping, and Planning for the Lunar Autonomy Challenge
We present a modular, full-stack autonomy system for lunar surface navigation and mapping developed for the Lunar Autonomy Challenge. Operating in a GNSS-denied, visually challenging environment, our pipeline integrates semantic segmentation, stereo visual odometry, pose graph SLAM with loop closures, and layered planning and control. We leverage lightweight learning-based perception models for real-time segmentation and feature tracking and use a factor-graph backend to maintain globally consistent localization. High-level waypoint planning is designed to promote mapping coverage while encouraging frequent loop closures, and local motion planning uses arc sampling with geometric obstacle checks for efficient, reactive control. We evaluate our approach in the competition's high-fidelity lunar simulator, demonstrating centimeter-level localization accuracy, high-fidelity map generation, and strong repeatability across random seeds and rock distributions. Our solution achieved first place in the final competition evaluation.
comment: Published in the Proceedings of the ION GNSS+ 2025 conference
☆ Visual SLAM with DEM Anchoring for Lunar Surface Navigation
Future lunar missions will require autonomous rovers capable of traversing tens of kilometers across challenging terrain while maintaining accurate localization and producing globally consistent maps. However, the absence of global positioning systems, extreme illumination, and low-texture regolith make long-range navigation on the Moon particularly difficult, as visual-inertial odometry pipelines accumulate drift over extended traverses. To address this challenge, we present a stereo visual simultaneous localization and mapping (SLAM) system that integrates learned feature detection and matching with global constraints from digital elevation models (DEMs). Our front-end employs learning-based feature extraction and matching to achieve robustness to illumination extremes and repetitive terrain, while the back-end incorporates DEM-derived height and surface-normal factors into a pose graph, providing absolute surface constraints that mitigate long-term drift. We validate our approach using both simulated lunar traverse data generated in Unreal Engine and real Moon/Mars analog data collected from Mt. Etna. Results demonstrate that DEM anchoring consistently reduces absolute trajectory error compared to baseline SLAM methods, lowering drift in long-range navigation even in repetitive or visually aliased terrain.
comment: Accepted to IEEE Aerospace Conference 2026
☆ Multi-material Direct Ink Writing and Embroidery for Stretchable Wearable Sensors
The development of wearable sensing systems for sports performance tracking, rehabilitation, and injury prevention has driven growing demand for smart garments that combine comfort, durability, and accurate motion detection. This paper presents a textile-compatible fabrication workflow that integrates multi-material direct ink writing with automated embroidery to create stretchable strain sensors directly embedded into garments. The process combines sequential multi-material printing of a silicone-carbon grease-silicone stack with automated embroidery that provides both mechanical fixation and electrical interfacing in a single step. The resulting hybrid sensor demonstrates stretchability up to 120% strain while maintaining electrical continuity, with approximately linear behaviour up to 60% strain (R^2 = 0.99), a gauge factor of 31.4, and hysteresis of 22.9%. Repeated loading-unloading tests over 80 cycles show baseline and peak drift of 0.135% and 0.236% per cycle, respectively, indicating moderate cycle-to-cycle stability. Mechanical testing further confirms that the silicone-fabric interface remains intact under large deformation, with failure occurring in the textile rather than at the stitched boundary. As a preliminary proof of concept, the sensor was integrated into wearable elbow and knee sleeves for joint angle monitoring, showing a clear correlation between normalised resistance change and bending angle. By addressing both mechanical fixation and electrical interfacing through embroidery-based integration, this approach provides a reproducible and scalable pathway for incorporating printed stretchable electronics into textile systems for motion capture and soft robotic applications.
comment: 6 pages, 8 figures, conference
☆ HRI-SA: A Multimodal Dataset for Online Assessment of Human Situational Awareness during Remote Human-Robot Teaming
Maintaining situational awareness (SA) is critical in human-robot teams. Yet, under high workload and dynamic conditions, operators often experience SA gaps. Automated detection of SA gaps could provide timely assistance for operators. However, conventional SA measures either disrupt task flow or cannot capture real-time fluctuations, limiting their operational utility. To the best of our knowledge, no publicly available dataset currently supports the systematic evaluation of online human SA assessment in human-robot teaming. To advance the development of online SA assessment tools, we introduce HRI-SA, a multimodal dataset from 30 participants in a realistic search-and-rescue human-robot teaming context, incorporating eye movements, pupil diameter, biosignals, user interactions, and robot data. The experimental protocol included predefined events requiring timely operator assistance, with ground truth SA latency of two types (perceptual and comprehension) systematically obtained by measuring the time between assistance need onset and resolution. We illustrate the utility of this dataset by evaluating standard machine learning models for detecting perceptual SA latencies using generic eye-tracking features and contextual features. Results show that eye-tracking features alone effectively classified perceptual SA latency (recall=88.91%, F1=67.63%) using leave-one-group-out cross-validation, with performance improved through contextual data fusion (recall=91.51%, F1=80.38%). This paper contributes the first public dataset supporting the systematic evaluation of SA throughout a human-robot teaming mission, while also demonstrating the potential of generic eye-tracking features for continuous perceptual SA latency detection in remote human-robot teaming.
comment: This work is currently under peer review
☆ Shifting Uncertainty to Critical Moments: Towards Reliable Uncertainty Quantification for VLA Model
Vision-Language-Action (VLA) models enable general-purpose robotic policies by mapping visual observations and language instructions to low-level actions, but they often lack reliable introspection. A common practice is to compute a token-level uncertainty signal and take its mean over a rollout. However, mean aggregation can dilute short-lived but safety-critical uncertainty spikes in continuous control. In particular, successful rollouts may contain localized high-entropy segments due to benign noise or non-critical micro-adjustments, while failure rollouts can appear low-entropy for most timesteps and only exhibit brief spikes near the onset of failure. We propose a unified uncertainty quantification approach for predicting rollout success versus failure that (1) uses max-based sliding window pooling to preserve transient risk signals, (2) applies motion-aware stability weighting to emphasize high-frequency action oscillations associated with unstable behaviors, and (3) performs DoF-adaptive calibration via Bayesian Optimization to prioritize kinematically critical axes. Experiments on the LIBERO benchmark show that our method substantially improves failure prediction accuracy and yields more reliable signals for failure detection, which can support downstream human-in-the-loop interventions.
☆ ManiDreams: An Open-Source Library for Robust Object Manipulation via Uncertainty-aware Task-specific Intuitive Physics
Dynamics models, whether simulators or learned world models, have long been central to robotic manipulation, but most focus on minimizing prediction error rather than confronting a more fundamental challenge: real-world manipulation is inherently uncertain. We argue that robust manipulation under uncertainty is fundamentally an integration problem: uncertainties must be represented, propagated, and constrained within the planning loop, not merely suppressed during training. We present and open-source ManiDreams, a modular framework for uncertainty-aware manipulation planning over intuitive physics models. It realizes this integration through composable abstractions for distributional state representation, backend-agnostic dynamics prediction, and declarative constraint specification for action optimization. The framework explicitly addresses three sources of uncertainty: perceptual, parametric, and structural. It wraps any base policy with a sample-predict-constrain loop that evaluates candidate actions against distributional outcomes, adding robustness without retraining. Experiments on ManiSkill tasks show that ManiDreams maintains robust performance under various perturbations where the RL baseline degrades significantly. Runnable examples on pushing, picking, catching, and real-world deployment demonstrate flexibility across different policies, optimizers, physics backends, and executors. The framework is publicly available at https://github.com/Rice-RobotPI-Lab/ManiDreams
comment: 9 pages, 10 figures. Project page at https://manidreams.github.io
☆ DriveVLM-RL: Neuroscience-Inspired Reinforcement Learning with Vision-Language Models for Safe and Deployable Autonomous Driving
Ensuring safe decision-making in autonomous vehicles remains a fundamental challenge despite rapid advances in end-to-end learning approaches. Traditional reinforcement learning (RL) methods rely on manually engineered rewards or sparse collision signals, which fail to capture the rich contextual understanding required for safe driving and make unsafe exploration unavoidable in real-world settings. Recent vision-language models (VLMs) offer promising semantic understanding capabilities; however, their high inference latency and susceptibility to hallucination hinder direct application to real-time vehicle control. To address these limitations, this paper proposes DriveVLM-RL, a neuroscience-inspired framework that integrates VLMs into RL through a dual-pathway architecture for safe and deployable autonomous driving. The framework decomposes semantic reward learning into a Static Pathway for continuous spatial safety assessment using CLIP-based contrasting language goals, and a Dynamic Pathway for attention-gated multi-frame semantic risk reasoning using a lightweight detector and a large VLM. A hierarchical reward synthesis mechanism fuses semantic signals with vehicle states, while an asynchronous training pipeline decouples expensive VLM inference from environment interaction. All VLM components are used only during offline training and are removed at deployment, ensuring real-time feasibility. Experiments in the CARLA simulator show significant improvements in collision avoidance, task success, and generalization across diverse traffic scenarios, including strong robustness under settings without explicit collision penalties. These results demonstrate that DriveVLM-RL provides a practical paradigm for integrating foundation models into autonomous driving without compromising real-time feasibility. Demo video and code are available at: https://zilin-huang.github.io/DriveVLM-RL-website/
comment: 32 pages, 15 figures. Code and demo available online
☆ Proprioceptive-only State Estimation for Legged Robots with Set-Coverage Measurements of Learned Dynamics
Proprioceptive-only state estimation is attractive for legged robots since it is computationally cheaper and is unaffected by perceptually degraded conditions. The history of joint-level measurements contains rich information that can be used to infer the dynamics of the system and subsequently produce navigational measurements. Recent approaches produce these estimates with learned measurement models and fuse with IMU data, under a Gaussian noise assumption. However, this assumption can easily break down with limited training data and render the estimates inconsistent and potentially divergent. In this work, we propose a proprioceptive-only state estimation framework for legged robots that characterizes the measurement noise using set-coverage statements that do not assume any distribution. We develop a practical and computationally inexpensive method to use these set-coverage measurements with a Gaussian filter in a systematic way. We validate the approach in both simulation and two real-world quadrupedal datasets. Comparison with the Gaussian baselines shows that our proposed method remains consistent and is not prone to drift under real noise scenarios.
☆ Sparse3DTrack: Monocular 3D Object Tracking Using Sparse Supervision
Monocular 3D object tracking aims to estimate temporally consistent 3D object poses across video frames, enabling autonomous agents to reason about scene dynamics. However, existing state-of-the-art approaches are fully supervised and rely on dense 3D annotations over long video sequences, which are expensive to obtain and difficult to scale. In this work, we address this fundamental limitation by proposing the first sparsely supervised framework for monocular 3D object tracking. Our approach decomposes the task into two sequential sub-problems: 2D query matching and 3D geometry estimation. Both components leverage the spatio-temporal consistency of image sequences to augment a sparse set of labeled samples and learn rich 2D and 3D representations of the scene. Leveraging these learned cues, our model automatically generates high-quality 3D pseudolabels across entire videos, effectively transforming sparse supervision into dense 3D track annotations. This enables existing fully-supervised trackers to effectively operate under extreme label sparsity. Extensive experiments on the KITTI and nuScenes datasets demonstrate that our method significantly improves tracking performance, achieving an improvement of up to 15.50 p.p. while using at most four ground truth annotations per track.
comment: 22 pages, 8 figures
☆ Offload or Overload: A Platform Measurement Study of Mobile Robotic Manipulation Workloads
Mobile robotic manipulation--the ability of robots to navigate spaces and interact with objects--is a core capability of physical AI. Foundation models have led to breakthroughs in their performance, but at a significant computational cost. We present the first measurement study of mobile robotic manipulation workloads across onboard, edge, and cloud GPU platforms. We find that the full workload stack is infeasible to run on smaller onboard GPUs, while larger onboard GPUs drain robot batteries several hours faster. Offloading alleviates these constraints but introduces its own challenges, as additional network latency degrades task accuracy, and the bandwidth requirement makes naive cloud offloading impractical. Finally, we quantify opportunities and pitfalls of sharing compute across robot fleets. We believe our measurement study will be crucial to designing inference systems for mobile robots.
comment: 15 pages, 17 figures
☆ SG-CoT: An Ambiguity-Aware Robotic Planning Framework using Scene Graph Representations
Ambiguity poses a major challenge to large language models (LLMs) used as robotic planners. In this letter, we present Scene Graph-Chain-of-Thought (SG-CoT), a two-stage framework where LLMs iteratively query a scene graph representation of the environment to detect and clarify ambiguities. First, a structured scene graph representation of the environment is constructed from input observations, capturing objects, their attributes, and relationships with other objects. Second, the LLM is equipped with retrieval functions to query portions of the scene graph that are relevant to the provided instruction. This grounds the reasoning process of the LLM in the observation, increasing the reliability of robotic planners under ambiguous situations. SG-CoT also allows the LLM to identify the source of ambiguity and pose a relevant disambiguation question to the user or another robot. Extensive experimentation demonstrates that SG-CoT consistently outperforms prior methods, with a minimum of 10% improvement in question accuracy and a minimum success rate increase of 4% in single-agent and 15% in multi-agent environments, validating its effectiveness for more generalizable robot planning.
comment: This work has been submitted to the IEEE Robotics and Automation Letters for possible publication
☆ Manufacturing Micro-Patterned Surfaces with Multi-Robot Systems
Applying micro-patterns to surfaces has been shown to impart useful physical properties such as drag reduction and hydrophobicity. However, current manufacturing techniques cannot produce micro-patterned surfaces at scale due to high-cost machinery and inefficient coverage techniques such as raster-scanning. In this work, we use multiple robots, each equipped with a patterning tool, to manufacture these surfaces. To allow these robots to coordinate during the patterning task, we use the ergodic control algorithm, which specifies coverage objectives using distributions. We demonstrate that robots can divide complicated coverage objectives by communicating compressed representations of their trajectory history both in simulations and experimental trials. Further, we show that robot-produced patterning can lower the coefficient of friction of metallic surfaces. This work demonstrates that distributed multi-robot systems can coordinate to manufacture products that were previously unrealizable at scale.
☆ Rapid Adaptation of Particle Dynamics for Generalized Deformable Object Mobile Manipulation ICRA 2026
We address the challenge of learning to manipulate deformable objects with unknown dynamics. In non-rigid objects, the dynamics parameters define how they react to interactions -- how they stretch, bend, compress, and move -- and they are critical to determining the optimal actions to perform a manipulation task successfully. In other robotic domains, such as legged locomotion and in-hand rigid object manipulation, state-of-the-art approaches can handle unknown dynamics using Rapid Motor Adaptation (RMA). Through a supervised procedure in simulation that encodes each rigid object's dynamics, such as mass and position, these approaches learn a policy that conditions actions on a vector of latent dynamic parameters inferred from sequences of state-actions. However, in deformable object manipulation, the object's dynamics not only includes its mass and position, but also how the shape of the object changes. Our key insight is that the recent ground-truth particle positions of a deformable object in simulation capture changes in the object's shape, making it possible to extend RMA to deformable object manipulation. This key insight allows us to develop RAPiD, a two-phase method that learns to perform real-robot deformable object mobile manipulation by: 1) learning a visuomotor policy conditioned on the object's dynamics embedding, which is encoded from the object's privileged information in simulation, such as its mass and ground-truth particle positions, and 2) learning to infer this embedding using non-privileged information instead, such as robot visual observations and actions, so that the learned policy can transfer to the real world. On a mobile manipulator with 22 degrees of freedom, RAPiD enables over 80%+ success rates across two vision-based deformable object mobile manipulation tasks in the real world, under various object dynamics, categories, and instances.
comment: 8 pages, ICRA 2026
☆ ReDAG-RT: Global Rate-Priority Scheduling for Real-Time Multi-DAG Execution in ROS 2
ROS 2 has become a dominant middleware for robotic systems, where perception, estimation, planning, and control pipelines are structured as directed acyclic graphs of callbacks executed under a shared executor. However, default ROS 2 executors use best-effort dispatch without cross-DAG priority enforcement, leading to callback contention, structural priority inversion, and deadline instability under concurrent workloads. These limitations restrict deployment in time-critical and safety-sensitive cyber-physical systems. This paper presents ReDAGRT, a user-space global scheduling framework for deterministic multi-DAG execution in unmodified ROS 2. The framework introduces a Rate-Priority driven global ready queue that orders callbacks by activation rate, enforces per-DAG concurrency bounds, and mitigates cross-graph priority inversion without modifying the ROS 2 API, executor interface, or underlying operating system scheduler. We formalize a multi-DAG task model for ROS 2 callback pipelines and analyze cross-DAG interference under Rate-Priority scheduling. Response-time recurrences and schedulability conditions are derived within classical Rate-Monotonic theory. Experiments in a ROS 2 Humble environment compare ReDAGRT against SingleThreadedExecutor and MultiThreadedExecutor using synthetic multi-DAG workloads. Results show up to 29.7 percent reduction in deadline miss rate, 42.9 percent reduction in 99th percentile response time, and 13.7 percent improvement over MultiThreadedExecutor under comparable utilization. Asymmetric per-DAG concurrency bounds further reduce interference by 40.8 percent. These results demonstrate that deterministic and analyzable multi-DAG scheduling can be achieved entirely in the ROS 2 user-space execution layer, providing a practical foundation for real-time robotic middleware in safety-critical systems.
comment: 12 pages, 6 figures
☆ Semantic Segmentation and Depth Estimation for Real-Time Lunar Surface Mapping Using 3D Gaussian Splatting
Navigation and mapping on the lunar surface require robust perception under challenging conditions, including poorly textured environments, high-contrast lighting, and limited computational resources. This paper presents a real-time mapping framework that integrates dense perception models with a 3D Gaussian Splatting (3DGS) representation. We first benchmark several models on synthetic datasets generated with the LuPNT simulator, selecting a stereo dense depth estimation model based on Gated Recurrent Units for its balance of speed and accuracy in depth estimation, and a convolutional neural network for its superior performance in detecting semantic segments. Using ground truth poses to decouple the local scene understanding from the global state estimation, our pipeline reconstructs a 120-meter traverse with a geometric height accuracy of approximately 3 cm, outperforming a traditional point cloud baseline without LiDAR. The resulting 3DGS map enables novel view synthesis and serves as a foundation for a full SLAM system, where its capacity for joint map and pose optimization would offer significant advantages. Our results demonstrate that combining semantic segmentation and dense depth estimation with learned map representations is an effective approach for creating detailed, large-scale maps to support future lunar surface missions.
☆ GoalVLM: VLM-driven Object Goal Navigation for Multi-Agent System
Object-goal navigation has traditionally been limited to ground robots with closed-set object vocabularies. Existing multi-agent approaches depend on precomputed probabilistic graphs tied to fixed category sets, precluding generalization to novel goals at test time. We present GoalVLM, a cooperative multi-agent framework for zero-shot, open-vocabulary object navigation. GoalVLM integrates a Vision-Language Model (VLM) directly into the decision loop, SAM3 for text-prompted detection and segmentation, and SpaceOM for spatial reasoning, enabling agents to interpret free-form language goals and score frontiers via zero-shot semantic priors without retraining. Each agent builds a BEV semantic map from depth-projected voxel splatting, while a Goal Projector back-projects detections through calibrated depth into the map for reliable goal localization. A constraint-guided reasoning layer evaluates frontiers through a structured prompt chain (scene captioning, room-type classification, perception gating, multi-frontier ranking), injecting commonsense priors into exploration. We evaluate GoalVLM on GOAT-Bench val_unseen (360 multi-subtask episodes, 1032 sequential object-goal subtasks, HM3D scenes), where each episode requires navigating to a chain of 5-7 open-vocabulary targets. GoalVLM with N=2 agents achieves 55.8% subtask SR and 18.3% SPL, competitive with state-of-the-art methods while requiring no task-specific training. Ablation studies confirm the contributions of VLM-guided frontier reasoning and depth-projected goal localization.
comment: 8 pages, 5 figures
☆ R2-Dreamer: Redundancy-Reduced World Models without Decoders or Augmentation ICLR 2026
A central challenge in image-based Model-Based Reinforcement Learning (MBRL) is to learn representations that distill essential information from irrelevant visual details. While promising, reconstruction-based methods often waste capacity on large task-irrelevant regions. Decoder-free methods instead learn robust representations by leveraging Data Augmentation (DA), but reliance on such external regularizers limits versatility. We propose R2-Dreamer, a decoder-free MBRL framework with a self-supervised objective that serves as an internal regularizer, preventing representation collapse without resorting to DA. The core of our method is a redundancy-reduction objective inspired by Barlow Twins, which can be easily integrated into existing frameworks. On DeepMind Control Suite and Meta-World, R2-Dreamer is competitive with strong baselines such as DreamerV3 and TD-MPC2 while training 1.59x faster than DreamerV3, and yields substantial gains on DMC-Subtle with tiny task-relevant objects. These results suggest that an effective internal regularizer can enable versatile, high-performance decoder-free MBRL. Code is available at https://github.com/NM512/r2dreamer.
comment: 20 pages, 12 figures, 2 tables. Published as a conference paper at ICLR 2026. Code available at https://github.com/NM512/r2dreamer
☆ GMT: Goal-Conditioned Multimodal Transformer for 6-DOF Object Trajectory Synthesis in 3D Scenes 3DV 2026
Synthesizing controllable 6-DOF object manipulation trajectories in 3D environments is essential for enabling robots to interact with complex scenes, yet remains challenging due to the need for accurate spatial reasoning, physical feasibility, and multimodal scene understanding. Existing approaches often rely on 2D or partial 3D representations, limiting their ability to capture full scene geometry and constraining trajectory precision. We present GMT, a multimodal transformer framework that generates realistic and goal-directed object trajectories by jointly leveraging 3D bounding box geometry, point cloud context, semantic object categories, and target end poses. The model represents trajectories as continuous 6-DOF pose sequences and employs a tailored conditioning strategy that fuses geometric, semantic, contextual, and goaloriented information. Extensive experiments on synthetic and real-world benchmarks demonstrate that GMT outperforms state-of-the-art human motion and human-object interaction baselines, such as CHOIS and GIMO, achieving substantial gains in spatial accuracy and orientation control. Our method establishes a new benchmark for learningbased manipulation planning and shows strong generalization to diverse objects and cluttered 3D environments. Project page: https://huajian- zeng.github. io/projects/gmt/.
comment: Accpeted by 3DV 2026. Project Page: https://huajian-zeng.github.io/projects/gmt/
☆ Final Report for the Workshop on Robotics & AI in Medicine
The CARE Workshop on Robotics and AI in Medicine, held on December 1, 2025 in Indianapolis, convened leading researchers, clinicians, industry innovators, and federal stakeholders to shape a national vision for advancing robotics and artificial intelligence in healthcare. The event highlighted the accelerating need for coordinated research efforts that bridge engineering innovation with real clinical priorities, emphasizing safety, reliability, and translational readiness with an emphasis on the use of robotics and AI to achieve this readiness goal. Across keynotes, panels, and breakout sessions, participants underscored critical gaps in data availability, standardized evaluation methods, regulatory pathways, and workforce training that hinder the deployment of intelligent robotic systems in surgical, diagnostic, rehabilitative, and assistive contexts. Discussions emphasized the transformative potential of AI enabled robotics to improve precision, reduce provider burden, expand access to specialized care, and enhance patient outcomes particularly in undeserved regions and high risk procedural domains. Special attention was given to austere settings, disaster and relief and military settings. The workshop demonstrated broad consensus on the urgency of establishing a national Center for AI and Robotic Excellence in medicine (CARE). Stakeholders identified priority research thrusts including human robot collaboration, trustworthy autonomy, simulation and digital twins, multi modal sensing, and ethical integration of generative AI into clinical workflows. Participants also articulated the need for high quality datasets, shared test beds, autonomous surgical systems, clinically grounded benchmarks, and sustained interdisciplinary training mechanisms.
comment: 51 pages, 5 figures
☆ A Single-Fiber Optical Frequency Domain Reflectometry (OFDR)-Based Shape Sensing of Concentric Tube Steerable Drilling Robots
This paper introduces a novel shape-sensing approach for Concentric Tube Steerable Drilling Robots (CT-SDRs) based on Optical Frequency Domain Reflectometry (OFDR). Unlike traditional FBG-based methods, OFDR enables continuous strain measurement along the entire fiber length with enhanced spatial resolution. In the proposed method, a Shape Sensing Assembly (SSA) is first fabricated by integrating a single OFDR fiber with a flat NiTi wire. The calibrated SSA is then routed through and housed within the internal channel of a flexible drilling instrument, which is guided by the pre-shaped NiTi tube of the CT-SDR. In this configuration, the drilling instrument serves as a protective sheath for the SSA during drilling, eliminating the need for integration or adhesion to the instrument surface that is typical of conventional optical sensor approaches. The performance of the proposed SSA, integrated within the cannulated CT-SDR, was thoroughly evaluated under free-bending conditions and during drilling along multiple J-shaped trajectories in synthetic Sawbones phantoms. Results demonstrate accurate and reliable shape-sensing capability, confirming the feasibility and robustness of this integration strategy.
comment: 8 pages, 7 figures
☆ Specification-Aware Distribution Shaping for Robotics Foundation Models
Robotics foundation models have demonstrated strong capabilities in executing natural language instructions across diverse tasks and environments. However, they remain largely data-driven and lack formal guarantees on safety and satisfaction of time-dependent specifications during deployment. In practice, robots often need to comply with operational constraints involving rich spatio-temporal requirements such as time-bounded goal visits, sequential objectives, and persistent safety conditions. In this work, we propose a specification-aware action distribution optimization framework that enforces a broad class of Signal Temporal Logic (STL) constraints during execution of a pretrained robotics foundation model without modifying its parameters. At each decision step, the method computes a minimally modified action distribution that satisfies a hard STL feasibility constraint by reasoning over the remaining horizon using forward dynamics propagation. We validate the proposed framework in simulation using a state-of-the-art robotics foundation model across multiple environments and complex specifications.
comment: 8 pages, 3 figures
☆ DexViTac: Collecting Human Visuo-Tactile-Kinematic Demonstrations for Contact-Rich Dexterous Manipulation
Large-scale, high-quality multimodal demonstrations are essential for robot learning of contact-rich dexterous manipulation. While human-centric data collection systems lower the barrier to scaling, they struggle to capture the tactile information during physical interactions. Motivated by this, we present DexViTac, a portable, human-centric data collection system tailored for contact-rich dexterous manipulation. The system enables the high-fidelity acquisition of first-person vision, high-density tactile sensing, end-effector poses, and hand kinematics within unstructured, in-the-wild environments. Building upon this hardware, we propose a kinematics-grounded tactile representation learning algorithm that effectively resolves semantic ambiguities within tactile signals. Leveraging the efficiency of DexViTac, we construct a multimodal dataset comprising over 2,400 visuo-tactile-kinematic demonstrations. Experiments demonstrate that DexViTac achieves a collection efficiency exceeding 248 demonstrations per hour and remains robust against complex visual occlusions. Real-world deployment confirms that policies trained with the proposed dataset and learning strategy achieve an average success rate exceeding 85% across four challenging tasks. This performance significantly outperforms baseline methods, thereby validating the substantial improvement the system provides for learning contact-rich dexterous manipulation. Project page: https://xitong-c.github.io/DexViTac/.
comment: 9 pages, 9 figures.Project page: https://xitong-c.github.io/DexViTac/
☆ ProbeFlow: Training-Free Adaptive Flow Matching for Vision-Language-Action Models
Recent Vision-Language-Action (VLA) models equipped with Flow Matching (FM) action heads achieve state-of-the-art performance in complex robot manipulation. However, the multi-step iterative ODE solving required by FM introduces inference latency that precludes responsive physical control. While current acceleration efforts optimize the Vision-Language Model (VLM) backbone, the action head bottleneck remains overlooked. To address this, we propose ProbeFlow, a training-free adaptive inference framework tai- lored for continuous robotic control. By evaluating geometric trajectory complexity via the cosine similarity between initial and lookahead velocity vectors, ProbeFlow dynamically sched- ules integration steps to prune redundant network evaluations. On the MetaWorld benchmark, it accelerates action decoding by 14.8x (reducing average steps from N = 50 to 2.6) and cuts end-to-end system latency by 2.8x without compromising the manipulation success rate. On the long-horizon LIBERO benchmark, the probe automatically allocates a denser schedule to navigate semantic bottlenecks, effectively resolving the flow solver delay. Real-world physical deployments confirm that ProbeFlow successfully mitigates action decoding latency while ensuring execution stability, offering a highly practical solution for low-latency continuous generative policies.
☆ Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control
Diffusion models and flow matching have become a cornerstone of robotic imitation learning, yet they suffer from a structural inefficiency where inference is often bound to a fixed integration schedule that is agnostic to state complexity. This paradigm forces the policy to expend the same computational budget on trivial motions as it does on complex tasks. We introduce Generative Control as Optimization (GeCO), a time-unconditional framework that transforms action synthesis from trajectory integration into iterative optimization. GeCO learns a stationary velocity field in the action-sequence space where expert behaviors form stable attractors. Consequently, test-time inference becomes an adaptive process that allocates computation based on convergence--exiting early for simple states while refining longer for difficult ones. Furthermore, this stationary geometry yields an intrinsic, training-free safety signal, as the field norm at the optimized action serves as a robust out-of-distribution (OOD) detector, remaining low for in-distribution states while significantly increasing for anomalies. We validate GeCO on standard simulation benchmarks and demonstrate seamless scaling to pi0-series Vision-Language-Action (VLA) models. As a plug-and-play replacement for standard flow-matching heads, GeCO improves success rates and efficiency with an optimization-native mechanism for safe deployment. Video and code can be found at https://hrh6666.github.io/GeCO/
comment: 10 pages, 6 figures
☆ EVA: Aligning Video World Models with Executable Robot Actions via Inverse Dynamics Rewards
Video generative models are increasingly used as world models for robotics, where a model generates a future visual rollout conditioned on the current observation and task instruction, and an inverse dynamics model (IDM) converts the generated frames into executable robot actions. However, current video world models lack explicit executability constraints. As a result, visually coherent rollouts may still violate rigid-body and kinematic consistency, producing unstable or infeasible control commands when decoded by an IDM. We refer to this mismatch between visual generation and physically executable control as the executability gap. While this gap can be mitigated at inference time using techniques such as rejection sampling, such approaches are inefficient due to the high cost of video generation. In this paper, we leverage the executability gap as a training signal and introduce Executable Video Alignment (EVA), a reinforcement-learning post-training framework for aligning video world models. EVA trains an inverse dynamics model on real robot trajectories and repurposes it as a reward model that evaluates generated videos through the action sequences they induce, encouraging smooth motions measured by velocity, acceleration, and jerk while penalizing actions that violate embodiment constraints. Importantly, the reward remains informative even when generated videos contain severe visual artifacts, since such artifacts typically translate into unstable or out-of-bound actions. Experiments on the RoboTwin benchmark and a real bimanual robot show that EVA reduces embodiment-specific artifacts in generated rollouts and improves downstream task execution success.
comment: Project page: https://eva-project-page.github.io/
☆ Huddle: Parallel Shape Assembly using Decentralized, Minimalistic Robots
We propose a novel algorithm for forming arbitrarily shaped assemblies using decentralized robots. By relying on local interactions, the algorithm ensures there are no unreachable states or gaps in the assembly, which are global properties. The in-assembly robots attract passing-by robots into expanding the assembly via a simple implementation of signaling and alignment. Our approach is minimalistic, requiring only communication between attached, immediate neighbors. It is motion-agnostic and requires no pose localization, enabling asynchronous and order-independent assembly. We prove the algorithm's correctness and demonstrate its effectiveness in forming a 107-robot assembly.
comment: 16 pages, 6 figures, submitted to DARS 2026
☆ Multi-Source Human-in-the-Loop Digital Twin Testbed for Connected and Autonomous Vehicles in Mixed Traffic Flow
In the emerging mixed traffic environments, Connected and Autonomous Vehicles (CAVs) have to interact with surrounding human-driven vehicles (HDVs). This paper introduces MSH-MCCT (Multi-Source Human-in-the-Loop Mixed Cloud Control Testbed), a novel CAV testbed that captures complex interactions between various CAVs and HDVs. Utilizing the Mixed Digital Twin concept, which combines Mixed Reality with Digital Twin, MSH-MCCT integrates physical, virtual, and mixed platforms, along with multi-source control inputs. Bridged by the mixed platform, MSH-MCCT allows human drivers and CAV algorithms to operate both physical and virtual vehicles within multiple fields of view. Particularly, this testbed facilitates the coexistence and real-time interaction of physical and virtual CAVs \& HDVs, significantly enhancing the experimental flexibility and scalability. Experiments on vehicle platooning in mixed traffic showcase the potential of MSH-MCCT to conduct CAV testing with multi-source real human drivers in the loop through driving simulators of diverse fidelity. The videos for the experiments are available at our project website: https://dongjh20.github.io/MSH-MCCT.
☆ VolumeDP: Modeling Volumetric Representation for Manipulation Policy Learning
Imitation learning is a prominent paradigm for robotic manipulation. However, existing visual imitation methods map 2D image observations directly to 3D action outputs, imposing a 2D-3D mismatch that hinders spatial reasoning and degrades robustness. We present VolumeDP, a policy architecture that restores spatial alignment by explicitly reasoning in 3D. VolumeDP first lifts image features into a Volumetric Representation via cross-attention. It then selects task-relevant voxels with a learnable module and converts them into a compact set of spatial tokens, markedly reducing computation while preserving action-critical geometry. Finally, a multi-token decoder conditions on the entire token set to predict actions, thereby avoiding lossy aggregation that collapses multiple spatial tokens into a single descriptor. VolumeDP achieves a state-of-the-art average success rate of 88.8% on the LIBERO simulation benchmark, outperforming the strongest baseline by a substantial 14.8% improvement. It also delivers large performance gains over prior methods on the ManiSkill and LIBERO-Plus benchmarks. Real-world experiments further demonstrate higher success rates and robust generalization to novel spatial layouts, camera viewpoints, and environment backgrounds. Code will be released.
☆ AERR-Nav: Adaptive Exploration-Recovery-Reminiscing Strategy for Zero-Shot Object Navigation
Zero-Shot Object Navigation (ZSON) in unknown multi-floor environments presents a significant challenge. Recent methods, mostly based on semantic value greedy waypoint selection, spatial topology-enhanced memory, and Multimodal Large Language Model (MLLM) as a decision-making framework, have led to improvements. However, these architectures struggle to balance exploration and exploitation for ZSON when encountering unseen environments, especially in multi-floor settings, such as robots getting stuck at narrow intersections, endlessly wandering, or failing to find stair entrances. To overcome these challenges, we propose AERR-Nav, a Zero-Shot Object Navigation framework that dynamically adjusts its state based on the robot's environment. Specifically, AERR-Nav has the following two key advantages: (1) An Adaptive Exploration-Recovery-Reminiscing Strategy, enables robots to dynamically transition between three states, facilitating specialized responses to diverse navigation scenarios. (2) An Adaptive Exploration State featuring Fast and Slow-Thinking modes helps robots better balance exploration, exploitation, and higher-level reasoning based on evolving environmental information. Extensive experiments on the HM3D and MP3D benchmarks demonstrate that our AERR-Nav achieves state-of-the-art performance among zero-shot methods. Comprehensive ablation studies further validate the efficacy of our proposed strategy and modules.
☆ Consistency-Driven Dual LSTM Models for Kinematic Control of a Wearable Soft Robotic Arm
In this paper, we introduce a consistency-driven dual LSTM framework for accurately learning both the forward and inverse kinematics of a pneumatically actuated soft robotic arm integrated into a wearable device. This approach effectively captures the nonlinear and hysteretic behaviors of soft pneumatic actuators while addressing the one-to-many mapping challenge between actuation inputs and end-effector positions. By incorporating a cycle consistency loss, we enhance physical realism and improve the stability of inverse predictions. Extensive experiments-including trajectory tracking, ablation studies, and wearable demonstrations-confirm the effectiveness of our method. Results indicate that the inclusion of the consistency loss significantly boosts prediction accuracy and promotes physical consistency over conventional approaches. Moreover, the wearable soft robotic arm demonstrates strong human-robot collaboration capabilities and adaptability in everyday tasks such as object handover, obstacle-aware pick-and-place, and drawer operation. This work underscores the promising potential of learning-based kinematic models for human-centric, wearable robotic systems.
☆ AgentVLN: Towards Agentic Vision-and-Language Navigation
Vision-and-Language Navigation (VLN) requires an embodied agent to ground complex natural-language instructions into long-horizon navigation in unseen environments. While Vision-Language Models (VLMs) offer strong 2D semantic understanding, current VLN systems remain constrained by limited spatial perception, 2D-3D representation mismatch, and monocular scale ambiguity. In this paper, we propose AgentVLN, a novel and efficient embodied navigation framework that can be deployed on edge computing platforms. We formulate VLN as a Partially Observable Semi-Markov Decision Process (POSMDP) and introduce a VLM-as-Brain paradigm that decouples high-level semantic reasoning from perception and planning via a plug-and-play skill library. To resolve multi-level representation inconsistency, we design a cross-space representation mapping that projects perception-layer 3D topological waypoints into the image plane, yielding pixel-aligned visual prompts for the VLM. Building on this bridge, we integrate a context-aware self-correction and active exploration strategy to recover from occlusions and suppress error accumulation over long trajectories. To further address the spatial ambiguity of instructions in unstructured environments, we propose a Query-Driven Perceptual Chain-of-Thought (QD-PCoT) scheme, enabling the agent with the metacognitive ability to actively seek geometric depth information. Finally, we construct AgentVLN-Instruct, a large-scale instruction-tuning dataset with dynamic stage routing conditioned on target visibility. Extensive experiments show that AgentVLN consistently outperforms prior state-of-the-art methods (SOTA) on long-horizon VLN benchmarks, offering a practical paradigm for lightweight deployment of next-generation embodied navigation models. Code: https://github.com/Allenxinn/AgentVLN.
comment: 19pages, 4 figures
☆ REAL: Robust Extreme Agility via Spatio-Temporal Policy Learning and Physics-Guided Filtering
Extreme legged parkour demands rapid terrain assessment and precise foot placement under highly dynamic conditions. While recent learning-based systems achieve impressive agility, they remain fundamentally fragile to perceptual degradation, where even brief visual noise or latency can cause catastrophic failure. To overcome this, we propose Robust Extreme Agility Learning (REAL), an end-to-end framework for reliable parkour under sensory corruption. Instead of relying on perfectly clean perception, REAL tightly couples vision, proprioceptive history, and temporal memory. We distill a cross-modal teacher policy into a deployable student equipped with a FiLM-modulated Mamba backbone to actively filter visual noise and build short-term terrain memory actively. Furthermore, a physics-guided Bayesian state estimator enforces rigid-body consistency during high-impact maneuvers. Validated on a Unitree Go2 quadruped, REAL successfully traverses extreme obstacles even with a 1-meter visual blind zone, while strictly satisfying real-time control constraints with a bounded 13.1 ms inference time.
☆ VectorWorld: Efficient Streaming World Model via Diffusion Flow on Vector Graphs
Closed-loop evaluation of autonomous-driving policies requires interactive simulation beyond log replay. However, existing generative world models often degrade in closed loop due to (i) history-free initialization that mismatches policy inputs, (ii) multi-step sampling latency that violates real-time budgets, and (iii) compounding kinematic infeasibility over long horizons. We propose VectorWorld, a streaming world model that incrementally generates ego-centric $64 \mathrm{m}\times 64\mathrm{m}$ lane--agent vector-graph tiles during rollout. VectorWorld aligns initialization with history-conditioned policies by producing a policy-compatible interaction state via a motion-aware gated VAE. It enables real-time outpainting via solver-free one-step masked completion with an edge-gated relational DiT trained with interval-conditioned MeanFlow and JVP-based large-step supervision. To stabilize long-horizon rollouts, we introduce $Δ$Sim, a physics-aligned non-ego (NPC) policy with hybrid discrete--continuous actions and differentiable kinematic logit shaping. On Waymo open motion and nuPlan, VectorWorld improves map-structure fidelity and initialization validity, and supports stable, real-time $1\mathrm{km}+$ closed-loop rollouts (\href{https://github.com/jiangchaokang/VectorWorld}{code}).
comment: Under Review
☆ Real-Time Online Learning for Model Predictive Control using a Spatio-Temporal Gaussian Process Approximation ICRA
Learning-based model predictive control (MPC) can enhance control performance by correcting for model inaccuracies, enabling more precise state trajectory predictions than traditional MPC. A common approach is to model unknown residual dynamics as a Gaussian process (GP), which leverages data and also provides an estimate of the associated uncertainty. However, the high computational cost of online learning poses a major challenge for real-time GP-MPC applications. This work presents an efficient implementation of an approximate spatio-temporal GP model, offering online learning at constant computational complexity. It is optimized for GP-MPC, where it enables improved control performance by learning more accurate system dynamics online in real-time, even for time-varying systems. The performance of the proposed method is demonstrated by simulations and hardware experiments in the exemplary application of autonomous miniature racing.
comment: to be published at 2026 IEEE International Conference on Robotics & Automation (ICRA)
☆ HeiSD: Hybrid Speculative Decoding for Embodied Vision-Language-Action Models with Kinematic Awareness
Vision-Language-Action (VLA) Models have become the mainstream solution for robot control, but suffer from slow inference speeds. Speculative Decoding (SD) is a promising acceleration method which can be divided into two categories: drafter-based SD and retrieval-based SD. Existing methods fail to analyze the advantages and disadvantages of these two types of SD in VLA models, leading to their sole application or optimization. In this paper, we analyze the trajectory patterns of robots controlled by the VLA model and derive a key insight: the two types of SD should be used in a hybrid manner. However, achieving hybrid SD in VLA models poses several challenges: (1) draft rejection and persistent errors in retrieval-based SD; (2) difficulty in determining the hybrid boundary. To address these, we propose the HeiSD framework. We propose a retrieval-based SD optimization method in HeiSD,which contains a verify-skip mechanism and a sequence-wise relaxed acceptance strategy. Moreover, we proposed a kinematic-based fused metric in HeiSD to automatically determine the hybrid boundary. Experimental results demonstrate that HeiSD attains a speedup of up to 2.45x in simulation benchmarks and 2.06x~2.41x in real-world scenarios, while sustaining a high task success rate.
☆ Action Draft and Verify: A Self-Verifying Framework for Vision-Language-Action Model
Vision-Language-Action (VLA) models have recently demonstrated strong performance across embodied tasks. Modern VLAs commonly employ diffusion action experts to efficiently generate high-precision continuous action chunks, while auto-regressive generation can be slower and less accurate at low-level control. Yet auto-regressive paradigms still provide complementary priors that can improve robustness and generalization in out-of-distribution environments. To leverage both paradigms, we propose Action-Draft-and-Verify (ADV): diffusion action expert drafts multiple candidate action chunks, and the VLM selects one by scoring all candidates in a single forward pass with a perplexity-style metric. Under matched backbones, training data, and action-chunk length, ADV improves success rate by +4.3 points in simulation and +19.7 points in real-world over diffusion-based baseline, with a single-pass VLM reranking overhead.
☆ Uncovering Latent Phase Structures and Branching Logic in Locomotion Policies: A Case Study on HalfCheetah
In locomotion control tasks, Deep Reinforcement Learning (DRL) has demonstrated high performance; however, the decision-making process of the learned policy remains a black box, making it difficult for humans to understand. On the other hand, in periodic motions such as walking, it is well known that implicit motion phases exist, such as the stance phase and the swing phase. Focusing on this point, this study hypothesizes that a policy trained for locomotion control may also represent a phase structure that is interpretable by humans. To examine this hypothesis in a controlled setting, we consider a locomotion task that is amenable to observing whether a policy autonomously acquires temporally structured phases through interaction with the environment. To verify this hypothesis, in the MuJoCo locomotion benchmark HalfCheetah-v5, the state transition sequences acquired by a policy trained for walking control through interaction with the environment were aggregated into semantic phases based on state similarity and consistency of subsequent transitions. As a result, we demonstrated that the state sequences generated by the trained policy exhibit periodic phase transition structures as well as phase branching. Furthermore, by approximating the states and actions corresponding to each semantic phase using Explainable Boosting Machines (EBMs), we analyzed phase-dependent decision making-namely, which state features the policy function attends to and how it controls action outputs in each phase. These results suggest that neural network-based policies, which are often regarded as black boxes, can autonomously acquire interpretable phase structures and logical branching mechanisms.
comment: Accepted at XAI-2026: The 4th World Conference on eXplainable Artificial Intelligence
♻ ☆ MG-Grasp: Metric-Scale Geometric 6-DoF Grasping Framework with Sparse RGB Observations
Single-view RGB-D grasp detection remains a common choice in 6-DoF robotic grasping systems, which typically requires a depth sensor. While RGB-only 6-DoF grasp methods has been studied recently, their inaccurate geometric representation is not directly suitable for physically reliable robotic manipulation, thereby hindering reliable grasp generation. To address these limitations, we propose MG-Grasp, a novel depth-free 6-DoF grasping framework that achieves high-quality object grasping. Leveraging two-view 3D foundation model with camera intrinsic/extrinsic, our method reconstructs metric-scale and multi-view consistent dense point clouds from sparse RGB images and generates stable 6-DoF grasp. Experiments on GraspNet-1Billion dataset and real world demonstrate that MG-Grasp achieves state-of-the-art (SOTA) grasp performance among RGB-based 6-DoF grasping methods.
comment: 8 pages, 5 figures
♻ ☆ Mimic Intent, Not Just Trajectories
While imitation learning (IL) has achieved impressive success in dexterous manipulation through generative modeling and pretraining, state-of-the-art approaches like Vision-Language-Action (VLA) models still struggle with adaptation to environmental changes and skill transfer. We argue this stems from mimicking raw trajectories without understanding the underlying intent. To address this, we propose explicitly disentangling behavior intent from execution details in end-2-end IL: Mimic Intent, Not just Trajectories(MINT). We achieve this via multi-scale frequency-space tokenization, which enforces a spectral decomposition of action chunk representation. We learn action tokens with a multi-scale coarse-to-fine structure, and force the coarsest token to capture low-frequency global structure and finer tokens to encode high-frequency details. This yields an abstract Intent token that facilitates planning and transfer, and multi-scale Execution tokens that enable precise adaptation to environmental dynamics. Building on this hierarchy, our policy generates trajectories through next-scale autoregression, performing progressive intent-to-execution reasoning, thus boosting learning efficiency and generalization. Crucially, this disentanglement enables one-shot transfer of skills, by simply injecting the Intent token from a demonstration into the autoregressive generation process. Experiments on several manipulation benchmarks and on a real robot demonstrate state-of-the-art success rates, superior inference efficiency, robust generalization against disturbances, and effective one-shot transfer.
comment: Under review
♻ ☆ Beyond Short-Horizon: VQ-Memory for Robust Long-Horizon Manipulation in Non-Markovian Simulation Benchmarks
The high cost of collecting real-robot data has made robotic simulation a scalable platform for both evaluation and data generation. Yet most existing benchmarks concentrate on simple manipulation tasks such as pick-and-place, failing to capture the non-Markovian characteristics of real-world tasks and the complexity of articulated object interactions. To address this limitation, we present RuleSafe, a new articulated manipulation benchmark built upon a scalable LLM-aided simulation framework. RuleSafe features safes with diverse unlocking mechanisms, such as key locks, password locks, and logic locks, which require different multi-stage reasoning and manipulation strategies. These LLM-generated rules produce non-Markovian and long-horizon tasks that require temporal modeling and memory-based reasoning. We further propose VQ-Memory, a compact and structured temporal representation that uses vector-quantized variational autoencoders (VQ-VAEs) to encode past proprioceptive states into discrete latent tokens. This representation filters low-level noise while preserving high-level task-phase context, providing lightweight yet robust temporal cues that are compatible with existing Vision-Language-Action models (VLA). Extensive experiments on state-of-the-art VLA models and diffusion policies show that VQ-Memory consistently improves long-horizon planning, enhances generalization to unseen configurations, and enables more efficient manipulation with reduced computational cost. Project page: vqmemory.github.io
comment: 9 pages
♻ ☆ Learning to See and Act: Task-Aware Virtual View Exploration for Robotic Manipulation CVPR 2026
Recent vision-language-action (VLA) models for multi-task robot manipulation often rely on fixed camera setups and shared visual encoders, which limit their performance under occlusions and during cross-task transfer. To address these challenges, we propose Task-aware Virtual View Exploration (TVVE), a framework that learns to select task-relevant virtual camera viewpoints and dynamically re-render observations from a reconstructed scene representation using the selected viewpoints. To enable efficient view selection, we train an exploration policy in a pseudo-environment. In addition, we introduce a Task-aware Mixture-of-Experts (TaskMoE) visual encoder that routes visual features to task-specialized experts, mitigating interference in multi-task learning. To evaluate robustness under distribution shifts, we construct RLBench-OG, an out-of-distribution benchmark with visual perturbations and camera pose variations. Experiments on RLBench and RLBench-OG demonstrate that TVVE achieves higher success rates than strong baselines, while real-robot experiments further confirm its robustness to visual disturbances and unseen instructions. Code and visualizations are available at: https://hcplab-sysu.github.io/TAVP.
comment: 24 pages, 15 figures, Project page: https://hcplab-sysu.github.io/TAVP, Code: https://github.com/HCPLab-SYSU/TAVP.git, Accepted at CVPR 2026
♻ ☆ Swarm Self Clustering for Communication denied Environments without Global Positioning
In this work, we investigate swarm self-clustering, where robots autonomously organize into spatially coherent groups using only local sensing and decision-making, without external commands, global positioning, or inter-robot communication. Each robot forms and maintains clusters by responding to relative distances from nearby neighbors detected through onboard range sensors with limited fields of view. The method is suited for GPS-denied and communication-constrained environments and requires no prior knowledge of cluster size, number, or membership. A mechanism enables robots to alternate between consensus-based and random goal assignment based on local neighborhood size, ensuring robustness, scalability, and untraceable clustering independent of initial conditions. Extensive simulations and real-robot experiments demonstrate empirical convergence, adaptability to dynamic additions, and improved performance over local-only baselines across standard cluster quality metrics.
comment: 36 Pages, 15 figures, 8 tables, pre-print version
♻ ☆ TwinTrack: Bridging Vision and Contact Physics for Real-Time Tracking of Unknown Objects in Contact-Rich Scenes ICRA
Real-time tracking of previously unseen, highly dynamic objects in contact-rich scenes, such as during dexterous in-hand manipulation, remains a major challenge. Pure vision-based approaches often fail under heavy occlusions due to frequent contact interactions and motion blur caused by abrupt impacts. We propose Twintrack, a physics-aware perception system that enables robust, real-time 6-DoF pose tracking of unknown dynamic objects in contact-rich scenes by leveraging contact physics cues. At its core, Twintrack integrates Real2Sim and Sim2Real. Real2Sim combines vision and contact physics to jointly estimate object geometry and physical properties: an initial reconstruction is obtained from vision, then refined by learning a geometry residual and simultaneously estimating physical parameters (e.g., mass, inertia, and friction) based on contact dynamics consistency. Sim2Real achieves robust pose estimation by adaptively fusing a visual tracker with predictions from the updated contact dynamics. Twintrack is implemented on a GPU-accelerated, customized MJX engine to guarantee real-time performance. We evaluate our method on two contact-rich scenarios: object falling with environmental contacts and multi-fingered in-hand manipulation. Results show that, compared to baselines, Twintrack delivers significantly more robust, accurate, and real-time tracking in these challenging settings, with tracking speeds above 20 Hz. Project page: https://irislab.tech/TwinTrack-webpage/
comment: Accepted by IEEE International Conference on Robotics & Automation (ICRA) 2026
♻ ☆ S-VAM: Shortcut Video-Action Model by Self-Distilling Geometric and Semantic Foresight
Video action models (VAMs) have emerged as a promising paradigm for robot learning, owing to their powerful visual foresight for complex manipulation tasks. However, current VAMs, typically relying on either slow multi-step video generation or noisy one-step feature extraction, cannot simultaneously guarantee real-time inference and high-fidelity foresight. To address this limitation, we propose S-VAM, a shortcut video-action model that foresees coherent geometric and semantic representations via a single forward pass. Serving as a stable blueprint, these foreseen representations significantly simplify the action prediction. To enable this efficient shortcut, we introduce a novel self-distillation strategy that condenses structured generative priors of multi-step denoising into one-step inference. Specifically, vision foundation model (VFM) representations extracted from the diffusion model's own multi-step generated videos provide teacher targets. Lightweight decouplers, as students, learn to directly map noisy one-step features to these targets. Extensive experiments in simulation and the real world demonstrate that our S-VAM outperforms state-of-the-art methods, enabling efficient and precise manipulation in complex environments. Our project page is https://haodong-yan.github.io/S-VAM/
♻ ☆ Echo Planning for Autonomous Driving: From Current Observations to Future Trajectories and Back
Modern end-to-end autonomous driving systems suffer from a critical limitation: their planners lack mechanisms to enforce temporal consistency between predicted trajectories and evolving scene dynamics. This absence of self-supervision allows early prediction errors to compound catastrophically over time. We introduce Echo Planning (EchoP), a new self-correcting framework that establishes an end-to-end Current - Future - Current (CFC) cycle to harmonize trajectory prediction with scene coherence. Our key insight is that plausible future trajectories should be bi-directionally consistent, i.e., not only generated from current observations but also capable of reconstructing them. The CFC mechanism first predicts future trajectories from the Bird's-Eye-View (BEV) scene representation, then inversely maps these trajectories back to estimate the current BEV state. By enforcing consistency between the original and reconstructed BEV representations through a cycle loss, the framework intrinsically penalizes physically implausible or misaligned trajectories. Experiments on nuScenes show that the proposed method yields competitive performance, reducing L2 error (Avg) by -0.04 m and collision rate by -0.12% compared to one-shot planners. Moreover, EchoP seamlessly extends to closed-loop evaluation, i.e., Bench2Drive, attaining a 26.54% success rate. Notably, EchoP requires no additional supervision: the CFC cycle acts as an inductive bias that stabilizes long-horizon planning. Overall, EchoP offers a simple, deployable pathway to improve reliability in safety-critical autonomous driving.
comment: 12 pages, 4 figures
♻ ☆ See, Plan, Cut: MPC-Based Autonomous Volumetric Robotic Laser Surgery with OCT Guidance
Robotic laser systems offer the potential for sub-millimeter, non-contact, high-precision tissue resection, yet existing platforms lack volumetric planning and intraoperative feedback. We present RATS (Robot-Assisted Tissue Surgery), an intelligent opto-mechanical, optical coherence tomography (OCT)-guided robotic platform designed for autonomous volumetric soft tissue resection in surgical applications. RATS integrates macro-scale RGB-D imaging, micro-scale OCT, and a fiber-coupled surgical laser, calibrated through a novel multistage alignment pipeline that achieves OCT-to-laser calibration accuracy of 0.161+-0.031mm on tissue phantoms and ex vivo porcine tissue. A super-Gaussian laser-tissue interaction (LTI) model characterizes ablation crater morphology with an average RMSE of 0.231+-0.121mm, outperforming Gaussian baselines. A sampling-based model predictive control (MPC) framework operates directly on OCT voxel data to generate constraint-aware resection trajectories with closed-loop feedback, achieving 0.842mm RMSE and improving intersection-over-union agreement by 64.8% compared to feedforward execution. With OCT, RATS detects subsurface structures and modifies the planner's objective to preserve them, demonstrating clinical feasibility.
comment: 9 pages, 8 figures
♻ ☆ AutoMoT: A Unified Vision-Language-Action Model with Asynchronous Mixture-of-Transformers for End-to-End Autonomous Driving
Integrating vision-language models (VLMs) into end-to-end (E2E) autonomous driving (AD) systems has shown promise in improving scene understanding. However, existing integration strategies suffer from several limitations: they either struggle to resolve distribution misalignment between reasoning and action spaces, underexploit the general reasoning capabilities of pretrained VLMs, or incur substantial inference latency during action policy generation, which degrades driving performance. To address these challenges, we propose \OURS in this work, an end-to-end AD framework that unifies reasoning and action generation within a single vision-language-action (VLA) model. Our approach leverages a mixture-of-transformer (MoT) architecture with joint attention sharing, which preserves the general reasoning capabilities of pre-trained VLMs while enabling efficient fast-slow inference through asynchronous execution at different task frequencies. Extensive experiments on multiple benchmarks, under both open- and closed-loop settings, demonstrate that \OURS achieves competitive performance compared to state-of-the-art methods. We further investigate the functional boundary of pre-trained VLMs in AD, examining when AD-tailored fine-tuning is necessary. Our results show that pre-trained VLMs can achieve competitive multi-task scene understanding performance through semantic prompting alone, while fine-tuning remains essential for action-level tasks such as decision-making and trajectory planning. We refer to \href{https://automot-website.github.io/}{Project Page} for the demonstration videos and qualitative results.
♻ ☆ IRIS-SLAM: Unified Geo-Instance Representations for Robust Semantic Localization and Mapping
Geometry foundation models have significantly advanced dense geometric SLAM, yet existing systems often lack deep semantic understanding and robust loop closure capabilities. Meanwhile, contemporary semantic mapping approaches are frequently hindered by decoupled architectures and fragile data association. We propose IRIS-SLAM, a novel RGB semantic SLAM system that leverages unified geometric-instance representations derived from an instance-extended foundation model. By extending a geometry foundation model to concurrently predict dense geometry and cross-view consistent instance embeddings, we enable a semantic-synergized association mechanism and instance-guided loop closure detection. Our approach effectively utilizes viewpoint-agnostic semantic anchors to bridge the gap between geometric reconstruction and open-vocabulary mapping. Experimental results demonstrate that IRIS-SLAM significantly outperforms state-of-the-art methods, particularly in map consistency and wide-baseline loop closure reliability.
comment: The reason for this withdrawal is that the current version was submitted without the final review and formal authorization of all co-authors. To ensure the academic consensus and integrity of our research group, we have decided to withdraw this submission from the repository
♻ ☆ ViSA: Visited-State Augmentation for Generalized Goal-Space Contrastive Reinforcement Learning
Goal-Conditioned Reinforcement Learning (GCRL) is a framework for learning a policy that can reach arbitrarily given goals. In particular, Contrastive Reinforcement Learning (CRL) provides a framework for policy updates using an approximation of the value function estimated via contrastive learning, achieving higher sample efficiency compared to conventional methods. However, since CRL treats the visited state as a pseudo-goal during learning, it can accurately estimate the value function only for limited goals. To address this issue, we propose a novel data augmentation approach for CRL called ViSA (Visited-State Augmentation). ViSA consists of two components: 1) generating augmented state samples, with the aim of augmenting hard-to-visit state samples during on-policy exploration, and 2) learning consistent embedding space, which uses an augmented state as auxiliary information to regularize the embedding space by reformulating the objective function of the embedding space based on mutual information. We evaluate ViSA in simulation and real-world robotic tasks and show improved goal-space generalization, which permits accurate value estimation for hard-to-visit goals. Further details can be found on the project page: https://issa-n.github.io/projectPage_ViSA/
comment: 8 pages, 7 figures, under Review
♻ ☆ DexGrasp-Zero: A Morphology-Aligned Policy for Zero-Shot Cross-Embodiment Dexterous Grasping
To meet the demands of increasingly diverse dexterous hand hardware, it is crucial to develop a policy that enables zero-shot cross-embodiment grasping without redundant re-learning. Cross-embodiment alignment is challenging due to heterogeneous hand kinematics and physical constraints. Existing approaches typically predict intermediate motion targets and retarget them to each embodiment, which may introduce errors and violate embodiment-specific limits, hindering transfer across diverse hands. To overcome these limitations, we propose DexGrasp-Zero, a policy that learns universal grasping skills from diverse embodiments, enabling zero-shot transfer to unseen hands. We first introduce a morphology-aligned graph representation that maps each hand's kinematic keypoints to anatomically grounded nodes and equips each node with tri-axial orthogonal motion primitives, enabling structural and semantic alignment across different morphologies. Relying on this graph-based representation, we design a Morphology-Aligned Graph Convolutional Network (MAGCN) to encode the graph for policy learning. MAGCN incorporates a Physical Property Injection mechanism that fuses hand-specific physical constraints into the graph features, enabling adaptive compensation for varying link lengths and actuation limits for precise and stable grasping. Our extensive simulation evaluations on the YCB dataset demonstrate that our policy, jointly trained on four heterogeneous hands (Allegro, Shadow, Schunk, Ability), achieves an 85% zero-shot success rate on unseen hardware (LEAP, Inspire), outperforming the state-of-the-art method by 59.5%. Real-world experiments further evaluate our policy on three robot platforms (LEAP, Inspire, Revo2), achieving an 82% average success rate on unseen objects.
♻ ☆ SimScale: Learning to Drive via Real-World Simulation at Scale CVPR 2026
Achieving fully autonomous driving systems requires learning rational decisions in a wide span of scenarios, including safety-critical and out-of-distribution ones. However, such cases are underrepresented in real-world corpus collected by human experts. To complement for the lack of data diversity, we introduce a novel and scalable simulation framework capable of synthesizing massive unseen states upon existing driving logs. Our pipeline utilizes advanced neural rendering with a reactive environment to generate high-fidelity multi-view observations controlled by the perturbed ego trajectory. Furthermore, we develop a pseudo-expert trajectory generation mechanism for these newly simulated states to provide action supervision. Upon the synthesized data, we find that a simple co-training strategy on both real-world and simulated samples can lead to significant improvements in both robustness and generalization for various planning methods on challenging real-world benchmarks, up to +8.6 EPDMS on navhard and +2.9 on navtest. More importantly, such policy improvement scales smoothly by increasing simulation data only, even without extra real-world data streaming in. We further reveal several crucial findings of such a sim-real learning system, which we term SimScale, including the design of pseudo-experts and the scaling properties for different policy architectures. Simulation data and code have been released at https://github.com/OpenDriveLab/SimScale.
comment: Accepted to CVPR 2026. Project page: https://opendrivelab.com/SimScale
♻ ☆ OGScene3D: Incremental Open-Vocabulary 3D Gaussian Scene Graph Mapping for Scene Understanding
Open-vocabulary scene understanding is crucial for robotic applications, enabling robots to comprehend complex 3D environmental contexts and supporting various downstream tasks such as navigation and manipulation. However, existing methods require pre-built complete 3D semantic maps to construct scene graphs for scene understanding, which limits their applicability in robotic scenarios where environments are explored incrementally. To address this challenge, we propose OGScene3D, an open-vocabulary scene understanding system that achieves accurate 3D semantic mapping and scene graph construction incrementally. Our system employs a confidence-based Gaussian semantic representation that jointly models semantic predictions and their reliability, enabling robust scene modeling. Building on this representation, we introduce a hierarchical 3D semantic optimization strategy that achieves semantic consistency through local correspondence establishment and global refinement, thereby constructing globally consistent semantic maps. Moreover, we design a long-term global optimization method that leverages temporal memory of historical observations to enhance semantic predictions. By integrating 2D-3D semantic consistency with Gaussian rendering contribution, this method continuously refines the semantic understanding of the entire scene. Furthermore, we develop a progressive graph construction approach that dynamically creates and updates both nodes and semantic relationships, allowing continuous updating of the 3D scene graphs. Extensive experiments on widely used datasets and real-world scenes demonstrate the effectiveness of our OGScene3D on open-vocabulary scene understanding.
♻ ☆ Latent Representations for Visual Proprioception in Inexpensive Robots
Robotic manipulation requires explicit or implicit knowledge of the robot's joint positions. Precise proprioception is standard in high-quality industrial robots but is often unavailable in inexpensive robots operating in unstructured environments. In this paper, we ask: to what extent can a fast, single-pass regression architecture perform visual proprioception from a single external camera image, available even in the simplest manipulation settings? We explore several latent representations, including CNNs, VAEs, ViTs, and bags of uncalibrated fiducial markers, using fine-tuning techniques adapted to the limited data available. We evaluate the achievable accuracy through experiments on an inexpensive 6-DoF robot.
♻ ☆ TiROD: Tiny Robotics Dataset and Benchmark for Continual Object Detection
Detecting objects with visual sensors is crucial for numerous mobile robotics applications, from autonomous navigation to inspection. However, robots often need to operate under significant domains shifts from those they were trained in, requiring them to adjust to these changes. Tiny mobile robots, subject to size, power, and computational constraints, face even greater challenges when running and adapting detection models on low-resolution and noisy images. Such adaptability, though, is crucial for real-world deployment, where robots must operate effectively in dynamic and unpredictable settings. In this work, we introduce a new vision benchmark to evaluate lightweight continual learning strategies tailored to the unique characteristics of tiny robotic platforms. Our contributions include: (i) Tiny Robotics Object Detection~(TiROD), a challenging video dataset collected using the onboard camera of a small mobile robot, designed to test object detectors across various domains and classes; (ii) a comprehensive benchmark of several continual learning strategies on different scenarios using NanoDet, a lightweight, real-time object detector for resource-constrained devices.. Our results highlight some key challenges in developing robust and efficient continual learning strategies for object detectors in tiny robotics.es; (ii) a benchmark of different continual learning strategies on this dataset using NanoDet, a lightweight object detector. Our results highlight key challenges in developing robust and efficient continual learning strategies for object detectors in tiny robotics.
♻ ☆ Learning Transferable Friction Models and LuGre Identification Via Physics-Informed Neural Networks
Accurately modeling friction in robotics remains a core challenge, as robotics simulators like MuJoCo and PyBullet use simplified friction models or heuristics to balance computational efficiency with accuracy, where these simplifications and approximations can lead to substantial differences between simulated and physical performance. In this paper, we present a physics-informed friction estimation framework that enables the integration of well-established friction models with learnable components, requiring only minimal, generic measurement data. Our approach enforces physical consistency yet retains the flexibility to capture complex friction phenomena. We demonstrate, on an underactuated and nonlinear system, that the learned friction models, trained solely on small and noisy datasets, accurately reproduce dynamic friction properties with significantly higher fidelity than the simplified models commonly used in robotics simulators. Crucially, we show that our approach enables the learned models to be transferable to systems they are not trained on. This ability to generalize across multiple systems streamlines friction modeling for complex, underactuated tasks, offering a scalable and interpretable path toward improving friction model accuracy in robotics and control.
comment: 7 pages, 8 figures, Accepted to 2026 American Control Conference (ACC)
♻ ☆ MLA: A Multisensory Language-Action Model for Multimodal Understanding and Forecasting in Robotic Manipulation
Vision-language-action models (VLAs) have shown generalization capabilities in robotic manipulation tasks by inheriting from vision-language models (VLMs) and learning action generation. Most VLA models focus on interpreting vision and language to generate actions, whereas robots must perceive and interact within the spatial-physical world. This gap highlights the need for a comprehensive understanding of robotic-specific multisensory information, which is crucial for achieving complex and contact-rich control. To this end, we introduce a multisensory language-action (MLA) model that collaboratively perceives heterogeneous sensory modalities and predicts future multisensory objectives to facilitate physical world modeling. Specifically, to enhance perceptual representations, we propose an encoder-free multimodal alignment scheme that innovatively repurposes the large language model itself as a perception module, directly interpreting multimodal cues by aligning 2D images, 3D point clouds, and tactile tokens through positional correspondence. To further enhance MLA's understanding of physical dynamics, we design a future multisensory generation post-training strategy that enables MLA to reason about semantic, geometric, and interaction information, providing more robust conditions for action generation. For evaluation, the MLA model outperforms the previous state-of-the-art 2D and 3D VLA methods by 12% and 24% in complex, contact-rich real-world tasks, respectively, while also demonstrating improved generalization to unseen configurations.
comment: Project page: https://robotic-mla.github.io/
♻ ☆ PLM-Net: Perception Latency Mitigation Network for Vision-Based Lateral Control of Autonomous Vehicles
This study introduces the Perception Latency Mitigation Network (PLM-Net), a modular deep learning framework designed to mitigate perception latency in vision-based imitation-learning lane-keeping systems. Perception latency, defined as the delay between visual sensing and steering actuation, can degrade lateral tracking performance and steering stability. While delay compensation has been extensively studied in classical predictive control systems, its treatment within vision-based imitation-learning architectures under constant and time-varying perception latency remains limited. Rather than reducing latency itself, PLM-Net mitigates its effect on control performance through a plug-in architecture that preserves the original control pipeline. The framework consists of a frozen Base Model (BM), representing an existing lane-keeping controller, and a Timed Action Prediction Model (TAPM), which predicts future steering actions corresponding to discrete latency conditions. Real-time mitigation is achieved by interpolating between model outputs according to the measured latency value, enabling adaptation to both constant and time-varying latency. The framework is evaluated in a closed-loop deterministic simulation environment under fixed-speed conditions to isolate the impact of perception latency. Results demonstrate significant reductions in steering error under multiple latency settings, achieving up to 62% and 78% reductions in Mean Absolute Error (MAE) for constant and time-varying latency cases, respectively. These findings demonstrate the architectural feasibility of modular latency mitigation for vision-based lateral control under controlled simulation settings. The project page including video demonstrations, code, and dataset is publicly released.
♻ ☆ Developing a Discrete-Event Simulator of School Shooter Behavior from VR Data
Virtual reality (VR) has emerged as a powerful tool for evaluating school security measures in high-risk scenarios such as school shootings, offering experimental control and high behavioral fidelity. However, assessing new interventions in VR requires recruiting new participant cohorts for each condition, making large-scale or iterative evaluation difficult. These limitations are especially restrictive when attempting to learn effective intervention strategies, which typically require many training episodes. To address this challenge, we develop a data-driven discrete-event simulator (DES) that models shooter movement and in-region actions as stochastic processes learned from participant behavior in VR studies. We use the simulator to examine the impact of a robot-based shooter intervention strategy. Once shown to reproduce key empirical patterns, the DES enables scalable evaluation and learning of intervention strategies that are infeasible to train directly with human subjects. Overall, this work demonstrates a high-to-mid fidelity simulation workflow that provides a scalable surrogate for developing and evaluating autonomous school-security interventions.
comment: Accepted for presentation at ANNSIM 2026. Camera-ready version. 13 pages, 4 figures, 4 tables
♻ ☆ PACE: Physics Augmentation for Coordinated End-to-end Reinforcement Learning toward Versatile Humanoid Table Tennis
Humanoid table tennis (TT) demands rapid perception, proactive whole-body motion, and agile footwork under strict timing--capabilities that remain difficult for end-to-end control policies. We propose a reinforcement learning (RL) framework that maps ball-position observations directly to whole-body joint commands for both arm striking and leg locomotion, strengthened by predictive signals and dense, physics-guided rewards. A lightweight learned predictor, fed with recent ball positions, estimates future ball states and augments the policy's observations for proactive decision-making. During training, a physics-based predictor supplies precise future states to construct dense, informative rewards that lead to effective exploration. The resulting policy attains strong performance across varied serve ranges (hit rate$\geq$96% and success rate$\geq$92%) in simulations. Ablation studies confirm that both the learned predictor and the predictive reward design are critical for end-to-end learning. Deployed zero-shot on a physical Booster T1 humanoid with 23 revolute joints, the policy produces coordinated lateral and forward-backward footwork with accurate, fast returns, suggesting a practical path toward versatile, competitive humanoid TT. We have open-sourced our RL training code at: https://github.com/purdue-tracelab/TTRL-ICRA2026
♻ ☆ Grounding Robot Generalization in Training Data via Retrieval-Augmented VLMs
Recent work on robot manipulation has advanced policy generalization to novel scenarios. However, it is often difficult to characterize how different evaluation settings actually represent generalization from the training distribution of a given policy. To work towards more precise evaluation of generalization in robotics, we propose RADAR, a scalable framework for directly comparing test-time evaluation tasks to policy training data, to determine what form of policy generalization is required. RADAR consists of a two-stage pipeline: first, retrieval using generalist policy embeddings identifies which training examples are relevant for a given evaluation task. Next, vision-language models (VLMs) analyze the evaluation task against the retrieved data, outputting interpretable analysis on how they compare along a variety of axes, and an overall classification of what type of policy generalization is required. Through controlled experiments, we demonstrate that VLMs are effective at analyzing data for generalization, and that our retrieval step effectively identifies examples needed to make accurate classifications with respect to the training data. Furthermore, we scale RADAR to large-scale datasets, where we observe agreement with human-defined benchmark conditions from prior work. We provide demonstrations at radar-analysis.github.io.
comment: 12 pages
♻ ☆ MOBODY: Model Based Off-Dynamics Offline Reinforcement Learning ICLR 2026
We study off-dynamics offline reinforcement learning, where the goal is to learn a policy from offline source and limited target datasets with mismatched dynamics. Existing methods either penalize the reward or discard source transitions occurring in parts of the transition space with high dynamics shift. As a result, they optimize the policy using data from low-shift regions, limiting exploration of high-reward states in the target domain that do not fall within these regions. Consequently, such methods often fail when the dynamics shift is significant or the optimal trajectories lie outside the low-shift regions. To overcome this limitation, we propose MOBODY, a Model-Based Off-Dynamics Offline RL algorithm that optimizes a policy using learned target dynamics transitions to explore the target domain, rather than only being trained with the low dynamics-shift transitions. For the dynamics learning, built on the observation that achieving the same next state requires taking different actions in different domains, MOBODY employs separate action encoders for each domain to encode different actions to the shared latent space while sharing a unified representation of states and a common transition function. We further introduce a target Q-weighted behavior cloning loss in policy optimization to avoid out-of-distribution actions, which push the policy toward actions with high target-domain Q-values, rather than high source domain Q-values or uniformly imitating all actions in the offline dataset. We evaluate MOBODY on a wide range of MuJoCo and Adroit benchmarks, demonstrating that it outperforms state-of-the-art off-dynamics RL baselines as well as policy learning methods based on different dynamics learning baselines, with especially pronounced improvements in challenging scenarios where existing methods struggle.
comment: Published at ICLR 2026
♻ ☆ LaS-Comp: Zero-shot 3D Completion with Latent-Spatial Consistency CVPR2026
This paper introduces LaS-Comp, a zero-shot and category-agnostic approach that leverages the rich geometric priors of 3D foundation models to enable 3D shape completion across diverse types of partial observations. Our contributions are threefold: First, \ourname{} harnesses these powerful generative priors for completion through a complementary two-stage design: (i) an explicit replacement stage that preserves the partial observation geometry to ensure faithful completion; and (ii) an implicit refinement stage ensures seamless boundaries between the observed and synthesized regions. Second, our framework is training-free and compatible with different 3D foundation models. Third, we introduce Omni-Comp, a comprehensive benchmark combining real-world and synthetic data with diverse and challenging partial patterns, enabling a more thorough and realistic evaluation. Both quantitative and qualitative experiments demonstrate that our approach outperforms previous state-of-the-art approaches. Our code and data will be available at \href{https://github.com/DavidYan2001/LaS-Comp}{LaS-Comp}.
comment: Accepted by CVPR2026
♻ ☆ Dynamic-ICP: Doppler-Aware Iterative Closest Point Registration for Dynamic Scenes
Reliable odometry in highly dynamic environments remains challenging when it relies on ICP-based registration: ICP assumes near-static scenes and degrades in repetitive or low-texture geometry. We introduce Dynamic-ICP, a Doppler-aware registration framework. The method (i) estimates ego motion from per-point Doppler velocity via robust regression and builds a velocity filter, (ii) clusters dynamic objects and reconstructs object-wise translational velocities from ego-compensated radial measurements, (iii) predicts dynamic points with a constant-velocity model, and (iv) aligns scans using a compact objective that combines point-to-plane geometry residual with a translation-invariant, rotation-only Doppler residual. The approach requires no external sensors or sensor-vehicle calibration and operates directly on FMCW LiDAR range and Doppler velocities. We evaluate Dynamic-ICP on three datasets-HeRCULES, HeLiPR, AevaScenes-focusing on highly dynamic scenes. Dynamic-ICP consistently improves rotational stability and translation accuracy over the state-of-the-art methods. Our approach is also simple to integrate into existing pipelines, runs in real time, and provides a lightweight solution for robust registration in dynamic environments. To encourage further research, the code is available at: https://github.com/JMUWRobotics/Dynamic-ICP.
comment: 8 pages, 5 figures
♻ ☆ CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions ICRA 2026
Reinforcement learning (RL), while powerful and expressive, can often prioritize performance at the expense of safety. Yet safety violations can lead to catastrophic outcomes in real-world deployments. Control Barrier Functions (CBFs) offer a principled method to enforce dynamic safety -- traditionally deployed online via safety filters. While the result is safe behavior, the fact that the RL policy does not have knowledge of the CBF can lead to conservative behaviors. This paper proposes CBF-RL, a framework for generating safe behaviors with RL by enforcing CBFs in training. CBF-RL has two key attributes: (1) minimally modifying a nominal RL policy to encode safety constraints via a CBF term, (2) and safety filtering of the policy rollouts in training. Theoretically, we prove that continuous-time safety filters can be deployed via closed-form expressions on discrete-time roll-outs. Practically, we demonstrate that CBF-RL internalizes the safety constraints in the learned policy -- both enforcing safer actions and biasing towards safer rewards -- enabling safe deployment without the need for an online safety filter. We validate our framework through ablation studies on navigation tasks and on the Unitree G1 humanoid robot, where CBF-RL enables safer exploration, faster convergence, and robust performance under uncertainty, enabling the humanoid robot to avoid obstacles and climb stairs safely in real-world settings without a runtime safety filter.
comment: To appear at ICRA 2026; sample code for the navigation example with CBF-RL reward core construction can be found at https://github.com/lzyang2000/cbf-rl-navigation-demo
♻ ☆ Simulation to Rules: A Dual-VLM Framework for Formal Visual Planning
Vision Language Models (VLMs) show strong potential for visual planning but struggle with precise spatial and long-horizon reasoning, while Planning Domain Definition Language (PDDL) planners excel at formal long-horizon planning but cannot interpret visual inputs. Recent works combine these complementary advantages by translating visual problems into PDDL. However, while VLMs can generate PDDL problem files satisfactorily, accurately generating PDDL domain files, which encode planning rules, remains challenging and typically requires human expertise or environment interaction. We propose VLMFP, a Dual-VLM-guided framework that autonomously generates both PDDL problem and domain files for formal visual planning. VLMFP combines a SimVLM that simulates action consequences with a GenVLM that generates and iteratively refines PDDL files by aligning symbolic execution with simulated outcomes, enabling multiple levels of generalization across unseen instances, visual appearances, and game rules. We evaluate VLMFP on 6 grid-world domains and demonstrate its generalization capability. On average, SimVLM achieves 87.3% and 86.0% scenario understanding and action simulation for seen and unseen appearances, respectively. With the guidance of SimVLM, VLMFP attains 70.0%, 54.1% planning success on unseen instances in seen and unseen appearances, respectively. We further demonstrate that VLMFP scales to complex long-horizon 3D planning tasks, including multi-robot collaboration and assembly scenarios with partial observability and diverse visual variations. Project page: https://sites.google.com/view/vlmfp.
comment: 40 pages, 6 figures, 13 tables
♻ ☆ NavThinker: Action-Conditioned World Models for Coupled Prediction and Planning in Social Navigation
Social navigation requires robots to act safely in dynamic human environments. Effective behavior demands thinking ahead: reasoning about how the scene and pedestrians evolve under different robot actions rather than reacting to current observations alone. This creates a coupled prediction-planning challenge, where robot actions and human motion mutually influence each other. To address this challenge, we propose NavThinker, a future-aware framework that couples an action-conditioned world model with on-policy reinforcement learning. The world model operates in the Depth Anything V2 patch feature space and performs autoregressive prediction of future scene geometry and human motion; multi-head decoders then produce future depth maps and human trajectories, yielding a future-aware state aligned with traversability and interaction risk. Crucially, we train the policy with DD-PPO while injecting world-model think-ahead signals via: (i) action-conditioned future features fused into the current observation embedding and (ii) social reward shaping from predicted human trajectories. Experiments on single- and multi-robot Social-HM3D show state-of-the-art navigation success, with zero-shot transfer to Social-MP3D and real-world deployment on a Unitree Go2, validating generalization and practical applicability. Webpage: https://hutslib.github.io/NavThinker.
♻ ☆ Beware Untrusted Simulators -- Reward-Free Backdoor Attacks in Reinforcement Learning ICLR 2026
Simulated environments are a key piece in the success of Reinforcement Learning (RL), allowing practitioners and researchers to train decision making agents without running expensive experiments on real hardware. Simulators remain a security blind spot, however, enabling adversarial developers to alter the dynamics of their released simulators for malicious purposes. Therefore, in this work we highlight a novel threat, demonstrating how simulator dynamics can be exploited to stealthily implant action-level backdoors into RL agents. The backdoor then allows an adversary to reliably activate targeted actions in an agent upon observing a predefined ``trigger'', leading to potentially dangerous consequences. Traditional backdoor attacks are limited in their strong threat models, assuming the adversary has near full control over an agent's training pipeline, enabling them to both alter and observe agent's rewards. As these assumptions are infeasible to implement within a simulator, we propose a new attack ``Daze'' which is able to reliably and stealthily implant backdoors into RL agents trained for real world tasks without altering or even observing their rewards. We provide formal proof of Daze's effectiveness in guaranteeing attack success across general RL tasks along with extensive empirical evaluations on both discrete and continuous action space domains. We additionally provide the first example of RL backdoor attacks transferring to real, robotic hardware. These developments motivate further research into securing all components of the RL training pipeline to prevent malicious attacks.
comment: 10 pages main body, ICLR 2026
♻ ☆ AsgardBench -- Evaluating Visually Grounded Interactive Planning Under Minimal Feedback
With AsgardBench we aim to evaluate visually grounded, high-level action sequence generation and interactive planning, focusing specifically on plan adaptation during execution based on visual observations rather than navigation or low-level manipulation. In the landscape of embodied AI benchmarks, AsgardBench targets the capability category of interactive planning, which is more sophisticated than offline high-level planning as it requires agents to revise plans in response to environmental feedback, yet remains distinct from low-level execution. Unlike prior embodied AI benchmarks that conflate reasoning with navigation or provide rich corrective feedback that substitutes for perception, AsgardBench restricts agent input to images, action history, and lightweight success/failure signals, isolating interactive planning in a controlled simulator without low-level control noise. The benchmark contains 108 task instances spanning 12 task types, each systematically varied through object state, placement, and scene configuration. These controlled variations create conditional branches in which a single instruction can require different action sequences depending on what the agent observes, emphasizing conditional branching and plan repair during execution. Our evaluations of leading vision language models show that performance drops sharply without visual input, revealing weaknesses in visual grounding and state tracking that ultimately undermine interactive planning. Our benchmark zeroes in on a narrower question: can a model actually use what it sees to adapt a plan when things do not go as expected?
comment: 19 figures, 6 tables, including appendix
♻ ☆ Aion: Towards Hierarchical 4D Scene Graphs with Temporal Flow Dynamics ICRA 2026
Autonomous navigation in dynamic environments requires spatial representations that capture both semantic structure and temporal evolution. 3D Scene Graphs (3DSGs) provide hierarchical multi-resolution abstractions that encode geometry and semantics, but existing extensions toward dynamics largely focus on individual objects or agents. In parallel, Maps of Dynamics (MoDs) model typical motion patterns and temporal regularities, yet are usually tied to grid-based discretizations that lack semantic awareness and do not scale well to large environments. In this paper we introduce Aion, a framework that embeds temporal flow dynamics directly within a hierarchical 3DSG, effectively incorporating the temporal dimension. Aion employs a graph-based sparse MoD representation to capture motion flows over arbitrary time intervals and attaches them to navigational nodes in the scene graph, yielding more interpretable and scalable predictions that improve planning and interaction in complex dynamic environments. We provide the code at https://github.com/IacopomC/aion
comment: Accepted at ICRA 2026, 8 pages
♻ ☆ SAATT Nav: a Socially Aware Autonomous Transparent Transportation Navigation Framework for Wheelchairs IROS 2026
While powered wheelchairs reduce physical fatigue as opposed to manual wheelchairs for individuals with mobility impairment, they demand high cognitive workload due to information processing, decision making and motor coordination. Current autonomous systems lack social awareness in navigation and transparency in decision-making, leading to decreased perceived safety and trust from the user and others in context. This work proposes Socially Aware Autonomous Transparent Transportation (SAATT) Navigation framework for wheelchairs as a potential solution. By implementing a Large Language Model (LLM) informed of user intent and capable of predicting other peoples' intent as a decision-maker for its local controller, it is able to detect and navigate social situations, such as passing pedestrians or a pair conversing. Furthermore, the LLM textually communicates its reasoning at each waypoint for transparency. In this experiment, it is compared against a standard global planner, a representative competing social navigation model, and an Ablation study in three simulated environments varied by social levels in eight metrics categorized under Safety, Social Compliance, Efficiency, and Comfort. Overall, SAATT Nav outperforms in most social situations and equivalently or only slightly worse in the remaining metrics, demonstrating the potential of a socially aware and transparent autonomous navigation system to assist wheelchair users.
comment: 8 pages, 4 figures, 2 tables, 1 algorithm. Submitted to IROS 2026
♻ ☆ World-Env: Leveraging World Model as a Virtual Environment for VLA Post-Training CVPR2026
Vision-Language-Action (VLA) models trained via imitation learning suffer from significant performance degradation in data-scarce scenarios due to their reliance on large-scale demonstration datasets. Although reinforcement learning (RL)-based post-training has proven effective in addressing data scarcity, its application to VLA models is hindered by the non-resettable nature of real-world environments. This limitation is particularly critical in high-risk domains such as industrial automation, where interactions often induce state changes that are costly or infeasible to revert. Furthermore, existing VLA approaches lack a reliable mechanism for detecting task completion, leading to redundant actions that reduce overall task success rates. To address these challenges, we propose RehearseVLA:, an RL-based post-training framework that replaces physical interaction with a low-cost world model-based virtual simulator. RehearseVLA: consists of two key components: (1) a physically-consistent world simulator that generates temporally consistent future visual observations, and (2) a vision-language model (VLM)-guided instant reflector that provides continuous reward signals and predicts action termination. This simulated environment enables VLA models to safely explore and generalize beyond their initial imitation learning distribution. Our method achieves notable performance gains with as few as five expert demonstrations per task. Experiments on complex robotic manipulation tasks demonstrate that RehearseVLA: effectively overcomes the data inefficiency, safety constraints, and inefficient execution of conventional VLA models that rely on real-world interaction, offering a practical and scalable solution for post-training in resource-constrained settings. Our code is available at https://github.com/amap-cvlab/world-env.
comment: Accepted to CVPR2026
♻ ☆ Safety Case Patterns for VLA-based driving systems: Insights from SimLingo
Vision-Language-Action (VLA)-based driving systems represent a significant paradigm shift in autonomous driving since, by combining traffic scene understanding, linguistic interpretation, and action generation, these systems enable more flexible, adaptive, and instruction-responsive driving behaviors. However, despite their growing adoption and potential to support socially responsible autonomous driving as well as understanding high-level human instructions, VLA-based driving systems may exhibit new types of hazardous behaviors. For instance, the integration of open-ended natural language inputs (e.g., user or navigation instructions) into the multimodal control loop, may lead to unpredictable and unsafe behaviors that could endanger vehicle occupants and pedestrians. Hence, assuring the safety of these systems is crucial to help build trust in their operations. To support this, we propose a novel safety case design approach called RAISE. Our approach introduces novel patterns tailored to instruction-based driving systems such as VLA-based driving systems, an extension of Hazard Analysis and Risk Assessment (HARA) detailing safe scenarios and their outcomes, and a design technique to create the safety cases of VLA-based driving systems. A case study on SimLingo illustrates how our approach can be used to construct rigorous, evidence-based safety claims for this emerging class of autonomous driving systems.
♻ ☆ ReTac-ACT: A State-Gated Vision-Tactile Fusion Transformer for Precision Assembly
Precision assembly requires sub-millimeter corrections in contact-rich "last-millimeter" regions where visual feedback fails due to occlusion from the end-effector and workpiece. We present ReTac-ACT (Reconstruction-enhanced Tactile ACT), a vision-tactile imitation learning policy that addresses this challenge through three synergistic mechanisms: (i) bidirectional cross-attention enabling reciprocal visuo-tactile feature enhancement before fusion, (ii) a proprioception-conditioned gating network that dynamically elevates tactile reliance when visual occlusion occurs, and (iii) a tactile reconstruction objective enforcing learning of manipulation-relevant contact information rather than generic visual textures. Evaluated on the standardized NIST Assembly Task Board M1 benchmark, ReTac-ACT achieves 90% peg-in-hole success, substantially outperforming vision-only and generalist baseline methods, and maintains 80% success at industrial-grade 0.1mm clearance. Ablation studies validate that each architectural component is indispensable. The ReTac-ACT codebase and a vision-tactile demonstration dataset covering various clearance levels with both visual and tactile features will be released to support reproducible research.
♻ ☆ Context-Nav: Context-Driven Exploration and Viewpoint-Aware 3D Spatial Reasoning for Instance Navigation CVPR 2026
Text-goal instance navigation (TGIN) asks an agent to resolve a single, free-form description into actions that reach the correct object instance among same-category distractors. We present \textit{Context-Nav}, which elevates long, contextual captions from a local matching cue to a global exploration prior and verifies candidates through 3D spatial reasoning. First, we compute dense text-image alignments for a value map that ranks frontiers -- guiding exploration toward regions consistent with the entire description rather than early detections. Second, upon observing a candidate, we perform a viewpoint-aware relation check: the agent samples plausible observer poses, aligns local frames, and accepts a target only if the spatial relations can be satisfied from at least one viewpoint. The pipeline requires no task-specific training or fine-tuning; we attain state-of-the-art performance on InstanceNav and CoIN-Bench. Ablations show that (i) encoding full captions into the value map avoids wasted motion and (ii) explicit, viewpoint-aware 3D verification prevents semantically plausible but incorrect stops. This suggests that geometry-grounded spatial reasoning is a scalable alternative to heavy policy training or human-in-the-loop interaction for fine-grained instance disambiguation in cluttered 3D scenes.
comment: Accepted to CVPR 2026. Code is available at https://github.com/AutoCompSysLab/ContextNav
♻ ☆ Thousand-GPU Large-Scale Training and Optimization Recipe for AI-Native Cloud Embodied Intelligence Infrastructure
Embodied intelligence is a key step towards Artificial General Intelligence (AGI), yet its development faces multiple challenges including data, frameworks, infrastructure, and evaluation systems. To address these issues, we have, for the first time in the industry, launched a cloud-based, thousand-GPU distributed training platform for embodied intelligence, built upon the widely adopted LeRobot framework, and have systematically overcome bottlenecks across the entire pipeline. At the data layer, we have restructured the data pipeline to optimize the flow of embodied training data. In terms of training, for the GR00T-N1.5 model, utilizing thousand-GPU clusters and data at the scale of hundreds of millions, the single-round training time has been reduced from 15 hours to just 22 minutes, achieving a 40-fold speedup. At the model layer, by combining variable-length FlashAttention and Data Packing, we have moved from sample redundancy to sequence integration, resulting in a 188% speed increase; π-0.5 attention optimization has accelerated training by 165%; and FP8 quantization has delivered a 140% speedup. On the infrastructure side, relying on high-performance storage, a 3.2T RDMA network, and a Ray-driven elastic AI data lake, we have achieved deep synergy among data, storage, communication, and computation. We have also built an end-to-end evaluation system, creating a closed loop from training to simulation to assessment. This framework has already been fully validated on thousand-GPU clusters, laying a crucial technical foundation for the development and application of next-generation autonomous intelligent robots, and is expected to accelerate the arrival of the era of human-machine integration.
Robotics 108
☆ Early-Terminable Energy-Safe Iterative Coupling for Parallel Simulation of Port-Hamiltonian Systems
Parallel simulation and control of large-scale robotic systems often rely on partitioned time stepping, yet finite-iteration coupling can inject spurious energy by violating power consistency--even when each subsystem is passive. This letter proposes a novel energy-safe, early-terminable iterative coupling for port-Hamiltonian subsystems by embedding a Douglas--Rachford (DR) splitting scheme in scattering (wave) coordinates. The lossless interconnection is enforced as an orthogonal constraint in the wave domain, while each subsystem contributes a discrete-time scattering port map induced by its one-step integrator. Under a discrete passivity condition on the subsystem time steps and a mild impedance-tuning condition, we prove an augmented-storage inequality certifying discrete passivity of the coupled macro-step for any finite inner-iteration budget, with the remaining mismatch captured by an explicit residual. As the inner budget increases, the partitioned update converges to the monolithic discrete-time update induced by the same integrators, yielding a principled, adaptive accuracy--compute trade-off, supporting energy-consistent real-time parallel simulation under varying computational budgets. Experiments on a coupled-oscillator benchmark validate the passivity certificates at numerical roundoff (on the order of 10e-14 in double precision) and show that the reported RMS state error decays monotonically with increasing inner-iteration budgets, consistent with the hard-coupling limit.
☆ Onboard MuJoCo-based Model Predictive Control for Shipboard Crane with Double-Pendulum Sway Suppression
Transferring heavy payloads in maritime settings relies on efficient crane operation, limited by hazardous double-pendulum payload sway. This sway motion is further exacerbated in offshore environments by external perturbations from wind and ocean waves. Manual suppression of these oscillations on an underactuated crane system by human operators is challenging. Existing control methods struggle in such settings, often relying on simplified analytical models, while deep reinforcement learning (RL) approaches tend to generalise poorly to unseen conditions. Deploying a predictive controller onto compute-constrained, highly non-linear physical systems without relying on extensive offline training or complex analytical models remains a significant challenge. Here we show a complete real-time control pipeline centered on the MuJoCo MPC framework that leverages a cross-entropy method planner to evaluate candidate action sequences directly within a physics simulator. By using simulated rollouts, this sampling-based approach successfully reconciles the conflicting objectives of dynamic target tracking and sway damping without relying on complex analytical models. We demonstrate that the controller can run effectively on a resource-constrained embedded hardware, while outperforming traditional PID and RL baselines in counteracting external base perturbations. Furthermore, our system demonstrates robustness even when subjected to unmodeled physical discrepancies like the introduction of a second payload.
comment: 8 pages, 5 figures
☆ Controlling Fish Schools via Reinforcement Learning of Virtual Fish Movement
This study investigates a method to guide and control fish schools using virtual fish trained with reinforcement learning. We utilize 2D virtual fish displayed on a screen to overcome technical challenges such as durability and movement constraints inherent in physical robotic agents. To address the lack of detailed behavioral models for real fish, we adopt a model-free reinforcement learning approach. First, simulation results show that reinforcement learning can acquire effective movement policies even when simulated real fish frequently ignore the virtual stimulus. Second, real-world experiments with live fish confirm that the learned policy successfully guides fish schools toward specified target directions. Statistical analysis reveals that the proposed method significantly outperforms baseline conditions, including the absence of stimulus and a heuristic "stay-at-edge" strategy. This study provides an early demonstration of how reinforcement learning can be used to influence collective animal behavior through artificial agents.
comment: English translation of the author's 2018 bachelor's thesis. Keywords: fish schooling, reinforcement learning, collective behavior, artificial agents, swarm-machine interaction
☆ Encoding Predictability and Legibility for Style-Conditioned Diffusion Policy
Striking a balance between efficiency and transparent motion is a core challenge in human-robot collaboration, as highly expressive movements often incur unnecessary time and energy costs. In collaborative environments, legibility allows a human observer a better understanding of the robot's actions, increasing safety and trust. However, these behaviors result in sub-optimal and exaggerated trajectories that are redundant in low-ambiguity scenarios where the robot's goal is already obvious. To address this trade-off, we propose Style-Conditioned Diffusion Policy (SCDP), a modular framework that constrains the trajectory generation of a pre-trained diffusion model toward either legibility or efficiency based on the environment's configuration. Our method utilizes a post-training pipeline that freezes the base policy and trains a lightweight scene encoder and conditioning predictor to modulate the diffusion process. At inference time, an ambiguity detection module activates the appropriate conditioning, prioritizing expressive motion only for ambiguous goals and reverting to efficient paths otherwise. We evaluate SCDP on manipulation and navigation tasks, and results show that it enhances legibility in ambiguous settings while preserving optimal efficiency when legibility is unnecessary, all without retraining the base policy.
comment: Submitted to the 18th International Conference on Social Robotics (ICSR 2026)
☆ Faulty Coffees: Barriers to Adoption of an In-the-wild Robo-Barista
We set out to study whether task-based narratives could influence long-term engagement with a service robot. To do so, we deployed a Robo-Barista for five weeks in an over-50's housing complex in Stockton, England. Residents received a free daily coffee by interacting with a Furhat robot assigned to either a narrative or non-narrative dialogue condition. Despite designing for sustained engagement, repeat interaction was low, and we encountered curiosity trials without retention, technical breakdowns, accessibility barriers, and the social dynamics of a housing complex setting. Rather than treating these as peripheral issues, we foreground them in this paper. We reflect on the in-the-wild realities of our experiment and offer lessons for conducting longitudinal Human-Robot Interaction research when studies unravel in practice.
comment: Accepted for publication in Failing Forward, Design and Deployment Lessons from Real-World Human-Robot Interaction Workshop at HRI 2026, March 16, 2026, Edinburgh, Scotland
☆ ADAPT: Adaptive Dual-projection Architecture for Perceptive Traversal
Agile humanoid locomotion in complex 3D en- vironments requires balancing perceptual fidelity with com- putational efficiency, yet existing methods typically rely on rigid sensing configurations. We propose ADAPT (Adaptive dual-projection architecture for perceptive traversal), which represents the environment using a horizontal elevation map for terrain geometry and a vertical distance map for traversable- space constraints. ADAPT further treats its spatial sensing range as a learnable action, enabling the policy to expand its perceptual horizon during fast motion and contract it in cluttered scenes for finer local resolution. Compared with voxel-based baselines, ADAPT drastically reduces observation dimensionality and computational overhead while substantially accelerating training. Experimentally, it achieves successful zero-shot transfer to a Unitree G1 Humanoid and signifi- cantly outperforms fixed-range baselines, yielding highly robust traversal across diverse 3D environtmental challenges.
☆ Toward Deep Representation Learning for Event-Enhanced Visual Autonomous Perception: the eAP Dataset
Recent visual autonomous perception systems achieve remarkable performances with deep representation learning. However, they fail in scenarios with challenging illumination.While event cameras can mitigate this problem, there is a lack of a large-scale dataset to develop event-enhanced deep visual perception models in autonomous driving scenes. To address the gap, we present the eAP (event-enhanced Autonomous Perception) dataset, the largest dataset with event cameras for autonomous perception. We demonstrate how eAP can facilitate the study of different autonomous perception tasks, including 3D vehicle detection and object time-to-contact (TTC) estimation, through deep representation learning. Based on eAP, we demonstrate the ffrst successful use of events to improve a popular 3D vehicle detection network in challenging illumination scenarios. eAP also enables a devoted study of the representation learning problem of object TTC estimation. We show how a geometryaware representation learning framework leads to the best eventbased object TTC estimation network that operates at 200 FPS. The dataset, code, and pre-trained models will be made publicly available for future research.
☆ OGScene3D: Incremental Open-Vocabulary 3D Gaussian Scene Graph Mapping for Scene Understanding
Open-vocabulary scene understanding is crucial for robotic applications, enabling robots to comprehend complex 3D environmental contexts and supporting various downstream tasks such as navigation and manipulation. However, existing methods require pre-built complete 3D semantic maps to construct scene graphs for scene understanding, which limits their applicability in robotic scenarios where environments are explored incrementally. To address this challenge, we propose OGScene3D, an open-vocabulary scene understanding system that achieves accurate 3D semantic mapping and scene graph construction incrementally. Our system employs a confidence-based Gaussian semantic representation that jointly models semantic predictions and their reliability, enabling robust scene modeling. Building on this representation, we introduce a hierarchical 3D semantic optimization strategy that achieves semantic consistency through local correspondence establishment and global refinement, thereby constructing globally consistent semantic maps. Moreover, we design a long-term global optimization method that leverages temporal memory of historical observations to enhance semantic predictions. By integrating 2D-3D semantic consistency with Gaussian rendering contribution, this method continuously refines the semantic understanding of the entire scene.Furthermore, we develop a progressive graph construction approach that dynamically creates and updates both nodes and semantic relationships, allowing continuous updating of the 3D scene graphs. Extensive experiments on widely used datasets and real-world scenes demonstrate the effectiveness of our OGScene3D on open-vocabulary scene understanding.
☆ Agile Interception of a Flying Target using Competitive Reinforcement Learning
This article presents a solution to intercept an agile drone by another agile drone carrying a catching net. We formulate the interception as a Competitive Reinforcement Learning problem, where the interceptor and the target drone are controlled by separate policies trained with Proximal Policy Optimization (PPO). We introduce a high-fidelity simulation environment that integrates a realistic quadrotor dynamics model and a low-level control architecture implemented in JAX, which allows for fast parallelized execution on GPUs. We train the agents using low-level control, collective thrust and body rates, to achieve agile flights both for the interceptor and the target. We compare the performance of the trained policies in terms of catch rate, time to catch, and crash rate, against common heuristic baselines and show that our solution outperforms these baselines for interception of agile targets. Finally, we demonstrate the performance of the trained policies in a scaled real-world scenario using agile drones inside an indoor flight arena.
☆ GenZ-LIO: Generalizable LiDAR-Inertial Odometry Beyond Indoor--Outdoor Boundaries
Light detection and ranging (LiDAR)-inertial odometry (LIO) enables accurate localization and mapping for autonomous navigation in various scenes. However, its performance remains sensitive to variations in spatial scale, which refers to the spatial extent of the scene reflected in the distribution of point ranges in a LiDAR scan. Transitions between confined indoor and expansive outdoor spaces induce substantial variations in point density, which may reduce robustness and computational efficiency. To address this issue, we propose GenZ-LIO, a LIO framework generalizable across both indoor and outdoor environments. GenZ-LIO comprises three key components. First, inspired by the principle of the proportional-integral-derivative (PID) controller, it adaptively regulates the voxel size for downsampling via feedback control, driving the voxelized point count toward a scale-informed setpoint while enabling stable and efficient processing across varying scene scales. Second, we formulate a hybrid-metric state update that jointly leverages point-to-plane and point-to-point residuals to mitigate LiDAR degeneracy arising from directionally insufficient geometric constraints. Third, to alleviate the computational burden introduced by point-to-point matching, we introduce a voxel-pruned correspondence search strategy that discards non-promising voxel candidates and reduces unnecessary computations. Experimental results demonstrate that GenZ-LIO achieves robust odometry estimation and improved computational efficiency across confined indoor, open outdoor, and transitional environments. Our code will be made publicly available upon publication.
comment: 19 pages, 11 figures
☆ MG-Grasp: Metric-Scale Geometric 6-DoF Grasping Framework with Sparse RGB Observations
Single-view RGB-D grasp detection remains a com- mon choice in 6-DoF robotic grasping systems, which typically requires a depth sensor. While RGB-only 6-DoF grasp methods has been studied recently, their inaccurate geometric repre- sentation is not directly suitable for physically reliable robotic manipulation, thereby hindering reliable grasp generation. To address these limitations, we propose MG-Grasp, a novel depth- free 6-DoF grasping framework that achieves high-quality object grasping. Leveraging two-view 3D foundation model with camera intrinsic/extrinsic, our method reconstructs metric- scale and multi-view consistent dense point clouds from sparse RGB images and generates stable 6-DoF grasp. Experiments on GraspNet-1Billion dataset and real world demonstrate that MG-Grasp achieves state-of-the-art (SOTA) grasp performance among RGB-based 6-DoF grasping methods.
comment: 8 pages, 5 figures
☆ Industrial cuVSLAM Benchmark & Integration
This work presents a comprehensive benchmark evaluation of visual odometry (VO) and visual SLAM (VSLAM) systems for mobile robot navigation in real-world logistical environments. We compare multiple visual odometry approaches across controlled trajectories covering translational, rotational, and mixed motion patterns, as well as a large-scale production facility dataset spanning approximately 1.7 km. Performance is evaluated using Absolute Pose Error (APE) against ground truth from a Vicon motion capture system and a LiDAR-based SLAM reference. Our results show that a hybrid stack combining the cuVSLAM front-end with a custom SLAM back-end achieves the strongest mapping accuracy, motivating a deeper integration of cuVSLAM as the core VO component in our robotics stack. We further validate this integration by deploying and testing the cuVSLAM-based VO stack on an NVIDIA Jetson platform.
☆ Ground Reaction Inertial Poser: Physics-based Human Motion Capture from Sparse IMUs and Insole Pressure Sensors
We propose Ground Reaction Inertial Poser (GRIP), a method that reconstructs physically plausible human motion using four wearable devices. Unlike conventional IMU-only approaches, GRIP combines IMU signals with foot pressure data to capture both body dynamics and ground interactions. Furthermore, rather than relying solely on kinematic estimation, GRIP uses a digital twin of a person, in the form of a synthetic humanoid in a physics simulator, to reconstruct realistic and physically plausible motion. At its core, GRIP consists of two modules: KinematicsNet, which estimates body poses and velocities from sensor data, and DynamicsNet, which controls the humanoid in the simulator using the residual between the KinematicsNet prediction and the simulated humanoid state. To enable robust training and fair evaluation, we introduce a large-scale dataset, Pressure and Inertial Sensing for Human Motion and Interaction (PRISM), that captures diverse human motions with synchronized IMUs and insole pressure sensors. Experimental results show that GRIP outperforms existing IMU-only and IMU-pressure fusion methods across all evaluated datasets, achieving higher global pose accuracy and improved physical consistency.
☆ Featurized Occupation Measures for Structured Global Search in Numerical Optimal Control
Numerical optimal control is commonly divided between globally structured but dimensionally intractable Hamilton-Jacobi-Bellman (HJB) methods and scalable but local trajectory optimization. We introduce the Featurized Occupation Measure (FOM), a finite-dimensional primal-dual interface for the occupation-measure formulation that unifies trajectory search and global HJB-type certification. FOM is broad yet numerically tractable, covering both explicit weak-form schemes and implicit simulator- or rollout-based sampling methods. Within this framework, approximate HJB subsolutions serve as intrinsic numerical certificates to directly evaluate and guide the primal search. We prove asymptotic consistency with the exact infinite-dimensional occupation-measure problem, and show that for block-organized feasible certificates, finite-dimensional approximation preserves certified lower bounds with blockwise error and complexity control. We also establish persistence of these lower bounds under time shifts and bounded model perturbations. Consequently, these structural properties render global certificates into flexible, reusable computational objects, establishing a systematic basis for certificate-guided optimization in nonlinear control.
☆ PA-LVIO: Real-Time LiDAR-Visual-Inertial Odometry and Mapping with Pose-Only Bundle Adjustment
Real-time LiDAR-visual-inertial odometry and mapping is crucial for navigation and planning tasks in intelligent transportation systems. This study presents a pose-only bundle adjustment (PA) LiDAR-visual-inertial odometry (LVIO), named PA-LVIO, to meet the urgent need for real-time navigation and mapping. The proposed PA framework for LiDAR and visual measurements is highly accurate and efficient, and it can derive reliable frame-to-frame constraints within multiple frames. A marginalization-free and frame-to-map (F2M) LiDAR measurement model is integrated into the state estimator to eliminate odometry drifts. Meanwhile, an IMU-centric online spatial-temporal calibration is employed to obtain a pixel-wise LiDAR-camera alignment. With accurate estimated odometry and extrinsics, a high-quality and RGB-rendered point-cloud map can be built. Comprehensive experiments are conducted on both public and private datasets collected by wheeled robot, unmanned aerial vehicle (UAV), and handheld devices with 28 sequences and more than 50 km trajectories. Sufficient results demonstrate that the proposed PA-LVIO yields superior or comparable performance to state-of-the-art LVIO methods, in terms of the odometry accuracy and mapping quality. Besides, PA-LVIO can run in real-time on both the desktop PC and the onboard ARM computer.
comment: 14 pages, 10 figures
☆ Enabling Dynamic Tracking in Vision-Language-Action Models via Time-Discrete and Time-Continuous Velocity Feedforward
While vision-language-action (VLA) models have shown great promise for robot manipulation, their deployment on rigid industrial robots remains challenging due to the inherent trade-off between compliance and responsiveness. Standard Behavior Cloning (BC) approaches predict discrete poses at low frequencies, omitting the velocity and acceleration feedforward terms typically used by low-level compliant controllers. This requires to rely on high stiffness for accurate tracking, thereby sacrificing safe contact dynamics. In this paper, we demonstrate the importance of integrating velocity feedforward terms into VLA policies to resolve this trade-off. We propose two methods for extracting velocity targets from VLAs: a time-discrete finite-difference approximation that serves as a highly effective bridge for existing models, and a continuous Cubic B-Spline action space that natively yields $C^2$ continuous trajectories for high-frequency control. Crucially, both approaches are strictly model-agnostic and compatible with any standard action-chunking architecture, requiring modifications only to teleoperation, data processing, and the low-level controller. We fine-tune the $π_{0.5}$ model and evaluate both of our approaches on a demanding, contact-rich cube-in-hole task. Our results indicate that incorporating the velocity feedforward term via finite differences significantly improves task execution speed, while the continuous B-Spline approach maintains high overall success rates and provides a foundation for smoother higher-order derivatives without compromising compliance.
☆ PanguMotion: Continuous Driving Motion Forecasting with Pangu Transformers
Motion forecasting is a core task in autonomous driving systems, aiming to accurately predict the future trajectories of surrounding agents to ensure driving safety. Existing methods typically process discrete driving scenes independently, neglecting the temporal continuity and historical context correlations inherent in real-world driving environments. This paper proposes PanguMotion, a motion forecasting framework for continuous driving scenarios that integrates Transformer blocks from the Pangu-1B large language model as feature enhancement modules into autonomous driving motion prediction architectures. We conduct experiments on the Argoverse 2 datasets processed by the RealMotion data reorganization strategy, transforming each independent scene into a continuous sequence to mimic real-world driving scenarios.
☆ S-VAM: Shortcut Video-Action Model by Self-Distilling Geometric and Semantic Foresight
Video action models (VAMs) have emerged as a promising paradigm for robot learning, owing to their powerful visual foresight for complex manipulation tasks. However, current VAMs, typically relying on either slow multi-step video generation or noisy one-step feature extraction, cannot simultaneously guarantee real-time inference and high-fidelity foresight. To address this limitation, we propose S-VAM, a shortcut video-action model that foresees coherent geometric and semantic representations via a single forward pass. Serving as a stable blueprint, these foreseen representations significantly simplify the action prediction. To enable this efficient shortcut, we introduce a novel self-distillation strategy that condenses structured generative priors of multi-step denoising into one-step inference. Specifically, vision foundation model (VFM) representations extracted from the diffusion model's own multi-step generated videos provide teacher targets. Lightweight decouplers, as students, learn to directly map noisy one-step features to these targets. Extensive experiments in simulation and the real world demonstrate that our S-VAM outperforms state-of-the-art methods, enabling efficient and precise manipulation in complex environments. Our project page is https://haodong-yan.github.io/S-VAM/
☆ Enforcing Task-Specified Compliance Bounds for Humanoids via Anisotropic Lipschitz-Constrained Policies
Reinforcement learning (RL) has demonstrated substantial potential for humanoid bipedal locomotion and the control of complex motions. To cope with oscillations and impacts induced by environmental interactions, compliant control is widely regarded as an effective remedy. However, the model-free nature of RL makes it difficult to impose task-specified and quantitatively verifiable compliance objectives, and classical model-based stiffness designs are not directly applicable. Lipschitz-Constrained Policies (LCP), which regularize the local sensitivity of a policy via gradient penalties, have recently been used to smooth humanoid motions. Nevertheless, existing LCP-based methods typically employ a single scalar Lipschitz budget and lack an explicit connection to physically meaningful compliance specifications in real-world systems. In this study, we propose an anisotropic Lipschitz-constrained policy (ALCP) that maps a task-space stiffness upper bound to a state-dependent Lipschitz-style constraint on the policy Jacobian. The resulting constraint is enforced during RL training via a hinge-squared spectral-norm penalty, preserving physical interpretability while enabling direction-dependent compliance. Experiments on humanoid robots show that ALCP improves locomotion stability and impact robustness, while reducing oscillations and energy usage.
comment: Submitted to IEEE for possible publication, under review
☆ SignNav: Leveraging Signage for Semantic Visual Navigation in Large-Scale Indoor Environments
Humans routinely leverage semantic hints provided by signage to navigate to destinations within novel Large-Scale Indoor (LSI) environments, such as hospitals and airport terminals. However, this capability remains underexplored within the field of embodied navigation. This paper introduces a novel embodied navigation task, SignNav, which requires the agent to interpret semantic hint from signage and reason about the subsequent action based on current observation. To facilitate research in this domain, we construct the LSI-Dataset for the training and evaluation of various SignNav agents. Dynamically changing semantic hints and sparse placement of signage in LSI environments present significant challenges to the SignNav task. To address these challenges, we propose the Spatial-Temporal Aware Transformer (START) model for end-to-end decision-making. The spatial-aware module grounds the semantic hint of signage into physical world, while the temporal-aware module captures long-range dependencies between historical states and current observation. Leveraging a two-stage training strategy with Dataset Aggregation (DAgger), our approach achieves state-of-the-art performance, recording an 80% Success Rate (SR) and 0.74 NDTW on val-unseen split. Real-world deployment further demonstrates the practicality of our method in physical environment without pre-built map.
☆ SE(3)-LIO: Smooth IMU Propagation With Jointly Distributed Poses on SE(3) Manifold for Accurate and Robust LiDAR-Inertial Odometry
In estimating odometry accurately, an inertial measurement unit (IMU) is widely used owing to its high-rate measurements, which can be utilized to obtain motion information through IMU propagation. In this paper, we address the limitations of existing IMU propagation methods in terms of motion prediction and motion compensation. In motion prediction, the existing methods typically represent a 6-DoF pose by separating rotation and translation and propagate them on their respective manifold, so that the rotational variation is not effectively incorporated into translation propagation. During motion compensation, the relative transformation between predicted poses is used to compensate motion-induced distortion in other measurements, while inherent errors in the predicted poses introduce uncertainty in the relative transformation. To tackle these challenges, we represent and propagate the pose on SE(3) manifold, where propagated translation properly accounts for rotational variation. Furthermore, we precisely characterize the relative transformation uncertainty by considering the correlation between predicted poses, and incorporate this uncertainty into the measurement noise during motion compensation. To this end, we propose a LiDAR-inertial odometry (LIO), referred to as SE(3)-LIO, that integrates the proposed IMU propagation and uncertainty-aware motion compensation (UAMC). We validate the effectiveness of SE(3)-LIO on diverse datasets. Our source code and additional material are available at: https://se3-lio.github.io/.
☆ Towards the Vision-Sound-Language-Action Paradigm: The HEAR Framework for Sound-Centric Manipulation
While recent Vision-Language-Action (VLA) models have begun to incorporate audio, they typically treat sound as static pre-execution prompts or focus exclusively on human speech. This leaves a significant gap in real-time, sound-centric manipulation where fleeting environmental acoustics provide critical state verification during task execution. Consequently, key sounds are easily missed due to low-frequency updates or system latency. This problem is exacerbated by action chunking with open-loop execution, which creates a Blind Execution Interval where acoustic events are lost between discrete audio observation windows. Recognizing the necessity of continuous auditory awareness, we formalize Vision-Sound-Language-Action (VSLA) as a continuous control paradigm conditioned on vision, streaming audio, language, and proprioception under delayed decision loops. As an instantiation, we introduce HEAR, a VSLA framework integrating four components: (i) a streaming Historizer to maintain a compact, causal audio context across execution gaps; (ii) an Envisioner adapted from omni foundation models to reason over multi-sensory inputs; (iii) an Advancer, formulated as an audio world model, to learn temporal dynamics by predicting near-future audio codes; and (iv) a flow-matching Realizer policy to generate smooth action chunks. To address the scarcity of pretraining data and evaluations for VSLA, we construct OpenX-Sound for pretraining, alongside HEAR-Bench, the first sound-centric manipulation benchmark with strict causal timing rules. Our results suggest that robust sound-centric manipulation necessitates causal persistence and explicit temporal learning. This framework provides a practical step toward multi-sensory foundation models for embodied agents, enabling robots to perceive and interact with dynamic environments. Code and videos are available at https://hear.irmv.top.
☆ Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models
Reinforcement Learning (RL) has shown great potential in refining robotic manipulation policies, yet its efficacy remains strongly bottlenecked by the difficulty of designing generalizable reward functions. In this paper, we propose a framework for online policy refinement by adapting foundation VLMs into online reward generators. We develop a robust, scalable reward model based on a state-of-the-art VLM, trained on a large-scale, multi-source dataset encompassing real-world robot trajectories, human-object interactions, and diverse simulated environments. Unlike prior approaches that evaluate entire trajectories post-hoc, our method leverages the VLM to formulate a multifaceted reward signal comprising process, completion, and temporal contrastive rewards based on current visual observations. Initializing with a base policy trained via Imitation Learning (IL), we employ these VLM rewards to guide the model to correct sub-optimal behaviors in a closed-loop manner. We evaluate our framework on challenging long-horizon manipulation benchmarks requiring sequential execution and precise control. Crucially, our reward model operates in a purely zero-shot manner within these test environments. Experimental results demonstrate that our method significantly improves the success rate of the initial IL policy within just 30 RL iterations, demonstrating remarkable sample efficiency. This empirical evidence highlights that VLM-generated signals can provide reliable feedback to resolve execution errors, effectively eliminating the need for manual reward engineering and facilitating efficient online refinement for robot learning.
☆ Ultrafast Sampling-based Kinodynamic Planning via Differential Flatness
Motion planning under dynamics constraints, i.e., kinodynamic planning, enables safe robot operation by generating dynamically feasible trajectories that the robot can accurately track. For high-\dof robots such as manipulators, sampling-based motion planners are commonly used, especially for complex tasks in cluttered environments. However, enforcing constraints on robot dynamics in such planners requires solving either challenging two-point boundary value problems (BVPs) or propagating robot dynamics over time, both of which are computational bottlenecks that drastically increase planning times. Meanwhile, recent efforts have shown that sampling-based motion planners can generate plans in microseconds using parallelization, but are limited to geometric paths. This paper develops AkinoPDF, a fast parallelized sampling-based kinodynamic motion planning technique for a broad class of differentially flat robot systems, including manipulators, ground and aerial vehicles, and more. Differential flatness allows us to transform the motion planning problem from the original state space to a flat output space, where an analytical time-parameterized solution of the BVP and dynamics integration can be obtained. A trajectory in the flat output space is then converted back to a closed-form dynamically feasible trajectory in the original state space, enabling fast validation via ``single instruction, multiple data" parallelism. Our method is fast, exact, and compatible with any sampling-based motion planner. We extensively verify the effectiveness of our approach in both simulated benchmarks and real experiments with cluttered and dynamic environments, requiring mere microseconds to milliseconds of planning time.
comment: 16 pages, 9 figures, under review
☆ The Era of End-to-End Autonomy: Transitioning from Rule-Based Driving to Large Driving Models
Autonomous driving is undergoing a shift from modular rule based pipelines toward end to end (E2E) learning systems. This paper examines this transition by tracing the evolution from classical sense perceive plan control architectures to large driving models (LDMs) capable of mapping raw sensor input directly to driving actions. We analyze recent developments including Tesla's Full Self Driving (FSD) V12 V14, Rivian's Unified Intelligence platform, NVIDIA Cosmos, and emerging commercial robotaxi deployments, focusing on architectural design, deployment strategies, safety considerations and industry implications. A key emerging product category is supervised E2E driving, often referred to as FSD (Supervised) or L2 plus plus, which several manufacturers plan to deploy from 2026 onwards. These systems can perform most of the Dynamic Driving Task (DDT) in complex environments while requiring human supervision, shifting the driver's role to safety oversight. Early operational evidence suggests E2E learning handles the long tail distribution of real world driving scenarios and is becoming a dominant commercial strategy. We also discuss how similar architectural advances may extend beyond autonomous vehicles (AV) to other embodied AI systems, including humanoid robotics.
☆ Compact Optical Single-axis Joint Torque Sensor Using Redundant Photo-Reflectors and Quadratic-Programming Calibration
This study proposes a non-contact photo-reflector-based joint torque sensor for precise joint-level torque control and safe physical interaction. Current-sensor-based torque estimation in many collaborative robots suffers from poor low-torque accuracy due to gearbox stiction/friction and current-torque nonlinearity, especially near static conditions. The proposed sensor optically measures micro-deformation of an elastic structure and employs a redundant array of photo-reflectors arranged in four directions to improve sensitivity and signal-to-noise ratio. We further present a quadratic-programming-based calibration method that exploits redundancy to suppress noise and enhance resolution compared to least-squares calibration. The sensor is implemented in a compact form factor (96 mm diameter, 12 mm thickness). Experiments demonstrate a maximum error of 0.083%FS and an RMS error of 0.0266 Nm for z-axis torque measurement. Calibration tests show that the proposed calibration achieves a 3 sigma resolution of 0.0224 Nm at 1 kHz without filtering, corresponding to a 2.14 times improvement over the least-squares baseline. Temperature chamber characterization and rational fitting based compensation mitigate zero drift induced by MCU self heating and motor heat. Motor-level validation via torque control and admittance control confirms improved low torque tracking and disturbance robustness relative to current-sensor-based control.
comment: 10 pages
☆ Geometry-Aligned LLM Fine-Tuning for Sequential Narrow-Opening Planning
We study rigid-body motion planning through multiple sequential narrow openings, which requires long-horizon geometric reasoning because the configuration used to traverse an early opening constrains the set of reachable configurations for subsequent ones. To achieve this, we propose a geometry-aligned large language model (LLM) fine-tuning framework that generates fixed-length, machine-readable waypoint sequences that are both geometrically feasible and coordinated across openings. Our approach uses a bi-level training pipeline. First, we perform failure-driven LoRA supervised fine-tuning (SFT) on human demonstrations, which incorporates structured failure feedback to teach the model common failure modes and enforce the output format. Second, we refine the same LoRA adapters using Group Relative Policy Optimization (GRPO) with geometric verification: each sampled waypoint sequence is densified by a model-based planner and scored with a deterministic geometry-derived reward to achieve continuous-motion feasibility. To validate the effectiveness of our proposed method, we provide both quantitative and qualitative results from simulations. Our method achieves the highest success rate in both in-distribution and out-of-distribution environments and qualitatively exhibits long-horizon geometric reasoning by selecting exit poses that facilitate entry into subsequent openings.
comment: 8 pages, 3 figures
☆ FastLoop: Parallel Loop Closing with GPU-Acceleration in Visual SLAM
Visual SLAM systems combine visual tracking with global loop closure to maintain a consistent map and accurate localization. Loop closure is a computationally expensive process as we need to search across the whole map for matches. This paper presents FastLoop, a GPU-accelerated loop closing module to alleviate this computational complexity. We identify key performance bottlenecks in the loop closing pipeline of visual SLAM and address them through parallel optimizations on the GPU. Specifically, we use task-level and data-level parallelism and integrate a GPU-accelerated pose graph optimization. Our implementation is built on top of ORB-SLAM3 and leverages CUDA for GPU programming. Experimental results show that FastLoop achieves an average speedup of 1.4x and 1.3x on the EuRoC dataset and 3.0x and 2.4x on the TUM-VI dataset for the loop closing module on desktop and embedded platforms, respectively, while maintaining the accuracy of the original system.
☆ Influence of Gripper Design on Human Demonstration Quality for Robot Learning
Opening sterile medical packaging is routine for healthcare workers but remains challenging for robots. Learning from demonstration enables robots to acquire manipulation skills directly from humans, and handheld gripper tools such as the Universal Manipulation Interface (UMI) offer a pathway for efficient data collection. However, the effectiveness of these tools depends heavily on their usability. We evaluated UMI in demonstrating a bandage opening task, a common manipulation task in hospital settings, by testing three conditions: distributed load grippers, concentrated load grippers, and bare hands. Eight participants performed timed trials, with task performance assessed by success rate, completion time, and damage, alongside perceived workload using the NASA-TLX questionnaire. Concentrated load grippers improved performance relative to distributed load grippers but remained substantially slower and less effective than hands. These results underscore the importance of ergonomic and mechanical refinements in handheld grippers to reduce user burden and improve demonstration quality, especially for applications in healthcare robotics.
comment: To be published in proceedings of 2026 IEEE International Conference on Robotics & Automation
☆ SLAM Adversarial Lab: An Extensible Framework for Visual SLAM Robustness Evaluation under Adverse Conditions
We present SAL (SLAM Adversarial Lab), a modular framework for evaluating visual SLAM systems under adversarial conditions such as fog and rain. SAL represents each adversarial condition as a perturbation that transforms an existing dataset into an adversarial dataset. When transforming a dataset, SAL supports severity levels using easily-interpretable real-world units such as meters for fog visibility. SAL's extensible architecture decouples datasets, perturbations, and SLAM algorithms through common interfaces, so users can add new components without rewriting integration code. Moreover, SAL includes a search procedure that finds the severity level of a perturbation at which a SLAM system fails. To showcase the capabilities of SAL, our evaluation integrates seven SLAM algorithms and evaluates them across three datasets under weather, camera, and video transport perturbations.
comment: 8 pages, 4 figures
☆ BEV-SLD: Self-Supervised Scene Landmark Detection for Global Localization with LiDAR Bird's-Eye View Images CVPR 2026
We present BEV-SLD, a LiDAR global localization method building on the Scene Landmark Detection (SLD) concept. Unlike scene-agnostic pipelines, our self-supervised approach leverages bird's-eye-view (BEV) images to discover scene-specific patterns at a prescribed spatial density and treat them as landmarks. A consistency loss aligns learnable global landmark coordinates with per-frame heatmaps, yielding consistent landmark detections across the scene. Across campus, industrial, and forest environments, BEV-SLD delivers robust localization and achieves strong performance compared to state-of-the-art methods.
comment: Accepted to CVPR 2026
☆ Shielded Reinforcement Learning Under Dynamic Temporal Logic Constraints
Reinforcement Learning (RL) has shown promise in various robotics applications, yet its deployment on real systems is still limited due to safety and operational constraints. The safe RL field has gained considerable attention in recent years, which focuses on imposing safety constraints throughout the learning process. However, real systems often require more complex constraints than just safety, such as periodic recharging or time-bounded visits to specific regions. Imposing such spatio-temporal tasks during learning still remains a challenge. Signal Temporal Logic (STL) is a formal language for specifying temporal properties of real-valued signals and provides a way to express such complex tasks. In this paper, we propose a framework that leverages sequential control barrier functions and model-free RL to ensure that the given STL tasks are satisfied throughout the learning process. Our method extends beyond traditional safety constraints by enforcing rich STL specifications, which can involve visits to dynamic targets with unknown trajectories. We also demonstrate the effectiveness of our framework through various simulations.
comment: 7 pages, 3 figures, 2026 IEEE American Control Conference (ACC)
☆ SLowRL: Safe Low-Rank Adaptation Reinforcement Learning for Locomotion
Sim-to-real transfer of locomotion policies often leads to performance degradation due to the inevitable sim-to-real gap. Naively fine-tuning these policies directly on hardware is problematic, as it poses risks of mechanical failure and suffers from high sample inefficiency. In this paper, we address the challenge of safely and efficiently fine-tuning reinforcement learning (RL) policies for dynamic locomotion tasks. Specifically, we focus on fine-tuning policies learned in simulation directly on hardware, while explicitly enforcing safety constraints. In doing so, we introduce SLowRL, a framework that combines Low-Rank Adaptation (LoRA) with training-time safety enforcement via a recovery policy. We evaluate our method both in simulation and on a real Unitree Go2 quadruped robot for jump and trot tasks. Experimental results show that our method achieves a $46.5\%$ reduction in fine-tuning time and near-zero safety violations compared to standard proximal policy optimization (PPO) baselines. Notably, we find that a rank-1 adaptation alone is sufficient to recover pre-trained performance in the real world, while maintaining stable and safe real-world fine-tuning. These results demonstrate the practicality of safe, efficient fine-tuning for dynamic real-world robotic applications.
☆ TrackDeform3D: Markerless and Autonomous 3D Keypoint Tracking and Dataset Collection for Deformable Objects
Structured 3D representations such as keypoints and meshes offer compact, expressive descriptions of deformable objects, jointly capturing geometric and topological information useful for downstream tasks such as dynamics modeling and motion planning. However, robustly extracting such representations remains challenging, as current perception methods struggle to handle complex deformations. Moreover, large-scale 3D data collection remains a bottleneck: existing approaches either require prohibitive data collection efforts, such as labor-intensive annotation or expensive motion capture setups, or rely on simplifying assumptions that break down in unstructured environments. As a result, large-scale 3D datasets and benchmarks for deformable objects remain scarce. To address these challenges, this paper presents an affordable and autonomous framework for collecting 3D datasets of deformable objects using only RGB-D cameras. The proposed method identifies 3D keypoints and robustly tracks their trajectories, incorporating motion consistency constraints to produce temporally smooth and geometrically coherent data. TrackDeform3D is evaluated against several state-of-the-art tracking methods across diverse object categories and demonstrates consistent improvements in both geometric and tracking accuracy. Using this framework, this paper presents a high-quality, large-scale dataset consisting of 6 deformable objects, totaling 110 minutes of trajectory data.
☆ TeleDex: Accessible Dexterous Teleoperation
Despite increasing dataset scale and model capacity, robot manipulation policies still struggle to generalize beyond their training distributions. As a result, deploying state-of-the-art policies in new environments, tasks, or robot embodiments often requires collecting additional demonstrations. Enabling this in real-world deployment settings requires tools that allow users to collect demonstrations quickly, affordably, and with minimal setup. We present TeleDex, an open-source system for intuitive teleoperation of dexterous hands and robotic manipulators using any readily available phone. The system streams low-latency 6-DoF wrist poses and articulated 21-DoF hand state estimates from the phone, which are retargeted to robot arms and multi-fingered hands without requiring external tracking infrastructure. TeleDex supports both a handheld phone-only mode and an optional 3D-printable hand-mounted interface for finger-level teleoperation. By lowering the hardware and setup barriers to dexterous teleoperation, TeleDex enables users to quickly collect demonstrations during deployment to support policy fine-tuning. We evaluate the system across simulation and real-world manipulation tasks, demonstrating its effectiveness as a unified scalable interface for robot teleoperation. All software and hardware designs, along with demonstration videos, are open-source and available at orayyan.com/teledex.
comment: For project website and videos, see https://www.orayyan.com/teledex
☆ Asymmetric Nash Seeking via Best Response Maps: Global Linear Convergence and Robustness to Inexact Reaction Models
Nash equilibria provide a principled framework for modeling interactions in multi-agent decision-making and control. However, many equilibrium-seeking methods implicitly assume that each agent has access to the other agents' objectives and constraints, an assumption that is often unrealistic in practice. This letter studies a class of asymmetric-information two-player constrained games with decoupled feasible sets, in which Player 1 knows its own objective and constraints while Player 2 is available only through a best-response map. For this class of games, we propose an asymmetric projected gradient descent-best response iteration that does not require full mutual knowledge of both players' optimization problems. Under suitable regularity conditions, we establish the existence and uniqueness of the Nash equilibrium and prove global linear convergence of the proposed iteration when the best-response map is exact. Recognizing that best-response maps are often learned or estimated, we further analyze the inexact case and show that, when the approximation error is uniformly bounded by $\varepsilon$, the iterates enter an explicit $O(\varepsilon)$ neighborhood of the true Nash equilibrium. Numerical results on a benchmark game corroborate the predicted convergence behavior and error scaling.
comment: 6 Pages, 2 Figures, Preprint submitted to IEEE L-CSS and CDC 2026
☆ Contingency-Aware Planning via Certified Neural Hamilton-Jacobi Reachability
Hamilton-Jacobi (HJ) reachability provides formal safety guarantees for dynamical systems, but solving high-dimensional HJ partial differential equations limits its use in real-time planning. This paper presents a contingency-aware multi-goal navigation framework that integrates learning-based reachability with sampling-based planning in unknown environments. We use Fourier Neural Operator (FNO) to approximate the solution operator of the Hamilton-Jacobi-Isaacs variational inequality under varying obstacle configurations. We first provide a theoretical under-approximation guarantee on the safe backward reach-avoid set, which enables formal safety certification of the learned reachable sets. Then, we integrate the certified reachable sets with an incremental multi-goal planner, which enforces reachable-set constraints and a recovery policy that guarantees finite-time return to a safe region. Overall, we demonstrate that the proposed framework achieves asymptotically optimal navigation with provable contingency behavior, and validate its performance through real-time deployment on KUKA's youBot in Webots simulation.
comment: 9 pages, 4 figures
☆ Efficient and Reliable Teleoperation through Real-to-Sim-to-Real Shared Autonomy
Fine-grained, contact-rich teleoperation remains slow, error-prone, and unreliable in real-world manipulation tasks, even for experienced operators. Shared autonomy offers a promising way to improve performance by combining human intent with automated assistance, but learning effective assistance in simulation requires a faithful model of human behavior, which is difficult to obtain in practice. We propose a real-to-sim-to-real shared autonomy framework that augments human teleoperation with learned corrective behaviors, using a simple yet effective k-nearest-neighbor (kNN) human surrogate to model operator actions in simulation. The surrogate is fit from less than five minutes of real-world teleoperation data and enables stable training of a residual copilot policy with model-free reinforcement learning. The resulting copilot is deployed to assist human operators in real-world fine-grained manipulation tasks. Through simulation experiments and a user study with sixteen participants on industry-relevant tasks, including nut threading, gear meshing, and peg insertion, we show that our system improves task success for novice operators and execution efficiency for experienced operators compared to direct teleoperation and shared-autonomy baselines that rely on expert priors or behavioral-cloning pilots. In addition, copilot-assisted teleoperation produces higher-quality demonstrations for downstream imitation learning.
comment: Project Page: https://residual-copilot.github.io/
☆ MessyKitchens: Contact-rich object-level 3D scene reconstruction
Monocular 3D scene reconstruction has recently seen significant progress. Powered by the modern neural architectures and large-scale data, recent methods achieve high performance in depth estimation from a single image. Meanwhile, reconstructing and decomposing common scenes into individual 3D objects remains a hard challenge due to the large variety of objects, frequent occlusions and complex object relations. Notably, beyond shape and pose estimation of individual objects, applications in robotics and animation require physically-plausible scene reconstruction where objects obey physical principles of non-penetration and realistic contacts. In this work we advance object-level scene reconstruction along two directions. First, we introduceMessyKitchens, a new dataset with real-world scenes featuring cluttered environments and providing high-fidelity object-level ground truth in terms of 3D object shapes, poses and accurate object contacts. Second, we build on the recent SAM 3D approach for single-object reconstruction and extend it with Multi-Object Decoder (MOD) for joint object-level scene reconstruction. To validate our contributions, we demonstrate MessyKitchens to significantly improve previous datasets in registration accuracy and inter-object penetration. We also compare our multi-object reconstruction approach on three datasets and demonstrate consistent and significant improvements of MOD over the state of the art. Our new benchmark, code and pre-trained models will become publicly available on our project website: https://messykitchens.github.io/.
☆ ManiTwin: Scaling Data-Generation-Ready Digital Object Dataset to 100K
Learning in simulation provides a useful foundation for scaling robotic manipulation capabilities. However, this paradigm often suffers from a lack of data-generation-ready digital assets, in both scale and diversity. In this work, we present ManiTwin, an automated and efficient pipeline for generating data-generation-ready digital object twins. Our pipeline transforms a single image into simulation-ready and semantically annotated 3D asset, enabling large-scale robotic manipulation data generation. Using this pipeline, we construct ManiTwin-100K, a dataset containing 100K high-quality annotated 3D assets. Each asset is equipped with physical properties, language descriptions, functional annotations, and verified manipulation proposals. Experiments demonstrate that ManiTwin provides an efficient asset synthesis and annotation workflow, and that ManiTwin-100K offers high-quality and diverse assets for manipulation data generation, random scene synthesis, and VQA data generation, establishing a strong foundation for scalable simulation data synthesis and policy learning. Our webpage is available at https://manitwin.github.io/.
comment: Website: https://manitwin.github.io/
☆ MolmoB0T: Large-Scale Simulation Enables Zero-Shot Manipulation
A prevailing view in robot learning is that simulation alone is not enough; effective sim-to-real transfer is widely believed to require at least some real-world data collection or task-specific fine-tuning to bridge the gap between simulated and physical environments. We challenge that assumption. With sufficiently large-scale and diverse simulated synthetic training data, we show that zero-shot transfer to the real world is not only possible, but effective for both static and mobile manipulation. We introduce MolmoBot-Engine, a fully open-source pipeline for procedural data generation across robots, tasks, and diverse simulated environments in MolmoSpaces. With it, we release MolmoBot-Data, a dataset of 1.8 million expert trajectories for articulated object manipulation and pick-and-place tasks. We train three policy classes: MolmoBot, a Molmo2-based multi-frame vision-language model with a flow-matching action head; MolmoBot-Pi0, which replicates the $π_0$ architecture to enable direct comparison; and MolmoBot-SPOC, a lightweight policy suitable for edge deployment and amenable to RL fine-tuning. We evaluate on two robotic platforms: the Franka FR3 for tabletop manipulation tasks and the Rainbow Robotics RB-Y1 mobile manipulator for door opening, drawer manipulation, cabinet interaction, and mobile pick-and-place. Without any real-world fine-tuning, our policies achieve zero-shot transfer to unseen objects and environments. On tabletop pick-and-place, MolmoBot achieves a success rate of 79.2% in real world evaluations across 4 settings, outperforming $π_{0.5}$ at 39.2%. Our results demonstrate that procedural environment generation combined with diverse articulated assets can produce robust manipulation policies that generalize broadly to the real world. Technical Blog: https://allenai.org/blog/molmobot-robot-manipulation
☆ DreamPlan: Efficient Reinforcement Fine-Tuning of Vision-Language Planners via Video World Models
Robotic manipulation requires sophisticated commonsense reasoning, a capability naturally possessed by large-scale Vision-Language Models (VLMs). While VLMs show promise as zero-shot planners, their lack of grounded physical understanding often leads to compounding errors and low success rates when deployed in complex real-world environments, particularly for challenging tasks like deformable object manipulation. Although Reinforcement Learning (RL) can adapt these planners to specific task dynamics, directly fine-tuning VLMs via real-world interaction is prohibitively expensive, unsafe, and sample-inefficient. To overcome this bottleneck, we introduce DreamPlan, a novel framework for the reinforcement fine-tuning of VLM planners via video world models. Instead of relying on costly physical rollouts, DreamPlan first leverages the zero-shot VLM to collect exploratory interaction data. We demonstrate that this sub-optimal data is sufficient to train an action-conditioned video generation model, which implicitly captures complex real-world physics. Subsequently, the VLM planner is fine-tuned entirely within the "imagination" of this video world model using Odds Ratio Policy Optimization (ORPO). By utilizing these virtual rollouts, physical and task-specific knowledge is efficiently injected into the VLM. Our results indicate that DreamPlan bridges the gap between semantic reasoning and physical grounding, significantly improving manipulation success rates without the need for large-scale real-world data collection. Our project page is https://psi-lab.ai/DreamPlan/.
☆ BrickSim: A Physics-Based Simulator for Manipulating Interlocking Brick Assemblies
Interlocking brick assemblies provide a standardized yet challenging testbed for contact-rich and long-horizon robotic manipulation, but existing rigid-body simulators do not faithfully capture snap-fit mechanics. We present BrickSim, the first real-time physics-based simulator for interlocking brick assemblies. BrickSim introduces a compact force-based mechanics model for snap-fit connections and solves the resulting internal force distribution using a structured convex quadratic program. Combined with a hybrid architecture that delegates rigid-body dynamics to the underlying physics engine while handling snap-fit mechanics separately, BrickSim enables real-time, high-fidelity simulation of assembly, disassembly, and structural collapse. On 150 real-world assemblies, BrickSim achieves 100% accuracy in static stability prediction with an average solve time of 5 ms. In dynamic drop tests, it also faithfully reproduces real-world structural collapse, precisely mirroring both the occurrence of breakage and the specific breakage locations. Built on Isaac Sim, BrickSim further supports seamless integration with a wide variety of robots and existing pipelines. We demonstrate robotic construction of brick assemblies using BrickSim, highlighting its potential as a foundation for research in dexterous, long-horizon robotic manipulation. BrickSim is open-source, and the code is available at https://github.com/intelligent-control-lab/BrickSim.
comment: 9 pages, 9 figures
☆ Real-Time Decoding of Movement Onset and Offset for Brain-Controlled Rehabilitation Exoskeleton ICRA 2026
Robot-assisted therapy can deliver high-dose, task-specific training after neurologic injury, but most systems act primarily at the limb level-engaging the impaired neural circuits only indirectly-which remains a key barrier to truly contingent, neuroplasticity-targeted rehabilitation. We address this gap by implementing online, dual-state motor imagery control of an upper-limb exoskeleton, enabling goal-directed reaches to be both initiated and terminated directly from non-invasive EEG. Eight participants used EEG to initiate assistance and then volitionally halt the robot mid-trajectory. Across two online sessions, group-mean hit rates were 61.5% for onset and 64.5% for offset, demonstrating reliable start-stop command delivery despite instrumental noise and passive arm motion. Methodologically, we reveal a systematic, class-driven bias induced by common task-based recentering using an asymmetric margin diagnostic, and we introduce a class-agnostic fixation-based recentering method that tracks drift without sampling command classes while preserving class geometry. This substantially improves threshold-free separability (AUC gains: onset +56%, p = 0.0117; offset +34%, p = 0.0251) and reduces bias within and across days. Together, these results help bridge offline decoding and practical, intention-driven start-stop control of a rehabilitation exoskeleton, enabling precisely timed, contingent assistance aligned with neuroplasticity goals while supporting future clinical translation.
comment: Accepted to ICRA 2026. 8 pages, 5 figures. Project page available at https://mitrakanishka.github.io/projects/startstop-bci/
☆ CABTO: Context-Aware Behavior Tree Grounding for Robot Manipulation
Behavior Trees (BTs) offer a powerful paradigm for designing modular and reactive robot controllers. BT planning, an emerging field, provides theoretical guarantees for the automated generation of reliable BTs. However, BT planning typically assumes that a well-designed BT system is already grounded -- comprising high-level action models and low-level control policies -- which often requires extensive expert knowledge and manual effort. In this paper, we formalize the BT Grounding problem: the automated construction of a complete and consistent BT system. We analyze its complexity and introduce CABTO (Context-Aware Behavior Tree grOunding), the first framework to efficiently solve this challenge. CABTO leverages pre-trained Large Models (LMs) to heuristically search the space of action models and control policies, guided by contextual feedback from BT planners and environmental observations. Experiments spanning seven task sets across three distinct robotic manipulation scenarios demonstrate CABTO's effectiveness and efficiency in generating complete and consistent behavior tree systems.
☆ Development of Low-Cost and Bidirectional Syringe Pumps for Soft Robotics Applications
Soft robotics leverages deformable materials to develop robots capable of navigating unstructured and dynamic environments. Silicone Voxel-Based Soft Robots (Silibots) are a type of pneumatically actuated soft robots that rely on the inflation and deflation of their voxels for shape-shifting behaviors. However, traditional pneumatic actuation methods (high pressure solenoids, medical diaphragm pumps, micro compressors, compressed fluid) pose significant challenges due to their limited efficacy, cost, complexity, or lack of precision. This work introduces a low cost and modular syringe pump system, constructed with off the shelf and 3D printed parts, designed to overcome these limitations. The syringe pump system also enhances actuation with the unique ability to pull a vacuum as well pump air into the soft robot. Furthermore, the syringe pump features modular hardware and customizable software, allowing for researchers to tailor the syringe pump to their requirements or operate multiple pumps simultaneously with unique pump parameters. This flexibility makes the syringe pump an accessible and scalable tool that paves the way for broader adoption of soft robotic technologies in research and education.
☆ Beyond Cybathlon: On-demand Quadrupedal Assistance for People with Limited Mobility
Background: Assistance robots have the potential to increase the independence of people who need daily care due to limited mobility or being wheelchair-bound. Current solutions of attaching robotic arms to motorized wheelchairs offer limited additional mobility at the cost of increased size and reduced wheelchair maneuverability. Methods: We present an on-demand quadrupedal assistance robot system controlled via a shared autonomy approach, which combines semi-autonomous task execution with human teleoperation. Due to the mobile nature of the system it can assist the operator whenever needed and perform autonomous tasks independently, without otherwise restricting their mobility. We automate pick-and-place tasks, as well as robot movement through the environment with semantic, collision-aware navigation. For teleoperation, we present a mouth-level joystick interface that enables an operator with reduced mobility to control the robot's end effector for precision manipulation. Results: We showcase our system in the \textit{Cybathlon 2024 Assistance Robot Race}, and validate it in an at-home experimental setup, where we measure task completion times and user satisfaction. We find our system capable of assisting in a broad variety of tasks, including those that require dexterous manipulation. The user study confirms the intuition that increased robot autonomy alleviates the operator's mental load. Conclusions: We present a flexible system that has the potential to help people in wheelchairs maintain independence in everyday life by enabling them to solve mobile manipulation problems without external support. We achieve results comparable to previous state-of-the-art on subjective metrics while allowing for more autonomy of the operator and greater agility for manipulation.
☆ Thermopneumatic Pixels for Fast, Localized, Low-Voltage Touch Feedback
We present thermopneumatic pixels (TPPs), which are tactile actuators designed for rapid fabrication and straightforward integration into compact wearable and surface-based haptic systems. Each TPP converts low-voltage ($\sim$10 V) electrical pulses into transient pressure increases within a sealed cavity, producing out-of-plane forces and displacements suitable for tactile stimulation. The architecture enables scalable fabrication and spatially distributed actuation while maintaining simple electrical interfacing. The TPPs are constructed from inexpensive, readily available materials using straightforward layer-based assembly, facilitating rapid prototyping and integration into interactive devices. Mechanical characterization demonstrates peak forces exceeding 1 N and millimeter displacements. We further present driving electronics for operating multiple TPP modules concurrently and report perceptual study results demonstrating the effectiveness of the resulting tactile feedback. Together, these results establish low-voltage thermopneumatic actuation as an accessible and high-performance approach for embedding tactile feedback into experimental and consumer-facing interfaces.
☆ vAccSOL: Efficient and Transparent AI Vision Offloading for Mobile Robots
Mobile robots are increasingly deployed for inspection, patrol, and search-and-rescue operations, relying on computer vision for perception, navigation, and autonomous decision-making. However, executing modern vision workloads onboard is challenging due to limited compute resources and strict energy constraints. While some platforms include embedded accelerators, these are typically tied to proprietary software stacks, leaving user-defined workloads to run on resource-constrained companion computers. We present vAccSOL, a framework for efficient and transparent execution of AI-based vision workloads across heterogeneous robotic and edge platforms. vAccSOL integrates two components: SOL, a neural network compiler that generates optimized inference libraries with minimal runtime dependencies, and vAccel, a lightweight execution framework that transparently dispatches inference locally on the robot or to nearby edge infrastructure. This combination enables hardware-optimized inference and flexible execution placement without requiring modifications to robot applications. We evaluate vAccSOL on a real-world testbed with a commercial quadruped robot and twelve deep learning models covering image classification, video classification, and semantic segmentation. Compared to a PyTorch compiler baseline, SOL achieves comparable or better inference performance. With edge offloading, vAccSOL reduces robot-side power consumption by up to 80% and edge-side power by up to 60% compared to PyTorch, while increasing vision pipeline frame rate by up to 24x, extending the operating lifetime of battery-powered robots.
☆ Learning Whole-Body Control for a Salamander Robot
Amphibious legged robots inspired by salamanders are promising in applications in complex amphibious environments. However, despite the significant success of training controllers that achieve diverse locomotion behaviors in conventional quadrupedal robots, most salamander robots relied on central-pattern-generator (CPG)-based and model-based coordination strategies for locomotion control. Learning unified joint-level whole-body control that reliably transfers from simulation to highly articulated physical salamander robots remains relatively underexplored. In addition, few legged robots have tried learning-based controllers in amphibious environments. In this work, we employ Reinforcement Learning to map proprioceptive observations and commanded velocities to joint-level actions, allowing coordinated locomotor behaviors to emerge. To deploy these policies on hardware, we adopt a system-level real-to-sim matching and sim-to-real transfer strategy. The learned controller achieves stable and coordinated walking on both flat and uneven terrains in the real world. Beyond terrestrial locomotion, the framework enables transitions between walking and swimming in simulation, highlighting a phenomenon of interest for understanding locomotion across distinct physical modes.
☆ When Should a Robot Think? Resource-Aware Reasoning via Reinforcement Learning for Embodied Robotic Decision-Making
Embodied robotic systems increasingly rely on large language model (LLM)-based agents to support high-level reasoning, planning, and decision-making during interactions with the environment. However, invoking LLM reasoning introduces substantial computational latency and resource overhead, which can interrupt action execution and reduce system reliability. Excessive reasoning may delay actions, while insufficient reasoning often leads to incorrect decisions and task failures. This raises a fundamental question for embodied agents: when should the agent reason, and when should it act? In this work, we propose RARRL (Resource-Aware Reasoning via Reinforcement Learning), a hierarchical framework for resource-aware orchestration of embodied agents. Rather than learning low-level control policies, RARRL learns a high-level orchestration policy that operates at the agent's decision-making layer. This policy enables the agent to adaptively determine whether to invoke reasoning, which reasoning role to employ, and how much computational budget to allocate based on current observations, execution history, and remaining resources. Extensive experiments, including evaluations with empirical latency profiles derived from the ALFRED benchmark, show that RARRL consistently improves task success rates while reducing execution latency and enhancing robustness compared with fixed or heuristic reasoning strategies. These results demonstrate that adaptive reasoning control is essential for building reliable and efficient embodied robotic agents.
☆ Kinema4D: Kinematic 4D World Modeling for Spatiotemporal Embodied Simulation
Simulating robot-world interactions is a cornerstone of Embodied AI. Recently, a few works have shown promise in leveraging video generations to transcend the rigid visual/physical constraints of traditional simulators. However, they primarily operate in 2D space or are guided by static environmental cues, ignoring the fundamental reality that robot-world interactions are inherently 4D spatiotemporal events that require precise interactive modeling. To restore this 4D essence while ensuring the precise robot control, we introduce Kinema4D, a new action-conditioned 4D generative robotic simulator that disentangles the robot-world interaction into: i) Precise 4D representation of robot controls: we drive a URDF-based 3D robot via kinematics, producing a precise 4D robot control trajectory. ii) Generative 4D modeling of environmental reactions: we project the 4D robot trajectory into a pointmap as a spatiotemporal visual signal, controlling the generative model to synthesize complex environments' reactive dynamics into synchronized RGB/pointmap sequences. To facilitate training, we curated a large-scale dataset called Robo4D-200k, comprising 201,426 robot interaction episodes with high-quality 4D annotations. Extensive experiments demonstrate that our method effectively simulates physically-plausible, geometry-consistent, and embodiment-agnostic interactions that faithfully mirror diverse real-world dynamics. For the first time, it shows potential zero-shot transfer capability, providing a high-fidelity foundation for advancing next-generation embodied simulation.
comment: Project page: https://mutianxu.github.io/Kinema4D-project-page/
☆ Reconciling distributed compliance with high-performance control in continuum soft robotics
High-performance closed-loop control of truly soft continuum manipulators has remained elusive. Experimental demonstrations have largely relied on sufficiently stiff, piecewise architectures in which each actuated segment behaves as a distributed yet effectively rigid element, while deformation modes beyond simple bending are suppressed. This strategy simplifies modeling and control, but sidesteps the intrinsic complexity of a fully compliant body and makes the system behave as a serial kinematic chain, much like a conventional articulated robot. An implicit conclusion has consequently emerged within the community: distributed softness and dynamic precision are incompatible. Here we show this trade-off is not fundamental. We present a highly compliant, fully continuum robotic arm - without hardware discretization or stiffness-based mode suppression - that achieves fast, precise task-space convergence under dynamic conditions. The platform integrates direct-drive actuation, a tendon routing scheme enabling coupled bending and twisting, and a structured nonlinear control architecture grounded in reduced-order strain modeling of underactuated systems. Modeling, actuation, and control are co-designed to preserve essential mechanical complexity while enabling high-bandwidth loop closure. Experiments demonstrate accurate, repeatable execution of dynamic Cartesian tasks, including fast positioning and interaction. The proposed system achieves the fastest reported task-execution speed among soft robots. At millimetric precision, execution speed increases nearly fourfold compared with prior approaches, while operating on a fully compliant continuum body. These results show that distributed compliance and high-performance dynamic control can coexist, opening a path toward truly soft manipulators approaching the operational capabilities of rigid robots without sacrificing morphological richness.
☆ Routing and Control for Marine Oil-Spill Cleanup with a Boom-Towing Vessel Fleet
Marine oil spills damage ecosystems, contaminate coastlines, and disrupt food webs, while imposing substantial economic losses on fisheries and coastal communities. Prior work has demonstrated the feasibility of containing and cleaning individual spills using a duo of autonomous surface vehicles (ASVs) equipped with a towed boom and skimmers. However, existing algorithmic approaches primarily address isolated slicks and individual ASV duos, lacking scalable methods for coordinating large robotic fleets across multiple spills representative of realistic oil-spill incidents. In this work, we propose an integrated multi-robot framework for coordinated oil-spill confinement and cleanup using autonomous ASV duos. We formulate multi-spill response as a risk-weighted minimum-latency problem, where spill-specific risk factors and service times jointly determine cumulative environmental damage. To solve this problem, we develop a hybrid optimization approach combining mixed-integer linear programming, and a tailored warm-start heuristic, enabling near-optimal routing plans for scenarios with tens of spills within minutes on commodity hardware. For physical execution, we design and analyze two tracking controllers for boom-towing ASV duos: a feedback-linearization controller with proven asymptotic stability, and a baseline PID controller. Simulation results under coupled vessel-boom dynamics demonstrate accurate path tracking for both controllers. Together, these components provide a scalable, holistic framework for rapid, risk-aware multi-robot response to large-scale oil spill disasters.
☆ Dexterous grasp data augmentation based on grasp synthesis with fingertip workspace cloud and contact-aware sampling
Robotic grasping is a fundamental yet crucial component of robotic applications, as effective grasping often serves as the starting point for various tasks. With the rapid advancement of neural networks, data-driven approaches for robotic grasping have become mainstream. However, efficiently generating grasp datasets for training remains a bottleneck. This is compounded by the diverse structures of robotic hands, making the design of generalizable grasp generation methods even more complex. In this work, we propose a teleoperation-based framework to collect a small set of grasp pose demonstrations, which are augmented using FSG--a Fingertip-contact-aware Sampling-based Grasp generator. Based on the demonstrated grasp poses, we propose AutoWS, which automatically generates structured workspace clouds of robotic fingertips, embedding the hand structure information directly into the clouds to eliminate the need for inverse kinematics calculations. Experiments on grasping the YCB objects show that our method significantly outperforms existing approaches in both speed and valid pose generation rate. Our framework enables real-time grasp generation for hands with arbitrary structures and produces human-like grasps when combined with demonstrations, providing an efficient and robust data augmentation tool for data-driven grasp training.
comment: Accepted to Advanced Robotics, GitHub: https://github.com/W567/FSG, YouTube: https://youtu.be/rFCDl9SxSSA
☆ Scalable Inspection Planning via Flow-based Mixed Integer Linear Programming
Inspection planning is concerned with computing the shortest robot path to inspect a given set of points of interest (POIs) using the robot's sensors. This problem arises in a wide range of applications from manufacturing to medical robotics. To alleviate the problem's complexity, recent methods rely on sampling-based methods to obtain a more manageable (discrete) graph inspection planning (GIP) problem. Unfortunately, GIP still remains highly difficult to solve at scale as it requires simultaneously satisfying POI-coverage and path-connectivity constraints, giving rise to a challenging optimization problem, particularly at scales encountered in real-world scenarios. In this work, we present highly scalable Mixed Integer Linear Programming (MILP) solutions for GIP that significantly advance the state-of-the-art in both runtime and solution quality. Our key insight is a reformulation of the problem's core constraints as a network flow, which enables effective MILP models and a specialized Branch-and-Cut solver that exploits the combinatorial structure of flows. We evaluate our approach on medical and infrastructure benchmarks alongside large-scale synthetic instances. Across all scenarios, our method produces substantially tighter lower bounds than existing formulations, reducing optimality gaps by 30-50% on large instances. Furthermore, our solver demonstrates unprecedented scalability: it provides non-trivial solutions for problems with up to 15,000 vertices and thousands of POIs, where prior state-of-the-art methods typically exhaust memory or fail to provide any meaningful optimality guarantees.
☆ ASCENT: Transformer-Based Aircraft Trajectory Prediction in Non-Towered Terminal Airspace ICRA 2026
Accurate trajectory prediction can improve General Aviation safety in non-towered terminal airspace, where high traffic density increases accident risk. We present ASCENT, a lightweight transformer-based model for multi-modal 3D aircraft trajectory forecasting, which integrates domain-aware 3D coordinate normalization and parameterized predictions. ASCENT employs a transformer-based motion encoder and a query-based decoder, enabling the generation of diverse maneuver hypotheses with low latency. Experiments on the TrajAir and TartanAviation datasets demonstrate that our model outperforms prior baselines, as the encoder effectively captures motion dynamics and the decoder aligns with structured aircraft traffic patterns. Furthermore, ablation studies confirm the contributions of the decoder design, coordinate-frame modeling, and parameterized outputs. These results establish ASCENT as an effective approach for real-time aircraft trajectory prediction in non-towered terminal airspace.
comment: ICRA 2026. Project Page at https://a-pru.github.io/ascent/
☆ A Pin-Array Structured Climbing Robot for Stable Locomotion on Steep Rocky Terrain ICRA
Climbing robots face significant challenges when navigating unstructured environments, where reliable attachment to irregular surfaces is critical. We present a novel mobile climbing robot equipped with compliant pin-array structured grippers that passively conform to surface irregularities, ensuring stable ground gripping without the need for complicated sensing or control. Each pin features a vertically split design, combining an elastic element with a metal spine to enable mechanical interlocking with microscale surface features. Statistical modeling and experimental validation indicate that variability in individual pin forces and contact numbers are the primary sources of grasping uncertainty. The robot demonstrated robust and stable locomotion in indoor tests on inclined walls (10-30 degrees) and in outdoor tests on natural rocky terrain. This work highlights that a design emphasizing passive compliance and mechanical redundancy provides a practical and robust solution for real-world climbing robots while minimizing control complexity.
comment: Author's version of a manuscript accepted at the 2026 IEEE International Conference on Robotics and Automation (ICRA). (c) IEEE
☆ Conservative Offline Robot Policy Learning via Posterior-Transition Reweighting
Offline post-training adapts a pretrained robot policy to a target dataset by supervised regression on recorded actions. In practice, robot datasets are heterogeneous: they mix embodiments, camera setups, and demonstrations of varying quality, so many trajectories reflect recovery behavior, inconsistent operator skill, or weakly informative supervision. Uniform post-training gives equal credit to all samples and can therefore average over conflicting or low-attribution data. We propose Posterior-Transition Reweighting (PTR), a reward-free and conservative post-training method that decides how much each training sample should influence the supervised update. For each sample, PTR encodes the observed post-action consequence as a latent target, inserts it into a candidate pool of mismatched targets, and uses a separate transition scorer to estimate a softmax identification posterior over target indices. The posterior-to-uniform ratio defines the PTR score, which is converted into a clipped-and-mixed weight and applied to the original action objective through self-normalized weighted regression. This construction requires no tractable policy likelihood and is compatible with both diffusion and flow-matching action heads. Rather than uniformly trusting all recorded supervision, PTR reallocates credit according to how attributable each sample's post-action consequence is under the current representation, improving conservative offline adaptation to heterogeneous robot data.
☆ Designing for Disagreement: Front-End Guardrails for Assistance Allocation in LLM-Enabled Robots
LLM-enabled robots prioritizing scarce assistance in social settings face pluralistic values and LLM behavioral variability: reasonable people can disagree about who is helped first, while LLM-mediated interaction policies vary across prompts, contexts, and groups in ways that are difficult to anticipate or verify at contact point. Yet user-facing guardrails for real-time, multi-user assistance allocation remain under-specified. We propose bounded calibration with contestability, a procedural front-end pattern that (i) constrains prioritization to a governance-approved menu of admissible modes, (ii) keeps the active mode legible in interaction-relevant terms at the point of deferral, and (iii) provides an outcome-specific contest pathway without renegotiating the global rule. Treating pluralism and LLM uncertainty as standing conditions, the pattern avoids both silent defaults that hide implicit value skews and wide-open user-configurable "value settings" that shift burden under time pressure. We illustrate the pattern with a public-concourse robot vignette and outline an evaluation agenda centered on legibility, procedural legitimacy, and actionability, including risks of automation bias and uneven usability of contest channels.
comment: Accepted at the Proceedings of the CHI 2026 Workshop: Ethics at the Front-End
☆ Kamino: GPU-based Massively Parallel Simulation of Multi-Body Systems with Challenging Topologies
We present Kamino, a GPU-based physics solver for massively parallel simulations of heterogeneous highly-coupled mechanical systems. Implemented in Python using NVIDIA Warp and integrated into the Newton framework, it enables the application of data-driven methods, such as large-scale reinforcement learning, to complex robotic systems that exhibit strongly coupled kinematic and dynamic constraints such as kinematic loops. The latter are often circumvented by practitioners; approximating the system topology as a kinematic tree and incorporating explicit loop-closure constraints or so-called mimic joints. Kamino aims at alleviating this burden by natively supporting these types of coupling. This capability facilitates high-throughput parallelized simulations that capture the true nature of mechanical systems that exploit closed kinematic chains for mechanical advantage. Moreover, Kamino supports heterogeneous worlds, allowing for batched simulation of structurally diverse robots on a single GPU. At its core lies a state-of-the-art constrained optimization algorithm that computes constraint forces by solving the constrained rigid multi-body forward dynamics transcribed as a nonlinear complementarity problem. This leads to high-fidelity simulations that can resolve contact dynamics without resorting to approximate models that simplify and/or convexify the problem. We demonstrate RL policy training on DR Legs, a biped with six nested kinematic loops, generating a feasible walking policy while simulating 4096 parallel environments on a single GPU.
☆ LIMBERO: A Limbed Climbing Exploration Robot Toward Traveling on Rocky Cliffs ICRA
In lunar and planetary exploration, legged robots have attracted significant attention as an alternative to conventional wheeled robots, which struggle to traverse rough and uneven terrain. To enable locomotion over highly irregular and steeply inclined surfaces, limbed climbing robots equipped with grippers on their feet have emerged as a promising solution. In this paper, we present LIMBERO, a 10 kg-class quadrupedal climbing robot that employs spine-type grippers for stable locomotion and climbing on rugged and steep terrain. We first introduce a novel gripper design featuring coupled finger-closing and spine-hooking motions, tightly actuated by a single motor, which achieves exceptional grasping performance (>150 N) despite its lightweight design (525 g). Furthermore, we develop an efficient algorithm to visualize a geometry-based graspability index on continuous rough terrain. Finally, we integrate these components into LIMBERO and demonstrate its ability to ascend steep rocky surfaces under a 1 G gravity condition, a performance not previously achieved yet for limbed climbing robots of this scale.
comment: Author's version of a manuscript accepted at the 2026 IEEE International Conference on Robotics and Automation (ICRA). (c) IEEE
☆ Rewarding DINO: Predicting Dense Rewards with Vision Foundation Models
Well-designed dense reward functions in robot manipulation not only indicate whether a task is completed but also encode progress along the way. Generally, designing dense rewards is challenging and usually requires access to privileged state information available only in simulation, not in real-world experiments. This makes reward prediction models that infer task state information from camera images attractive. A common approach is to predict rewards from expert demonstrations based on visual similarity or sequential frame ordering. However, this biases the resulting reward function towards a specific solution and leaves it undefined in states not covered by the demonstrations. In this work, we introduce Rewarding DINO, a method for language-conditioned reward modeling that learns actual reward functions rather than specific trajectories. The model's compact size allows it to serve as a direct replacement for analytical reward functions with comparatively low computational overhead. We train our model on data sampled from 24 Meta-World+ tasks using a rank-based loss and evaluate pairwise accuracy, rank correlation, and calibration. Rewarding DINO achieves competitive performance in tasks from the training set and generalizes to new settings in simulation and the real world, indicating that it learns task semantics. We also test the model with off-the-shelf reinforcement learning algorithms to solve tasks from our Meta-World+ training set.
comment: 10 pages, 5 figures, submitted to IEEE
☆ When Rolling Gets Weird: A Curved-Link Tensegrity Robot for Non-Intuitive Behavior ICRA
Conventional mobile tensegrity robots constructed with straight links offer mobility at the cost of locomotion speed. While spherical robots provide highly effective rolling behavior, they often lack the stability required for navigating unstructured terrain common in many space exploration environments. This research presents a solution with a semi-circular, curved-link tensegrity robot that strikes a balance between efficient rolling locomotion and controlled stability, enabled by discontinuities present at the arc endpoints. Building upon an existing geometric static modeling framework [1], this work presents the system design of an improved Tensegrity eXploratory Robot 2 (TeXploR2). Internal shifting masses instantaneously roll along each curved-link, dynamically altering the two points of contact with the ground plane. Simulations of quasistatic, piecewise continuous locomotion sequences reveal new insights into the positional displacement between inertial and body frames. Non-intuitive rolling behaviors are identified and experimentally validated using a tetherless prototype, demonstrating successful dynamic locomotion. A preliminary impact test highlights the tensegrity structure's inherent shock absorption capabilities and conformability. Future work will focus on finalizing a dynamic model that is experimentally validated with extended testing in real-world environments as well as further refinement of the prototype to incorporate additional curved-links and subsequent ground contact points for increased controllability.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
☆ Coverage First Next Best View for Inspection of Cluttered Pipe Networks Using Mobile Manipulators
Robotic inspection of radioactive areas enables operators to be removed from hazardous environments; however, planning and operating in confined, cluttered environments remain challenging. These systems must autonomously reconstruct the unknown environment and cover its surfaces, whilst estimating and avoiding collisions with objects in the environment. In this paper, we propose a new planning approach based on next-best-view that enables simultaneous exploration and exploitation of the environment by reformulating the coverage path planning problem in terms of information gain. To handle obstacle avoidance under uncertainty, we extend the vector-field-inequalities framework to explicitly account for stochastic measurements of geometric primitives in the environment via chance constraints in a constrained optimal control law. The stochastic constraints were evaluated experimentally alongside the planner on a mobile manipulator in a confined environment to inspect a pipe network. These experiments demonstrate that the system can autonomously plan and execute inspection and coverage paths to reconstruct and fully cover the simplified pipe network. Moreover, the system successfully estimated geometric primitives online and avoided collisions during motion between viewpoints.
comment: 8 pages, 9 figures, 1 table. Submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems 2026
♻ ☆ Crowd-FM: Learned Optimal Selection of Conditional Flow Matching-generated Trajectories for Crowd Navigation ICRA 2026
Safe and computationally efficient local planning for mobile robots in dense, unstructured human crowds remains a fundamental challenge. Moreover, ensuring that robot trajectories are similar to how a human moves will increase the acceptance of the robot in human environments. In this paper, we present Crowd-FM, a learning-based approach to address both safety and human-likeness challenges. Our approach has two novel components. First, we train a Conditional Flow-Matching (CFM) policy over a dataset of optimally controlled trajectories to learn a set of collision-free primitives that a robot can choose at any given scenario. The chosen optimal control solver can generate multi-modal collision-free trajectories, allowing the CFM policy to learn a diverse set of maneuvers. Secondly, we learn a score function over a dataset of human demonstration trajectories that provides a human-likeness score for the flow primitives. At inference time, computing the optimal trajectory requires selecting the one with the highest score. Our approach improves the state-of-the-art by showing that our CFM policy alone can produce collision-free navigation with a higher success rate than existing learning-based baselines. Furthermore, when augmented with inference-time refinement, our approach can outperform even expensive optimisation-based planning approaches. Finally, we validate that our scoring network can select trajectories closer to the expert data than a manually designed cost function.
comment: Accepted at IEEE ICRA 2026. Authors Antareep Singha and Laksh Nanwani have equal contributions
♻ ☆ Stein Variational Ergodic Surface Coverage with SE(3) Constraints
Surface manipulation tasks require robots to generate trajectories that comprehensively cover complex 3D surfaces while maintaining precise end-effector poses. Existing ergodic trajectory optimization (TO) methods demonstrate success in coverage tasks, while struggling with point-cloud targets due to the nonconvex optimization landscapes and the inadequate handling of SE(3) constraints in sampling-as-optimization (SAO) techniques. In this work, we introduce a preconditioned SE(3) Stein Variational Gradient Descent (SVGD) approach for SAO ergodic trajectory generation. Our proposed approach comprises multiple innovations. First, we reformulate point-cloud ergodic coverage as a manifold-aware sampling problem. Second, we derive SE(3)-specific SVGD particle updates, and, third, we develop a preconditioner to accelerate TO convergence. Our sampling-based framework consistently identifies superior local optima compared to strong optimization-based and SAO baselines while preserving the SE(3) geometric structure. Experiments on a 3D point-cloud surface coverage benchmark and robotic surface drawing tasks demonstrate that our method achieves superior coverage quality with tractable computation in our setting relative to existing TO and SAO approaches, and is validated in real-world robot experiments.
♻ ☆ DreamFlow: Local Navigation Beyond Observation via Conditional Flow Matching in the Latent Space
Local navigation in cluttered environments often suffers from dense obstacles and frequent local minima. Conventional local planners rely on heuristics and are prone to failure, while deep reinforcement learning(DRL)based approaches provide adaptability but are constrained by limited onboard sensing. These limitations lead to navigation failures because the robot cannot perceive structures outside its field of view. In this paper, we propose DreamFlow, a DRL-based local navigation framework that extends the robot's perceptual horizon through conditional flow matching(CFM). The proposed CFM based prediction module learns probabilistic mapping between local height map latent representation and broader spatial representation conditioned on navigation context. This enables the navigation policy to predict unobserved environmental features and proactively avoid potential local minima. Experimental results demonstrate that DreamFlow outperforms existing methods in terms of latent prediction accuracy and navigation performance in simulation. The proposed method was further validated in cluttered real world environments with a quadrupedal robot. The project page is available at https://dreamflow-icra.github.io.
♻ ☆ MSGNav: Unleashing the Power of Multi-modal 3D Scene Graph for Zero-Shot Embodied Navigation CVPR 2026
Embodied navigation is a fundamental capability for robotic agents operating. Real-world deployment requires open vocabulary generalization and low training overhead, motivating zero-shot methods rather than task-specific RL training. However, existing zero-shot methods that build explicit 3D scene graphs often compress rich visual observations into text-only relations, leading to high construction cost, irreversible loss of visual evidence, and constrained vocabularies. To address these limitations, we introduce the Multi-modal 3D Scene Graph (M3DSG), which preserves visual cues by replacing textual relational edges with dynamically assigned images. Built on M3DSG, we propose MSGNav, a zero-shot navigation system that includes a Key Subgraph Selection module for efficient reasoning, an Adaptive Vocabulary Update module for open vocabulary support, and a Closed-Loop Reasoning module for accurate exploration reasoning. Additionally, we further identify the last mile problem in zero-shot navigation determining the feasible target location with a suitable final viewpoint, and propose a Visibility-based Viewpoint Decision module to explicitly resolve it. Comprehensive experimental results demonstrate that MSGNav achieves state-of-the-art performance on the challenging GOAT-Bench and HM3D-ObjNav benchmark. The code will be publicly available at https://github.com/ylwhxht/MSGNav.
comment: 18 pages, Accepted by CVPR 2026
♻ ☆ CloSE: A Geometric Shape-Agnostic Cloth State Representation ICRA 2026
Cloth manipulation is a difficult problem mainly because of the non-rigid nature of cloth, which makes a good representation of deformation essential. We present a new representation for the deformation-state of clothes. First, we propose the dGLI disk representation based on topological indices computed for edge segments of the cloth border that are arranged on a circular grid. The heat-map of the dGLI disk uncovers patterns that correspond to features of the cloth state that are consistent for different shapes, sizes or orientation of the cloth. We then abstract these important features from the dGLI disk into a circle, calling it the Cloth StatE representation (CloSE). This representation is compact, continuous, and general for different shapes. We show that this representation is able to accurately predict the fold locations for several simulation clothing datasets. Finally, we also show the strengths of this representation in two relevant applications: semantic labeling and high- and low-level planning. The code and the dataset can be accessed from: https://close-representation.github.io/
comment: Accepted at ICRA 2026 (8 pages, 11 figures, 1 table). Project page: https://close-representation.github.io/
♻ ☆ DefVINS: Visual-Inertial Odometry for Deformable Scenes
Deformable scenes violate the rigidity assumptions underpinning classical visual--inertial odometry (VIO), often leading to over-fitting to local non-rigid motion or to severe camera pose drift when deformation dominates visual parallax. In this paper, we introduce DefVINS, the first visual-inertial odometry pipeline designed to operate in deformable environments. Our approach models the odometry state by decomposing it into a rigid, IMU-anchored component and a non-rigid scene warp represented by an embedded deformation graph. As a second contribution, we present VIMandala, the first benchmark containing real images and ground-truth camera poses for visual-inertial odometry in deformable scenes. In addition, we augment the synthetic Drunkard's benchmark with simulated inertial measurements to further evaluate our pipeline under controlled conditions. We also provide an observability analysis of the visual-inertial deformable odometry problem, characterizing how inertial measurements constrain camera motion and render otherwise unobservable modes identifiable in the presence of deformation. This analysis motivates the use of IMU anchoring and leads to a conditioning-based activation strategy that avoids ill-posed updates under poor excitation. Experimental results on both the synthetic Drunkard's and our real VIMandala benchmarks show that DefVINS outperforms rigid visual--inertial and non-rigid visual odometry baselines. Our source code and data will be released upon acceptance.
comment: 4 figures, 2 tables. Submitted to RA-L
♻ ☆ Traj2Action: A Co-Denoising Framework for Trajectory-Guided Human-to-Robot Skill Transfer
Learning diverse manipulation skills for real-world robots is severely bottlenecked by the reliance on costly and hard-to-scale teleoperated demonstrations. While human videos offer a scalable alternative, effectively transferring manipulation knowledge is fundamentally hindered by the significant morphological gap between human and robotic embodiments. To address this challenge and facilitate skill transfer from human to robot, we introduce Traj2Action, a novel framework that bridges this embodiment gap by using the 3D trajectory of the operational endpoint as a unified intermediate representation, and then transfers the manipulation knowledge embedded in this trajectory to the robot's actions. Our policy first learns to generate a coarse trajectory, which forms a high-level motion plan by leveraging both human and robot data. This plan then conditions the synthesis of precise, robot-specific actions (e.g., orientation and gripper state) within a co-denoising framework. Our work centers on two core objectives: first, the systematic verification of the Traj2Action framework's effectiveness-spanning architectural design, cross-task generalization, and data efficiency and second, the revelation of key laws that govern robot policy learning during the integration of human hand demonstration data. This research focus enables us to provide a scalable paradigm tailored to address human-to-robot skill transfer across morphological gaps. Extensive real-world experiments on a Franka robot demonstrate that Traj2Action boosts the performance by up to 27% and 22.25% over $π_0$ baseline on short- and long-horizon real-world tasks, and achieves significant gains as human data scales in robot policy learning.
♻ ☆ $χ_{0}$: Resource-Aware Robust Manipulation via Taming Distributional Inconsistencies
High-reliability long-horizon robotic manipulation has traditionally relied on large-scale data and compute to understand complex real-world dynamics. However, we identify that the primary bottleneck to real-world robustness is not resource scale alone, but the distributional shift among the human demonstration distribution, the inductive bias learned by the policy, and the test-time execution distribution -- a systematic inconsistency that causes compounding errors in multi-stage tasks. To mitigate these inconsistencies, we propose $χ_{0}$, a resource-efficient framework with effective modules designated to achieve production-level robustness in robotic manipulation. Our approach builds off three technical pillars: (i) Model Arithmetic, a weight-space merging strategy that efficiently soaks up diverse distributions of different demonstrations, varying from object appearance to state variations; (ii) Stage Advantage, a stage-aware advantage estimator that provides stable, dense progress signals, overcoming the numerical instability of prior non-stage approaches; and (iii) Train-Deploy Alignment, which bridges the distribution gap via spatio-temporal augmentation, heuristic DAgger corrections, and temporal chunk-wise smoothing. $χ_{0}$ enables two sets of dual-arm robots to collaboratively orchestrate long-horizon garment manipulation, spanning tasks from flattening, folding, to hanging different clothes. Our method exhibits high-reliability autonomy; we are able to run the system from arbitrary initial state for consecutive 24 hours non-stop. Experiments validate that $χ_{0}$ surpasses the state-of-the-art $π_{0.5}$ in success rate by nearly 250%, with only 20-hour data and 8 A100 GPUs. Code, data and models will be released to facilitate the community.
♻ ☆ UGotMe: An Embodied System for Affective Human-Robot Interaction ICRA
Equipping humanoid robots with the capability to understand emotional states of human interactants and express emotions appropriately according to situations is essential for affective human-robot interaction. However, enabling current vision-aware multimodal emotion recognition models for affective human-robot interaction in the real-world raises embodiment challenges: addressing the environmental noise issue and meeting real-time requirements. First, in multiparty conversation scenarios, the noises inherited in the visual observation of the robot, which may come from either 1) distracting objects in the scene or 2) inactive speakers appearing in the field of view of the robot, hinder the models from extracting emotional cues from vision inputs. Secondly, realtime response, a desired feature for an interactive system, is also challenging to achieve. To tackle both challenges, we introduce an affective human-robot interaction system called UGotMe designed specifically for multiparty conversations. Two denoising strategies are proposed and incorporated into the system to solve the first issue. Specifically, to filter out distracting objects in the scene, we propose extracting face images of the speakers from the raw images and introduce a customized active face extraction strategy to rule out inactive speakers. As for the second issue, we employ efficient data transmission from the robot to the local server to improve realtime response capability. We deploy UGotMe on a human robot named Ameca to validate its real-time inference capabilities in practical scenarios. Videos demonstrating real-world deployment are available at https://lipzh5.github.io/HumanoidVLE/.
comment: Accepted to the 2025 IEEE International Conference on Robotics and Automation (ICRA)
♻ ☆ EfficientNav: Towards On-Device Object-Goal Navigation with Navigation Map Caching and Retrieval NeurIPS 2025
Object-goal navigation (ObjNav) tasks an agent with navigating to the location of a specific object in an unseen environment. Embodied agents equipped with large language models (LLMs) and online constructed navigation maps can perform ObjNav in a zero-shot manner. However, existing agents heavily rely on giant LLMs on the cloud, e.g., GPT-4, while directly switching to small LLMs, e.g., LLaMA3.2-11b, suffer from significant success rate drops due to limited model capacity for understanding complex navigation maps, which prevents deploying ObjNav on local devices. At the same time, the long prompt introduced by the navigation map description will cause high planning latency on local devices. In this paper, we propose EfficientNav to enable on-device efficient LLM-based zero-shot ObjNav. To help the smaller LLMs better understand the environment, we propose semantics-aware memory retrieval to prune redundant information in navigation maps. To reduce planning latency, we propose discrete memory caching and attention-based memory clustering to efficiently save and re-use the KV cache. Extensive experimental results demonstrate that EfficientNav achieves 11.1% improvement in success rate on HM3D benchmark over GPT-4-based baselines, and demonstrates 6.7x real-time latency reduction and 4.7x end-to-end latency reduction over GPT-4 planner. Our code is available on https://github.com/PKU-SEC-Lab/EfficientNav.
comment: NeurIPS 2025
♻ ☆ WorldVLM: Combining World Model Forecasting and Vision-Language Reasoning
Autonomous driving systems depend on on models that can reason about high-level scene contexts and accurately predict the dynamics of their surrounding environment. Vision- Language Models (VLMs) have recently emerged as promising tools for decision-making and scene understanding, offering strong capabilities in contextual reasoning. However, their limited spatial comprehension constrains their effectiveness as end-to-end driving models. World Models (WM) internalize environmental dynamics to predict future scene evolution. Recently explored as ego-motion predictors and foundation models for autonomous driving, they represent a promising direction for addressing key challenges in the field, particularly enhancing generalization while maintaining dynamic prediction. To leverage the complementary strengths of context-based decision making and prediction, we propose WorldVLM: A hybrid architecture that unifies VLMs and WMs. In our design, the high-level VLM generates behavior commands to guide the driving WM, enabling interpretable and context-aware actions. We evaluate conditioning strategies and provide insights into the hybrid design challenges.
comment: 8 pages, 6 figures, 5 tables; submitted to IEEE
♻ ☆ KEEP: A KV-Cache-Centric Memory Management System for Efficient Embodied Planning
Memory-augmented Large Language Models (LLMs) have demonstrated remarkable capability for complex and long-horizon embodied planning. By keeping track of past experiences and environmental states, memory enables LLMs to maintain a global view, thereby avoiding repetitive exploration. However, existing approaches often store the memory as raw text, leading to excessively long prompts and high prefill latency. While it is possible to store and reuse the KV caches, the efficiency benefits are greatly undermined due to frequent KV cache updates. In this paper, we propose KEEP, a KV-cache-centric memory management system for efficient embodied planning. KEEP features 3 key innovations: (1) a Static-Dynamic Memory Construction algorithm that reduces KV cache recomputation by mixed-granularity memory group; (2) a Multi-hop Memory Re-computation algorithm that dynamically identifies important cross-attention among different memory groups and reconstructs memory interactions iteratively; (3) a Layer-balanced Memory Loading that eliminates unbalanced KV cache loading and cross-attention computation across different layers. Extensive experimental results have demonstrated that KEEP achieves 2.68x speedup with negligible accuracy loss compared with text-based memory methods on ALFRED dataset. Compared with the KV re-computation method CacheBlend (EuroSys'25), KEEP shows 4.13% success rate improvement and 1.90x time-to-first-token (TTFT) reduction. Our code is available on https://github.com/PKU-SEC-Lab/KEEP_Embodied_Memory.
comment: DAC 2026
♻ ☆ DySL-VLA: Efficient Vision-Language-Action Model Inference via Dynamic-Static Layer-Skipping for Robot Manipulation
Vision-Language-Action (VLA) models have shown remarkable success in robotic tasks like manipulation by fusing a language model's reasoning with a vision model's 3D understanding. However, their high computational cost remains a major obstacle for real-world applications that require real-time performance. We observe that the actions within a task have varying levels of importance: critical steps demand high precision, while less important ones can tolerate more variance. Leveraging this insight, we propose DySL-VLA, a novel framework that addresses computational cost by dynamically skipping VLA layers based on each action's importance. DySL-VLA categorizes its layers into two types: informative layers, which are consistently executed, and incremental layers, which can be selectively skipped. To intelligently skip layers without sacrificing accuracy, we invent a prior-post skipping guidance mechanism to determine when to initiate layer-skipping. We also propose a skip-aware two-stage knowledge distillation algorithm to efficiently train a standard VLA into a DySL-VLA. Our experiments indicate that DySL-VLA achieves 2.1% improvement in success length over Deer-VLA on the Calvin dataset, while simultaneously reducing trainable parameters by a factor of 85.7 and providing a 3.75x speedup relative to the RoboFlamingo baseline at iso-accuracy. Our code is available on https://github.com/PKU-SEC-Lab/DYSL_VLA.
comment: DAC 2026
♻ ☆ CLAIM: Camera-LiDAR Alignment with Intensity and Monodepth IROS 2025
In this paper, we unleash the potential of the powerful monodepth model in camera-LiDAR calibration and propose CLAIM, a novel method of aligning data from the camera and LiDAR. Given the initial guess and pairs of images and LiDAR point clouds, CLAIM utilizes a coarse-to-fine searching method to find the optimal transformation minimizing a patched Pearson correlation-based structure loss and a mutual information-based texture loss. These two losses serve as good metrics for camera-LiDAR alignment results and require no complicated steps of data processing, feature extraction, or feature matching like most methods, rendering our method simple and adaptive to most scenes. We validate CLAIM on public KITTI, Waymo, and MIAS-LCEC datasets, and the experimental results demonstrate its superior performance compared with the state-of-the-art methods. The code is available at https://github.com/Tompson11/claim.
comment: Accepted by IROS 2025
♻ ☆ An Intention-driven Lane Change Framework Considering Heterogeneous Dynamic Cooperation in Mixed-traffic Environment
In mixed-traffic environments, autonomous vehicles (AVs) must interact with heterogeneous human-driven vehicles (HVs) whose intentions and driving styles vary across individuals and scenarios. Such variability introduces uncertainty into lane change interactions, where safety and efficiency critically depend on accurately anticipating surrounding drivers' cooperative responses. Existing methods often oversimplify these interactions by assuming uniform or fixed behavioral patterns. To address this limitation, we propose an intention-driven lane change framework that integrates driving-style recognition with cooperation-aware decision-making and motion-planning. A deep learning-based classifier identifies distinct human driving styles in real time. We then introduce a dual-perspective cooperation score composed of intrinsic style-dependent tendencies and interactive dynamic components, enabling interpretable and adaptive intention prediction and quantitative inference. A decision-making module combines behavior cloning (BC) and inverse reinforcement learning (IRL) to determine lane change feasibility. Later, a coordinated motion-planning architecture integrating IRL-based intention inference with model predictive control (MPC) is established to generate collision-free and socially compliant trajectories. Experiments on the NGSIM dataset show that the proposed decision-making model outperforms representative rule-based and learning-based baselines, achieving 96.98% accuracy in lane change classification. Motion-planning evaluations further demonstrate improved maneuver success and execution stability in mixed-traffic environments. These results validate the effectiveness of structured cooperation modeling for intention-driven autonomous lane changes.
♻ ☆ Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matching
Stereo foundation models achieve strong zero-shot generalization but remain computationally prohibitive for real-time applications. Efficient stereo architectures, on the other hand, sacrifice robustness for speed and require costly per-domain fine-tuning. To bridge this gap, we present Fast-FoundationStereo, a family of architectures that achieve, for the first time, strong zero-shot generalization at real-time frame rate. We employ a divide-and-conquer acceleration strategy with three components: (1) knowledge distillation to compress the hybrid backbone into a single efficient student; (2) blockwise neural architecture search for automatically discovering optimal cost filtering designs under latency budgets, reducing search complexity exponentially; and (3) structured pruning for eliminating redundancy in the iterative refinement module. Furthermore, we introduce an automatic pseudo-labeling pipeline used to curate 1.4M in-the-wild stereo pairs to supplement synthetic training data and facilitate knowledge distillation. The resulting model can run over 10x faster than FoundationStereo while closely matching its zero-shot accuracy, thus establishing a new state-of-the-art among real-time methods. Project page: https://nvlabs.github.io/Fast-FoundationStereo/
♻ ☆ Haptic Light-Emitting Diodes: Miniature, Luminous Tactile Actuators
We present Haptic Light-Emitting Diodes (HLEDs), luminous thermopneumatic actuators that directly convert pulsed light into mechanical forces and displacements. Each device packages a miniature surface-mount LED in a gas-filled cavity that contains a low-inertia graphite photoabsorber. The cavity is sealed by an elastic membrane, which functions as a working diaphragm. Brief optical pulses heat the photoabsorber, which heats the gas. The resulting rapid pressure increases generate forces and displacements at the working diaphragm. Millimeter-scale HLEDs produce forces exceeding 0.4 N and displacements of 0.9 mm at low voltages, with 5 to 100 ms response times, making them attractive as actuators providing tactile feedback in human-machine interfaces. Unusually, these actuators are also light-emitting, as a fraction of optical energy is transmitted through the membrane. These photomechanical actuators have many potential applications in tactile displays, human interface engineering, wearable computing, and other areas.
♻ ☆ Vision-Language Models for Infrared Industrial Sensing in Additive Manufacturing Scene Description
Many manufacturing environments operate in low-light conditions or within enclosed machines where conventional vision systems struggle. Infrared cameras provide complementary advantages in such environments. Simultaneously, supervised AI systems require large labeled datasets, which makes zero-shot learning frameworks more practical for applications including infrared cameras. Recent advances in vision-language foundation models (VLMs) offer a new path in zero-shot predictions from paired image-text representations. However, current VLMs cannot understand infrared camera data since they are trained on RGB data. This work introduces VLM-IRIS (Vision-Language Models for InfraRed Industrial Sensing), a zero-shot framework that adapts VLMs to infrared data by preprocessing infrared images captured by a FLIR Boson sensor into RGB-compatible inputs suitable for CLIP-based encoders. We demonstrate zero-shot workpiece presence detection on a 3D printer bed where temperature differences between the build plate and workpieces make the task well-suited for thermal imaging. VLM-IRIS converts the infrared images to magma representation and applies centroid prompt ensembling with a CLIP ViT-B/32 encoder to achieve high accuracy on infrared images without any model retraining. These findings demonstrate that the proposed improvements to VLMs can be effectively extended to thermal applications for label-free monitoring.
♻ ☆ REFINE-DP: Diffusion Policy Fine-tuning for Humanoid Loco-manipulation via Reinforcement Learning
Humanoid loco-manipulation requires coordinated high-level motion plans with stable, low-level whole-body execution under complex robot-environment dynamics and long-horizon tasks. While diffusion policies (DPs) show promise for learning from demonstrations, deploying them on humanoids poses critical challenges: the motion planner trained offline is decoupled from the low-level controller, leading to poor command tracking, compounding distribution shift, and task failures. The common approach of scaling demonstration data is prohibitively expensive for high-dimensional humanoid systems. To address this challenge, we present REFINE-DP (REinforcement learning FINE-tuning of Diffusion Policy), a hierarchical framework that jointly optimizes a DP high-level planner and an RL-based low-level loco-manipulation controller. The DP is fine-tuned via a PPO-based diffusion policy gradient to improve task success rate, while the controller is simultaneously updated to accurately track the planner's evolving command distribution, reducing the distributional mismatch that degrades motion quality. We validate REFINE-DP on a humanoid robot performing loco-manipulation tasks, including door traversal and long-horizon object transport. REFINE-DP achieves an over $90\%$ success rate in simulation, even in out-of-distribution cases not seen in the pre-trained data, and enables smooth autonomous task execution in real-world dynamic environments. Our proposed method substantially outperforms pre-trained DP baselines and demonstrates that RL fine-tuning is key to reliable humanoid loco-manipulation. https://refine-dp.github.io/REFINE-DP/
♻ ☆ Volumetric Ergodic Control ICRA
Ergodic control synthesizes optimal coverage behaviors over spatial distributions for nonlinear systems. However, existing formulations model the robot as a non-volumetric point, whereas in practice a robot interacts with the environment through its body and sensors with physical volume. In this work, we introduce a new ergodic control formulation that optimizes spatial coverage using a volumetric state representation. Our method preserves the asymptotic coverage guarantees of ergodic control, adds minimal computational overhead for real-time control, and supports arbitrary sample-based volumetric models. We evaluate our method across search and manipulation tasks -- with multiple robot dynamics and end-effector geometries or sensor models -- and show that it improves coverage efficiency by more than a factor of two while maintaining a 100% task completion rate across all experiments, outperforming the standard ergodic control method. Finally, we demonstrate the effectiveness of our method on a robot arm performing mechanical erasing tasks. Project website: https://murpheylab.github.io/vec/
comment: 8 pages, 8 figures; Accepted to 2026 IEEE International Conference on Robotics and Automation (ICRA); Project website: https://murpheylab.github.io/vec/
♻ ☆ SO-Bench: A Structural Output Evaluation of Multimodal LLMs
Multimodal large language models (MLLMs) are increasingly deployed in real-world, agentic settings where outputs must not only be correct, but also conform to predefined data schemas. Despite recent progress in structured generation in textual domain, there is still no benchmark that systematically evaluates schema-grounded information extraction and reasoning over visual inputs. In this work, we conduct a comprehensive study of visual structural output capabilities for MLLMs with our carefully designed SO-Bench benchmark. Covering four visual domains, including UI screens, natural images, documents, and charts, SO-Bench is built from over 6.5K diverse JSON schemas and 1.8K curated image-schema pairs with human-verified quality. Benchmarking experiments on open-sourced and frontier proprietary models reveal persistent gaps in predicting accurate, schema compliant outputs, highlighting the need for better multimodal structured reasoning. Beyond benchmarking, we further conduct training experiments to largely improve the model's structured output capability. We make the benchmark and evaluation publicly available at https://github.com/apple/ml-sobench
comment: v3 preprint. Added the link to the public benchmark
♻ ☆ Lyapunov Constrained Soft Actor-Critic (LC-SAC) using Koopman Operator Theory for Quadrotor Trajectory Tracking
Reinforcement Learning (RL) has achieved remarkable success in solving complex sequential decision-making problems. However, its application to safety-critical physical systems remains constrained by the lack of stability guarantees. Standard RL algorithms prioritize reward maximization, often yielding policies that may induce oscillations or unbounded state divergence. There has been significant work in incorporating Lyapunov-based stability guarantees in RL algorithms with key challenges being selecting a candidate Lyapunov function, computational complexity by using excessive function approximators and conservative policies by incorporating stability criterion in the learning process. In this work we propose a novel Lyapunov-constrained Soft Actor-Critic (LC-SAC) algorithm using Koopman operator theory. We propose use of extended dynamic mode decomposition (EDMD) to produce a linear approximation of the system and use this approximation to derive a closed form solution for candidate Lyapunov function. This derived Lyapunov function is incorporated in the SAC algorithm to further provide guarantees for a policy that stabilizes the nonlinear system. The results are evaluated trajectory tracking of a 2D Quadrotor environment based on safe-control-gym. The proposed algorithm shows training convergence and decaying violations for Lyapunov stability criterion compared to baseline vanilla SAC algorithm. GitHub Repository: https://github.com/DhruvKushwaha/LC-SAC-Quadrotor-Trajectory-Tracking
comment: 11 pages, 7 Figures, submitted to IEEE RA-L
♻ ☆ AgriChrono: A Multi-modal Dataset Capturing Crop Growth and Lighting Variability with a Field Robot
Advances in AI and Robotics have accelerated significant initiatives in agriculture, particularly in the areas of robot navigation and 3D digital twin creation. A significant bottleneck impeding this progress is the critical lack of "in-the-wild" datasets that capture the full complexities of real farmland, including non-rigid motion from wind, drastic illumination variance, and morphological changes resulting from growth. This data gap fundamentally limits research on robust AI models for autonomous field navigation and scene-level dynamic 3D reconstruction. In this paper, we present AgriChrono, a modular robotic data collection platform and multi-modal dataset designed to capture these dynamic farmland conditions. Our platform integrates multiple sensors, enabling remote, time-synchronized acquisition of RGB, Depth, LiDAR, IMU, and Pose data for efficient and repeatable long-term data collection in real-world agricultural environments. We successfully collected 18TB of data over one month, documenting the entire growth cycle of Canola under diverse illumination conditions. We benchmark state-of-the-art 3D reconstruction methods on AgriChrono, revealing the profound challenge of reconstructing high-fidelity, dynamic non-rigid scenes in such farmland settings. This benchmark validates AgriChrono as a critical asset for advancing model generalization, and its public release is expected to significantly accelerate research and development in precision agriculture. The code and dataset are publicly available at: https://github.com/StructuresComp/agri-chrono
comment: Keywords: Agricultural Robotics, In-the-wild Dataset, 3D Reconstruction
♻ ☆ Push, Press, Slide: Mode-Aware Planar Contact Manipulation via Reduced-Order Models IROS 2026
Non-prehensile planar manipulation, including pushing and press-and-slide, is critical for diverse robotic tasks, but notoriously challenging due to hybrid contact mechanics, under-actuation, and asymmetric friction limits that traditionally necessitate computationally expensive iterative control. In this paper, we propose a mode-aware framework for planar manipulation with one or two robotic arms based on contact topology selection and reduced-order kinematic modeling. Our core insight is that complex wrench-twist limit surface mechanics can be abstracted into a discrete library of physically intuitive models. We systematically map various single-arm and bimanual contact topologies to simple non-holonomic formulations, e.g. unicycle for simplified press-and-slide motion. By anchoring trajectory generation to these reduced-order models, our framework computes the required object wrench and distributes feasible, friction-bounded contact forces via a direct algebraic allocator. We incorporate manipulator kinematics to ensure long-horizon feasibility and demonstrate our fast, optimization-free approach in simulation across diverse single-arm and bimanual manipulation tasks. Supplementary videos and additional information are available at: https://sites.google.com/view/pushpressslide
comment: 8 pages, 13 figures. Submitted to IEEE IROS 2026
♻ ☆ Dual Quaternion Based Contact Modeling for Fast and Smooth Collision Recovery of Quadrotors
Unmanned aerial vehicles (UAVs) operating in cluttered environments require accurate impact modeling to maintain stability post collisions. However, conventional contact models decouple linear and angular impulses, risking manifold inconsistency during rapid state transitions. This letter presents a dual quaternion reset map that resolves rigid-body impacts directly on the SE(3) manifold. By operating on the unified spatial twist (linear and angular velocities as a single dual entity), the proposed formulation is shown to be algebraically equivalent to the classical Newton impulse model while preserving manifold consistency during discrete state jumps. Building on this framework, a hybrid recovery controller is designed that couples linear and angular momentum to ensure strict energy dissipation across impacts. Hardware-in-the-loop benchmarks demonstrate a 24% reduction in execution latency compared to an optimized matrix-based implementation. High-fidelity MuJoCo simulations validate the controller's response to complex contact dynamics, with Monte Carlo trials showing a 56.3% reduction in post-impact root-mean-square error (RMSE) and a 61.1% decrease in peak kinetic energy compared to decoupled baseline controllers.
comment: 7 pages, 5 figures
♻ ☆ TurboMap: GPU-Accelerated Local Mapping for Visual SLAM
In real-time Visual SLAM systems, local mapping must operate under strict latency constraints, as delays degrade map quality and increase the risk of tracking failure. GPU parallelization offers a promising way to reduce latency. However, parallelizing local mapping is challenging due to synchronized shared-state updates and the overhead of transferring large map data structures to the GPU. This paper presents TurboMap, a GPU-parallelized and CPU-optimized local mapping backend that holistically addresses these challenges. We restructure Map Point Creation to enable parallel Keypoint Correspondence Search on the GPU, redesign and parallelize Map Point Fusion, optimize Redundant Keyframe Culling on the CPU, and integrate a fast GPU-based Local Bundle Adjustment solver. To minimize data transfer and synchronization costs, we introduce persistent GPU-resident keyframe storage. Experiments on the EuRoC and TUM-VI datasets show average local mapping speedups of 1.3x and 1.6x, respectively, while preserving accuracy.
♻ ☆ Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning
Continual learning is a long-standing challenge in robot policy learning, where a policy must acquire new skills over time without catastrophically forgetting previously learned ones. While prior work has extensively studied continual learning in relatively small behavior cloning (BC) policy models trained from scratch, its behavior in modern large-scale pretrained Vision-Language-Action (VLA) models remains underexplored. In this work, we found that pretrained VLAs are remarkably resistant to forgetting compared with smaller policy models trained from scratch. Simple Experience Replay (ER) works surprisingly well on VLAs, sometimes achieving zero forgetting even with a small replay data size. Our analysis reveals that pretraining plays a critical role in downstream continual learning performance: large pretrained models mitigate forgetting with a small replay buffer size while maintaining strong forward learning capabilities. Furthermore, we found that VLAs can retain relevant knowledge from prior tasks despite performance degradation during learning new tasks. This knowledge retention enables rapid recovery of seemingly forgotten skills through finetuning. Together, these insights imply that large-scale pretraining fundamentally changes the dynamics of continual learning, enabling models to continually acquire new skills over time with simple replay. Code and more information can be found at https://continual-vlas.github.io/forget-me-not/
comment: Project website: https://continual-vlas.github.io/forget-me-not/
♻ ☆ Bundle Adjustment in the Eager Mode
Bundle adjustment (BA) is a critical technique in various robotic applications such as simultaneous localization and mapping (SLAM), augmented reality (AR), and photogrammetry. BA optimizes parameters such as camera poses and 3D landmarks to align them with observations. With the growing importance of deep learning in perception systems, there is an increasing need to integrate BA with deep learning frameworks for enhanced reliability and performance. However, widely-used C++-based BA libraries, such as GTSAM, g$^2$o, and Ceres Solver, lack native integration with modern deep learning libraries like PyTorch. This limitation affects their flexibility, ease of debugging, and overall implementation efficiency. To address this gap, we introduce an eager-mode BA library seamlessly integrated with PyTorch with high efficiency. Our approach includes a sparsity-aware auto-differentiation design and GPU-accelerated sparse operations designed for 2nd-order optimization. Our eager-mode BA on GPU demonstrates substantial runtime efficiency, achieving an average speedup of 18.5$\times$, 22$\times$, and 23$\times$ across all benchmarks compared to GTSAM, g$^2$o, and Ceres, respectively.
♻ ☆ SHaRe-RL: Structured, Interactive Reinforcement Learning for Contact-Rich Industrial Assembly Tasks ICRA
High-mix low-volume (HMLV) industrial assembly, common in small and medium-sized enterprises (SMEs), requires the same precision, safety, and reliability as high-volume automation while remaining flexible to product variation and environmental uncertainty. Current robotic systems struggle to meet these demands. Manual programming is brittle and costly to adapt, while learning-based methods suffer from poor sample efficiency and unsafe exploration in contact-rich tasks. To address this, we present SHaRe-RL, a reinforcement learning framework that leverages multiple sources of prior knowledge. By (i) structuring skills into manipulation primitives, (ii) incorporating human demonstrations and online corrections, and (iii) bounding interaction forces with per-axis compliance, SHaRe-RL enables efficient and safe online learning for long-horizon, contact-rich industrial assembly tasks. Experiments on the insertion of industrial Harting connector modules with 0.2-0.4 mm clearance demonstrate that SHaRe-RL achieves reliable performance within practical time budgets. Our results show that process expertise, without requiring robotics or RL knowledge, can meaningfully contribute to learning, enabling safer, more robust, and more economically viable deployment of RL for industrial assembly.
comment: 8 pages, 8 figures, accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
♻ ☆ Ontological foundations for contrastive explanatory narration of robot plans
Mutual understanding of artificial agents' decisions is key to ensuring a trustworthy and successful human-robot interaction. Hence, robots are expected to make reasonable decisions and communicate them to humans when needed. In this article, the focus is on an approach to modeling and reasoning about the comparison of two competing plans, so that robots can later explain the divergent result. First, a novel ontological model is proposed to formalize and reason about the differences between competing plans, enabling the classification of the most appropriate one (e.g., the shortest, the safest, the closest to human preferences, etc.). This work also investigates the limitations of a baseline algorithm for ontology-based explanatory narration. To address these limitations, a novel algorithm is presented, leveraging divergent knowledge between plans and facilitating the construction of contrastive narratives. Through empirical evaluation, it is observed that the explanations excel beyond the baseline method.
♻ ☆ CompliantVLA-adaptor: VLM-Guided Variable Impedance Action for Safe Contact-Rich Manipulation
We propose a CompliantVLA-adaptor that augments the state-of-the-art Vision-Language-Action (VLA) models with vision-language model (VLM)-informed context-aware variable impedance control (VIC) to improve the safety and effectiveness of contact-rich robotic manipulation tasks. Existing VLA systems (e.g., RDT, Pi0.5, OpenVLA-oft) typically output position, but lack force-aware adaptation, leading to unsafe or failed interactions in physical tasks involving contact, compliance, or uncertainty. In the proposed CompliantVLA-adaptor, a VLM interprets task context from images and natural language to adapt the stiffness and damping parameters of a VIC controller. These parameters are further regulated using real-time force/torque feedback to ensure interaction forces remain within safe thresholds. We demonstrate that our method outperforms the VLA baselines on a suite of complex contact-rich tasks, both in simulation and the real world, with improved success rates and reduced force violations. This work presents a promising path towards a safe foundation model for physical contact-rich manipulation. We release our code, prompts, and force-torque-impedance-scenario context datasets at https://sites.google.com/view/compliantvla.
comment: under review
♻ ☆ Real-Time Quasi-Static Modeling of UAV Tether Aerodynamics
One of the main limitations of multirotor UAVs is their short flight time due to battery constraints. A practical solution for continuous operation is to power the drone from the ground via a tether. While this approach has been demonstrated for stationary systems, scenarios with a fast-moving base vehicle or strong wind conditions require modeling the tether forces, including aerodynamic effects. In this work, we propose two complementary approaches for real-time quasi-static tether modeling with aerodynamics. The first is an analytical method based on catenary theory with a uniform drag assumption, achieving very fast solve times below 1~ms. The second is a numerical method that discretizes the tether into segments and lumped masses, solving the equilibrium equations using CasADi and IPOPT. By leveraging initialization strategies, such as warm starting and analytical initialization, real-time performance was achieved with a solve time of 5~ms, while allowing for flexible force formulations. Both approaches were validated in real-world tests using a load cell to measure the tether force. The results show that the analytical method provides sufficient accuracy for most tethered UAV applications with minimal computational cost, while the numerical method offers higher flexibility and physical accuracy when required. These approaches form a lightweight and extensible framework for real-time tether simulation, applicable to both offline optimization and online tasks such as simulation, control, and trajectory planning.
♻ ☆ System Design of the Ultra Mobility Vehicle: A Driving, Balancing, and Jumping Bicycle Robot
Trials cyclists and mountain bike riders can hop, jump, balance, and drive on one or both wheels. This versatility allows them to achieve speed and energy-efficiency on smooth terrain and agility over rough terrain. Inspired by these athletes, we present the design and control of a robotic platform, Ultra Mobility Vehicle (UMV), which combines a bicycle and a reaction mass to move dynamically with minimal actuated degrees of freedom. We employ a simulation-driven design optimization process to synthesize a spatial linkage topology with a focus on vertical jump height and momentum-based balancing on a single wheel contact. Using a constrained Reinforcement Learning (RL) framework, we demonstrate zero-shot transfer of diverse athletic behaviors, including track-stands, jumps, wheelies, rear wheel hopping, and front flips. This 23.5 kg robot is capable of high speeds (8 m/s) and jumping on and over large obstacles (1 m tall, or 130% of the robot's nominal height).
comment: 17 Pages, 11 figures, 3 movies, 2 tables
♻ ☆ Contraction Theory for Nonlinear Stability Analysis and Learning-based Control: A Tutorial Overview
Contraction theory is an analytical tool to study differential dynamics of a non-autonomous (i.e., time-varying) nonlinear system under a contraction metric defined with a uniformly positive definite matrix, the existence of which results in a necessary and sufficient characterization of incremental exponential stability of multiple solution trajectories with respect to each other. By using a squared differential length as a Lyapunov-like function, its nonlinear stability analysis boils down to finding a suitable contraction metric that satisfies a stability condition expressed as a linear matrix inequality, indicating that many parallels can be drawn between well-known linear systems theory and contraction theory for nonlinear systems. Furthermore, contraction theory takes advantage of a superior robustness property of exponential stability used in conjunction with the comparison lemma. This yields much-needed safety and stability guarantees for neural network-based control and estimation schemes, without resorting to a more involved method of using uniform asymptotic stability for input-to-state stability. Such distinctive features permit the systematic construction of a contraction metric via convex optimization, thereby obtaining an explicit exponential bound on the distance between a time-varying target trajectory and solution trajectories perturbed externally due to disturbances and learning errors. The objective of this paper is, therefore, to present a tutorial overview of contraction theory and its advantages in nonlinear stability analysis of deterministic and stochastic systems, with an emphasis on deriving formal robustness and stability guarantees for various learning-based and data-driven automatic control methods. In particular, we provide a detailed review of techniques for finding contraction metrics and associated control and estimation laws using deep neural networks.
comment: Annual Reviews in Control, Preprint Version, Accepted, Oct. 1st
♻ ☆ BiGraspFormer: End-to-End Bimanual Grasp Transformer
Bimanual grasping is essential for robots to handle large and complex objects. However, existing methods either focus solely on single-arm grasping or employ separate grasp generation and bimanual evaluation stages, leading to coordination problems including collision risks and unbalanced force distribution. To address these limitations, we propose BiGraspFormer, a unified end-to-end transformer framework that directly generates coordinated bimanual grasps from object point clouds. Our key idea is the Single-Guided Bimanual (SGB) strategy, which first generates diverse single grasp candidates using a transformer decoder, then leverages their learned features through specialized attention mechanisms to jointly predict bimanual poses and quality scores. This conditioning strategy reduces the complexity of the 12-DoF search space while ensuring coordinated bimanual manipulation. Comprehensive simulation experiments and real-world validation demonstrate that BiGraspFormer consistently outperforms existing methods while maintaining efficient inference speed (<0.05s), confirming the effectiveness of our framework. Code and supplementary materials are available at https://sites.google.com/view/bigraspformer
comment: 8 pages, 5 figures
♻ ☆ Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy ICRA2026
Diffusion-based policies have achieved remarkable results in robotic manipulation but often struggle to adapt rapidly in dynamic scenarios, leading to delayed responses or task failures. We present DCDP, a Dynamic Closed-Loop Diffusion Policy framework that integrates chunk-based action generation with real-time correction. DCDP integrates a self-supervised dynamic feature encoder, cross-attention fusion, and an asymmetric action encoder-decoder to inject environmental dynamics before action execution, achieving real-time closed-loop action correction and enhancing the system's adaptability in dynamic scenarios. In dynamic PushT simulations, DCDP improves adaptability by 19\% without retraining while requiring only 5\% additional computation. Its modular design enables plug-and-play integration, achieving both temporal coherence and real-time responsiveness in dynamic robotic scenarios, including real-world manipulation tasks. The project page is at: https://github.com/wupengyuan/dcdp
comment: Accepted by ICRA2026
♻ ☆ Minimal Intervention Shared Control with Guaranteed Safety under Non-Convex Constraints ICRA
Shared control combines human intention with autonomous decision-making. At the low level, the primary goal is to maintain safety regardless of the user's input to the system. However, existing shared control methods-based on, e.g., Model Predictive Control, Control Barrier Functions, or learning-based control-often face challenges with feasibility, scalability, and mixed constraints. To address these challenges, we propose a Constraint-Aware Assistive Controller that computes control actions online while ensuring recursive feasibility, strict constraint satisfaction, and minimal deviation from the user's intent. It also accommodates a structured class of non-convex constraints common in real-world settings. We leverage Robust Controlled Invariant Sets for recursive feasibility and a Mixed-Integer Quadratic Programming formulation to handle non-convex constraints. We validate the approach through a large-scale user study with 66 participants-one of the most extensive in shared control research-using a simulated environment to assess task load, trust, and perceived control, in addition to performance. The results show consistent improvements across all these aspects without compromising safety and user intent. Additionally, a real-world experiment on a robotic manipulator demonstrates the framework's applicability under bounded disturbances, ensuring safety and collision-free operation.
comment: Accepted for publication at the 2026 IEEE International Conference on Robotics and Automation (ICRA)
♻ ☆ When a Robot is More Capable than a Human: Learning from Constrained Demonstrators
Learning from demonstrations enables experts to teach robots complex tasks using interfaces such as kinesthetic teaching, joystick control, and sim-to-real transfer. However, these interfaces often constrain the expert's ability to demonstrate optimal behavior due to indirect control, setup restrictions, and hardware safety. For example, a joystick can move a robotic arm only in a 2D plane, even though the robot operates in a higher-dimensional space. As a result, the demonstrations collected by constrained experts lead to suboptimal performance of the learned policies. This raises a key question: Can a robot learn a better policy than the one demonstrated by a constrained expert? We address this by allowing the agent to go beyond direct imitation of expert actions and explore shorter and more efficient trajectories. We use the demonstrations to infer a state-only reward signal that measures task progress, and self-label reward for unknown states using temporal interpolation. Our approach outperforms common imitation learning in both sample efficiency and task completion time. On a real WidowX robotic arm, it completes the task in 12 seconds, 10x faster than behavioral cloning, as shown in real-robot videos on https://sites.google.com/view/constrainedexpert .
♻ ☆ One-Shot Badminton Shuttle Detection for Mobile Robots
This paper presents a robust one-shot badminton shuttlecock detection framework for non-stationary robots. To address the lack of egocentric shuttlecock detection datasets, we introduce a dataset of 20,510 semi-automatically annotated frames captured across 11 distinct backgrounds in diverse indoor and outdoor environments, and categorize each frame into one of three difficulty levels. For labeling, we present a novel semi-automatic annotation pipeline, that enables efficient labeling from stationary camera footage. We propose a metric suited to our downstream use case and fine-tune a YOLOv8 network optimized for real-time shuttlecock detection, achieving an F1-score of 0.86 under our metric in test environments similar to training, and 0.70 in entirely unseen environments. Our analysis reveals that detection performance is critically dependent on shuttlecock size and background texture complexity. Qualitative experiments confirm their applicability to robots with moving cameras. Unlike prior work with stationary camera setups, our detector is specifically designed for the egocentric, dynamic viewpoints of mobile robots, providing a foundational building block for downstream tasks, including tracking, trajectory estimation, and system (re)-initialization.
♻ ☆ Metamorphic Testing of Vision-Language Action-Enabled Robots
Vision-Language-Action (VLA) models are multimodal robotic task controllers that, given an instruction and visual inputs, produce a sequence of low-level control actions (or motor commands) enabling a robot to execute the requested task in the physical environment. These systems face the test oracle problem from multiple perspectives. On the one hand, a test oracle must be defined for each instruction prompt, which is a complex and non-generalizable approach. On the other hand, current state-of-the-art oracles typically capture symbolic representations of the world (e.g., robot and object states), enabling the correctness evaluation of a task, but fail to assess other critical aspects, such as the quality with which VLA-enabled robots perform a task. In this paper, we explore whether Metamorphic Testing (MT) can alleviate the test oracle problem in this context. To do so, we propose two metamorphic relation patterns and five metamorphic relations to assess whether changes to the test inputs impact the original trajectory of the VLA-enabled robots. An empirical study involving five VLA models, two simulated robots, and four robotic tasks shows that MT can effectively alleviate the test oracle problem by automatically detecting diverse types of failures, including, but not limited to, uncompleted tasks. More importantly, the proposed MRs are generalizable, making the proposed approach applicable across different VLA models, robots, and tasks, even in the absence of test oracles.
♻ ☆ Trust in Autonomous Human--Robot Collaboration: Effects of Responsive Interaction Policies
Trust plays a central role in human--robot collaboration, yet its formation is rarely examined under the constraints of fully autonomous interaction. This pilot study investigated how interaction policy influences trust during in-person collaboration with a social robot operating without Wizard-of-Oz control or scripted repair. Participants completed a multi-stage collaborative task with a mobile robot that autonomously managed spoken-language dialogue, affect inference, and task progression. Two interaction policies were compared: a responsive policy, in which the robot proactively adapted its dialogue and assistance based on inferred interaction state, and a neutral, reactive policy, in which the robot provided only direct, task-relevant responses when prompted. Responsive interaction was associated with significantly higher post-interaction trust under viable communication conditions, despite no reliable differences in overall task accuracy. Sensitivity analyses indicated that affective and experiential components of trust were more sensitive to communication breakdown than evaluative judgments of reliability, and that as language-mediated interaction degraded, the trust advantage associated with responsiveness attenuated and ratings became less clearly interpretable as calibrated evaluations of collaborative competence. These findings suggest that trust in autonomous human--robot interaction emerges from process-level interaction dynamics and operates within constraints imposed by communication viability, highlighting the importance of evaluating trust under real autonomy conditions when designing interactive robotic systems.
♻ ☆ Dual-Agent Reinforcement Learning for Adaptive and Cost-Aware Visual-Inertial Odometry CVPR 2026
Visual-Inertial Odometry (VIO) is a critical component for robust ego-motion estimation, enabling foundational capabilities such as autonomous navigation in robotics and real-time 6-DoF tracking for augmented reality. Existing methods face a well-known trade-off: filter-based approaches are efficient but prone to drift, while optimization-based methods, though accurate, rely on computationally prohibitive Visual-Inertial Bundle Adjustment (VIBA) that is difficult to run on resource-constrained platforms. Rather than removing VIBA altogether, we aim to reduce how often and how heavily it must be invoked. To this end, we cast two key design choices in modern VIO, when to run the visual frontend and how strongly to trust its output, as sequential decision problems, and solve them with lightweight reinforcement learning (RL) agents. Our framework introduces a lightweight, dual-pronged RL policy that serves as our core contribution: (1) a Select Agent intelligently gates the entire VO pipeline based only on high-frequency IMU data; and (2) a composite Fusion Agent that first estimates a robust velocity state via a supervised network, before an RL policy adaptively fuses the full (p, v, q) state. Experiments on the EuRoC MAV and TUM-VI datasets show that, in our unified evaluation, the proposed method achieves a more favorable accuracy-efficiency-memory trade-off than prior GPU-based VO/VIO systems: it attains the best average ATE while running up to 1.77 times faster and using less GPU memory. Compared to classical optimization-based VIO systems, our approach maintains competitive trajectory accuracy while substantially reducing computational load.
comment: Accepted to the CVPR 2026 Main Track
♻ ☆ Optimal Solutions for the Moving Target Vehicle Routing Problem via Branch-and-Price with Relaxed Continuity ICAPS 2026
The Moving Target Vehicle Routing Problem (MT-VRP) seeks trajectories for several agents that intercept a set of moving targets, subject to speed, time window, and capacity constraints. We introduce an exact algorithm, Branch-and-Price with Relaxed Continuity (BPRC), for the MT-VRP. The main challenge in a branch-and-price approach for the MT-VRP is the pricing subproblem, which is complicated by moving targets and time-dependent travel costs between targets. Our key contribution is a new labeling algorithm that solves this subproblem by means of a novel dominance criterion tailored for problems with moving targets. Numerical results on instances with up to 25 targets show that our algorithm finds optimal solutions more than an order of magnitude faster than a baseline based on previous work, showing particular strength in scenarios with limited agent capacities.
comment: Accepted to ICAPS 2026
Robotics 103
☆ FlatLands: Generative Floormap Completion From a Single Egocentric View
A single egocentric image typically captures only a small portion of the floor, yet a complete metric traversability map of the surroundings would better serve applications such as indoor navigation. We introduce FlatLands, a dataset and benchmark for single-view bird's-eye view (BEV) floor completion. The dataset contains 270,575 observations from 17,656 real metric indoor scenes drawn from six existing datasets, with aligned observation, visibility, validity, and ground-truth BEV maps, and the benchmark includes both in- and out-of-distribution evaluation protocols. We compare training-free approaches, deterministic models, ensembles, and stochastic generative models. Finally, we instantiate the task as an end-to-end monocular RGB-to-floormaps pipeline. FlatLands provides a rigorous testbed for uncertainty-aware indoor mapping and generative completion for embodied navigation.
comment: Under review
☆ Safety Case Patterns for VLA-based driving systems: Insights from SimLingo
Vision-Language-Action (VLA)-based driving systems represent a significant paradigm shift in autonomous driving since, by combining traffic scene understanding, linguistic interpretation, and action generation, these systems enable more flexible, adaptive, and instruction-responsive driving behaviors. However, despite their growing adoption and potential to support socially responsible autonomous driving while understanding high-level human instructions, VLA-based driving systems may exhibit new types of hazardous behaviors. Such as the addition of natural language inputs (e.g., user or navigation instructions) into the multimodal control loop, which may lead to unpredictable and unsafe behaviors that could endanger vehicle occupants and pedestrians. Hence, assuring the safety of these systems is crucial to help build trust in their operations. To support this, we propose a novel safety case design approach called RAISE. Our approach introduces novel patterns tailored to instruction-based driving systems such as VLA-based driving systems, an extension of Hazard Analysis and Risk Assessment (HARA) detailing safe scenarios and their outcomes, and a design technique to create the safety cases of VLA-based driving systems. A case study on SimLingo illustrates how our approach can be used to construct rigorous, evidence-based safety claims for this emerging class of autonomous driving systems.
☆ ExpertGen: Scalable Sim-to-Real Expert Policy Learning from Imperfect Behavior Priors
Learning generalizable and robust behavior cloning policies requires large volumes of high-quality robotics data. While human demonstrations (e.g., through teleoperation) serve as the standard source for expert behaviors, acquiring such data at scale in the real world is prohibitively expensive. This paper introduces ExpertGen, a framework that automates expert policy learning in simulation to enable scalable sim-to-real transfer. ExpertGen first initializes a behavior prior using a diffusion policy trained on imperfect demonstrations, which may be synthesized by large language models or provided by humans. Reinforcement learning is then used to steer this prior toward high task success by optimizing the diffusion model's initial noise while keep original policy frozen. By keeping the pretrained diffusion policy frozen, ExpertGen regularizes exploration to remain within safe, human-like behavior manifolds, while also enabling effective learning with only sparse rewards. Empirical evaluations on challenging manipulation benchmarks demonstrate that ExpertGen reliably produces high-quality expert policies with no reward engineering. On industrial assembly tasks, ExpertGen achieves a 90.5% overall success rate, while on long-horizon manipulation tasks it attains 85% overall success, outperforming all baseline methods. The resulting policies exhibit dexterous control and remain robust across diverse initial configurations and failure states. To validate sim-to-real transfer, the learned state-based expert policies are further distilled into visuomotor policies via DAgger and successfully deployed on real robotic hardware.
☆ Gaze-Aware Task Progression Detection Framework for Human-Robot Interaction Using RGB Cameras
In human-robot interaction (HRI), detecting a human's gaze helps robots interpret user attention and intent. However, most gaze detection approaches rely on specialized eye-tracking hardware, limiting deployment in everyday settings. Appearance-based gaze estimation methods remove this dependency by using standard RGB cameras, but their practicality in HRI remains underexplored. We present a calibration-free framework for detecting task progression when information is conveyed via integrated display interfaces. The framework uses only the robot's built-in monocular RGB camera (640x480 resolution) and state-of-the-art gaze estimation to monitor attention patterns. It leverages natural behavior, where users shift focus from task interfaces to the robot's face to signal task completion, formalized through three Areas of Interest (AOI): tablet, robot face, and elsewhere. Systematic parameter optimization identifies configurations that balance detection accuracy and interaction latency. We validate our framework in a "First Day at Work" scenario, comparing it to button-based interaction. Results show a task completion detection accuracy of 77.6%. Compared to button-based interaction, the proposed system exhibits slightly higher response latency but preserves information retention and significantly improves comfort, social presence, and perceived naturalness. Notably, most participants reported that they did not consciously use eye movements to guide the interaction, underscoring the intuitive role of gaze as a communicative cue. This work demonstrates the feasibility of intuitive, low-cost, RGB-only gaze-based HRI for natural and engaging interactions.
comment: 9 pages, 7 figures. This article has been accepted for publication in IEEE Robotics and Automation Letters
☆ AsgardBench - Evaluating Visually Grounded Interactive Planning Under Minimal Feedback
With AsgardBench we aim to evaluate visually grounded, high-level action sequence generation and interactive planning, focusing specifically on plan adaptation during execution based on visual observations rather than navigation or low-level manipulation. In the landscape of embodied AI benchmarks, AsgardBench targets the capability category of interactive planning, which is more sophisticated than offline high-level planning as it requires agents to revise plans in response to environmental feedback, yet remains distinct from low-level execution. Unlike prior embodied AI benchmarks that conflate reasoning with navigation or provide rich corrective feedback that substitutes for perception, AsgardBench restricts agent input to images, action history, and lightweight success/failure signals, isolating interactive planning in a controlled simulator without low-level control noise. The benchmark contains 108 task instances spanning 12 task types, each systematically varied through object state, placement, and scene configuration. These controlled variations create conditional branches in which a single instruction can require different action sequences depending on what the agent observes, emphasizing conditional branching and plan repair during execution. Our evaluations of leading vision language models show that performance drops sharply without visual input, revealing weaknesses in visual grounding and state tracking that ultimately undermine interactive planning. Our benchmark zeroes in on a narrower question: can a model actually use what it sees to adapt a plan when things do not go as expected?
comment: 19 figures, 6 tables, including appendix
☆ Resilience Meets Autonomy: Governing Embodied AI in Critical Infrastructure
Critical infrastructure increasingly incorporates embodied AI for monitoring, predictive maintenance, and decision support. However, AI systems designed to handle statistically representable uncertainty struggle with cascading failures and crisis dynamics that exceed their training assumptions. This paper argues that Embodied AIs resilience depends on bounded autonomy within a hybrid governance architecture. We outline four oversight modes and map them to critical infrastructure sectors based on task complexity, risk level, and consequence severity. Drawing on the EU AI Act, ISO safety standards, and crisis management research, we argue that effective governance requires a structured allocation of machine capability and human judgement.
comment: 6 pages
☆ Regularized Latent Dynamics Prediction is a Strong Baseline For Behavioral Foundation Models ICLR 2026
Behavioral Foundation Models (BFMs) produce agents with the capability to adapt to any unknown reward or task. These methods, however, are only able to produce near-optimal policies for the reward functions that are in the span of some pre-existing state features, making the choice of state features crucial to the expressivity of the BFM. As a result, BFMs are trained using a variety of complex objectives and require sufficient dataset coverage, to train task-useful spanning features. In this work, we examine the question: are these complex representation learning objectives necessary for zero-shot RL? Specifically, we revisit the objective of self-supervised next-state prediction in latent space for state feature learning, but observe that such an objective alone is prone to increasing state-feature similarity, and subsequently reducing span. We propose an approach, Regularized Latent Dynamics Prediction (RLDP), that adds a simple orthogonality regularization to maintain feature diversity and can match or surpass state-of-the-art complex representation learning methods for zero-shot RL. Furthermore, we empirically show that prior approaches perform poorly in low-coverage scenarios where RLDP still succeeds.
comment: ICLR 2026
☆ FEEL (Force-Enhanced Egocentric Learning): A Dataset for Physical Action Understanding
We introduce FEEL (Force-Enhanced Egocentric Learning), the first large-scale dataset pairing force measurements gathered from custom piezoresistive gloves with egocentric video. Our gloves enable scalable data collection, and FEEL contains approximately 3 million force-synchronized frames of natural unscripted manipulation in kitchen environments, with 45% of frames involving hand-object contact. Because force is the underlying cause that drives physical interaction, it is a critical primitive for physical action understanding. We demonstrate the utility of force for physical action understanding through application of FEEL to two families of tasks: (1) contact understanding, where we jointly perform temporal contact segmentation and pixel-level contacted object segmentation; and, (2) action representation learning, where force prediction serves as a self-supervised pretraining objective for video backbones. We achieve state-of-the-art temporal contact segmentation results and competitive pixel-level segmentation results without any need for manual contacted object segmentation annotations. Furthermore we demonstrate that action representation learning with FEEL improves transfer performance on action understanding tasks without any manual labels over EPIC-Kitchens, SomethingSomething-V2, EgoExo4D and Meccano.
comment: 14 pages, 7 figures
☆ Robust Dynamic Object Detection in Cluttered Indoor Scenes via Learned Spatiotemporal Cues
Reliable dynamic object detection in cluttered environments remains a critical challenge for autonomous navigation. Purely geometric LiDAR pipelines that rely on clustering and heuristic filtering can miss dynamic obstacles when they move in close proximity to static structure or are only partially observed. Vision-augmented approaches can provide additional semantic cues, but are often limited by closed-set detectors and camera field-of-view constraints, reducing robustness to novel obstacles and out-of-frustum events. In this work, we present a LiDAR-only framework that fuses temporal occupancy-grid-based motion segmentation with a learned bird's-eye-view (BEV) dynamic prior. A fusion module prioritizes 3D detections when available, while using the learned dynamic grid to recover detections that would otherwise be lost due to proximity-induced false negatives. Experiments with motion-capture ground truth show our method achieves 28.67% higher recall and 18.50% higher F1 score than the state-of-the-art in substantially cluttered environments while maintaining comparable precision and position error.
☆ Emergent Dexterity via Diverse Resets and Large-Scale Reinforcement Learning
Reinforcement learning in massively parallel physics simulations has driven major progress in sim-to-real robot learning. However, current approaches remain brittle and task-specific, relying on extensive per-task engineering to design rewards, curricula, and demonstrations. Even with this engineering, they often fail on long-horizon, contact-rich manipulation tasks and do not meaningfully scale with compute, as performance quickly saturates when training revisits the same narrow regions of state space. We introduce \Method, a simple and scalable framework that enables on-policy reinforcement learning to robustly solve a broad class of dexterous manipulation tasks using a single reward function, fixed algorithm hyperparameters, no curricula, and no human demonstrations. Our key insight is that long-horizon exploration can be dramatically simplified by using simulator resets to systematically expose the RL algorithm to the diverse set of robot-object interactions which underlie dexterous manipulation. \Method\ programmatically generates such resets with minimal human input, converting additional compute directly into broader behavioral coverage and continued performance gains. We show that \Method\ gracefully scales to long-horizon dexterous manipulation tasks beyond the capabilities of existing approaches and is able to learn robust policies over significantly wider ranges of initial conditions than baselines. Finally, we distill \Method \ into visuomotor policies which display robust retrying behavior and substantially higher success rates than baselines when transferred to the real world zero-shot. Project webpage: https://omnireset.github.io
☆ CorrectionPlanner: Self-Correction Planner with Reinforcement Learning in Autonomous Driving
Autonomous driving requires safe planning, but most learning-based planners lack explicit self-correction ability: once an unsafe action is proposed, there is no mechanism to correct it. Thus, we propose CorrectionPlanner, an autoregressive planner with self-correction that models planning as motion-token generation within a propose, evaluate, and correct loop. At each planning step, the policy proposes an action, namely a motion token, and a learned collision critic predicts whether it will induce a collision within a short horizon. If the critic predicts a collision, we retain the sequence of historical unsafe motion tokens as a self-correction trace, generate the next motion token conditioned on it, and repeat this process until a safe motion token is proposed or the safety criterion is met. This self-correction trace, consisting of all unsafe motion tokens, represents the planner's correction process in motion-token space, analogous to a reasoning trace in language models. We train the planner with imitation learning followed by model-based reinforcement learning using rollouts from a pretrained world model that realistically models agents' reactive behaviors. Closed-loop evaluations show that CorrectionPlanner reduces collision rate by over 20% on Waymax and achieves state-of-the-art planning scores on nuPlan.
☆ Simulation Distillation: Pretraining World Models in Simulation for Rapid Real-World Adaptation
Simulation-to-real transfer remains a central challenge in robotics, as mismatches between simulated and real-world dynamics often lead to failures. While reinforcement learning offers a principled mechanism for adaptation, existing sim-to-real finetuning methods struggle with exploration and long-horizon credit assignment in the low-data regimes typical of real-world robotics. We introduce Simulation Distillation (SimDist), a sim-to-real framework that distills structural priors from a simulator into a latent world model and enables rapid real-world adaptation via online planning and supervised dynamics finetuning. By transferring reward and value models directly from simulation, SimDist provides dense planning signals from raw perception without requiring value learning during deployment. As a result, real-world adaptation reduces to short-horizon system identification, avoiding long-horizon credit assignment and enabling fast, stable improvement. Across precise manipulation and quadruped locomotion tasks, SimDist substantially outperforms prior methods in data efficiency, stability, and final performance. Project website and code: https://sim-dist.github.io/
comment: Project website: https://sim-dist.github.io/
☆ You've Got a Golden Ticket: Improving Generative Robot Policies With A Single Noise Vector
What happens when a pretrained generative robot policy is provided a constant initial noise as input, rather than repeatedly sampling it from a Gaussian? We demonstrate that the performance of a pretrained, frozen diffusion or flow matching policy can be improved with respect to a downstream reward by swapping the sampling of initial noise from the prior distribution (typically isotropic Gaussian) with a well-chosen, constant initial noise input -- a golden ticket. We propose a search method to find golden tickets using Monte-Carlo policy evaluation that keeps the pretrained policy frozen, does not train any new networks, and is applicable to all diffusion/flow matching policies (and therefore many VLAs). Our approach to policy improvement makes no assumptions beyond being able to inject initial noise into the policy and calculate (sparse) task rewards of episode rollouts, making it deployable with no additional infrastructure or models. Our method improves the performance of policies in 38 out of 43 tasks across simulated and real-world robot manipulation benchmarks, with relative improvements in success rate by up to 58% for some simulated tasks, and 60% within 50 search episodes for real-world tasks. We also show unique benefits of golden tickets for multi-task settings: the diversity of behaviors from different tickets naturally defines a Pareto frontier for balancing different objectives (e.g., speed, success rates); in VLAs, we find that a golden ticket optimized for one task can also boost performance in other related tasks. We release a codebase with pretrained policies and golden tickets for simulation benchmarks using VLAs, diffusion policies, and flow matching policies.
comment: 13 pages, 9 figures
☆ Towards Generalizable Robotic Manipulation in Dynamic Environments
Vision-Language-Action (VLA) models excel in static manipulation but struggle in dynamic environments with moving targets. This performance gap primarily stems from a scarcity of dynamic manipulation datasets and the reliance of mainstream VLAs on single-frame observations, restricting their spatiotemporal reasoning capabilities. To address this, we introduce DOMINO, a large-scale dataset and benchmark for generalizable dynamic manipulation, featuring 35 tasks with hierarchical complexities, over 110K expert trajectories, and a multi-dimensional evaluation suite. Through comprehensive experiments, we systematically evaluate existing VLAs on dynamic tasks, explore effective training strategies for dynamic awareness, and validate the generalizability of dynamic data. Furthermore, we propose PUMA, a dynamics-aware VLA architecture. By integrating scene-centric historical optical flow and specialized world queries to implicitly forecast object-centric future states, PUMA couples history-aware perception with short-horizon prediction. Results demonstrate that PUMA achieves state-of-the-art performance, yielding a 6.3% absolute improvement in success rate over baselines. Moreover, we show that training on dynamic data fosters robust spatiotemporal representations that transfer to static tasks. All code and data are available at https://github.com/H-EmbodVis/DOMINO.
☆ HSImul3R: Physics-in-the-Loop Reconstruction of Simulation-Ready Human-Scene Interactions
We present HSImul3R, a unified framework for simulation-ready 3D reconstruction of human-scene interactions (HSI) from casual captures, including sparse-view images and monocular videos. Existing methods suffer from a perception-simulation gap: visually plausible reconstructions often violate physical constraints, leading to instability in physics engines and failure in embodied AI applications. To bridge this gap, we introduce a physically-grounded bi-directional optimization pipeline that treats the physics simulator as an active supervisor to jointly refine human dynamics and scene geometry. In the forward direction, we employ Scene-targeted Reinforcement Learning to optimize human motion under dual supervision of motion fidelity and contact stability. In the reverse direction, we propose Direct Simulation Reward Optimization, which leverages simulation feedback on gravitational stability and interaction success to refine scene geometry. We further present HSIBench, a new benchmark with diverse objects and interaction scenarios. Extensive experiments demonstrate that HSImul3R produces the first stable, simulation-ready HSI reconstructions and can be directly deployed to real-world humanoid robots.
comment: https://yukangcao.github.io/HSImul3R/
☆ Perception-Aware Autonomous Exploration in Feature-Limited Environments
Autonomous exploration in unknown environments typically relies on onboard state estimation for localisation and mapping. Existing exploration methods primarily maximise coverage efficiency, but often overlook that visual-inertial odometry (VIO) performance strongly depends on the availability of robust visual features. As a result, exploration policies can drive a robot into feature-sparse regions where tracking degrades, leading to odometry drift, corrupted maps, and mission failure. We propose a hierarchical perception-aware exploration framework for a stereo-equipped unmanned aerial vehicle (UAV) that explicitly couples exploration progress with feature observability. Our approach (i) associates each candidate frontier with an expected feature quality using a global feature map, and prioritises visually informative subgoals, and (ii) optimises a continuous yaw trajectory along the planned motion to maintain stable feature tracks. We evaluate our method in simulation across environments with varying texture levels and in real-world indoor experiments with largely textureless walls. Compared to baselines that ignore feature quality and/or do not optimise continuous yaw, our method maintains more reliable feature tracking, reduces odometry drift, and achieves on average 30\% higher coverage before the odometry error exceeds specified thresholds.
☆ EAAE: Energy-Aware Autonomous Exploration for UAVs in Unknown 3D Environments
Battery-powered multirotor unmanned aerial vehicles (UAVs) can rapidly map unknown environments, but mission performance is often limited by energy rather than geometry alone. Standard exploration policies that optimise for coverage or time can therefore waste energy through manoeuvre-heavy trajectories. In this paper, we address energy-aware autonomous 3D exploration for multirotor UAVs in initially unknown environments. We propose Energy-Aware Autonomous Exploration (EAAE), a modular frontier-based framework that makes energy an explicit decision variable during frontier selection. EAAE clusters frontiers into view-consistent regions, plans dynamically feasible candidate trajectories to the most informative clusters, and predicts their execution energy using an offline power estimation loop. The next target is then selected by minimising predicted trajectory energy while preserving exploration progress through a dual-layer planning architecture for safe execution. We evaluate EAAE in a full exploration pipeline with a rotor-speed-based power model across simulated 3D environments of increasing complexity. Compared to representative distance-based and information gain-based frontier baselines, EAAE consistently reduces total energy consumption while maintaining competitive exploration time and comparable map quality, providing a practical drop-in energy-aware layer for frontier exploration.
☆ S2Act: Simple Spiking Actor
Spiking neural networks (SNNs) and biologically-inspired learning mechanisms are attractive in mobile robotics, where the size and performance of onboard neural network policies are constrained by power and computational budgets. Existing SNN approaches, such as population coding, reward modulation, and hybrid artificial neural network (ANN)-SNN architectures, have shown promising results; however, they face challenges in complex, highly stochastic environments due to SNN sensitivity to hyperparameters and inconsistent gradient signals. To address these challenges, we propose simple spiking actor (S2Act), a computationally lightweight framework that deploys an RL policy using an SNN in three steps: (1) architect an actor-critic model based on an approximated network of rate-based spiking neurons, (2) train the network with gradients using compatible activation functions, and (3) transfer the trained weights into physical parameters of rate-based leaky integrate-and-fire (LIF) neurons for inference and deployment. By globally shaping LIF neuron parameters such that their rate-based responses approximate ReLU activations, S2Act effectively mitigates the vanishing gradient problem, while pre-constraining LIF response curves reduces reliance on complex SNN-specific hyperparameter tuning. We demonstrate our method in two multi-agent stochastic environments (capture-the-flag and parking) that capture the complexity of multi-robot interactions, and deploy our trained policies on physical TurtleBot platforms using Intel's Loihi neuromorphic hardware. Our experimental results show that S2Act outperforms relevant baselines in task performance and real-time inference in nearly all considered scenarios, highlighting its potential for rapid prototyping and efficient real-world deployment of SNN-based RL policies.
comment: This work has been submitted to the IEEE for possible publication
☆ From Passive Observer to Active Critic: Reinforcement Learning Elicits Process Reasoning for Robotic Manipulation
Accurate process supervision remains a critical challenge for long-horizon robotic manipulation. A primary bottleneck is that current video MLLMs, trained primarily under a Supervised Fine-Tuning (SFT) paradigm, function as passive "Observers" that recognize ongoing events rather than evaluating the current state relative to the final task goal. In this paper, we introduce PRIMO R1 (Process Reasoning Induced Monitoring), a 7B framework that transforms video MLLMs into active "Critics". We leverage outcome-based Reinforcement Learning to incentivize explicit Chain-of-Thought generation for progress estimation. Furthermore, our architecture constructs a structured temporal input by explicitly anchoring the video sequence between initial and current state images. Supported by the proposed PRIMO Dataset and Benchmark, extensive experiments across diverse in-domain environments and out-of-domain real-world humanoid scenarios demonstrate that PRIMO R1 achieves state-of-the-art performance. Quantitatively, our 7B model achieves a 50% reduction in the mean absolute error of specialized reasoning baselines, demonstrating significant relative accuracy improvements over 72B-scale general MLLMs. Furthermore, PRIMO R1 exhibits strong zero-shot generalization on difficult failure detection tasks. We establish state-of-the-art performance on RoboFail benchmark with 67.0% accuracy, surpassing closed-source models like OpenAI o1 by 6.0%.
comment: 31 pages
☆ Panoramic Affordance Prediction
Affordance prediction serves as a critical bridge between perception and action in embodied AI. However, existing research is confined to pinhole camera models, which suffer from narrow Fields of View (FoV) and fragmented observations, often missing critical holistic environmental context. In this paper, we present the first exploration into Panoramic Affordance Prediction, utilizing 360-degree imagery to capture global spatial relationships and holistic scene understanding. To facilitate this novel task, we first introduce PAP-12K, a large-scale benchmark dataset containing over 1,000 ultra-high-resolution (12k, 11904 x 5952) panoramic images with over 12k carefully annotated QA pairs and affordance masks. Furthermore, we propose PAP, a training-free, coarse-to-fine pipeline inspired by the human foveal visual system to tackle the ultra-high resolution and severe distortion inherent in panoramic images. PAP employs recursive visual routing via grid prompting to progressively locate targets, applies an adaptive gaze mechanism to rectify local geometric distortions, and utilizes a cascaded grounding pipeline to extract precise instance-level masks. Experimental results on PAP-12K reveal that existing affordance prediction methods designed for standard perspective images suffer severe performance degradation and fail due to the unique challenges of panoramic vision. In contrast, PAP framework effectively overcomes these obstacles, significantly outperforming state-of-the-art baselines and highlighting the immense potential of panoramic perception for robust embodied intelligence.
☆ Kimodo: Scaling Controllable Human Motion Generation
High-quality human motion data is becoming increasingly important for applications in robotics, simulation, and entertainment. Recent generative models offer a potential data source, enabling human motion synthesis through intuitive inputs like text prompts or kinematic constraints on poses. However, the small scale of public mocap datasets has limited the motion quality, control accuracy, and generalization of these models. In this work, we introduce Kimodo, an expressive and controllable kinematic motion diffusion model trained on 700 hours of optical motion capture data. Our model generates high-quality motions while being easily controlled through text and a comprehensive suite of kinematic constraints including full-body keyframes, sparse joint positions/rotations, 2D waypoints, and dense 2D paths. This is enabled through a carefully designed motion representation and two-stage denoiser architecture that decomposes root and body prediction to minimize motion artifacts while allowing for flexible constraint conditioning. Experiments on the large-scale mocap dataset justify key design decisions and analyze how the scaling of dataset size and model size affect performance.
comment: Project page: https://research.nvidia.com/labs/sil/projects/kimodo/
☆ Optimal control of differentially flat underactuated planar robots in the perspective of oscillation mitigation
Underactuated robots are characterized by a larger number of degrees of freedom than actuators and if they are designed with a specific mass distribution, they can be controlled by means of differential flatness theory. This structural property enables the development of lightweight and cost-effective robotic systems with enhanced dexterity. However, a key challenge lies in managing the passive joints, whose control demands precise and comprehensive dynamic modeling of the system. To simplify dynamic models, particularly for low-speed trajectories, friction is often neglected. While this assumption simplifies analysis and control design, it introduces residual oscillations of the end-effector about the target position. In this paper, the possibility of using optimal control along with differential flatness control is investigated to improve the tracking of the planned trajectories. First, the study was carried out through formal analysis, and then, it was validated by means of numerical simulations. Results highlight that optimal control can be used to plan the flat variables considering different (quadratic) performance indices: control effort, i.e. motor torque, and potential energy of the considered underactuated joint. Moreover, the minimization of potential energy can be used to design motion laws that are robust against variation of the stiffness and damping of the underactuated joint, thus reducing oscillations in the case of stiffness/damping mismatch.
comment: Accepted to European Control Conference (ECC 2026)
☆ Seeing Beyond: Extrapolative Domain Adaptive Panoramic Segmentation CVPR 2026
Cross-domain panoramic semantic segmentation has attracted growing interest as it enables comprehensive 360° scene understanding for real-world applications. However, it remains particularly challenging due to severe geometric Field of View (FoV) distortions and inconsistent open-set semantics across domains. In this work, we formulate an open-set domain adaptation setting, and propose Extrapolative Domain Adaptive Panoramic Segmentation (EDA-PSeg) framework that trains on local perspective views and tests on full 360° panoramic images, explicitly tackling both geometric FoV shifts across domains and semantic uncertainty arising from previously unseen classes. To this end, we propose the Euler-Margin Attention (EMA), which introduces an angular margin to enhance viewpoint-invariant semantic representation, while performing amplitude and phase modulation to improve generalization toward unseen classes. Additionally, we design the Graph Matching Adapter (GMA), which builds high-order graph relations to align shared semantics across FoV shifts while effectively separating novel categories through structural adaptation. Extensive experiments on four benchmark datasets under camera-shift, weather-condition, and open-set scenarios demonstrate that EDA-PSeg achieves state-of-the-art performance, robust generalization to diverse viewing geometries, and resilience under varying environmental conditions. The code is available at https://github.com/zyfone/EDA-PSeg.
comment: Accepted to CVPR 2026. The code is available at https://github.com/zyfone/EDA-PSeg
☆ On the Derivation of Tightly-Coupled LiDAR-Inertial Odometry with VoxelMap
This note presents a concise mathematical formulation of tightly-coupled LiDAR-Inertial Odometry within an iterated error-state Kalman filter framework using a VoxelMap representation. Rather than proposing a new algorithm, it provides a clear and self-contained derivation that unifies the geometric modeling and probabilistic state estimation through consistent notation and explicit formulations. The document is intended to serve both as a technical reference and as an accessible entry point for a foundational understanding of the system architecture and estimation principles.
☆ RoCo Challenge at AAAI 2026: Benchmarking Robotic Collaborative Manipulation for Assembly Towards Industrial Automation
Embodied Artificial Intelligence (EAI) is rapidly developing, gradually subverting previous autonomous systems' paradigms from isolated perception to integrated, continuous action. This transition is highly significant for industrial robotic manipulation, promising to free human workers from repetitive, dangerous daily labor. To benchmark and advance this capability, we introduce the Robotic Collaborative Assembly Assistance (RoCo) Challenge with a dataset towards simulation and real-world assembly manipulation. Set against the backdrop of human-centered manufacturing, this challenge focuses on a high-precision planetary gearbox assembly task, a demanding yet highly representative operation in modern industry. Built upon a self-developed data collection, training, and evaluation system in Isaac Sim, and utilizing a dual-arm robot for real-world deployment, the challenge operates in two phases. The Simulation Round defines fine-grained task phases for step-wise scoring to handle the long-horizon nature of the assembly. The Real-World Round mirrors this evaluation with physical gearbox components and high-quality teleoperated datasets. The core tasks require assembling an epicyclic gearbox from scratch, including mounting three planet gears, a sun gear, and a ring gear. Attracting over 60 teams and 170+ participants from more than 10 countries, the challenge yielded highly effective solutions, most notably ARC-VLA and RoboCola. Results demonstrate that a dual-model framework for long-horizon multi-task learning is highly effective, and the strategic utilization of recovery-from-failure curriculum data is a critical insight for successful deployment. This report outlines the competition setup, evaluation approach, key findings, and future directions for industrial EAI. Our dataset, CAD files, code, and evaluation results can be found at: https://rocochallenge.github.io/RoCo2026/.
comment: 16 pages, 8 figures
☆ Zero-Shot Generalization from Motion Demonstrations to New Tasks
Learning motion policies from expert demonstrations is an essential paradigm in modern robotics. While end-to-end models aim for broad generalization, they require large datasets and computationally heavy inference. Conversely, learning dynamical systems (DS) provides fast, reactive, and provably stable control from very few demonstrations. However, existing DS learning methods typically model isolated tasks and struggle to reuse demonstrations for novel behaviors. In this work, we formalize the problem of combining isolated demonstrations within a shared workspace to enable generalization to unseen tasks. The Gaussian Graph is introduced, which reinterprets spatial components of learned motion primitives as discrete vertices with connections to one another. This formulation allows us to bridge continuous control with discrete graph search. We propose two frameworks leveraging this graph: Stitching, for constructing time-invariant DSs, and Chaining, giving a sequence-based DS for complex motions while retaining convergence guarantees. Simulations and real-robot experiments show that these methods successfully generalize to new tasks where baseline methods fail.
☆ Formalisms for Robotic Mission Specification and Execution: A Comparative Analysis
Robots are increasingly deployed across diverse domains and designed for multi-purpose operation. As robotic systems grow in complexity and operate in dynamic environments, the need for structured, expressive, and scalable mission-specification approaches becomes critical, with mission specifications often defined in the field by domain experts rather than robotics specialists. However, there is no standard or widely accepted formalism for specifying missions in single- or multi-robot systems. A variety of formalisms, such as Behavior Trees, State Machines, Hierarchical Task Networks, and Business Process Model and Notation, have been adopted in robotics to varying degrees, each providing different levels of abstraction, expressiveness, and support for integration with human workflows and external devices. This paper presents a systematic analysis of these four formalisms with respect to their suitability for robot mission specification. Our study focuses on mission-level descriptions rather than robot software development. We analyze their underlying control structures and mission concepts, evaluate their expressiveness and limitations in modeling real-world missions, and assess the extent of available tool support. By comparing the formalisms and validating our findings with experts, we provide insights into their applicability, strengths, and shortcomings in robotic system modeling. The results aim to support practitioners and researchers in selecting appropriate modeling approaches for designing robust and adaptable robot and multi-robot missions.
☆ MA-VLCM: A Vision Language Critic Model for Value Estimation of Policies in Multi-Agent Team Settings
Multi-agent reinforcement learning (MARL) commonly relies on a centralized critic to estimate the value function. However, learning such a critic from scratch is highly sample-inefficient and often lacks generalization across environments. At the same time, large vision-language-action models (VLAs) trained on internet-scale data exhibit strong multimodal reasoning and zero-shot generalization capabilities, yet directly deploying them for robotic execution remains computationally prohibitive, particularly in heterogeneous multi-robot systems with diverse embodiments and resource constraints. To address these challenges, we propose Multi-Agent Vision-Language-Critic Models (MA-VLCM), a framework that replaces the learned centralized critic in MARL with a pretrained vision-language model fine-tuned to evaluate multi-agent behavior. MA-VLCM acts as a centralized critic conditioned on natural language task descriptions, visual trajectory observations, and structured multi-agent state information. By eliminating critic learning during policy optimization, our approach significantly improves sample efficiency while producing compact execution policies suitable for deployment on resource-constrained robots. Results show good zero-shot return estimation on models with differing VLM backbones on in-distribution and out-of-distribution scenarios in multi-agent team settings
comment: 7 pages, 6 figures
☆ End-to-End Dexterous Grasp Learning from Single-View Point Clouds via a Multi-Object Scene Dataset
Dexterous grasping in multi-object scene constitutes a fundamental challenge in robotic manipulation. Current mainstream grasping datasets predominantly focus on single-object scenarios and predefined grasp configurations, often neglecting environmental interference and the modeling of dexterous pre-grasp gesture, thereby limiting their generalizability in real-world applications. To address this, we propose DGS-Net, an end-to-end grasp prediction network capable of learning dense grasp configurations from single-view point clouds in multi-object scene. Furthermore, we propose a two-stage grasp data generation strategy that progresses from dense single-object grasp synthesis to dense scene-level grasp generation. Our dataset comprises 307 objects, 240 multi-object scenes, and over 350k validated grasps. By explicitly modeling grasp offsets and pre-grasp configurations, the dataset provides more robust and accurate supervision for dexterous grasp learning. Experimental results show that DGS-Net achieves grasp success rates of 88.63\% in simulation and 78.98\% on a real robotic platform, while exhibiting lower penetration with a mean penetration depth of 0.375 mm and penetration volume of 559.45 mm^3, outperforming existing methods and demonstrating strong effectiveness and generalization capability. Our dataset is available at https://github.com/4taotao8/DGS-Net.
comment: 10 pages, 6 figures. Submitted to IEEE Transactions on Automation Science and Engineering (T-ASE)
☆ Efficient Morphology-Control Co-Design via Stackelberg Proximal Policy Optimization
Morphology-control co-design concerns the coupled optimization of an agent's body structure and control policy. This problem exhibits a bi-level structure, where the control dynamically adapts to the morphology to maximize performance. Existing methods typically neglect the control's adaptation dynamics by adopting a single-level formulation that treats the control policy as fixed when optimizing morphology. This can lead to inefficient optimization, as morphology updates may be misaligned with control adaptation. In this paper, we revisit the co-design problem from a game-theoretic perspective, modeling the intrinsic coupling between morphology and control as a novel variant of a Stackelberg game. We propose Stackelberg Proximal Policy Optimization (Stackelberg PPO), which explicitly incorporates the control's adaptation dynamics into morphology optimization. By modeling this intrinsic coupling, our method aligns morphology updates with control adaptation, thereby stabilizing training and improving learning efficiency. Experiments across diverse co-design tasks demonstrate that Stackelberg PPO outperforms standard PPO in both stability and final performance, opening the way for dramatically more efficient robotics designs.
comment: presented at the Fourteenth International Conference on Learning Representations; 11 pages in main text + 3 pages of references + 23 pages of appendices, 5 figures in main text + 11 figures in appendices, 16 tables in appendices; accompanying website available at https://yanningdai.github.io/stackelberg-ppo-co-design/ ; source code available at https://github.com/YanningDai/StackelbergPPO
☆ NavThinker: Action-Conditioned World Models for Coupled Prediction and Planning in Social Navigation
Social navigation requires robots to act safely in dynamic human environments. Effective behavior demands thinking ahead: reasoning about how the scene and pedestrians evolve under different robot actions rather than reacting to current observations alone. This creates a coupled prediction-planning challenge, where robot actions and human motion mutually influence each other. To address this challenge, we propose NavThinker, a future-aware framework that couples an action-conditioned world model with on-policy reinforcement learning. The world model operates in the Depth Anything V2 patch feature space and performs autoregressive prediction of future scene geometry and human motion; multi-head decoders then produce future depth maps and human trajectories, yielding a future-aware state aligned with traversability and interaction risk. Crucially, we train the policy with DD-PPO while injecting world-model think-ahead signals via: (i) action-conditioned future features fused into the current observation embedding and (ii) social reward shaping from predicted human trajectories. Experiments on single- and multi-robot Social-HM3D show state-of-the-art navigation success, with zero-shot transfer to Social-MP3D and real-world deployment on a Unitree Go2, validating generalization and practical applicability. Webpage: https://github.com/hutslib/NavThinker.
☆ User-Tailored Learning to Forecast Walking Modes for Exosuits
Assistive robotic devices, like soft lower-limb exoskeletons or exosuits, are widely spreading with the promise of helping people in everyday life. To make such systems adaptive to the variety of users wearing them, it is desirable to endow exosuits with advanced perception systems. However, exosuits have little sensory equipment because they need to be light and easy to wear. This paper presents a perception module based on machine learning that aims at estimating 3 walking modes (i.e., ascending or descending stairs and walking on level ground) of users wearing an exosuit. We tackle this perception problem using only inertial data from two sensors. Our approach provides an estimate for both future and past timesteps that supports control and enables a self-labeling procedure for online model adaptation. Indeed, we show that our estimate can label data acquired online and refine the model for new users. A thorough analysis carried out on real-life datasets shows the effectiveness of our user-tailored perception module. Finally, we integrate our system with the exosuit in a closed-loop controller, validating its performance in an online single-subject experiment.
☆ GNIO: Gated Neural Inertial Odometry
Inertial navigation using low-cost MEMS sensors is plagued by rapid drift due to sensor noise and bias instability. While recent data-driven approaches have made significant strides, they often struggle with micro-drifts during stationarity and mode fusion during complex motion transitions due to their reliance on fixed-window regression. In this work, we introduce Gated Neural Inertial Odometry (GNIO), a novel learning-based framework that explicitly models motion validity and context. We propose two key architectural innovations: \ding{182} a learnable Motion Bank that queries a global dictionary of motion patterns to provide semantic context beyond the local receptive field, and \ding{183} a Gated Prediction Head that decomposes displacement into magnitude and direction. This gating mechanism acts as a soft, differentiable Zero-Velocity Update (ZUPT), dynamically suppressing sensor noise during stationary periods while scaling predictions during dynamic motion. Extensive experiments across four public benchmarks demonstrate that GNIO significantly reduces position drift compared to state-of-the-art CNN and Transformer-based baselines. Notably, GNIO achieves a $60.21\%$ reduction in trajectory error on the OxIOD dataset and exhibits superior generalization in challenging scenarios involving frequent stops and irregular motion speeds.
comment: Submitted to IEEE Robotics and Automation Letters
☆ Encirclement Guaranteed Finite-Time Capture against Unknown Evader Strategies
We consider a pursuit-evasion scenario involving a group of pursuers and a single evader in a two-dimensional unbounded environment. The pursuers aim to capture the evader in finite time while ensuring the evader remains enclosed within the convex hull of their positions until capture, without knowledge of the evader's heading angle. Prior works have addressed the problem of encirclement and capture separately in different contexts. In this paper, we present a class of strategies for the pursuers that guarantee capture in finite time while maintaining encirclement, irrespective of the evader's strategy. Furthermore, we derive an upper bound on the time to capture. Numerical results highlight the effectiveness of the proposed framework against a range of evader strategies.
☆ MoE-ACT: Scaling Multi-Task Bimanual Manipulation with Sparse Language-Conditioned Mixture-of-Experts Transformers
The ability of robots to handle multiple tasks under a unified policy is critical for deploying embodied intelligence in real-world household and industrial applications. However, out-of-distribution variation across tasks often causes severe task interference and negative transfer when training general robotic policies. To address this challenge, we propose a lightweight multi-task imitation learning framework for bimanual manipulation, termed Mixture-of-Experts-Enhanced Action Chunking Transformer (MoE-ACT), which integrates sparse Mixture-of-Experts (MoE) modules into the Transformer encoder of ACT. The MoE layer decomposes a unified task policy into independently invoked expert components. Through adaptive activation, it naturally decouples multi-task action distributions in latent space. During decoding, Feature-wise Linear Modulation (FiLM) dynamically modulates action tokens to improve consistency between action generation and task instructions. In parallel, multi-scale cross-attention enables the policy to simultaneously focus on both low-level and high-level semantic features, providing rich visual information for robotic manipulation. We further incorporate textual information, transitioning the framework from a purely vision-based model to a vision-centric, language-conditioned action generation system. Experimental validation in both simulation and a real-world dual-arm setup shows that MoE-ACT substantially improves multi-task performance. Specifically, MoE-ACT outperforms vanilla ACT by an average of 33% in success rate. These results indicate that MoE-ACT provides stronger robustness and generalization in complex multi-task bimanual manipulation environments. Our open-source project page can be found at https://j3k7.github.io/MoE-ACT/.
☆ HapticVLA: Contact-Rich Manipulation via Vision-Language-Action Model without Inference-Time Tactile Sensing
Tactile sensing is a crucial capability for Vision-Language-Action (VLA) architectures, as it enables dexterous and safe manipulation in contact-rich tasks. However, reliance on dedicated tactile hardware increases cost and reduces reproducibility across robotic platforms. We argue that tactile-aware manipulation can be learned offline and deployed without direct haptic feedback at inference. To this end, we present HapticVLA, which proceeds in two tightly coupled stages: Safety-Aware Reward-Weighted Flow Matching (SA-RWFM) and Tactile Distillation (TD). SA-RWFM trains a flow-matching action expert that incorporates precomputed, safety-aware tactile rewards penalizing excessive grasping force and suboptimal grasping trajectories. TD further transfers this tactile-aware capability into a conventional VLA: we distill a compact tactile token from the SA-RWFM teacher and train a student VLA to predict that token from vision and state modalities, enabling tactile-aware action generation at inference without requiring on-board tactile sensors. This design preserves contact-rich tactile-aware reasoning within VLA while removing the need for on-board tactile sensors during deployment. On real-world experiments, HapticVLA achieves a mean success rate of 86.7%, consistently outperforming baseline VLAs - including versions provided with direct tactile feedback during inference.
☆ A Methodology for Dynamic Parameters Identification of 3-DOF Parallel Robots in Terms of Relevant Parameters
The identification of dynamic parameters in mechanical systems is important for improving model-based control as well as for performing realistic dynamic simulations. Generally, when identification techniques are applied only a subset of so-called base parameters can be identified. More even, some of these parameters cannot be identified properly given that they have a small contribution to the robot dynamics and hence in the presence of noise in measurements and discrepancy in modeling, their quality of being identifiable decreases. For this reason, a strategy for dynamic parameter identification of fully parallel robots in terms of a subset called relevant parameters is put forward. The objective of the proposed methodology is to start from a full dynamic model, then simplification concerning the geometry of each link and, the symmetry due to legs of fully parallel robots, are carried out. After that, the identification is done by Weighted Least Squares. Then, with statistical considerations the model is reduced until the physical feasibility conditions are met. The application of the propose strategy has been experimentally tested on two difierent configurations of actual 3-DOF parallel robots. The response of the inverse and forward dynamics of the identified models agrees with experiments. In order to evaluate the forward dynamics response, an approach for obtaining the forward dynamics in terms of the relevant parameters is also proposed.
☆ Coupled Particle Filters for Robust Affordance Estimation ICRA
Robotic affordance estimation is challenging due to visual, geometric, and semantic ambiguities in sensory input. We propose a method that disambiguates these signals using two coupled recursive estimators for sub-aspects of affordances: graspable and movable regions. Each estimator encodes property-specific regularities to reduce uncertainty, while their coupling enables bidirectional information exchange that focuses attention on regions where both agree, i.e., affordances. Evaluated on a real-world dataset, our method outperforms three recent affordance estimators (Where2Act, Hands-as-Probes, and HRP) by 308%, 245%, and 257% in precision, and remains robust under challenging conditions such as low light or cluttered environments. Furthermore, our method achieves a 70% success rate in our real-world evaluation. These results demonstrate that coupling complementary estimators yields precise, robust, and embodiment-appropriate affordance predictions.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
☆ NavGSim: High-Fidelity Gaussian Splatting Simulator for Large-Scale Navigation
Simulating realistic environments for robots is widely recognized as a critical challenge in robot learning, particularly in terms of rendering and physical simulation. This challenge becomes even more pronounced in navigation tasks, where trajectories often extend across multiple rooms or entire floors. In this work, we present NavGSim, a Gaussian Splatting-based simulator designed to generate high-fidelity, large-scale navigation environments. Built upon a hierarchical 3D Gaussian Splatting framework, NavGSim enables photorealistic rendering in expansive scenes spanning hundreds of square meters. To simulate navigation collisions, we introduce a Gaussian Splatting-based slice technique that directly extracts navigable areas from reconstructed Gaussians. Additionally, for ease of use, we provide comprehensive NavGSim APIs supporting multi-GPU development, including tools for custom scene reconstruction, robot configuration, policy training, and evaluation. To evaluate NavGSim's effectiveness, we train a Vision-Language-Action (VLA) model using trajectories collected from NavGSim and assess its performance in both simulated and real-world environments. Our results demonstrate that NavGSim significantly enhances the VLA model's scene understanding, enabling the policy to handle diverse navigation queries effectively.
☆ What Matters for Scalable and Robust Learning in End-to-End Driving Planners? CVPR
End-to-end autonomous driving has gained significant attention for its potential to learn robust behavior in interactive scenarios and scale with data. Popular architectures often build on separate modules for perception and planning connected through latent representations, such as bird's eye view feature grids, to maintain end-to-end differentiability. This paradigm emerged mostly on open-loop datasets, with evaluation focusing not only on driving performance, but also intermediate perception tasks. Unfortunately, architectural advances that excel in open-loop often fail to translate to scalable learning of robust closed-loop driving. In this paper, we systematically re-examine the impact of common architectural patterns on closed-loop performance: (1) high-resolution perceptual representations, (2) disentangled trajectory representations, and (3) generative planning. Crucially, our analysis evaluates the combined impact of these patterns, revealing both unexpected limitations as well as underexplored synergies. Building on these insights, we introduce BevAD, a novel lightweight and highly scalable end-to-end driving architecture. BevAD achieves 72.7% success rate on the Bench2Drive benchmark and demonstrates strong data-scaling behavior using pure imitation learning. Our code and models are publicly available here: https://dmholtz.github.io/bevad/
comment: To be published in CVPR Findings 2026
☆ KiRAS: Keyframe Guided Self-Imitation for Robust and Adaptive Skill Learning in Quadruped Robots ICRA
With advances in reinforcement learning and imitation learning, quadruped robots can acquire diverse skills within a single policy by imitating multiple skill-specific datasets. However, the lack of datasets on complex terrains limits the ability of such multi-skill policies to generalize effectively in unstructured environments. Inspired by animation, we adopt keyframes as minimal and universal skill representations, relaxing dataset constraints and enabling the integration of terrain adaptability with skill diversity. We propose Keyframe Guided Self-Imitation for Robust and Adaptive Skill Learning (KiRAS), an end-to-end framework for acquiring and transitioning between diverse skill primitives on complex terrains. KiRAS first learns diverse skills on flat terrain through keyframe-guided self-imitation, eliminating the need for expert datasets; then continues training the same policy network on rough terrains to enhance robustness. To eliminate catastrophic forgetting, a proficiency-based Skill Initialization Technique is introduced. Experiments on Solo-8 and Unitree Go1 robots show that KiRAS enables robust skill acquisition and smooth transitions across challenging terrains. This framework demonstrates its potential as a lightweight platform for multi-skill generation and dataset collection. It further enables flexible skill transitions that enhance locomotion on challenging terrains.
comment: Received by 2026 IEEE International Conference on Robotics and Automation (ICRA)
☆ ForceVLA2: Unleashing Hybrid Force-Position Control with Force Awareness for Contact-Rich Manipulation CVPR 2026
Embodied intelligence for contact-rich manipulation has predominantly relied on position control, while explicit awareness and regulation of interaction forces remain under-explored, limiting stability, precision, and robustness in real-world tasks. We propose ForceVLA2, an end-to-end vision-language-action framework that equips robots with hybrid force-position control and explicit force awareness. ForceVLA2 introduces force-based prompts into the VLM expert to construct force-aware task concepts across stages, and employs a Cross-Scale Mixture-of-Experts (MoE) in the action expert to adaptively fuse these concepts with real-time interaction forces for closed-loop hybrid force-position regulation. To support learning and evaluation, we construct ForceVLA2-Dataset, containing 1,000 trajectories over 5 contact-rich tasks, including wiping, pressing, and assembling, with multi-view images, task prompts, proprioceptive state, and force signals. Extensive experiments show that ForceVLA2 substantially improves success rates and reliability in contact-rich manipulation, outperforming pi0 and pi0.5 by 48.0% and 35.0%, respectively, across the 5 tasks, and mitigating common failure modes such as arm overload and unstable contact, thereby actively advancing force-aware interactive physical intelligence in VLAs. The project page is available at https://sites.google.com/view/force-vla2/home.
comment: Accepted by CVPR 2026
☆ Master Micro Residual Correction with Adaptive Tactile Fusion and Force-Mixed Control for Contact-Rich Manipulation
Robotic contact-rich and fine-grained manipulation remains a significant challenge due to complex interaction dynamics and the competing requirements of multi-timescale control. While current visual imitation learning methods excel at long-horizon planning, they often fail to perceive critical interaction cues like friction variations or incipient slip, and struggle to balance global task coherence with local reactive feedback. To address these challenges, we propose M2-ResiPolicy, a novel Master-Micro residual control architecture that synergizes high-level action guidance with low-level correction. The framework consists of a Master-Guidance Policy (MGP) operating at 10 Hz, which generates temporally consistent action chunks via a diffusion-based backbone and employs a tactile-intensity-driven adaptive fusion mechanism to dynamically modulate perceptual weights between vision and touch. Simultaneously, a high-frequency (60 Hz) Micro-Residual Corrector (MRC) utilizes a lightweight GRU to provide real-time action compensation based on TCP wrench feedback. This policy is further integrated with a force-mixed PBIC execution layer, effectively regulating contact forces to ensure interaction safety. Experiments across several demanding tasks including fragile object grasping and precision insertion, demonstrate that M2-ResiPolicy significantly outperforms standard Diffusion Policy (DP) and state-of-the-art Reactive Diffusion Policy (RDP), achieving a 93\% damage-free success rate in chip grasping and superior force regulation stability.
☆ Confusion-Aware In-Context-Learning for Vision-Language Models in Robotic Manipulation SC
Vision-language models (VLMs) have significantly improved the generalization capabilities of robotic manipulation. However, VLM-based systems often suffer from a lack of robustness, leading to unpredictable errors, particularly in scenarios involving confusable objects. Our preliminary analysis reveals that these failures are mainly caused by shortcut learning problem inherently in VLMs, limiting their ability to accurately distinguish between confusable features. To this end, we propose Confusion-Aware In-Context Learning (CAICL), a method that enhances VLM performance in confusable scenarios for robotic manipulation. The approach begins with confusion localization and analysis, identifying potential sources of confusion. This information is then used as a prompt for the VLM to focus on features most likely to cause misidentification. Extensive experiments on the VIMA-Bench show that CAICL effectively addresses the shortcut learning issue, achieving a 85.5\% success rate and showing good stability across tasks with different degrees of generalization.
comment: Accepted by the 29th International Conference on Computer Supported Cooperative Work in Design (CSCWD 2026)
☆ A Novel Camera-to-Robot Calibration Method for Vision-Based Floor Measurements SP
A novel hand-eye calibration method for ground-observing mobile robots is proposed. While cameras on mobile robots are com- mon, they are rarely used for ground-observing measurement tasks. Laser trackers are increasingly used in robotics for precise localization. A referencing plate is designed to combine the two measurement modalities of laser-tracker 3D metrology and camera- based 2D imaging. It incorporates reflector nests for pose acquisition using a laser tracker and a camera calibration target that is observed by the robot-mounted camera. The procedure comprises estimating the plate pose, the plate-camera pose, and the robot pose, followed by computing the robot-camera transformation. Experiments indicate sub-millimeter repeatability.
comment: 8 pages; accepted for publication in the ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
☆ BodyGuards: Escorting by Multiple Robots in Unknown Environment under Limited Communication ICRA 2026
Multi-robot systems are increasingly deployed in high-risk missions such as reconnaissance, disaster response, and subterranean operations. Protecting a human operator while navigating unknown and adversarial environments remains a critical challenge, especially when the communication among the operator and robots is restricted. Unlike existing collaborative exploration methods that aim for complete coverage, this work focuses on task-oriented exploration to minimize the navigation time of the operator to reach its goal while ensuring safety under adversarial threats. A novel escorting framework BodyGuards, is proposed to explicitly integrate seamlessly collaborative exploration, inter-robot-operator communication and escorting. The framework consists of three core components: (I) a dynamic movement strategy for the operator that maintains a local map with risk zones for proactive path planning; (II) a dual-mode robotic strategy combining frontier based exploration with optimized return events to balance exploration, threat detection, and intermittent communication; and (III) multi-robot coordination protocols that jointly plan exploration and information sharing for efficient escorting. Extensive human-in-the-loop simulations and hardware experiments demonstrate that the method significantly reduces operator risk and mission time, outperforming baselines in adversarial and constrained environments.
comment: Accept by ICRA 2026
☆ AeroGrab: A Unified Framework for Aerial Grasping in Cluttered Environments
Reliable aerial grasping in cluttered environments remains challenging due to occlusions and collision risks. Existing aerial manipulation pipelines largely rely on centroid-based grasping and lack integration between the grasp pose generation models, active exploration, and language-level task specification, resulting in the absence of a complete end-to-end system. In this work, we present an integrated pipeline for reliable aerial grasping in cluttered environments. Given a scene and a language instruction, the system identifies the target object and actively explores it to gain better views of the object. During exploration, a grasp generation network predicts multiple 6-DoF grasp candidates for each view. Each candidate is evaluated using a collision-aware feasibility framework, and the overall best grasp is selected and executed using standard trajectory generation and control methods. Experiments in cluttered real-world scenarios demonstrate robust and reliable grasp execution, highlighting the effectiveness of combining active perception with feasibility-aware grasp selection for aerial manipulation.
☆ HALO:Closing Sim-to-Real Gap for Heavy-loaded Humanoid Agile Motion Skills via Differentiable Simulation
Humanoid robots deployed in real-world scenarios often need to carry unknown payloads, which introduce significant mismatch and degrade the effectiveness of simulation-to-reality reinforcement learning methods. To address this challenge, we propose a two-stage gradient-based system identification framework built on the differentiable simulator MuJoCo XLA. The first stage calibrates the nominal robot model using real-world data to reduce intrinsic sim-to-real discrepancies, while the second stage further identifies the mass distribution of the unknown payload. By explicitly reducing structured model bias prior to policy training, our approach enables zero-shot transfer of reinforcement learning policies to hardware under heavy-load conditions. Extensive simulation and real-world experiments demonstrate more precise parameter identification, improved motion tracking accuracy, and substantially enhanced agility and robustness compared to existing baselines. Project Page: https://mwondering.github.io/halo-humanoid/
comment: 9 pages, 5 figures, conference
☆ Multi-Mode Pneumatic Artificial Muscles Driven by Hybrid Positive-Negative Pressure
Artificial muscles embody human aspirations for engineering lifelike robotic movements. This paper introduces an architecture for Inflatable Fluid-Driven Origami-Inspired Artificial Muscles (IN-FOAMs). A typical IN-FOAM consists of an inflatable skeleton enclosed within an outer skin, which can be driven using a combination of positive and negative pressures (e.g., compressed air and vacuum). IN-FOAMs are manufactured using low-cost heat-sealable sheet materials through heat-pressing and heat-sealing processes. Thus, they can be ultra-thin when not actuated, making them flexible, lightweight, and portable. The skeleton patterns are programmable, enabling a variety of motions, including contracting, bending, twisting, and rotating, based on specific skeleton designs. We conducted comprehensive experimental, theoretical, and numerical studies to investigate IN-FOAM's basic mechanical behavior and properties. The results show that IN-FOAM's output force and contraction can be tuned through multiple operation modes with the applied hybrid positive-negative pressure. Additionally, we propose multilayer skeleton structures to enhance the contraction ratio further, and we demonstrate a multi-channel skeleton approach that allows the integration of multiple motion modes into a single IN-FOAM. These findings indicate that IN-FOAMs hold great potential for future applications in flexible wearable devices and compact soft robotic systems.
comment: 20 pages, 17 figures. Published in IEEE Transactions on Robotics
☆ AnoleVLA: Lightweight Vision-Language-Action Model with Deep State Space Models for Mobile Manipulation
In this study, we address the problem of language-guided robotic manipulation, where a robot is required to manipulate a wide range of objects based on visual observations and natural language instructions. This task is essential for service robots that operate in human environments, and requires safety, efficiency, and task-level generality. Although Vision-Language-Action models (VLAs) have demonstrated strong performance for this task, their deployment in resource-constrained environments remains challenging because of the computational cost of standard transformer backbones. To overcome this limitation, we propose AnoleVLA, a lightweight VLA that uses a deep state space model to process multimodal sequences efficiently. The model leverages its lightweight and fast sequential state modeling to process visual and textual inputs, which allows the robot to generate trajectories efficiently. We evaluated the proposed method in both simulation and physical experiments. Notably, in real-world evaluations, AnoleVLA outperformed a representative large-scale VLA by 21 points for the task success rate while achieving an inference speed approximately three times faster.
☆ CycleRL: Sim-to-Real Deep Reinforcement Learning for Robust Autonomous Bicycle Control
Autonomous bicycles offer a promising agile solution for urban mobility and last-mile logistics, however, conventional control strategies often struggle with their underactuated nonlinear dynamics, suffering from sensitivity to model mismatches and limited adaptability to real-world uncertainties. To address this, this paper presents CycleRL, the first sim-to-real deep reinforcement learning framework designed for robust autonomous bicycle control. Our approach trains an end-to-end neural control policy within the high-fidelity NVIDIA Isaac Sim environment, leveraging Proximal Policy Optimization (PPO) to circumvent the need for an explicit dynamics model. The framework features a composite reward function tailored for concurrent balance maintenance, velocity tracking, and steering control. Crucially, systematic domain randomization is employed to bridge the simulation-to-reality gap and facilitate direct transfer. In simulation, CycleRL achieves considerable performance, including a 99.90% balance success rate, a low steering tracking error of 1.15°, and a velocity tracking error of 0.18 m/s. These quantitative results, coupled with successful hardware transfer, validate DRL as an effective paradigm for autonomous bicycle control, offering superior adaptability over traditional methods. Video demonstrations are available at https://anony6f05.github.io/CycleRL/.
comment: 10 pages, 7 figures, 9 tables
☆ Thermal Image Refinement with Depth Estimation using Recurrent Networks for Monocular ORB-SLAM3
Autonomous navigation in GPS-denied and visually degraded environments remains challenging for unmanned aerial vehicles (UAVs). To this end, we investigate the use of a monocular thermal camera as a standalone sensor on a UAV platform for real-time depth estimation and simultaneous localization and mapping (SLAM). To extract depth information from thermal images, we propose a novel pipeline employing a lightweight supervised network with recurrent blocks (RBs) integrated to capture temporal dependencies, enabling more robust predictions. The network combines lightweight convolutional backbones with a thermal refinement network (T-RefNet) to refine raw thermal inputs and enhance feature visibility. The refined thermal images and predicted depth maps are integrated into ORB-SLAM3, enabling thermal-only localization. Unlike previous methods, the network is trained on a custom non-radiometric dataset, obviating the need for high-cost radiometric thermal cameras. Experimental results on datasets and UAV flights demonstrate competitive depth accuracy and robust SLAM performance under low-light conditions. On the radiometric VIVID++ (indoor-dark) dataset, our method achieves an absolute relative error of approximately 0.06, compared to baselines exceeding 0.11. In our non-radiometric indoor set, baseline errors remain above 0.24, whereas our approach remains below 0.10. Thermal-only ORB-SLAM3 maintains a mean trajectory error under 0.4 m.
comment: 8 pages, 8 figures, 2 table
☆ ReMAP-DP: Reprojected Multi-view Aligned PointMaps for Diffusion Policy
Generalist robot policies built upon 2D visual representations excel at semantic reasoning but inherently lack the explicit 3D spatial awareness required for high-precision tasks. Existing 3D integration methods struggle to bridge this gap due to the structural irregularity of sparse point clouds and the geometric distortion introduced by multi-view orthographic rendering. To overcome these barriers, we present ReMAP-DP, a novel framework synergizing standardized perspective reprojection with a structure-aware dual-stream diffusion policy. By coupling the re-projected views with pixel-aligned PointMaps, our dual-stream architecture leverages learnable modality embeddings to fuse frozen semantic features and explicit geometric descriptors, ensuring precise implicit patch-level alignment. Extensive experiments across simulation and real-world environments demonstrate ReMAP-DP's superior performance in diverse manipulation tasks. On RoboTwin 2.0, it attains a 59.3% average success rate, outperforming the DP3 baseline by +6.6%. On ManiSkill 3, our method yields a 28% improvement over DP3 on the geometrically challenging Stack Cube task. Furthermore, ReMAP-DP exhibits remarkable real-world robustness, executing high-precision and dynamic manipulations with superior data efficiency from only a handful of demonstrations. Project page is available at: https://icr-lab.github.io/ReMAP-DP/
☆ Voronoi-based Second-order Descriptor with Whitened Metric in LiDAR Place Recognition ICRA 26
The pooling layer plays a vital role in aggregating local descriptors into the metrizable global descriptor in the LiDAR Place Recognition (LPR). In particular, the second-order pooling is capable of capturing higher-order interactions among local descriptors. However, its existing methods in the LPR adhere to conventional implementations and post-normalization, and incur the descriptor unsuitable for Euclidean distancing. Based on the recent interpretation that associates NetVLAD with the second-order statistics, we propose to integrate second-order pooling with the inductive bias from Voronoi cells. Our novel pooling method aggregates local descriptors to form the second-order matrix and whitens the global descriptor to implicitly measure the Mahalanobis distance while conserving the cluster property from Voronoi cells, addressing its numerical instability during learning with diverse techniques. We demonstrate its performance gains through the experiments conducted on the Oxford Robotcar and Wild-Places benchmarks and analyze the numerical effect of the proposed whitening algorithm.
comment: Accepted at ICRA 26
☆ Learning from Mistakes: Post-Training for Driving VLA with Takeover Data
Current Vision-Language-Action (VLA) paradigms in end-to-end autonomous driving rely on offline training from static datasets, leaving them vulnerable to distribution shift. Recent post-training methods use takeover data to mitigate this by augmenting the dataset with high-quality expert takeover samples, yet they suffer from two key limitations: supervision restricted to the period after the takeover moments leads to policies with limited safety margins, and passive preference optimization lacks active exploration for optimal performance. In this paper, we propose TakeVLA, a novel VLA post-training framework that overcomes these shortcomings through two complementary innovations. First, we introduce pre-takeover language supervision, which allows the VLA to learn from mistakes proactively. By explicitly teaching the model about what to do in error-prone situations, we cultivate a precautionary mindset that anticipates hazards early and substantially enlarges safety margins. Second, we propose Scenario Dreaming, a reinforcement fine-tuning paradigm that operates in reconstruceted takeover scenarios, encouraging active exploration beyond mere preference fitting. Experiments on the Bench2Drive benchmark demonstrate that TakeVLA achieves state-of-the-art closed-loop performance, surpassing the strong VLA baseline SimLingo by 4.93 in driving score, with an enhanced safety margin as evidenced by an 11.76% increase in average TTC.
☆ Intelligent Control of Differential Drive Robots Subject to Unmodeled Dynamics with EKF-based State Estimation
Reliable control and state estimation of differential drive robots (DDR) operating in dynamic and uncertain environments remains a challenge, particularly when system dynamics are partially unknown and sensor measurements are prone to degradation. This work introduces a unified control and state estimation framework that combines a Lyapunov-based nonlinear controller and Adaptive Neural Networks (ANN) with Extended Kalman Filter (EKF)-based multi-sensor fusion. The proposed controller leverages the universal approximation property of neural networks to model unknown nonlinearities in real time. An online adaptation scheme updates the weights of the radial basis function (RBF), the architecture chosen for the ANN. The learned dynamics are integrated into a feedback linearization (FBL) control law, for which theoretical guarantees of closed-loop stability and asymptotic convergence in a trajectory-tracking task are established through a Lyapunov-like stability analysis. To ensure robust state estimation, the EKF fuses inertial measurement unit (IMU) and odometry from monocular, 2D-LiDAR and wheel encoders. The fused state estimate drives the intelligent controller, ensuring consistent performance even under drift, wheel slip, sensor noise and failure. Gazebo simulations and real-world experiments are done using DDR, demonstrating the effectiveness of the approach in terms of improved velocity tracking performance with reduction in linear and angular velocity errors up to $53.91\%$ and $29.0\%$ in comparison to the baseline FBL.
Transformers As Generalizable Optimal Controllers
We study whether optimal state-feedback laws for a family of heterogeneous Multiple-Input, Multiple-Output (MIMO) Linear Time-Invariant (LTI) systems can be captured by a single learned controller. We train one transformer policy on LQR-generated trajectories from systems with different state and input dimensions, using a shared representation with standardization, padding, dimension encoding, and masked loss. The policy maps recent state history to control actions without requiring plant matrices at inference time. Across a broad set of systems, it achieves empirically small sub-optimality relative to Linear Quadratic Regulator (LQR), remains stabilizing under moderate parameter perturbations, and benefits from lightweight fine-tuning on unseen systems. These results support transformer policies as practical approximators of near-optimal feedback laws over structured linear-system families.
comment: 6 pages
☆ PerlAD: Towards Enhanced Closed-loop End-to-end Autonomous Driving with Pseudo-simulation-based Reinforcement Learning
End-to-end autonomous driving policies based on Imitation Learning (IL) often struggle in closed-loop execution due to the misalignment between inadequate open-loop training objectives and real driving requirements. While Reinforcement Learning (RL) offers a solution by directly optimizing driving goals via reward signals, the rendering-based training environments introduce the rendering gap and are inefficient due to high computational costs. To overcome these challenges, we present a novel Pseudo-simulation-based RL method for closed-loop end-to-end autonomous driving, PerlAD. Based on offline datasets, PerlAD constructs a pseudo-simulation that operates in vector space, enabling efficient, rendering-free trial-and-error training. To bridge the gap between static datasets and dynamic closed-loop environments, PerlAD introduces a prediction world model that generates reactive agent trajectories conditioned on the ego vehicle's plan. Furthermore, to facilitate efficient planning, PerlAD utilizes a hierarchical decoupled planner that combines IL for lateral path generation and RL for longitudinal speed optimization. Comprehensive experimental results demonstrate that PerlAD achieves state-of-the-art performance on the Bench2Drive benchmark, surpassing the previous E2E RL method by 10.29% in Driving Score without requiring expensive online interactions. Additional evaluations on the DOS benchmark further confirm its reliability in handling safety-critical occlusion scenarios.
comment: Accepted by IEEE RA-L. Submitted: 2025.12.2; Revised: 2026.2.4; Accepeted: 2026.3.7
☆ From Folding Mechanics to Robotic Function: A Unified Modeling Framework for Compliant Origami
Origami inspired architectures offer a powerful route toward lightweight, reconfigurable, and programmable robotic systems. Yet, a unified mechanics framework capable of seamlessly bridging rigid folding, elastic deformation, and stability driven transitions in compliant origami remains lacking. Here, we introduce a geometry consistent modeling framework based on discrete differential geometry (DDG) that unifies panel elasticity and crease rotation within a single variational formulation. By embedding crease panel coupling directly into a mid edge geometric discretization, the framework naturally captures rigid folding limits, distributed bending, multistability, and nonlinear dynamic snap through within one mechanically consistent structure. This unified description enables programmable control of stability and deformation across rigid and compliant regimes, allowing origami structures to transition from static folding mechanisms to active robotic modules. An implicit dynamic formulation incorporating gravity, contact, friction, and magnetic actuation further supports strongly coupled multiphysics simulations. Through representative examples spanning single fold bifurcation, deployable Miura membranes, bistable Waterbomb modules, and Kresling based crawling robots, we demonstrate how geometry driven mechanics directly informs robotic functionality. This work establishes discrete differential geometry as a foundational design language for intelligent origami robotics, enabling predictive modeling, stability programming, and mechanics guided robotic actuation within a unified computational platform.
comment: 24 pages, 7 figures
☆ ViSA: Visited-State Augmentation for Generalized Goal-Space Contrastive Reinforcement Learning
Goal-Conditioned Reinforcement Learning (GCRL) is a framework for learning a policy that can reach arbitrarily given goals. In particular, Contrastive Reinforcement Learning (CRL) provides a framework for policy updates using an approximation of the value function estimated via contrastive learning, achieving higher sample efficiency compared to conventional methods. However, since CRL treats the visited state as a pseudo-goal during learning, it can accurately estimate the value function only for limited goals. To address this issue, we propose a novel data augmentation approach for CRL called ViSA (Visited-State Augmentation). ViSA consists of two components: 1) generating augmented state samples, with the aim of augmenting hard-to-visit state samples during on-policy exploration, and 2) learning consistent embedding space, which uses an augmented state as auxiliary information to regularize the embedding space by reformulating the objective function of the embedding space based on mutual information. We evaluate ViSA in simulation and real-world robotic tasks and show improved goal-space generalization, which permits accurate value estimation for hard-to-visit goals. Further details can be found on the project page: \href{https://issa-n.github.io/projectPage_ViSA/}{\texttt{https://issa-n.github.io/projectPage\_ViSA/}}
comment: 8 pages, 7 figures, under Review
☆ Surgical Robot, Path Planning, Joint Space, Riemannian Manifolds
Robotic surgery for minimally invasive surgery can reduce the surgeon's workload by autonomously guiding robotic forceps. Movement of the robot is restricted around a fixed insertion port. The robot often encounters angle limitations during operation. Also, the surface of the abdominal cavity is non-concave, making it computationally expensive to find the desired path.In this work, to solve these problems, we propose a method for path planning in joint space by transforming the position into a Riemannian manifold. An edge cost function is defined to search for a desired path in the joint space and reduce the range of motion of the joints. We found that the organ is mostly non-concave, making it easy to find the optimal path using gradient descent method. Experimental results demonstrated that the proposed method reduces the range of joint angle movement compared to calculations in position space.
comment: 11 pages, 8 figures
☆ AutoMoT: A Unified Vision-Language-Action Model with Asynchronous Mixture-of-Transformers for End-to-End Autonomous Driving
Integrating vision-language models (VLMs) into end-to-end (E2E) autonomous driving (AD) systems has shown promise in improving scene understanding. However, existing integration strategies suffer from several limitations: they either struggle to resolve distribution misalignment between reasoning and action spaces, underexploit the general reasoning capabilities of pretrained VLMs, or incur substantial inference latency during action policy generation, which degrades driving performance. To address these challenges, we propose \OURS in this work, an end-to-end AD framework that unifies reasoning and action generation within a single vision-language-action (VLA) model. Our approach leverages a mixture-of-transformer (MoT) architecture with joint attention sharing, which preserves the general reasoning capabilities of pre-trained VLMs while enabling efficient fast-slow inference through asynchronous execution at different task frequencies. Extensive experiments on multiple benchmarks, under both open- and closed-loop settings, demonstrate that \OURS achieves competitive performance compared to state-of-the-art methods. We further investigate the functional boundary of pre-trained VLMs in AD, examining when AD-tailored fine-tuning is necessary. Our results show that pre-trained VLMs can achieve competitive multi-task scene understanding performance through semantic prompting alone, while fine-tuning remains essential for action-level tasks such as decision-making and trajectory planning. We refer to \href{https://automot-website.github.io/}{Project Page} for the demonstration videos and qualitative results.
☆ Ego to World: Collaborative Spatial Reasoning in Embodied Systems via Reinforcement Learning
Understanding the world from distributed, partial viewpoints is a fundamental challenge for embodied multi-agent systems. Each agent perceives the environment through an ego-centric view that is often limited by occlusion and ambiguity. To study this problem, we introduce the Ego-to-World (E2W) benchmark, which evaluates a vision-language model's ability to fuse heterogeneous viewpoints across three tasks: (i) global counting, (ii) relational location reasoning, and (iii) action-oriented grasping that requires predicting view-specific image coordinates. To address this setting, we propose CoRL, a two-stage framework that combines Chain-of-Thought supervised fine-tuning with reinforcement learning using Group-Relative Policy Optimization. Its core component, the Cross-View Spatial Reward (CVSR), provides dense task-aligned feedback by linking reasoning steps to visual evidence, ensuring coherent cross-view entity resolution, and guiding the model toward correct final predictions. Experiments on E2W show that CoRL consistently surpasses strong proprietary and open-source baselines on both reasoning and perception-grounding metrics, while ablations further confirm the necessity of each CVSR component. Beyond that, CoRL generalizes to external spatial reasoning benchmarks and enables effective real-world multi-robot manipulation with calibrated multi-camera rigs, demonstrating cross-view localization and successful grasp-and-place execution. Together, E2W and CoRL provide a principled foundation for learning world-centric scene understanding from distributed, ego-centric observations, advancing collaborative embodied AI.
☆ A Unified Calibration Framework for Coordinate and Kinematic Parameters in Dual-Arm Robots
Precise collaboration in vision-based dual-arm robot systems requires accurate system calibration. Recent dual-robot calibration methods have achieved strong performance by simultaneously solving multiple coordinate transformations. However, these methods either treat kinematic errors as implicit noise or handle them through separated error modeling, resulting in non-negligible accumulated errors. In this paper, we present a novel framework for unified calibration of the coordinate transformations and kinematic parameters in both robot arms. Our key idea is to unify all the tightly coupled parameters within a single Lie-algebraic formulation. To this end, we construct a consolidated error model grounded in the product-of-exponentials formula, which naturally integrates the coordinate and kinematic parameters in twist forms. Our model introduces no artificial error separation and thus greatly mitigates the error propagation. In addition, we derive a closed-form analytical Jacobian from this model using Lie derivatives. By exploring the Jacobian rank property, we analyze the identifiability of all calibration parameters and show that our joint optimization is well-posed under mild conditions. This enables off-the-shelf iterative solvers to stably optimize these parameters on the manifold space. Besides, to ensure robust convergence of our joint optimization, we develop a certifiably correct algorithm for initializing the unknown coordinates. Relying on semidefinite relaxation, our algorithm can yield a reliable estimate whose near-global optimality can be verified a posteriori. Extensive experiments validate the superior accuracy of our approach over previous baselines under identical visual measurements. Meanwhile, our certifiable initialization consistently outperforms several coordinate-only baselines, proving its reliability as a starting point for joint optimization.
comment: 21 pages, 12 figures
☆ HiMemVLN: Enhancing Reliability of Open-Source Zero-Shot Vision-and-Language Navigation with Hierarchical Memory System
LLM-based agents have demonstrated impressive zero-shot performance in vision-language navigation (VLN) tasks. However, most zero-shot methods primarily rely on closed-source LLMs as navigators, which face challenges related to high token costs and potential data leakage risks. Recent efforts have attempted to address this by using open-source LLMs combined with a spatiotemporal CoT framework, but they still fall far short compared to closed-source models. In this work, we identify a critical issue, Navigation Amnesia, through a detailed analysis of the navigation process. This issue leads to navigation failures and amplifies the gap between open-source and closed-source methods. To address this, we propose HiMemVLN, which incorporates a Hierarchical Memory System into a multimodal large model to enhance visual perception recall and long-term localization, mitigating the amnesia issue and improving the agent's navigation performance. Extensive experiments in both simulated and real-world environments demonstrate that HiMemVLN achieves nearly twice the performance of the open-source state-of-the-art method. The code is available at https://github.com/lvkailin0118/HiMemVLN.
comment: 9 pages, 7 figures
☆ Global Truncated Loss Minimization for Robust and Threshold-Resilient Geometric Estimation
To achieve outlier-robust geometric estimation, robust objective functions are generally employed to mitigate the influence of outliers. The widely used consensus maximization(CM) is highly robust when paired with global branch-and-bound(BnB) search. However, CM relies solely on inlier counts and is sensitive to the inlier threshold. Besides, the discrete nature of CM leads to loose bounds, necessitating extensive BnB iterations and computation cost. Truncated losses(TL), another continuous alternative, leverage residual information more effectively and could potentially overcome these issues. But to our knowledge, no prior work has systematically explored globally minimizing TL with BnB and its potential for enhanced threshold resilience or search efficiency. In this work, we propose GTM, the first unified BnB-based framework for globally-optimal TL loss minimization across diverse geometric problems. GTM involves a hybrid solving design: given an n-dimensional problem, it performs BnB search over an (n-1)-dimensional subspace while the remaining 1D variable is solved by bounding the objective function. Our hybrid design not only reduces the search space, but also enables us to derive Lipschitz-continuous bounding functions that are general, tight, and can be efficiently solved by a classic global Lipschitz solver named DIRECT, which brings further acceleration. We conduct a systematic evaluation on various BnB-based methods for CM and TL on the robust linear regression problem, showing that GTM enjoys remarkable threshold resilience and the highest efficiency compared to baseline methods. Furthermore, we apply GTM on different geometric estimation problems with diverse residual forms. Extensive experiments demonstrate that GTM achieves state-of-the-art outlier-robustness and threshold-resilience while maintaining high efficiency across these estimation tasks.
comment: 19 pages, 10 figures
☆ GraspALL: Adaptive Structural Compensation from Illumination Variation for Robotic Garment Grasping in Any Low-Light Conditions
Achieving accurate garment grasping under dynamically changing illumination is crucial for all-day operation of service robots.However, the reduced illumination in low-light scenes severely degrades garment structural features, leading to a significant drop in grasping robustness.Existing methods typically enhance RGB features by exploiting the illumination-invariant properties of non-RGB modalities, yet they overlook the varying dependence on non-RGB features under varying lighting conditions, which can introduce misaligned non-RGB cues and thereby weaken the model's adaptability to illumination changes when utilizing multimodal information.To address this problem, we propose GraspALL, an illumination-structure interactive compensation model.The innovation of GraspALL lies in encoding continuous illumination changes into quantitative references to guide adaptive feature fusion between RGB and non-RGB modalities according to varying lighting intensities, thereby generating illumination-consistent grasping representations.Experiments on the self-built garment grasping dataset demonstrate that GraspALL improves grasping accuracy by 32-44% over baselines under diverse illumination conditions.
☆ Exploring the dynamic properties and motion reproducibility of a small upper-body humanoid robot with 13-DOF pneumatic actuation for data-driven control
Pneumatically-actuated anthropomorphic robots with high degrees of freedom (DOF) offer significant potential for physical human-robot interaction. However, precise control of pneumatic actuators is challenging due to their inherent nonlinearities. This paper presents the development of a compact 13-DOF upper-body humanoid robot. To assess the feasibility of an effective controller, we first investigate its key dynamic properties, such as actuation time delays, and confirm that the system exhibits highly reproducible behavior. Leveraging this reproducibility, we implement a preliminary data-driven controller for a 4-DOF arm subsystem based on a multilayer perceptron with explicit time delay compensation. The network was trained on random movement data to generate pressure commands for tracking arbitrary trajectories. Comparative evaluations with a traditional PID controller demonstrate superior trajectory tracking performance, highlighting the potential of data-driven approaches for controlling complex, high-DOF pneumatic robots.
comment: 24 pages, 21 figures. Submitted to Advanced Robotics
☆ CORAL: COntextual Reasoning And Local Planning in A Hierarchical VLM Framework for Underwater Monitoring IROS 2026
Oyster reefs are critical ecosystem species that sustain biodiversity, filter water, and protect coastlines, yet they continue to decline globally. Restoring these ecosystems requires regular underwater monitoring to assess reef health, a task that remains costly, hazardous, and limited when performed by human divers. Autonomous underwater vehicles (AUVs) offer a promising alternative, but existing AUVs rely on geometry-based navigation that cannot interpret scene semantics. Recent vision-language models (VLMs) enable semantic reasoning for intelligent exploration, but existing VLM-driven systems adopt an end-to-end paradigm, introducing three key limitations. First, these systems require the VLM to generate every navigation decision, forcing frequent waits for inference. Second, VLMs cannot model robot dynamics, causing collisions in cluttered environments. Third, limited self-correction allows small deviations to accumulate into large path errors. To address these limitations, we propose CORAL, a framework that decouples high-level semantic reasoning from low-level reactive control. The VLM provides high-level exploration guidance by selecting waypoints, while a dynamics-based planner handles low-level collision-free execution. A geometric verification module validates waypoints and triggers replanning when needed. Compared with the previous state-of-the-art, CORAL improves coverage by 14.28% percentage points, or 17.85% relatively, reduces collisions by 100%, and requires 57% fewer VLM calls.
comment: Submitted to IROS 2026
☆ LiDAR-EVS: Enhance Extrapolated View Synthesis for 3D Gaussian Splatting with Pseudo-LiDAR Supervision
3D Gaussian Splatting (3DGS) has emerged as a powerful technique for real-time LiDAR and camera synthesis in autonomous driving simulation. However, simulating LiDAR with 3DGS remains challenging for extrapolated views beyond the training trajectory, as existing methods are typically trained on single-traversal sensor scans, suffer from severe overfitting and poor generalization to novel ego-vehicle paths. To enable reliable simulation of LiDAR along unseen driving trajectories without external multi-pass data, we present LiDAR-EVS, a lightweight framework for robust extrapolated-view LiDAR simulation in autonomous driving. Designed to be plug-and-play, LiDAR-EVS readily extends to diverse LiDAR sensors and neural rendering baselines with minimal modification. Our framework comprises two key components: (1) pseudo extrapolated-view point cloud supervision with multi-frame LiDAR fusion, view transformation, occlusion curling, and intensity adjustment; (2) spatially-constrained dropout regularization that promotes robustness to diverse trajectory variations encountered in real-world driving. Extensive experiments demonstrate that LiDAR-EVS achieves SOTA performance on extrapolated-view LiDAR synthesis across three datasets, making it a promising tool for data-driven simulation, closed-loop evaluation, and synthetic data generation in autonomous driving systems.
comment: 22 pages, 8 figures
☆ Efficient Event Camera Volume System ICRA 2026
Event cameras promise low latency and high dynamic range, yet their sparse output challenges integration into standard robotic pipelines. We introduce \nameframew (Efficient Event Camera Volume System), a novel framework that models event streams as continuous-time Dirac impulse trains, enabling artifact-free compression through direct transform evaluation at event timestamps. Our key innovation combines density-driven adaptive selection among DCT, DTFT, and DWT transforms with transform-specific coefficient pruning strategies tailored to each domain's sparsity characteristics. The framework eliminates temporal binning artifacts while automatically adapting compression strategies based on real-time event density analysis. On EHPT-XC and MVSEC datasets, our framework achieves superior reconstruction fidelity with DTFT delivering the lowest earth mover distance. In downstream segmentation tasks, EECVS demonstrates robust generalization. Notably, our approach demonstrates exceptional cross-dataset generalization: when evaluated with EventSAM segmentation, EECVS achieves mean IoU 0.87 on MVSEC versus 0.44 for voxel grids at 24 channels, while remaining competitive on EHPT-XC. Our ROS2 implementation provides real-time deployment with DCT processing achieving 1.5 ms latency and 2.7X higher throughput than alternative transforms, establishing the first adaptive event compression framework that maintains both computational efficiency and superior generalization across diverse robotic scenarios.
comment: Accepted to ICRA 2026
☆ A Dual Quaternion Framework for Collision Recovery of Quadrotor
Unmanned aerial vehicles (UAVs) operating in cluttered environments require accurate impact modeling to maintain stability. However, conventional contact models decouple linear and angular impulses, risking manifold inconsistency during rapid state transitions. This article presents a dual quaternion reset map that resolves rigid-body impacts directly on the SE(3) manifold. By operating on the unified spatial twist (linear and angular velocities as a single dual entity), our formulation is algebraically equivalent to the classical Newton impulse model while preserving manifold consistency during discrete state jumps. Building on this framework, we design a hybrid recovery controller that couples linear and angular momentum to ensure strict energy dissipation across impacts. Hardware-in-the-loop benchmarks demonstrate a 24% reduction in execution latency compared to an optimized matrix-based implementation. High-fidelity MuJoCo simulations validate the controller's robustness to complex contact dynamics, showing a 56.6% reduction in post-impact root-mean-square error (RMSE) and a 41.2% decrease in peak kinetic energy compared to decoupled recovery methods.
comment: 7 pages, 5 figures
☆ Embodied Foundation Models at the Edge: A Survey of Deployment Constraints and Mitigation Strategies
Deploying foundation models in embodied edge systems is fundamentally a systems problem, not just a problem of model compression. Real-time control must operate within strict size, weight, and power constraints, where memory traffic, compute latency, timing variability, and safety margins interact directly. The Deployment Gauntlet organizes these constraints into eight coupled barriers that determine whether embodied foundation models can run reliably in practice. Across representative edge workloads, autoregressive Vision-Language-Action policies are constrained primarily by memory bandwidth, whereas diffusion-based controllers are limited more by compute latency and sustained execution cost. Reliable deployment therefore depends on system-level co-design across memory, scheduling, communication, and model architecture, including decompositions that separate fast control from slower semantic reasoning.
♻ ☆ No More Blind Spots: Learning Vision-Based Omnidirectional Bipedal Locomotion for Challenging Terrain
Effective bipedal locomotion in dynamic environments, such as cluttered indoor spaces or uneven terrain, requires agile and adaptive movement in all directions. This necessitates omnidirectional terrain sensing and a controller capable of processing such input. We present a learning framework for vision-based omnidirectional bipedal locomotion, enabling seamless movement using depth images. A key challenge is the high computational cost of rendering omnidirectional depth images in simulation, making traditional sim-to-real reinforcement learning (RL) impractical. Our method combines a robust blind controller with a teacher policy that supervises a vision-based student policy, trained on noise-augmented terrain data to avoid rendering costs during RL and ensure robustness. We also introduce a data augmentation technique for supervised student training, accelerating training by up to 10 times compared to conventional methods. Our framework is validated through simulation and real-world tests, demonstrating effective omnidirectional locomotion with minimal reliance on expensive rendering. This is, to the best of our knowledge, the first demonstration of vision-based omnidirectional bipedal locomotion, showcasing its adaptability to diverse terrains.
World Models for Learning Dexterous Hand-Object Interactions from Human Videos
Modeling dexterous hand-object interactions is challenging as it requires understanding how subtle finger motions influence the environment through contact with objects. While recent world models address interaction modeling, they typically rely on coarse action spaces that fail to capture fine-grained dexterity. We, therefore, introduce DexWM, a Dexterous Interaction World Model that predicts future latent states of the environment conditioned on past states and dexterous actions. To overcome the scarcity of finely annotated dexterous datasets, DexWM represents actions using finger keypoints extracted from egocentric videos, enabling training on over 900 hours of human and non-dexterous robot data. Further, to accurately model dexterity, we find that predicting visual features alone is insufficient; therefore, we incorporate an auxiliary hand consistency loss that enforces accurate hand configurations. DexWM outperforms prior world models conditioned on text, navigation, or full-body actions in future-state prediction and demonstrates strong zero-shot transfer to unseen skills on a Franka Panda arm with an Allegro gripper, surpassing Diffusion Policy by over 50% on average across grasping, placing, and reaching tasks.
♻ ☆ Real-World Deployment of Cloud-based Autonomous Mobility Systems for Outdoor and Indoor Environments
Autonomous mobility systems increasingly operate in dense and dynamic environments where perception occlusions, limited sensing coverage, and multi-agent interactions pose major challenges. While onboard sensors provide essential local perception, they often struggle to maintain reliable situational awareness in crowded urban or indoor settings. This article presents the Cloud-based Autonomous Mobility (CAM) framework, a generalized architecture that integrates infrastructure-based intelligent sensing with cloud-level coordination to enhance autonomous operations. The system deploys distributed Intelligent Sensor Nodes (ISNs) equipped with cameras, LiDAR, and edge computing to perform multi-modal perception and transmit structured information to a cloud platform via high-speed wireless communication. The cloud aggregates observations from multiple nodes to generate a global scene representation for other autonomous modules, such as decision making, motion planning, etc. Real-world deployments in an urban roundabout and a hospital-like indoor environment demonstrate improved perception robustness, safety, and coordination for future intelligent mobility systems.
comment: This paper has been submitted to IEEE Robotics and Automation Magazine
♻ ☆ GeoFIK: A Fast and Reliable Geometric Solver for the IK of the Franka Arm based on Screw Theory Enabling Multiple Redundancy Parameters
Modern robotics applications require an inverse kinematics (IK) solver that is fast, robust and consistent, and that provides all possible solutions. Currently, the Franka robot arm is the most widely used manipulator in robotics research. With 7 DOFs, the IK of this robot is not only complex due to its 1-DOF redundancy, but also due to the link offsets at the wrist and elbow. Due to this complexity, none of the Franka IK solvers available in the literature provide satisfactory results when used in real-world applications. Therefore, in this paper we introduce GeoFIK (Geometric Franka IK), an analytical IK solver that allows the use of different joint variables to resolve the redundancy. The approach uses screw theory to describe the entire geometry of the robot, allowing the computation of the Jacobian matrix prior to computation of joint angles. All singularities are identified and handled. As an example of how the geometric elements obtained by the IK can be exploited, a solver with the swivel angle as the free variable is provided. Several experiments are carried out to validate the speed, robustness and reliability of the GeoFIK against two state-of-the-art solvers.
♻ ☆ Real-time Capable Learning-based Visual Tool Pose Correction via Differentiable Simulation
Autonomy in robot-assisted minimally invasive surgery has the potential to reduce surgeon cognitive and task load, thereby increasing procedural efficiency. However, implementing accurate autonomous control can be difficult due to poor end-effector proprioception. Joint encoder readings are typically inaccurate due to kinematic non-idealities in their cable-driven transmissions. Vision-based pose estimation approaches are highly effective, but lack real-time capability, generalizability, or can be hard to train. In this work, we demonstrate a real-time capable, Vision Transformer-based pose estimation approach that is trained using end-to-end differentiable kinematics and rendering. We demonstrate the potential of this approach to correct for noisy pose estimates through a real robot dataset and the potential real-time processing ability. Our approach is able to reduce more than 50% of hand-eye translation errors in the dataset, reaching the same performance level as an existing optimization-based method. Our approach is four times faster, and capable of near real-time inference at 22 Hz. A zero-shot prediction on an unseen dataset shows good generalization ability, and can be further finetuned for increased performance without human labeling.
♻ ☆ GoalSwarm: Multi-UAV Semantic Coordination for Open-Vocabulary Object Navigation
Cooperative visual semantic navigation is a foundational capability for aerial robot teams operating in unknown environments. However, achieving robust open-vocabulary object-goal navigation remains challenging due to the computational constraints of deploying heavy perception models onboard and the complexity of decentralized multi-agent coordination. We present GoalSwarm, a fully decentralized multi-UAV framework for zero-shot semantic object-goal navigation. Each UAV collaboratively constructs a shared, lightweight 2D top-down semantic occupancy map by projecting depth observations from aerial vantage points, eliminating the computational burden of full 3D representations while preserving essential geometric and semantic structure. The core contributions of GoalSwarm are threefold: (1) integration of zero-shot foundation model -- SAM3 for open vocabulary detection and pixel-level segmentation, enabling open-vocabulary target identification without task-specific training; (2) a Bayesian Value Map that fuses multi-viewpoint detection confidences into a per-pixel goal-relevance distribution, enabling informed frontier scoring via Upper Confidence Bound (UCB) exploration; and (3) a decentralized coordination strategy combining semantic frontier extraction, cost-utility bidding with geodesic path costs, and spatial separation penalties to minimize redundant exploration across the swarm.
comment: 6 pages, 2 figures
♻ ☆ On transferring safety certificates across dynamical systems
Control barrier functions (CBFs) provide a powerful tool for enforcing safety constraints in control systems, but their direct application to complex, high-dimensional dynamics is often challenging. In many settings, safety certificates are more naturally designed for simplified or alternative system models that do not exactly match the dynamics of interest. This paper addresses the problem of transferring safety guarantees between dynamical systems with mismatched dynamics. We propose a transferred control barrier function (tCBF) framework that enables safety constraints defined on one system to be systematically enforced on another system using a simulation function and an explicit margin term. The resulting transferred barrier accounts for model mismatch and induces a safety condition that can be enforced on the target system via a quadratic-program-based safety filter. The proposed approach is general and does not require the two systems to share the same state dimension or dynamics. We demonstrate the effectiveness of the framework on a quadrotor navigation task with the transferred barrier ensuring collision avoidance for the target system, while remaining minimally invasive to a nominal controller. These results highlight the potential of transferred control barrier functions as a general mechanism for enforcing safety across heterogeneous dynamical systems.
♻ ☆ Optimization-Based Robust Permissive Synthesis for Interval MDPs
We present an optimization-based framework for robust permissive synthesis for Interval Markov Decision Processes (IMDPs), motivated by robotic decision-making under transition uncertainty. In many robotic systems, model inaccuracies and sensing noise lead to interval-valued transition probabilities. While robust IMDP synthesis typically yields a single policy and permissive synthesis assumes exact models, we show that robust permissive synthesis under interval uncertainty can be cast as a global mixed-integer linear program (MILP) that directly encodes robust Bellman constraints. The formulation maximizes a quantitative permissiveness metric (the number of enabled state-action pairs), while guaranteeing that every compliant strategy satisfies probabilistic reachability or expected reward specifications under all admissible transition realizations. To address the exponential complexity of vertex-based uncertainty representations, we derive a dualization-based encoding that eliminates explicit vertex enumeration and scales linearly with the number of successors. Experimental evaluation on four representative robotic benchmark domains demonstrates scalability to IMDPs with hundreds of thousands of states. The proposed framework provides a practical and general foundation for uncertainty-aware, flexibility-preserving controller synthesis in robotic systems.
♻ ☆ PhysMoDPO: Physically-Plausible Humanoid Motion with Preference Optimization
Recent progress in text-conditioned human motion generation has been largely driven by diffusion models trained on large-scale human motion data. Building on this progress, recent methods attempt to transfer such models for character animation and real robot control by applying a Whole-Body Controller (WBC) that converts diffusion-generated motions into executable trajectories. While WBC trajectories become compliant with physics, they may expose substantial deviations from original motion. To address this issue, we here propose PhysMoDPO, a Direct Preference Optimization framework. Unlike prior work that relies on hand-crafted physics-aware heuristics such as foot-sliding penalties, we integrate WBC into our training pipeline and optimize diffusion model such that the output of WBC becomes compliant both with physics and original text instructions. To train PhysMoDPO we deploy physics-based and task-specific rewards and use them to assign preference to synthesized trajectories. Our extensive experiments on text-to-motion and spatial control tasks demonstrate consistent improvements of PhysMoDPO in both physical realism and task-related metrics on simulated robots. Moreover, we demonstrate that PhysMoDPO results in significant improvements when applied to zero-shot motion transfer in simulation and for real-world deployment on a G1 humanoid robot.
comment: Project page: https://mael-zys.github.io/PhysMoDPO/
♻ ☆ sim2art: Accurate Articulated Object Modeling from a Single Video using Synthetic Training Data Only
Understanding articulated objects from monocular video is a crucial yet challenging task in robotics and digital twin creation. Existing methods often rely on complex multi-view setups, high-fidelity object scans, or fragile long-term point tracks that frequently fail in casual real-world captures. In this paper, we present sim2art, a data-driven framework that recovers the 3D part segmentation and joint parameters of articulated objects from a single monocular video captured by a freely moving camera. Our core insight is a robust representation based on per-frame surface point sampling, which we augment with short-term scene flow and DINOv3 semantic features. Unlike previous works that depend on error-prone long-term correspondences, our representation is easy to obtain and exhibits a negligible difference between simulation and reality without requiring domain adaptation. Also, by construction, our method relies on single-viewpoint visibility, ensuring that the geometric representation remains consistent across synthetic and real data despite noise and occlusions. Leveraging a suitable Transformer-based architecture, sim2art is trained exclusively on synthetic data yet generalizes strongly to real-world sequences. To address the lack of standardized benchmarks in the field, we introduce two datasets featuring a significantly higher diversity of object categories and instances than prior work. Our evaluations show that sim2art effectively handles large camera motions and complex articulations, outperforming state-of-the-art optimization-based and tracking-dependent methods. sim2art offers a scalable solution that can be easily extended to new object categories without the need for cumbersome real-world annotations. Project webpage: https://aartykov.github.io/sim2art/
♻ ☆ Lightweight 3D LiDAR-Based UAV Tracking: An Adaptive Extended Kalman Filtering Approach
Accurate relative positioning is crucial for swarm aerial robotics, enabling coordinated flight and collision avoidance. Although vision-based tracking has been extensively studied, 3D LiDAR-based methods remain underutilized despite their robustness under varying lighting conditions. Existing systems often rely on bulky, power-intensive sensors, making them impractical for small UAVs with strict payload and energy constraints. This paper presents a lightweight LiDAR-based UAV tracking system incorporating an Adaptive Extended Kalman Filter (AEKF) framework. Our approach effectively addresses the challenges posed by sparse, noisy, and nonuniform point cloud data generated by non-repetitive scanning 3D LiDARs, ensuring reliable tracking while remaining suitable for small drones with strict payload constraints. Unlike conventional filtering techniques, the proposed method dynamically adjusts the noise covariance matrices using innovation and residual statistics, thereby enhancing tracking accuracy under real-world conditions. Additionally, a recovery mechanism ensures continuity of tracking during temporary detection failures caused by scattered LiDAR returns or occlusions. Experimental validation was performed using a Livox Mid-360 LiDAR mounted on a DJI F550 UAV in real-world flight scenarios. The proposed method demonstrated robust UAV tracking performance under sparse LiDAR returns and intermittent detections, consistently outperforming both standard Kalman filtering and particle filtering approaches during aggressive maneuvers. These results confirm that the framework enables reliable relative positioning in GPS-denied environments without the need for multi-sensor arrays or external infrastructure.
comment: Presented at the 19th International Conference on Intelligent Autonomous Systems, IAS-19, Genoa, Italy, June 30 to July 4, 2025. To appear in the Springer post-proceedings of the conference
♻ ☆ RAG-3DSG: Enhancing 3D Scene Graphs with Re-Shot Guided Retrieval-Augmented Generation
Open-vocabulary 3D Scene Graph (3DSG) can enhance various downstream tasks in robotics by leveraging structured semantic representations, yet current 3DSG construction methods suffer from semantic inconsistencies caused by noisy cross-image aggregation under occlusions and constrained viewpoints. To mitigate the impact of such inconsistency, we propose RAG-3DSG, which introduces re-shot guided uncertainty estimation. By measuring the semantic consistency between original limited viewpoints and re-shot optimal viewpoints, this method quantifies the underlying semantic ambiguity of each graph object. Based on this quantification, we devise an Object-level Retrieval-Augmented Generation (RAG) that leverages low-uncertainty objects as semantic anchors to retrieve more reliable contextual knowledge, enabling a Vision-Language Model to rectify the predictions of uncertain objects and optimize the final 3DSG. Extensive evaluations across three challenging benchmarks and real-world robot trials demonstrate that RAG-3DSG achieves superior recall and precision, effectively mitigating semantic noise to provide highly reliable scene representations for robotics tasks.
♻ ☆ Persistent Autoregressive Mapping with Traffic Rules for Autonomous Driving AAAI2026
Safe autonomous driving requires both accurate HD map construction and persistent awareness of traffic rules, even when their associated signs are no longer visible. However, existing methods either focus solely on geometric elements or treat rules as temporary classifications, failing to capture their persistent effectiveness across extended driving sequences. In this paper, we present PAMR (Persistent Autoregressive Mapping with Traffic Rules), a novel framework that performs autoregressive co-construction of lane vectors and traffic rules from visual observations. Our approach introduces two key mechanisms: Map-Rule Co-Construction for processing driving scenes in temporal segments, and Map-Rule Cache for maintaining rule consistency across these segments. To properly evaluate continuous and consistent map generation, we develop MapDRv2, featuring improved lane geometry annotations. Extensive experiments demonstrate that PAMR achieves superior performance in joint vector-rule mapping tasks, while maintaining persistent rule effectiveness throughout extended driving sequences.
comment: AAAI2026
♻ ☆ Barrier-Riccati Synthesis for Nonlinear Safe Control with Expanded Region of Attraction
We present a Riccati-based framework for safety-critical nonlinear control that integrates the barrier states (BaS) methodology with the State-Dependent Riccati Equation (SDRE) approach. The BaS formulation embeds safety constraints into the system dynamics via auxiliary states, enabling safety to be treated as a control objective. To overcome the limited region of attraction in linear BaS controllers, we extend the framework to nonlinear systems using SDRE synthesis applied to the barrier-augmented dynamics and derive a matrix inequality condition that certifies forward invariance of a large region of attraction and guarantees asymptotic safe stabilization. The resulting controller is computed online via pointwise Riccati solutions. We validate the method on an unstable constrained system and cluttered quadrotor navigation tasks, demonstrating improved constraint handling, scalability, and robustness near safety boundaries. This framework offers a principled and computationally tractable solution for synthesizing nonlinear safe feedback in safety-critical environments.
comment: This work has been accepted for publication in the proceedings of the 2026 American Control Conference (ACC), New Orleans, Louisiana, USA
♻ ☆ Open-World Motion Forecasting
Motion forecasting aims to predict the future trajectories of dynamic agents in the scene, enabling autonomous vehicles to effectively reason about scene evolution. Existing approaches operate under the closed-world regime and assume fixed object taxonomy as well as access to high-quality perception. Therefore, they struggle in real-world settings where perception is imperfect and object taxonomy evolves over time. In this work, we bridge this fundamental gap by introducing open-world motion forecasting, a novel setting in which new object classes are sequentially introduced over time and future object trajectories are estimated directly from camera images. We tackle this setting by proposing the first end-to-end class-incremental motion forecasting framework to mitigate catastrophic forgetting while simultaneously learning to forecast newly introduced classes. When a new class is introduced, our framework employs a pseudo-labeling strategy to first generate motion forecasting pseudo-labels for all known classes which are then processed by a vision-language model to filter inconsistent and over-confident predictions. Parallelly, our approach further mitigates catastrophic forgetting by using a novel replay sampling strategy that leverages query feature variance to sample previous sequences with informative motion patterns. Extensive evaluation on the nuScenes and Argoverse 2 datasets demonstrates that our approach successfully resists catastrophic forgetting and maintains performance on previously learned classes while improving adaptation to novel ones. Further, we demonstrate that our approach supports zero-shot transfer to real-world driving and naturally extends to end-to-end class-incremental planning, enabling continual adaptation of the full autonomous driving system. We provide the code at https://omen.cs.uni-freiburg.de.
comment: V2: Adapt author affiliation
♻ ☆ MoRoCo: An Online Topology-Adaptive Framework for Multi-Operator Multi-Robot Coordination under Restricted Communication
Fleets of autonomous robots are increasingly deployed with multiple human operators in communication-restricted environments for exploration and intervention tasks such as subterranean inspection, reconnaissance, and search-and-rescue. In these settings, communication is often limited to short-range ad-hoc links, making it difficult to coordinate exploration while supporting online human-fleet interactions. Existing work on multi-robot exploration largely focuses on information gathering itself, but pays limited attention to the fact that operators and robots issue time-critical requests during execution. These requests may require different communication structures, ranging from intermittent status delivery to sustained video streaming and teleoperation. To address this challenge, this paper presents MoRoCo, an online topology-adaptive framework for multi-operator multi-robot coordination under restricted communication. MoRoCo is built on a latency-bounded intermittent communication backbone that guarantees a prescribed delay for information collected by any robot to reach an operator, together with a detach-and-rejoin mechanism that enables online team resizing and topology reconfiguration. On top of this backbone, the framework instantiates request-consistent communication subgraphs to realize different modes of operator-robot interaction by jointly assigning robot roles, positions, and communication topology. It further supports the online decomposition and composition of these subgraphs using only local communication, allowing multiple requests to be serviced during exploration. The framework extends to heterogeneous fleets, multiple teams, and robot failures. Extensive human-in-the-loop simulations and hardware experiments demonstrate effective and reliable coordination under restricted communication.
comment: 20 pages, 19 figures. Submitted to IEEE Transactions on Robotics (TRO)
♻ ☆ Learning Dexterous Manipulation with Quantized Hand State ICRA 2026
Dexterous robotic hands enable robots to perform complex manipulations that require fine-grained control and adaptability. Achieving such manipulation is challenging because the high degrees of freedom tightly couple hand and arm motions, making learning and control difficult. Successful dexterous manipulation relies not only on precise hand motions, but also on accurate spatial positioning of the arm and coordinated arm-hand dynamics. However, most existing visuomotor policies represent arm and hand actions in a single combined space, which often causes high-dimensional hand actions to dominate the coupled action space and compromise arm control. To address this, we propose DQ-RISE, which quantizes hand states to simplify hand motion prediction while preserving essential patterns, and applies a continuous relaxation that allows arm actions to diffuse jointly with these compact hand states. This design enables the policy to learn arm-hand coordination from data while preventing hand actions from overwhelming the action space. Experiments show that DQ-RISE achieves more balanced and efficient learning, paving the way toward structured and generalizable dexterous manipulation. Project website: http://rise-policy.github.io/DQ-RISE/
comment: accepted by ICRA 2026
♻ ☆ History-Aware Visuomotor Policy Learning via Point Tracking ICRA 2026
Many manipulation tasks require memory beyond the current observation, yet most visuomotor policies rely on the Markov assumption and thus struggle with repeated states or long-horizon dependencies. Existing methods attempt to extend observation horizons but remain insufficient for diverse memory requirements. To this end, we propose an object-centric history representation based on point tracking, which abstracts past observations into a compact and structured form that retains only essential task-relevant information. Tracked points are encoded and aggregated at the object level, yielding a compact history representation that can be seamlessly integrated into various visuomotor policies. Our design provides full history-awareness with high computational efficiency, leading to improved overall task performance and decision accuracy. Through extensive evaluations on diverse manipulation tasks, we show that our method addresses multiple facets of memory requirements - such as task stage identification, spatial memorization, and action counting, as well as longer-term demands like continuous and pre-loaded memory - and consistently outperforms both Markovian baselines and prior history-based approaches. Project website: http://tonyfang.net/history
comment: accepted by ICRA 2026
♻ ☆ EMMA: Generalizing Real-World Robot Manipulation via Generative Visual Transfer
The generalization of vision-language-action (VLA) models heavily relies on diverse training data. However, acquiring large-scale data for robot manipulation across varied object appearances is costly and labor-intensive. To address this limitation, we introduce Embodied Manipulation Media Adaptation (EMMA), a framework for augmenting VLA policies that combines a generative data engine with an effective training pipeline. We introduce DreamTransfer, a diffusion Transformer-based architecture for generating multi-view consistent and geometrically grounded embodied manipulation videos. DreamTransfer enables visual editing of robot videos through prompts, allowing for changes to the foreground, background, and lighting while preserving their 3D structure and geometric validity. We also utilize a hybrid training set of real and generated data and propose AdaMix to enhance the training process. AdaMix is a training strategy that adaptively weights samples according to policy performance to emphasize challenging samples. Comprehensive evaluations demonstrate that videos created by DreamTransfer yield substantial improvements over previous video generation techniques in multi-view consistency, geometric accuracy, and text-conditioning precision. We conduct extensive evaluations with a total of more than 1800 trials in both simulated and real-world robotic environments. In real-world robotic tasks with zero-shot visual settings, our framework achieves a relative performance increase of over 92% compared to training with real data alone, and improves by an additional 17% with AdaMix, demonstrating its efficacy in enhancing policy generalization.
♻ ☆ RoboMD: Uncovering Robot Vulnerabilities through Semantic Potential Fields
Robot manipulation policies, while central to the promise of physical AI, are highly vulnerable in the presence of external variations in the real world. Diagnosing these vulnerabilities is hindered by two key challenges: (i) the relevant variations to test against are often unknown, and (ii) direct testing in the real world is costly and unsafe. We introduce a framework that tackles both issues by learning a separate deep reinforcement learning (deep RL) policy for vulnerability prediction through virtual runs on a continuous vision-language embedding trained with limited success-failure data. By treating this embedding space, which is rich in semantic and visual variations, as a potential field, the policy learns to move toward vulnerable regions while being repelled from success regions. This vulnerability prediction policy, trained on virtual rollouts, enables scalable and safe vulnerability analysis without expensive physical trials. By querying this policy, our framework builds a probabilistic vulnerability-likelihood map. Experiments across simulation benchmarks and a physical robot arm show that our framework uncovers up to 23% more unique vulnerabilities than state-of-the-art vision-language baselines, revealing subtle vulnerabilities overlooked by heuristic testing. Additionally, we show that fine-tuning the manipulation policy with the vulnerabilities discovered by our framework improves manipulation performance with much less fine-tuning data.
comment: 26 Pages, 20 figures
♻ ☆ MARVL: Multi-Stage Guidance for Robotic Manipulation via Vision-Language Models
Designing dense reward functions is pivotal for efficient robotic Reinforcement Learning (RL). However, most dense rewards rely on manual engineering, which fundamentally limits the scalability and automation of reinforcement learning. While Vision-Language Models (VLMs) offer a promising path to reward design, naive VLM rewards often misalign with task progress, struggle with spatial grounding, and show limited understanding of task semantics. To address these issues, we propose MARVL-Multi-stAge guidance for Robotic manipulation via Vision-Language models. MARVL fine-tunes a VLM for spatial and semantic consistency and decomposes tasks into multi-stage subtasks with task direction projection for trajectory sensitivity. Empirically, MARVL significantly outperforms existing VLM-reward methods on the Meta-World benchmark, demonstrating superior sample efficiency and robustness on sparse-reward manipulation tasks.
♻ ☆ TurboMap: GPU-Accelerated Local Mapping for Visual SLAM IROS 2026
In real-time Visual SLAM systems, local mapping must operate under strict latency constraints, as delays degrade map quality and increase the risk of tracking failure. GPU parallelization offers a promising way to reduce latency. However, parallelizing local mapping is challenging due to synchronized shared-state updates and the overhead of transferring large map data structures to the GPU. This paper presents TurboMap, a GPU-parallelized and CPU-optimized local mapping backend that holistically addresses these challenges. We restructure Map Point Creation to enable parallel Keypoint Correspondence Search on the GPU, redesign and parallelize Map Point Fusion, optimize Redundant Keyframe Culling on the CPU, and integrate a fast GPU-based Local Bundle Adjustment solver. To minimize data transfer and synchronization costs, we introduce persistent GPU-resident keyframe storage. Experiments on the EuRoC and TUM-VI datasets show average local mapping speedups of 1.3x and 1.6x, respectively, while preserving accuracy.
comment: Submitted to IROS 2026
♻ ☆ H2R: A Human-to-Robot Data Augmentation for Robot Pre-training from Videos
Large-scale pre-training using egocentric human videos has proven effective for robot learning. However, the models pre-trained on such data can be suboptimal for robot learning due to the significant visual gap between human hands and those of different robots. To remedy this, we propose H2R, a human-to-robot data augmentation pipeline that converts egocentric human videos into robot-centric visual data. H2R estimates human hand pose from videos, retargets the motion to simulated robotic arms, removes human limbs via segmentation and inpainting, and composites rendered robot embodiments into the original frames with camera-aligned geometry. This process explicitly bridges the visual gap between human and robot embodiments during pre-training. We apply H2R to augment large-scale egocentric human video datasets such as Ego4D and SSv2. To verify the effectiveness of the augmentation pipeline, we introduce a CLIP-based image-text similarity metric that quantitatively evaluates the semantic fidelity of robot-rendered frames to the original human actions. We evaluate H2R through comprehensive experiments in both simulation and real-world settings. In simulation, H2R consistently improves downstream success rates across four benchmark suites-Robomimic, RLBench, PushT, and CortexBench-yielding gains of 1.3%-10.2% across different visual encoders and policy learning methods. In real-world experiments, H2R improves performance on UR5 and dual-arm Franka/UR5 manipulation platforms, achieving 3.3%-23.3% success rate gains across gripper-based, dexterous, and bimanual tasks. We further demonstrate the potential of H2R in cross-embodiment generalization and its compatibility with vision-language-action models. These results indicate that H2R improves the generalization ability of robotic policies by mitigating the visual discrepancies between human and robot domains.
♻ ☆ DynaFlow: Dynamics-embedded Flow Matching for Physically Consistent Motion Generation from State-only Demonstrations
This paper introduces DynaFlow, a novel framework that embeds a differentiable simulator directly into a flow matching model. By generating trajectories in the action space and mapping them to dynamically feasible state trajectories via the simulator, DynaFlow ensures all outputs are physically consistent by construction. This end-to-end differentiable architecture enables training on state-only demonstrations, allowing the model to simultaneously generate physically consistent state trajectories while inferring the underlying action sequences required to produce them. We demonstrate the effectiveness of our approach through quantitative evaluations and showcase its real-world applicability by deploying the generated actions onto a physical Go1 quadruped robot. The robot successfully reproduces diverse gait present in the dataset, executes long-horizon motions in open-loop control and translates infeasible kinematic demonstrations into dynamically executable, stylistic behaviors. These hardware experiments validate that DynaFlow produces deployable, highly effective motions on real-world hardware from state-only demonstrations, effectively bridging the gap between kinematic data and real-world execution.
comment: 8 pages
♻ ☆ TinyIO: Lightweight Reparameterized Inertial Odometry
Inertial odometry (IO) is a widely used approach for localization on mobile devices; however, obtaining a lightweight IO model that also achieves high accuracy remains challenging. To address this issue, we propose TinyIO, a lightweight IO method. During training, we adopt a multi-branch architecture to extract diverse motion features more effectively. At inference time, the trained multi-branch model is converted into an equivalent single-path architecture to reduce computational complexity. We further propose a Dual-Path Adaptive Attention mechanism (DPAA), which enhances TinyIO's perception of contextual motion along both channel and temporal dimensions with negligible additional parameters. Extensive experiments on public datasets demonstrate that our method attains a favorable trade-off between accuracy and model size. On the RoNIN dataset, TinyIO reduces the ATE by 23.53% compared with R-ResNet and decreases the parameter count by 3.68%.
♻ ☆ SToRM: Supervised Token Reduction for Multi-modal LLMs toward efficient end-to-end autonomous driving ICRA 2026
In autonomous driving, end-to-end (E2E) driving systems that predict control commands directly from sensor data have achieved significant advancements. For safe driving in unexpected scenarios, these systems may additionally rely on human interventions such as natural language instructions. Using a multi-modal large language model (MLLM) facilitates human-vehicle interaction and can improve performance in such scenarios. However, this approach requires substantial computational resources due to its reliance on an LLM and numerous visual tokens from sensor inputs, which are limited in autonomous vehicles. Many MLLM studies have explored reducing visual tokens, but often suffer end-task performance degradation compared to using all tokens. To enable efficient E2E driving while maintaining performance comparable to using all tokens, this paper proposes the first Supervised Token Reduction framework for multi-modal LLMs (SToRM). The proposed framework consists of three key elements. First, a lightweight importance predictor with short-term sliding windows estimates token importance scores. Second, a supervised training approach uses an auxiliary path to obtain pseudo-supervision signals from an all-token LLM pass. Third, an anchor-context merging module partitions tokens into anchors and context tokens, and merges context tokens into relevant anchors to reduce redundancy while minimizing information loss. Experiments on the LangAuto benchmark show that SToRM outperforms state-of-the-art E2E driving MLLMs under the same reduced-token budget, maintaining all-token performance while reducing computational cost by up to 30x, and enabling real-time E2E driving on a standard GPU.
comment: Accepted to ICRA 2026
♻ ☆ Pose Estimation of a Thruster-Driven Bioinspired Multi-Link Robot
This work demonstrates simultaneous pose (position and orientation) and shape estimation for a free-floating, bioinspired multi-link robot with unactuated joints, link-mounted thrusters for control, and a single gyroscope per link, resulting in an underactuated, minimally sensed platform. Because the inter-link joint angles are constrained, translation and rotation of the multi-link system requires cyclic, reciprocating actuation of the thrusters, referred to as a gait. Through a proof-of-concept hardware experiment and offline analysis, we show that the robot's shape can be reliably estimated using an Unscented Kalman Filter augmented with Gaussian process residual models to compensate for non-zero-mean, non-Gaussian noise, while the pose exhibits drift expected from gyroscope integration in the absence of absolute position measurements. Experimental results demonstrate that a Gaussian process model trained on a multi-gait dataset (forward, backward, left, right, and turning) performs comparably to one trained exclusively on forward-gait data, revealing an overlap in the gait input space, which can be exploited to reduce per-gait training data requirements while enhancing the filter's generalizability across multiple gaits. Lastly, we introduce a heuristic derived from the observability Gramian to correlate joint angle estimate quality with gait periodicity and thruster inputs, highlighting how control affects estimation quality.
comment: 8 pages, 8 figures
♻ ☆ VLAD-Grasp: Zero-shot Grasp Detection via Vision-Language Models
Robotic grasping is a fundamental capability for enabling autonomous manipulation, with usually infinite solutions. State-of-the-art approaches for grasping rely on learning from large-scale datasets comprising expert annotations of feasible grasps. Curating such datasets is challenging, and hence, learning-based methods are limited by the solution coverage of the dataset, and require retraining to handle novel objects. Towards this, we present VLAD-Grasp, a Vision-Language model Assisted zero-shot approach for Detecting Grasps. Our method (1) prompts a large vision-language model to generate a goal image where a virtual cylindrical proxy intersects the object's geometry, explicitly encoding an antipodal grasp axis in image space, then (2) predicts depth and segmentation to lift this generated image into 3D, and (3) aligns generated and observed object point clouds via principal components and correspondence-free optimization to recover an executable grasp pose. Unlike prior work, our approach is training-free and does not require curated grasp datasets, while achieving performance competitive with the state-of-the-art methods on the Cornell and Jacquard datasets. Furthermore, we demonstrate zero-shot generalization to real-world objects on a Franka Research 3 robot, highlighting vision-language models as powerful priors for robotic manipulation.
comment: 8 pages, 4 figures, under review
♻ ☆ A Deconfounding Framework for Human Behavior Prediction: Enhancing Robotic Systems in Dynamic Environments
Accurate prediction of human behavior is crucial for effective human-robot interaction (HRI) systems, especially in dynamic environments where real-time decisions are essential. This paper addresses the challenge of forecasting future human behavior using multivariate time series data from wearable sensors, which capture various aspects of human movement. The presence of hidden confounding factors in this data often leads to biased predictions, limiting the reliability of traditional models. To overcome this, we propose a robust predictive model that integrates deconfounding techniques with advanced time series prediction methods, enhancing the model's ability to isolate true causal relationships and improve prediction accuracy. Evaluation on real-world datasets demonstrates that our approach significantly outperforms traditional methods, providing a more reliable foundation for responsive and adaptive HRI systems.
comment: 7 pages, Under review
♻ ☆ Register Any Point: Scaling 3D Point Cloud Registration by Flow Matching
Point cloud registration aligns multiple unposed point clouds into a common reference frame and is a core step for 3D reconstruction and robot localization without initial guess. In this work, we cast registration as conditional generation: a learned, continuous point-wise velocity field transports noisy points to a registered scene, from which the pose of each view is recovered. Unlike prior methods that perform correspondence matching to estimate pairwise transformations and then optimize a pose graph for multi-view registration, our model directly generates the registered point cloud, yielding both efficiency and point-level global consistency. By scaling the training data and conducting test-time rigidity enforcement, our approach achieves state-of-the-art results on existing pairwise registration benchmarks and on our proposed cross-domain multi-view registration benchmark. The superior zero-shot performance on this benchmark shows that our method generalizes across view counts, scene scales, and sensor modalities even with low overlap. Source code available at: https://github.com/PRBonn/RAP.
Robotics 3
☆ Coordinate-Independent Robot Model Identification
Robot model identification is commonly performed by least-squares regression on inverse dynamics, but existing formulations measure residuals directly in coordinate force space and therefore depend on the chosen coordinate chart, units, and scaling. This paper proposes a coordinate-independent identification method that weights inverse-dynamics residuals by the dual metric induced by the system Riemannian metric. Using the force--velocity vector--covector duality, the dual metric provides a physically meaningful normalization of generalized forces, pulling coordinate residuals back into the ambient mechanical space and eliminating coordinate-induced bias. The resulting objective remains convex through an affine-metric and Schur-complement reformulation, and is compatible with physical-consistency constraints and geometric regularization. Experiments on an inertia-dominated Crazyflie--pendulum system and a drag-dominated LandSalp robot show improved identification accuracy, especially on shape coordinates, in both low-data and high-data settings.
comment: 8 pages, 7 figures, supplementary video: https://youtu.be/w2bBBV9t1fk?si=iCoJ4l51wumwvCIo
☆ Seeing Where to Deploy: Metric RGB-Based Traversability Analysis for Aerial-to-Ground Hidden Space Inspection
Inspection of confined infrastructure such as culverts often requires accessing hidden spaces whose entrances are reachable primarily from elevated viewpoints. Aerial-ground cooperation enables a UAV to deploy a compact UGV for interior exploration, but selecting a suitable deployment region from aerial observations requires metric terrain reasoning involving scale ambiguity, reconstruction uncertainty, and terrain semantics. We present a metric RGB-based geometric-semantic reconstruction and traversability analysis framework for aerial-to-ground hidden space inspection. A feed-forward multi-view RGB reconstruction backbone produces dense geometry, while temporally consistent semantic segmentation yields a 3D semantic map. To enable deployment-relevant measurements without LiDAR-based dense mapping, we introduce an embodied motion prior that recovers metric scale by enforcing consistency between predicted camera motion and onboard platform egomotion. From the metrically grounded reconstruction, we construct a confidence-aware geometric-semantic traversability map and evaluate candidate deployment zones under explicit reachability constraints. Experiments on a tethered UAV-UGV platform demonstrate reliable deployment-zone identification in hidden space scenarios.
☆ Physically Accurate Rigid-Body Dynamics in Particle-Based Simulation IROS 2026
Robotics demands simulation that can reason about the diversity of real-world physical interactions, from rigid to deformable objects and fluids. Current simulators address this by stitching together multiple subsolvers for different material types, resulting in a compositional architecture that complicates physical reasoning. Particle-based simulators offer a compelling alternative, representing all materials through a single unified formulation that enables seamless cross-material interactions. Among particle-based simulators, position-based dynamics (PBD) is a popular solver known for its computational efficiency and visual plausibility. However, its lack of physical accuracy has limited its adoption in robotics. To leverage the benefits of particle-based solvers while meeting the physical fidelity demands of robotics, we introduce PBD-R, a revised PBD formulation that enforces physically accurate rigid-body dynamics through a novel momentum-conservation constraint and a modified velocity update. Additionally, we introduce a solver-agnostic benchmark with analytical solutions to evaluate physical accuracy. Using this benchmark, we show that PBD-R significantly outperforms PBD and achieves competitive accuracy with MuJoCo while requiring less computation.
comment: Submitted to IROS 2026
Robotics 50
☆ H-RINS: Hierarchical Tightly-coupled Radar-Inertial Navigation via Smoothing and Mapping
Millimeter-wave radar provides robust perception in visually degraded environments. However, radar-inertial state estimation is inherently susceptible to drift. Because radar yields only sparse, body-frame velocity measurements, it provides weak constraints on absolute orientation. Consequently, IMU biases remain poorly observable over the short time horizons typical of sliding-window filters. To address this fundamental observability challenge, we propose a tightly coupled, hierarchical radar-inertial factor graph framework. Our architecture decouples the estimation problem into a high-rate resetting graph and a persistent global graph. The resetting graph fuses IMU preintegration, radar velocities, and adaptive Zero-Velocity Updates (ZUPT) to generate the smooth, low-latency odometry required for real-time control. Concurrently, the persistent graph is a full-state factor graph maintaining the complete information of poses, velocities, and biases by fusing inertial data with keyframe-based geometric mapping and loop closures. Leveraging Incremental Smoothing and Mapping, the persistent graph can operate without explicit marginalization of variables, preserving their information while ensuring long-term bias observability. The cornerstone of our approach is a probabilistic tight-coupling mechanism: fully observable, optimized biases and their exact covariances are continuously injected from the persistent graph into the resetting graph's prior, effectively anchoring the high-rate estimator against integration drift. Extensive evaluations demonstrate our system achieves high accuracy with drift-reduced estimation at 27x real-time execution speeds. We release the implementation code and datasets upon the acceptance of the paper.
comment: 8 pages, 5 figures, Submitted to conference
☆ GelSphere: An Omnidirectional Rolling Vision-Based Tactile Sensor for Online 3D Reconstruction and Normal Force Estimation
We present GelSphere, a spherical vision-based tactile sensor designed for real-time continuous surface scanning. Unlike traditional vision-based tactile sensors that can only sense locally and are damaged when slid across surfaces, and cylindrical tactile sensors that can only roll along a fixed direction, our design enables omnidirectional rolling on surfaces. We accomplish this through our novel sensing system design, which has steel balls inside the sensor, forming a bearing layer between the gel and the rigid housing that allows rolling motion in all axes. The sensor streams tactile images through Wi-Fi, with online large-surface reconstruction capabilities. We present quantitative results for both reconstruction accuracy and image fusion performance. The results show that our sensor maintains geometric fidelity and high reconstruction accuracy even under multi-directional rolling, enabling uninterrupted surface scanning.
☆ Stiffness Copilot: An Impedance Policy for Contact-Rich Teleoperation
In teleoperation of contact-rich manipulation tasks, selecting robot impedance is critical but difficult. The robot must be compliant to avoid damaging the environment, but stiff to remain responsive and to apply force when needed. In this paper, we present Stiffness Copilot, a vision-based policy for shared-control teleoperation in which the operator commands robot pose and the policy adjusts robot impedance online. To train Stiffness Copilot, we first infer direction-dependent stiffness matrices in simulation using privileged contact information. We then use these matrices to supervise a lightweight vision policy that predicts robot stiffness from wrist-camera images and transfers zero-shot to real images at runtime. In a human-subject study, Stiffness Copilot achieved safety comparable to using a constant low stiffness while matching the efficiency of using a constant high stiffness.
comment: Project website: https://stiffness-copilot.github.io
☆ Amortizing Trajectory Diffusion with Keyed Drift Fields
Diffusion-based trajectory planners can synthesize rich, multimodal action sequences for offline reinforcement learning, but their iterative denoising incurs substantial inference-time cost, making closed-loop planning slow under tight compute budgets. We study the problem of achieving diffusion-like trajectory planning behavior with one-step inference, while retaining the ability to sample diverse candidate plans and condition on the current state in a receding-horizon control loop. Our key observation is that conditional trajectory generation fails under naïve distribution-matching objectives when the similarity measure used to align generated trajectories with the dataset is dominated by unconstrained future dimensions. In practice, this causes attraction toward average trajectories, collapses action diversity, and yields near-static behavior. Our key insight is that conditional generative planning requires a conditioning-aware notion of neighborhood: trajectory updates should be computed using distances in a compact key space that reflects the condition, while still applying updates in the full trajectory space. Building on this, we introduce Keyed Drifting Policies (KDP), a one-step trajectory generator trained with a drift-field objective that attracts generated trajectories toward condition-matched dataset windows and repels them from nearby generated samples, using a stop-gradient drifted target to amortize iterative refinement into training. At inference, the resulting policy produces a full trajectory window in a single forward pass. Across standard RL benchmarks and real-time hardware deployments, KDP achieves strong performance with one-step inference and substantially lower planning latency than diffusion sampling. Project website, code and videos: https://keyed-drifting.github.io/
☆ Distributional Uncertainty and Adaptive Decision-Making in System
Complex engineered systems require coordinated design choices across heterogeneous components under multiple conflicting objectives and uncertain specifications. Monotone co-design provides a compositional framework for such problems by modeling each subsystem as a design problem: a feasible relation between provided functionalities and required resources in partially ordered sets. Existing uncertain co-design models rely on interval bounds, which support worst-case reasoning but cannot represent probabilistic risk or multi-stage adaptive decisions. We develop a distributional extension of co-design that models uncertain design outcomes as distributions over design problems and supports adaptive decision processes through Markov-kernel re-parameterizations. Using quasi-measurable and quasi-universal spaces, we show that the standard co-design interconnection operations remain compositional under this richer notion of uncertainty. We further introduce queries and observations that extract probabilistic design trade-offs, including feasibility probabilities, confidence bounds, and distributions of minimal required resources. A task-driven unmanned aerial vehicle case study illustrates how the framework captures risk-sensitive and information-dependent design choices that interval-based models cannot express.
☆ URDF-Anything+: Autoregressive Articulated 3D Models Generation for Physical Simulation
Articulated objects are fundamental for robotics, simulation of physics, and interactive virtual environments. However, reconstructing them from visual input remains challenging, as it requires jointly inferring both part geometry and kinematic structure. We present, an end-to-end autoregressive framework that directly generates executable articulated object models from visual observations. Given image and object-level 3D cues, our method sequentially produces part geometries and their associated joint parameters, resulting in complete URDF models without reliance on multi-stage pipelines. The generation proceeds until the model determines that all parts have been produced, automatically inferring complete geometry and kinematics. Building on this capability, we enable a new Real-Follow-Sim paradigm, where high-fidelity digital twins constructed from visual observations allow policies trained and tested purely in simulation to transfer to real robots without online adaptation. Experiments on large-scale articulated object benchmarks and real-world robotic tasks demonstrate that outperforms prior methods in geometric reconstruction quality, joint parameter accuracy, and physical executability.
☆ Vision-guided Autonomous Dual-arm Extraction Robot for Bell Pepper Harvesting
Agricultural robotics has emerged as a critical solution to the labor shortages and rising costs associated with manual crop harvesting. Bell pepper harvesting, in particular, is a labor-intensive task, accounting for up to 50% of total production costs. While automated solutions have shown promise in controlled greenhouse environments, harvesting in unstructured outdoor farms remains an open challenge due to environmental variability and occlusion. This paper presents VADER (Vision-guided Autonomous Dual-arm Extraction Robot), a dual-arm mobile manipulation system designed specifically for the autonomous harvesting of bell peppers in outdoor environments. The system integrates a robust perception pipeline coupled with a dual-arm planning framework that coordinates a gripping arm and a cutting arm for extraction. We validate the system through trials in various realistic conditions, demonstrating a harvest success rate exceeding 60% with a cycle time of under 100 seconds per fruit, while also featuring a teleoperation fail-safe based on the GELLO teleoperation framework to ensure robustness. To support robust perception, we contribute a hierarchically structured dataset of over 3,200 images spanning indoor and outdoor domains, pairing wide-field scene images with close-up pepper images to enable a coarse-to-fine training strategy from fruit detection to high-precision pose estimation. The code and dataset will be made publicly available upon acceptance.
comment: 9 pages; first four authors have equal contribution
☆ ToMPC: Task-oriented Model Predictive Control via ADMM for Safe Robotic Manipulation
This paper proposes a task-oriented model predictive control (ToMPC) framework for safe and efficient robotic manipulation in open workspaces. The framework unifies collision-free motion and robot-environment interaction to address diverse scenarios. Additionally, it introduces task-oriented obstacle avoidance that leverages kinematic redundancy to enhance manipulation efficiency in obstructed environments. This complex optimization problem is solved by the alternating direction method of multipliers (ADMM), which decomposes the problem into two subproblems tackled by differential dynamic programming (DDP) and quadratic programming (QP), respectively. The effectiveness of this approach is validated in simulation and hardware experiments on a Franka Panda robotic manipulator. Results demonstrate that the framework can plan motion and/or force trajectories in real time, maximize the manipulation range while avoiding obstacles, and strictly adhere to safety-related hard constraints.
comment: 8 pages, 10 figures, accepted by IEEE Robotics and Automation Letters (RAL)
☆ SmoothVLA: Aligning Vision-Language-Action Models with Physical Constraints via Intrinsic Smoothness Optimization
Vision-Language-Action (VLA) models have emerged as a powerful paradigm for robotic manipulation. However, existing post-training methods face a dilemma between stability and exploration: Supervised Fine-Tuning (SFT) is constrained by demonstration quality and lacks generalization, whereas Reinforcement Learning (RL) improves exploration but often induces erratic, jittery trajectories that violate physical constraints. To bridge this gap, we propose SmoothVLA, a novel reinforcement learning fine-tuning framework that synergistically optimizes task performance and motion smoothness. The technical core is a physics-informed hybrid reward function that integrates binary sparse task rewards with a continuous dense term derived from trajectory jerk. Crucially, this reward is intrinsic, that computing directly from policy rollouts, without requiring extrinsic environment feedback or laborious reward engineering. Leveraging the Group Relative Policy Optimization (GRPO), SmoothVLA establishes trajectory smoothness as an explicit optimization prior, guiding the model toward physically feasible and stable control. Extensive experiments on the LIBERO benchmark demonstrate that SmoothVLA outperforms standard RL by 13.8\% in smoothness and significantly surpasses SFT in generalization across diverse tasks. Our work offers a scalable approach to aligning VLA models with physical-world constraints through intrinsic reward optimization.
☆ Data-Driven Autoregressive Power Prediction for GTernal Robots in the Robotarium
Energy-aware algorithms for multi-robot systems require accurate power consumption models, yet existing approaches rely on kinematic approximations that fail to capture the complex dynamics of real hardware. We present a lightweight autoregressive predictor for the GTernal mobile robot platform deployed in the Georgia Tech Robotarium. Through analysis of 48,000 samples collected across six motion trials, we discover that power consumption exhibits strong temporal autocorrelation ($ρ_1 = 0.95$) that dominates kinematic effects. A 7,041-parameter multi-layer perceptron (MLP) achieves $R^2 = 0.90$ on held-out motion patterns by conditioning on recent power history, reaching the theoretical prediction ceiling imposed by measurement noise. Physical validation across seven robots in a collision avoidance scenario yields mean $R^2 = 0.87$, demonstrating zero-shot transfer to unseen robots and behaviors. The predictor runs in 224 $μ$s per inference, enabling real-time deployment at 150$\times$ the platform's 30 Hz control rate. We release the trained model and dataset to support energy-aware multi-robot algorithm development.
comment: 8 pages, 5 figures
☆ LineMaster Pro: A Low-Cost Intelligent Line Following Robot with PID Control and Ultrasonic Obstacle Avoidance for Educational Robotics
Line following robots are fundamental platforms in robotics education, yet commercially available solutions remain prohibitively expensive ($150-300$) while lacking integrated obstacle detection capabilities essential for real-world applications. This paper presents LineMaster Pro, an intelligent low-cost line following robot implemented on an Arduino Nano platform that integrates dual TCRT5000 infrared sensors for precision line tracking, an HC-SR04 ultrasonic sensor for real-time obstacle detection, a digitally tuned PID controller with Ziegler-Nichols optimization, and a hierarchical finite state machine for robust obstacle avoidance. A systematic four-phase sensor calibration methodology ensures reliable operation across varying lighting and surface conditions. Experimental validation through 200 controlled trials and 72-hour continuous operation demonstrates mean tracking accuracy of 1.18 cm at 0.4 m/s (95\% CI [1.06, 1.30]), obstacle detection reliability of 96.7\% within 10-40 cm range with 0.7\% false positive rate, and 94\% successful recovery from path deviations. The PID implementation achieves 43\% improvement over conventional on-off control ($p<0.001$). At a total hardware cost of \$28.50 based on verified Bangladesh market prices, LineMaster Pro achieves a 94\% cost reduction compared to commercial alternatives, establishing a practical benchmark for accessible robotics education in resource-constrained environments.
☆ Pixel-level Scene Understanding in One Token: Visual States Need What-is-Where Composition
For robotic agents operating in dynamic environments, learning visual state representations from streaming video observations is essential for sequential decision making. Recent self-supervised learning methods have shown strong transferability across vision tasks, but they do not explicitly address what a good visual state should encode. We argue that effective visual states must capture what-is-where by jointly encoding the semantic identities of scene elements and their spatial locations, enabling reliable detection of subtle dynamics across observations. To this end, we propose CroBo, a visual state representation learning framework based on a global-to-local reconstruction objective. Given a reference observation compressed into a compact bottleneck token, CroBo learns to reconstruct heavily masked patches in a local target crop from sparse visible cues, using the global bottleneck token as context. This learning objective encourages the bottleneck token to encode a fine-grained representation of scene-wide semantic entities, including their identities, spatial locations, and configurations. As a result, the learned visual states reveal how scene elements move and interact over time, supporting sequential decision making. We evaluate CroBo on diverse vision-based robot policy learning benchmarks, where it achieves state-of-the-art performance. Reconstruction analyses and perceptual straightness experiments further show that the learned representations preserve pixel-level scene composition and encode what-moves-where across observations.
comment: Preprint
☆ Path-conditioned Reinforcement Learning-based Local Planning for Long-Range Navigation
Long-range navigation is commonly addressed through hierarchical pipelines in which a global planner generates a path, decomposed into waypoints, and followed sequentially by a local planner. These systems are sensitive to global path quality, as inaccurate remote sensing data can result in locally infeasible waypoints, which degrade local execution. At the same time, the limited global context available to the local planner hinders long-range efficiency. To address this issue, we propose a reinforcement learning-based local navigation policy that leverages path information as contextual guidance. The policy is conditioned on reference path observations and trained with a reward function mainly based on goal-reaching objectives, without any explicit path-following reward. Through this implicit conditioning, the policy learns to opportunistically exploit path information while remaining robust to misleading or degraded guidance. Experimental results show that the proposed approach significantly improves navigation efficiency when high-quality paths are available and maintains baseline-level performance when path observations are severely degraded or even non-existent. These properties make the method particularly well-suited for long-range navigation scenarios in which high-level plans are approximate and local execution must remain adaptive to uncertainty.
☆ Benchmarking the Energy Cost of Assurance in Neuromorphic Edge Robotics
Deploying trustworthy artificial intelligence on edge robotics imposes a difficult trade-off between high-assurance robustness and energy sustainability. Traditional defense mechanisms against adversarial attacks typically incur significant computational overhead, threatening the viability of power-constrained platforms in environments such as cislunar space. This paper quantifies the energy cost of assurance in event-driven neuromorphic systems. We benchmark the Hierarchical Temporal Defense (HTD) framework on the BrainChip Akida AKD1000 processor against a suite of adversarial temporal attacks. We demonstrate that unlike traditional deep learning defenses which often degrade efficiency significantly with increased robustness, the event-driven nature of the proposed architecture achieves a superior trade-off. The system reduces gradient-based adversarial success rates from 82.1% to 18.7% and temporal jitter success rates from 75.8% to 25.1%, while maintaining an energy consumption of approximately 45 microjoules per inference. We report a counter-intuitive reduction in dynamic power consumption in the fully defended configuration, attributed to volatility-gated plasticity mechanisms that induce higher network sparsity. These results provide empirical evidence that neuromorphic sparsity enables sustainable and high-assurance edge autonomy.
comment: 6 pages, 4 figures. Accepted and presented at the STEAR 2026 Workshop on Sustainable and Trustworthy Edge AI for Robotics, HiPEAC 2026, Krakow, Poland
☆ TransDex: Pre-training Visuo-Tactile Policy with Point Cloud Reconstruction for Dexterous Manipulation of Transparent Objects
Dexterous manipulation enables complex tasks but suffers from self-occlusion, severe depth noise, and depth information loss when manipulating transparent objects. To solve this problem, this paper proposes TransDex, a 3D visuo-tactile fusion motor policy based on point cloud reconstruction pre-training. Specifically, we first propose a self-supervised point cloud reconstruction pre-training approach based on Transformer. This method accurately recovers the 3D structure of objects from interactive point clouds of dexterous hands, even when random noise and large-scale masking are added. Building on this, TransDex is constructed in which perceptual encoding adopts a fine-grained hierarchical scheme and multi-round attention mechanisms adaptively fuse features of the robotic arm and dexterous hand to enable differentiated motion prediction. Results from transparent object manipulation experiments conducted on a real robotic system demonstrate that TransDex outperforms existing baseline methods. Further analysis validates the generalization capabilities of TransDex and the effectiveness of its individual components.
comment: Project page: https://transdex.github.io/
☆ LDHP: Library-Driven Hierarchical Planning for Non-prehensile Dexterous Manipulation
Non-prehensile manipulation is essential for handling thin, large, or otherwise ungraspable objects in unstructured settings. Prior planning and search-based methods often rely on ad-hoc manual designs or generate physically unrealizable motions by ignoring critical gripper properties, while training-based approaches are data-intensive and struggle to generalize to novel, out-of-distribution tasks. We propose a library-driven hierarchical planner (LDHP) that makes executability a first-class design goal: a top-tier contact-state planner proposes object-pose paths using MoveObject primitives, and a bottom-tier grasp planner synthesizes feasible grasp sequences with AdjustGrasp primitives; feasibility is certified by collision checks and quasi-static mechanics, and contact-sensitive segments are recovered via a bounded dichotomy refinement. This gripper-aware decomposition decouples object motion from grasp realizability, yields a task-agnostic pipeline that transfers across manipulation tasks and geometric variations without re-design, and exposes clean hooks for optional learned priors. Real-robot studies on zero-mobility lifting and slot insertion demonstrate consistent execution and robustness to shape and environment changes.
comment: 9 pages
☆ Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving
End-to-end autonomous driving is typically built upon imitation learning (IL), yet its performance is constrained by the quality of human demonstrations. To overcome this limitation, recent methods incorporate reinforcement learning (RL) through sequential fine-tuning. However, such a paradigm remains suboptimal: sequential RL fine-tuning can introduce policy drift and often leads to a performance ceiling due to its dependence on the pretrained IL policy. To address these issues, we propose PaIR-Drive, a general Parallel framework for collaborative Imitation and Reinforcement learning in end-to-end autonomous driving. During training, PaIR-Drive separates IL and RL into two parallel branches with conflict-free training objectives, enabling fully collaborative optimization. This design eliminates the need to retrain RL when applying a new IL policy. During inference, RL leverages the IL policy to further optimize the final plan, allowing performance beyond prior knowledge of IL. Furthermore, we introduce a tree-structured trajectory neural sampler to group relative policy optimization (GRPO) in the RL branch, which enhances exploration capability. Extensive analysis on NAVSIMv1 and v2 benchmark demonstrates that PaIR-Drive achieves Competitive performance of 91.2 PDMS and 87.9 EPDMS, building upon Transfuser and DiffusionDrive IL baselines. PaIR-Drive consistently outperforms existing RL fine-tuning methods, and could even correct human experts' suboptimal behaviors. Qualitative results further confirm that PaIR-Drive can effectively explore and generate high-quality trajectories.
comment: 8 pages, 7 figures, 6 tables
☆ ImagiNav: Scalable Embodied Navigation via Generative Visual Prediction and Inverse Dynamics
Enabling robots to navigate open-world environments via natural language is critical for general-purpose autonomy. Yet, Vision-Language Navigation has relied on end-to-end policies trained on expensive, embodiment-specific robot data. While recent foundation models trained on vast simulation data show promise, the challenge of scaling and generalizing due to the limited scene diversity and visual fidelity in simulation persists. To address this gap, we propose ImagiNav, a novel modular paradigm that decouples visual planning from robot actuation, enabling the direct utilization of diverse in-the-wild navigation videos. Our framework operates as a hierarchy: a Vision-Language Model first decomposes instructions into textual subgoals; a finetuned generative video model then imagines the future video trajectory towards that subgoal; finally, an inverse dynamics model extracts the trajectory from the imagined video, which can then be tracked via a low-level controller. We additionally develop a scalable data pipeline of in-the-wild navigation videos auto-labeled via inverse dynamics and a pretrained Vision-Language Model. ImagiNav demonstrates strong zero-shot transfer to robot navigation without requiring robot demonstrations, paving the way for generalist robots that learn navigation directly from unlabeled, open-world data.
☆ GraspADMM: Improving Dexterous Grasp Synthesis via ADMM Optimization
Synthesizing high-quality dexterous grasps is a fundamental challenge in robot manipulation, requiring adherence to diversity, kinematic feasibility (valid hand-object contact without penetration), and dynamic stability (secure multi-contact forces). The recent framework Dexonomy successfully ensures broad grasp diversity through dense sampling and improves kinematic feasibility via a simulator-based refinement method that excels at resolving exact collisions. However, its reliance on fixed contact points restricts the hand's reachability and prevents the optimization of grasp metrics for dynamic stability. Conversely, purely gradient-based optimizers can maximize dynamic stability but rely on simplified contact approximations that inevitably cause physical penetrations. To bridge this gap, we propose GraspADMM, a novel grasp synthesis framework that preserves sampling-based diversity while improving kinematic feasibility and dynamic stability. By formulating the refinement stage using the Alternating Direction Method of Multipliers (ADMM), we decouple the target contact points on the object from the actual contact locations on the hand. This decomposition allows the pipeline to alternate between updating the target object points to directly maximize dynamic grasp metrics, and adjusting the hand pose to physically reach these targets while strictly respecting collision boundaries. Extensive experiments demonstrate that GraspADMM significantly outperforms state-of-the-art baselines, achieving a nearly 15\% absolute improvement in grasp success rate for type-unaware synthesis and roughly a 100\% relative improvement in type-aware synthesis. Furthermore, our approach maintains robust, physically plausible grasp generation even under extreme low-friction conditions.
☆ ArrayTac: A tactile display for simultaneous rendering of shape, stiffness and friction
Human-computer interaction in the visual and auditory domains has achieved considerable maturity, yet machine-to-human tactile feedback remains underdeveloped. Existing tactile displays struggle to simultaneously render multiple tactile dimensions, such as shape, stiffness, and friction, which limits the realism of haptic simulation. Here, we present ArrayTac, a piezoelectric-driven tactile display capable of simultaneously rendering shape, stiffness, and friction to reproduce realistic haptic signals. The system comprises a 4x4 array of 16 actuator units, each employing a three-stage micro-lever mechanism to amplify the micrometer-scale displacement of the piezoelectric element, with Hall sensor-based closed-loop control at the end effector to enhance response speed and precision. We further implement two end-to-end pipelines: 1) a vision-to-touch framework that converts visual inputs into tactile signals using multimodal foundation models, and 2) a real-time tele-palpation system operating over distances of several thousand kilometers. In user studies, first-time participants accurately identify object shapes and physical properties with high success rates. In a tele-palpation experiment over 1,000km, untrained volunteers correctly identified both the number and type of tumors in a breast phantom with 100% accuracy and precisely localized their positions. The system pioneers a new pathway for high-fidelity haptic feedback by introducing the unprecedented capability to simultaneously render an object's shape, stiffness, and friction, delivering a holistic tactile experience that was previously unattainable.
☆ Building Explicit World Model for Zero-Shot Open-World Object Manipulation
Open-world object manipulation remains a fundamental challenge in robotics. While Vision-Language-Action (VLA) models have demonstrated promising results, they rely heavily on large-scale robot action demonstrations, which are costly to collect and can hinder out-of-distribution generalization. In this paper, we propose an explicit-world-model-based framework for open-world manipulation that achieves zero-shot generalization by constructing a physically grounded digital twin of the environment. The framework integrates open-set perception, digital-twin reconstruction, sampling and evaluation of interaction strategies. By constructing a digital twin of the environment, our approach efficiently explores and evaluates manipulation strategies in physic-enabled simulator and reliably deploys the chosen strategy to the real world. Experimentally, the proposed framework is able to perform multiple open-set manipulation tasks without any task-specific action demonstrations, proving strong zero-shot generalization on both the task and object levels. Project Page: https://bojack-bj.github.io/projects/thesis/
☆ ST-VLA: Enabling 4D-Aware Spatiotemporal Understanding for General Robot Manipulation
Robotic manipulation in open-world environments requires reasoning across semantics, geometry, and long-horizon action dynamics. Existing hierarchical Vision-Language-Action (VLA) frameworks typically use 2D representations to connect high-level reasoning with low-level control, but lack depth awareness and temporal consistency, limiting robustness in complex 3D scenes. We propose ST-VLA, a hierarchical VLA framework using a unified 3D-4D representation to bridge perception and action. ST-VLA converts 2D guidance into 3D trajectories and generates smooth spatial masks that capture 4D spatio-temporal context, providing a stable interface between semantic reasoning and continuous control. To enable effective learning of such representations, we introduce ST-Human, a large-scale human manipulation dataset with 14 tasks and 300k episodes, annotated with 2D, 3D, and 4D supervision via a semi-automated pipeline. Using ST-Human, we train ST-VLM, a spatio-temporal vision-language model that generates spatially grounded and temporally coherent 3D representations to guide policy execution. The smooth spatial masks focus on task-relevant geometry and stabilize latent representations, enabling online replanning and long-horizon reasoning. Experiments on RLBench and real-world manipulation tasks show that \method significantly outperforms state-of-the-art baselines, improving zero-shot success rates by 44.6% and 30.3%. These results demonstrate that offloading spatio-temporal reasoning to VLMs with unified 3D-4D representations substantially improves robustness and generalization for open-world robotic manipulation. Project website: https://oucx117.github.io/ST-VLA/.
comment: 25 pages, under review
☆ Robust Sim-to-Real Cloth Untangling through Reduced-Resolution Observations via Adaptive Force-Difference Quantization
Robotic cloth untangling requires progressively disentangling fabric by adapting pulling actions to changing contact and tension conditions. Because large-scale real-world training is impractical due to cloth damage and hardware wear, sim-to-real policy transfer is a promising solution. However, cloth manipulation is highly sensitive to interaction dynamics, and policies that depend on precise force magnitudes often fail after transfer because similar force responses cannot be reproduced due to the reality gap. We observe that untangling is largely characterized by qualitative tension transitions rather than exact force values. This indicates that directly minimizing the sim-to-real gap in raw force measurements does not necessarily align with the task structure. We therefore hypothesize that emphasizing coarse force-change patterns while suppressing fine environment-dependent variations can improve robustness of sim-to-real transfer. Based on this insight, we propose Adaptive Force-Difference Quantization (ADQ), which reduces observation resolution by representing force inputs as discretized temporal differences and learning state-dependent quantization thresholds adaptively. This representation mitigates overfitting to environment-specific force characteristics and facilitates direct sim-to-real transfer. Experiments in both simulation and real-world cloth untangling demonstrate that ADQ achieves higher success rates and exhibits greater robustness in sim-to-real transfer than policies using raw force inputs. Supplementary video is available at https://youtu.be/ZeoBs-t0AWc
comment: under review
☆ Your Vision-Language-Action Model Already Has Attention Heads For Path Deviation Detection
Vision-Language-Action (VLA) models have demonstrated strong potential for predicting semantic actions in navigation tasks, demonstrating the ability to reason over complex linguistic instructions and visual contexts. However, they are fundamentally hindered by visual-reasoning hallucinations that lead to trajectory deviations. Addressing this issue has conventionally required training external critic modules or relying on complex uncertainty heuristics. In this work, we discover that monitoring a few attention heads within a frozen VLA model can accurately detect path deviations without incurring additional computational overhead. We refer to these heads, which inherently capture the spatiotemporal causality between historical visual sequences and linguistic instructions, as Navigation Heads. Using these heads, we propose an intuitive, training-free anomaly-detection framework that monitors their signals to detect hallucinations in real time. Surprisingly, among over a thousand attention heads, a combination of just three is sufficient to achieve a 44.6 % deviation detection rate with a low false-positive rate of 11.7 %. Furthermore, upon detecting a deviation, we bypass the heavy VLA model and trigger a lightweight Reinforcement Learning (RL) policy to safely execute a shortest-path rollback. By integrating this entire detection-to-recovery pipeline onto a physical robot, we demonstrate its practical robustness. All source code will be publicly available.
comment: Keywords: Vision-Language Action (VLA), Reinforcement Learning (RL), Navigation Path Recovery, Robot Operating System (ROS)
☆ KoopmanFlow: Spectrally Decoupled Generative Control Policy via Koopman Structural Bias
Generative Control Policies (GCPs) show immense promise in robotic manipulation but struggle to simultaneously model stable global motions and high-frequency local corrections. While modern architectures extract multi-scale spatial features, their underlying Probability Flow ODEs apply a uniform temporal integration schedule. Compressed to a single step for real-time Receding Horizon Control (RHC), uniform ODE solvers mathematically smooth over sparse, high-frequency transients entangled within low-frequency steady states. To decouple these dynamics without accumulating pipelined errors, we introduce KoopmanFlow, a parameter-efficient generative policy guided by a Koopman-inspired structural inductive bias. Operating in a unified multimodal latent space with visual context, KoopmanFlow bifurcates generation at the terminal stage. Because visual conditioning occurs before spectral decomposition, both branches are visually guided yet temporally specialized. A macroscopic branch anchors slow-varying trajectories via single-step Consistency Training, while a transient branch uses Flow Matching to isolate high-frequency residuals stimulated by sudden visual cues (e.g., contacts or occlusions). Guided by an explicit spectral prior and optimized via a novel asymmetric consistency objective, KoopmanFlow establishes a fused co-training mechanism. This allows the variant branch to absorb localized dynamics without multi-stage error accumulation. Extensive experiments show KoopmanFlow significantly outperforms state-of-the-art baselines in contact-rich tasks requiring agile disturbance rejection. By trading a surplus latency buffer for a richer structural prior, KoopmanFlow achieves superior control fidelity and parameter efficiency within real-time deployment limits.
☆ Exploration-assisted Bottleneck Transition Toward Robust and Data-efficient Deformable Object Manipulation
Imitation learning has demonstrated impressive results in robotic manipulation but fails under out-of-distribution (OOD) states. This limitation is particularly critical in Deformable Object Manipulation (DOM), where the near-infinite possible configurations render comprehensive data collection infeasible. Although several methods address OOD states, they typically require exhaustive data or highly precise perception. Such requirements are often impractical for DOM owing to its inherent complexities, including self-occlusion. To address the OOD problem in DOM, we propose a novel framework, Exploration-assisted Bottleneck Transition for Deformable Object Manipulation (ExBot), which addresses the OOD challenge through two key advantages. First, we introduce bottleneck states, standardized configurations that serve as starting points for task execution. This enables the reconceptualization of OOD challenges as the problem of transitioning diverse initial states to these bottleneck states, significantly reducing demonstration requirements. Second, to account for imperfect perception, we partition the OOD state space based on recognizability and employ dual action primitives. This approach enables ExBot to manipulate even unrecognizable states without requiring accurate perception. By concentrating demonstrations around bottleneck states and leveraging exploration to alter perceptual conditions, ExBot achieves both data efficiency and robustness to severe OOD scenarios. Real-world experiments on rope and cloth manipulation demonstrate successful task completion from diverse OOD states, including severe self-occlusions.
☆ Multi-Robot Coordination for Planning under Context Uncertainty
Real-world robots often operate in settings where objective priorities depend on the underlying context of operation. When the underlying context is unknown apriori, multiple robots may have to coordinate to gather informative observations to infer the context, since acting based on an incorrect context can lead to misaligned and unsafe behavior. Once the underlying true context is inferred, the robots optimize their task-specific objectives in the preference order induced by the context. We formalize this problem as a Multi-Robot Context-Uncertain Stochastic Shortest Path (MR-CUSSP), which captures context-relevant information at landmark states through joint observations. Our two-stage solution approach is composed of: (1) CIMOP (Coordinated Inference for Multi-Objective Planning) to compute plans that guide robots toward informative landmarks to efficiently infer the true context, and (2) LCBS (Lexicographic Conflict-Based Search) for collision-free multi-robot path planning with lexicographic objective preferences, induced by the context. We evaluate the algorithms using three simulated domains and demonstrate its practical applicability using five mobile robots in the salp domain setup.
comment: 8 pages, 6 figures
☆ Implicit Maximum Likelihood Estimation for Real-time Generative Model Predictive Control ICRA
Diffusion-based models have recently shown strong performance in trajectory planning, as they are capable of capturing diverse, multimodal distributions of complex behaviors. A key limitation of these models is their slow inference speed, which results from the iterative denoising process. This makes them less suitable for real-time applications such as closed-loop model predictive control (MPC), where plans must be generated quickly and adapted continuously to a changing environment. In this paper, we investigate Implicit Maximum Likelihood Estimation (IMLE) as an alternative generative modeling approach for planning. IMLE offers strong mode coverage while enabling inference that is two orders of magnitude faster, making it particularly well suited for real-time MPC tasks. Our results demonstrate that IMLE achieves competitive performance on standard offline reinforcement learning benchmarks compared to the standard diffusion-based planner, while substantially improving planning speed in both open-loop and closed-loop settings. We further validate IMLE in a closed-loop human navigation scenario, operating in real-time, demonstrating how it enables rapid and adaptive plan generation in dynamic environments.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026. Project page: https://kir-.github.io/GMPC-IMLE/
☆ LPV-MPC for Lateral Control in Full-Scale Autonomous Racing
Autonomous racing has attracted significant attention recently, presenting challenges in selecting an optimal controller that operates within the onboard system's computational limits and meets operational constraints such as limited track time and high costs. This paper introduces a Linear Parameter-Varying Model Predictive Controller (LPV-MPC) for lateral control. Implemented on an IAC AV-24, the controller achieved stable performance at speeds exceeding 160 mph (71.5 m/s). We detail the controller design, the methodology for extracting model parameters, and key system-level and implementation considerations. Additionally, we report results from our final race run, providing a comprehensive analysis of both vehicle dynamics and controller performance. A Python implementation of the framework is available at: https://tinyurl.com/LPV-MPC-acados
☆ REFINE-DP: Diffusion Policy Fine-tuning for Humanoid Loco-manipulation via Reinforcement Learning
Humanoid loco-manipulation requires coordinated high-level motion plans with stable, low-level whole-body execution under complex robot-environment dynamics and long-horizon tasks. While diffusion policies (DPs) show promise for learning from demonstrations, deploying them on humanoids poses critical challenges: the motion planner trained offline is decoupled from the low-level controller, leading to poor command tracking, compounding distribution shift, and task failures. The common approach of scaling demonstration data is prohibitively expensive for high-dimensional humanoid systems. To address this challenge, we present REFINE-DP (REinforcement learning FINE-tuning of Diffusion Policy), a hierarchical framework that jointly optimizes a DP high-level planner and an RL-based low-level loco-manipulation controller. The DP is fine-tuned via a PPO-based diffusion policy gradient to improve task success rate, while the controller is simultaneously updated to accurately track the planner's evolving command distribution, reducing the distributional mismatch that degrades motion quality. We validate REFINE-DP on a humanoid robot performing loco-manipulation tasks, including door traversal and long-horizon object transport. REFINE-DP achieves an over $90\%$ success rate in simulation, even in out-of-distribution cases not seen in the pre-trained data, and enables smooth autonomous task execution in real-world dynamic environments. Our proposed method substantially outperforms pre-trained DP baselines and demonstrates that RL fine-tuning is key to reliable humanoid loco-manipulation. https://refine-dp.github.io/REFINE-DP/
☆ D-Compress: Detail-Preserving LiDAR Range Image Compression for Real-Time Streaming on Resource-Constrained Robots ICRA 2026
Efficient 3D LiDAR point cloud compression (LPCC) and streaming are critical for edge server-assisted robotic systems, enabling real-time communication with compact data representations. A widely adopted approach represents LiDAR point clouds as range images, enabling the direct use of mature image and video compression codecs. However, because these codecs are designed with human visual perception in mind, they often compromise geometric details, which downgrades the performance of downstream robotic tasks such as mapping and object detection. Furthermore, rate-distortion optimization (RDO)-based rate control remains largely underexplored for range image compression (RIC) under dynamic bandwidth conditions. To address these limitations, we propose D-Compress, a new detail-preserving and fast RIC framework tailored for real-time streaming. D-Compress integrates both intra- and inter-frame prediction with an adaptive discrete wavelet transform approach for precise residual compression. Additionally, we introduce a new RDO-based rate control algorithm for RIC through new rate-distortion modeling. Extensive evaluations on various datasets demonstrate the superiority of D-Compress, which outperforms state-of-the-art (SOTA) compression methods in both geometric accuracy and downstream task performance, particularly at compression ratios exceeding 100x, while maintaining real-time execution on resource-constrained hardware. Moreover, evaluations under dynamic bandwidth conditions validate the robustness of its rate control mechanism.
comment: To appear in IEEE ICRA 2026
☆ SAATT Nav: a Socially Aware Autonomous Transparent Transportation Navigation Framework for Wheelchairs IROS 2026
While powered wheelchairs reduce physical fatigue as opposed to manual wheelchairs for individuals with mobility impairment, they demand high cognitive workload due to information processing, decision making and motor coordination. Current autonomous systems lack social awareness in navigation and transparency in decision-making, leading to decreased perceived safety and trust from the user and others in context. This work proposes Socially Aware Autonomous Transparent Transportation (SAATT) Navigation framework for wheelchairs as a potential solution. By implementing a Large Language Model (LLM) informed of user intent and capable of predicting other peoples' intent as a decision-maker for its local controller, it is able to detect and navigate social situations, such as passing pedestrians or a pair conversing. Furthermore, the LLM textually communicates its reasoning at each waypoint for transparency. In this experiment, it is compared against a standard global planner, a representative competing social navigation model, and an Ablation study in three simulated environments varied by social levels in eight metrics categorized under Safety, Social Compliance, Efficiency, and Comfort. Overall, SAATT Nav outperforms in most social situations and equivalently or only slightly worse in the remaining metrics, demonstrating the potential of a socially aware and transparent autonomous navigation system to assist wheelchair users.
comment: 8 pages, 4 figures, 2 tables, 1 algorithm. Submitted to IROS 2026
♻ ☆ From Fold to Function: Simulation-Driven Design of Origami Mechanisms
Origami-inspired mechanisms can transform flat sheets into functional three-dimensional dynamic structures that are lightweight, compact, and capable of complex motion. These properties make origami increasingly valuable in robotic and deployable systems. However, accurately simulating their folding behavior and interactions with the environment remains challenging. To address this, we present a design framework for origami mechanism simulation that utilizes MuJoCo's deformable-body capabilities. In our approach, origami sheets are represented as graphs of interconnected deformable elements with user-specified constraints such as creases and actuation, defined through an intuitive graphical user interface (GUI). This framework allows users to generate physically consistent simulations that capture both the geometric structure of origami mechanisms and their interactions with external objects and surfaces. We demonstrate our method's utility through a case study on an origami catapult, where design parameters are optimized in simulation using the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) and validated experimentally on physical prototypes. The optimized structure achieves improved throwing performance, illustrating how our system enables rapid, simulation-driven origami design, optimization, and analysis.
comment: 8 Pages, 9 Figures, Submitted to IEEE RoboSoft
♻ ☆ Multi-Robot Navigation in Social Mini-Games: Definitions, Taxonomy, and Algorithms
The "Last Mile Challenge" has long been considered an important, yet unsolved, challenge for autonomous vehicles, public service robots, and delivery robots. A central issue in this challenge is the ability of robots to navigate constrained and cluttered environments that have high agency (e.g., doorways, hallways, corridor intersections), often while competing for space with other robots and humans. We refer to these environments as "Social Mini-Games" (SMGs). Traditional navigation approaches designed for MRN do not perform well in SMGs, which has led to focused research on dedicated SMG solvers. However, publications on SMG navigation research make different assumptions, and have different objective functions (safety versus liveness). These assumptions and objectives are sometimes implicitly assumed or described informally. This makes it difficult to establish appropriate baselines for comparison in research papers, as well as making it difficult for practitioners to find the papers relevant to their concrete application. Such ad-hoc representation of the field also presents a barrier to new researchers wanting to start research in this area. SMG navigation research requires its own taxonomy, definitions, and evaluation protocols to guide effective research moving forward. This survey is the first to catalog SMG solvers using a well-defined and unified taxonomy and to classify existing methods accordingly. It also discusses the essential properties of SMG solvers, defines what SMGs are and how they appear in practice, outlines how to evaluate SMG solvers, and highlights the differences between SMG solvers and general navigation systems. The survey concludes with an overview of future directions and open challenges in the field. Our project is open-sourced at https://socialminigames.github.io/{https://socialminigames.github.io/.
comment: Accepted for publication in Autonomous Robots 2026
♻ ☆ SERFN: Sample-Efficient Real-World Dexterous Policy Fine-Tuning via Action-Chunked Critics and Normalizing Flows
Real-world fine-tuning of dexterous manipulation policies remains challenging due to limited real-world interaction budgets and highly multimodal action distributions. Diffusion-based policies, while expressive, do not permit conservative likelihood-based updates during fine-tuning because action probabilities are intractable. In contrast, conventional Gaussian policies collapse under multimodality, particularly when actions are executed in chunks, and standard per-step critics fail to align with chunked execution, leading to poor credit assignment. We present SERFN, a sample-efficient off-policy fine-tuning framework with normalizing flow (NF) to address these challenges. The normalizing flow policy yields exact likelihoods for multimodal action chunks, allowing conservative, stable policy updates through likelihood regularization and thereby improving sample efficiency. An action-chunked critic evaluates entire action sequences, aligning value estimation with the policy's temporal structure and improving long-horizon credit assignment. To our knowledge, this is the first demonstration of a likelihood-based, multimodal generative policy combined with chunk-level value learning on real robotic hardware. We evaluate SERFN on two challenging dexterous manipulation tasks in the real world: cutting tape with scissors retrieved from a case, and in-hand cube rotation with a palm-down grasp -- both of which require precise, dexterous control over long horizons. On these tasks, SERFN achieves stable, sample-efficient adaptation where standard methods struggle.
comment: https://srl-ethz.github.io/SERNF/
♻ ☆ ComFree-Sim: A GPU-Parallelized Analytical Contact Physics Engine for Scalable Contact-Rich Robotics Simulation and Control
Physics simulation for contact-rich robotics is often bottlenecked by contact resolution: mainstream engines enforce non-penetration and Coulomb friction via complementarity constraints or constrained optimization, requiring per-step iterative solves whose cost grows superlinearly with contact density. We present ComFree-Sim, a GPU-parallelized analytical contact physics engine built on complementarity-free contact modeling. ComFree-Sim computes contact impulses in closed form via an impedance-style prediction--correction update in the dual cone of Coulomb friction. Contact computation decouples across contact pairs and becomes separable across cone facets, mapping naturally to GPU kernels and yielding near-linear runtime scaling with the number of contacts. We further extend the formulation to a unified 6D contact model capturing tangential, torsional, and rolling friction, and introduce a practical dual-cone impedance heuristic. ComFree-Sim is implemented in Warp and exposed through a MuJoCo-compatible interface as a drop-in backend alternative to MuJoCo Warp (MJWarp). Experiments benchmark penetration, friction behaviors, stability, and simulation runtime scaling against MJWarp, demonstrating near-linear scaling and 2--3 times higher throughput in dense contact scenes with comparable physical fidelity. We deploy ComFree-Sim in real-time MPC for in-hand dexterous manipulation on a real-world multi-fingered LEAP hand and in dynamics-aware motion retargeting, demonstrating that low-latency simulation yields higher closed-loop success rates and enables practical high-frequency control in contact-rich tasks.
comment: 9 pages
♻ ☆ UniPrototype: Humn-Robot Skill Learning with Uniform Prototypes
Data scarcity remains a fundamental challenge in robot learning. While human demonstrations benefit from abundant motion capture data and vast internet resources, robotic manipulation suffers from limited training examples. To bridge this gap between human and robot manipulation capabilities, we propose UniPrototype, a novel framework that enables effective knowledge transfer from human to robot domains via shared motion primitives. ur approach makes three key contributions: (1) We introduce a compositional prototype discovery mechanism with soft assignments, enabling multiple primitives to co-activate and thus capture blended and hierarchical skills; (2) We propose an adaptive prototype selection strategy that automatically adjusts the number of prototypes to match task complexity, ensuring scalable and efficient representation; (3) We demonstrate the effectiveness of our method through extensive experiments in both simulation environments and real-world robotic systems. Our results show that UniPrototype successfully transfers human manipulation knowledge to robots, significantly improving learning efficiency and task performance compared to existing approaches.The code and dataset will be released upon acceptance at an anonymous repository.
comment: This submission was uploaded in error and has been withdrawn. A substantial revision will need to be completed
♻ ☆ Social Robots for People Living with Dementia: A Scoping Review on Deception from Design to Perception
As social robots are increasingly introduced into dementia care, their embodied and interactive design may blur the boundary between artificial and lifelike entities, raising ethical concerns about robotic deception. However, it remains unclear which specific design cues of social robots might lead to social robotic deception (SRD) in people living with dementia (PLwD), and which perceptions and responses of PLwD might indicate that SRD is taking place. To address these questions, we conducted a scoping review of 26 empirical studies reporting PLwD interacting with social robots. We identified three key design cue categories that might contribute to SRD and one that might break the illusion. However, the available literature does not provide sufficient evidence to determine which specific design cues lead to SRD. Thematic analysis of user responses reveals six recurring patterns in how PLwD perceive and respond to social robots. However, conceptual limitations in existing definitions of robotic deception make it difficult to identify when and to what extent deception actually occurs. Building on the results, we propose a dual-process interpretation that clarifies the cognitive basis of false beliefs in human-robot interaction and distinguishes SRD from anthropomorphism or emotional engagement.
♻ ☆ Using VLM Reasoning to Constrain Task and Motion Planning IROS 2026
In task and motion planning, high-level task planning is done over an abstraction of the world to enable efficient search in long-horizon robotics problems. However, the feasibility of these task-level plans relies on the downward refinability of the abstraction into continuous motion. When a domain's refinability is poor, task-level plans that appear valid may ultimately fail during motion planning, requiring replanning and resulting in slower overall performance. Prior works mitigate this by encoding refinement issues as constraints to prune infeasible task plans. However, these approaches only add constraints upon refinement failure, expending significant search effort on infeasible branches. We propose VIZ-COAST, a method of leveraging the common-sense spatial reasoning of large pretrained Vision-Language Models to identify issues with downward refinement a priori, bypassing the need to fix these failures during planning. Experiments on three challenging TAMP domains show that our approach is able to extract plausible constraints from images and domain descriptions, drastically reducing planning times and, in some cases, eliminating downward refinement failures altogether, generalizing to a diverse range of instances from the broader domain.
comment: 9 pages, 7 figures, 1 table. Submitted to IROS 2026
♻ ☆ Dribble Master: Learning Agile Humanoid Dribbling through Legged Locomotion
Humanoid soccer dribbling is a highly challenging task that demands dexterous ball manipulation while maintaining dynamic balance. Traditional rule-based methods often struggle to achieve accurate ball control due to their reliance on fixed walking patterns and limited adaptability to real-time ball dynamics. To address these challenges, we propose a two-stage curriculum learning framework that enables a humanoid robot to acquire dribbling skills without explicit dynamics or predefined trajectories. In the first stage, the robot learns basic locomotion skills; in the second stage, we fine-tune the policy for agile dribbling maneuvers. We further introduce a virtual camera model in simulation that simulates the field of view and perception constraints of the real robot, enabling realistic ball perception during training. We also design heuristic rewards to encourage active sensing, promoting a broader visual range for continuous ball perception. The policy is trained in simulation and successfully transferred to a physical humanoid robot. Experiment results demonstrate that our method enables effective ball manipulation, achieving flexible and visually appealing dribbling behaviors across multiple environments. This work highlights the potential of reinforcement learning in developing agile humanoid soccer robots. Additional details and videos are available at https://zhuoheng0910.github.io/dribble-master/.
♻ ☆ DyQ-VLA: Temporal-Dynamic-Aware Quantization for Embodied Vision-Language-Action Models
Vision-Language-Action (VLA) models are dominant in embodied intelligence but are constrained by inference overheads. While model quantization alleviates these bottlenecks for edge deployment, static quantization approaches remain suboptimal for VLAs due to two critical challenges: (1) Temporal-dynamic sensitivity, where fixed precision wastes resources by ignoring stage-varying error tolerances; and (2) Real-time allocation, where identifying real-time sensitivity to guide bit allocation remains unsolved. To address these challenges, we propose DyQ-VLA, a dynamic quantization framework for VLAs. Specifically, a sensitivity-aware switching strategy leverages real-time kinematic proxies to trigger the bit-width switch, while a kinematic-guided module dynamically allocates the optimal bit-width. Experiments show that DyQ-VLA requires only 30.9% of the original memory footprint while maintaining 99.5% of its original performance, achieving 1.49x simulation and up to 1.43x real-world speedups.
♻ ☆ IRIS-SLAM: Unified Geo-Instance Representations for Robust Semantic Localization and Mapping
Geometry foundation models have significantly advanced dense geometric SLAM, yet existing systems often lack deep semantic understanding and robust loop closure capabilities. Meanwhile, contemporary semantic mapping approaches are frequently hindered by decoupled architectures and fragile data association. We propose IRIS-SLAM, a novel RGB semantic SLAM system that leverages unified geometric-instance representations derived from an instance-extended foundation model. By extending a geometry foundation model to concurrently predict dense geometry and cross-view consistent instance embeddings, we enable a semantic-synergized association mechanism and instance-guided loop closure detection. Our approach effectively utilizes viewpoint-agnostic semantic anchors to bridge the gap between geometric reconstruction and open-vocabulary mapping. Experimental results demonstrate that IRIS-SLAM significantly outperforms state-of-the-art methods, particularly in map consistency and wide-baseline loop closure reliability.
comment: This version is being withdrawn because it was submitted without the final review and formal approval of all co-authors. The authors plan to resubmit a revised version once all internal approvals are secured
♻ ☆ Humanoid Goalkeeper: Learning from Position Conditioned Task-Motion Constraints
We present a reinforcement learning framework for autonomous goalkeeping with humanoid robots in real-world scenarios. While prior work has demonstrated similar capabilities on quadrupedal platforms, humanoid goalkeeping introduces two critical challenges: (1) generating natural, human-like whole-body motions, and (2) covering a wider guarding range with an equivalent response time. Unlike existing approaches that rely on separate teleoperation or fixed motion tracking for whole-body control, our method learns a single end-to-end RL policy, enabling fully autonomous, highly dynamic, and human-like robot-object interactions. To achieve this, we integrate multiple human motion priors conditioned on perceptual inputs into the RL training via an adversarial scheme. We demonstrate the effectiveness of our method through real-world experiments, where the humanoid robot successfully performs agile, autonomous, and naturalistic interceptions of fast-moving balls. In addition to goalkeeping, we demonstrate the generalization of our approach through tasks such as ball escaping and grabbing. Our work presents a practical and scalable solution for enabling highly dynamic interactions between robots and moving objects, advancing the field toward more adaptive and lifelike robotic behaviors.
♻ ☆ VLD: Visual Language Goal Distance for Reinforcement Learning Navigation
Training end-to-end policies from image data to directly predict navigation actions for robotic systems has proven inherently difficult. Existing approaches often suffer from either the sim-to-real gap during policy transfer or a limited amount of training data with action labels. To address this problem, we introduce Vision-Language Distance (VLD) learning, a scalable framework for goal-conditioned navigation that decouples perception learning from policy learning. Instead of relying on raw sensory inputs during policy training, we first train a self-supervised distance-to-goal predictor on internet-scale video data. This predictor generalizes across both image- and text-based goals, providing a distance signal that can be minimized by a reinforcement learning (RL) policy. The RL policy can be trained entirely in simulation using privileged geometric distance signals, with injected noise to mimic the uncertainty of the trained distance predictor. At deployment, the policy consumes VLD predictions, inheriting semantic goal information-"where to go"-from large-scale visual training while retaining the robust low-level navigation behaviors learned in simulation. We propose using ordinal consistency to assess distance functions directly and demonstrate that VLD outperforms prior temporal distance approaches, such as ViNT and VIP. Experiments show that our decoupled design achieves competitive navigation performance in simulation with strong sim-to-real transfer, providing an alternative and, most importantly, scalable path toward reliable, multimodal navigation policies.
♻ ☆ Balancing Safety and Optimality in Robot Path Planning: Algorithm and Metric
Path planning for autonomous robots faces a fundamental trade-off between path length and obstacle clearance. While existing algorithms typically prioritize a single objective, we introduce the Unified Path Planner (UPP), a graph-search algorithm that dynamically balances safety and optimality via adaptive heuristic weighting. UPP employs a local inverse-distance safety field and auto-tunes its parameters based on real-time search progress, achieving provable suboptimality bounds while maintaining superior clearance. To enable rigorous evaluation, we introduce the OptiSafe index, a normalized metric that quantifies the trade-off between safety and optimality. Extensive evaluation across 10 environments shows that UPP achieves a 0.94 OptiSafe score in cluttered environments, compared with 0.22-0.85 for existing methods, with only 0.5-1% path-length overhead in simulation and a 100% success rate. Hardware validation on TurtleBot confirms practical advantages despite sim-to-real gaps.
comment: 26 pages
♻ ☆ GM3: A General Physical Model for Micro-Mobility Vehicles
Modeling the dynamics of micro-mobility vehicles (MMV) is becoming increasingly important for training autonomous vehicle systems and building urban traffic simulations. However, mainstream tools rely on variants of the Kinematic Bicycle Model (KBM) or mode-specific physics that miss tire slip, load transfer, and rider/vehicle lean. To our knowledge, no unified, physics-based model captures these dynamics across the full range of common MMVs and wheel layouts. We propose the "Generalized Micro-mobility Model" (GM3), a tire-level formulation based on the tire brush representation that supports arbitrary wheel configurations, including single/double track and multi-wheel platforms. We introduce an interactive model-agnostic simulation framework that decouples vehicle/layout specification from dynamics to compare the GM3 with the KBM and other models, consisting of fixed step RK4 integration, human-in-the-loop and scripted control, real-time trajectory traces and logging for analysis. We also empirically validate the GM3 on the Stanford Drone Dataset's deathCircle (roundabout) scene for biker, skater, and cart classes.
♻ ☆ Decoupled Action Expert: Confining Task Knowledge to the Conditioning Pathway
Many recent Vision-Language-Action models employ diffusion or flow-matching backbones with hundreds of millions of parameters for action generation. However, unlike image synthesis where the output spans millions of diverse pixels, a manipulation policy generates only short sequences of low-dimensional, physically correlated action values, a far simpler target that should not demand such capacity. We confirm this intuition and show that task-specific knowledge in these policies can be fully confined to the conditioning pathway, leaving the action backbone task-agnostic. To establish this, we introduce a decoupled training recipe: a general-purpose action head is first pretrained on observation-free forward-kinematics data, then frozen while only the conditioning pathway is trained for downstream tasks. Using Diffusion Policy as a testbed, we show that on both MimicGen and LIBERO, a single frozen backbone shared across all tasks matches normally trained counterparts. This confirms that the action expert encodes little task-specific knowledge. Ablations show that the specific pretraining signal (joint positions, end-effector poses, or no conditioning at all) has no effect on downstream performance, indicating that the backbone learns only general trajectory structure. Pushing this finding further, we replace the 244M U-Net in Diffusion Policy with a 5M-parameter MLP backbone that matches or exceeds its performance, calling into question the large capacity budgets allocated to action generation in current VLA designs.
♻ ☆ Hierarchical Diffusion Motion Planning with Task-Conditioned Uncertainty-Aware Priors
We propose a novel hierarchical diffusion planner that embeds task and motion structure directly into the noise model. Unlike standard diffusion-based planners that rely on zero-mean, isotropic Gaussian corruption, we introduce task-conditioned structured Gaussians whose means and covariances are derived from Gaussian Process Motion Planning (GPMP), explicitly encoding trajectory smoothness and task semantics in the prior. We first generalize the standard diffusion process to biased, non-isotropic corruption with closed-form forward and posterior expressions. Building on this formulation, our hierarchical design separates prior instantiation from trajectory denoising. At the upper level, the model predicts sparse, task-centric key states and their associated timings, which instantiate a structured Gaussian prior (mean and covariance). At the lower level, the full trajectory is denoised under this fixed prior, treating the upper-level outputs as noisy observations. Experiments on Maze2D goal-reaching and KUKA block stacking show consistently higher success rates and smoother trajectories than isotropic baselines, achieving dataset-level smoothness substantially earlier during training. Ablation studies further show that explicitly structuring the corruption process provides benefits beyond neural conditioning the denoising network alone. Overall, our approach concentrates the prior's probability mass near feasible and semantically meaningful trajectories. Our project page is available at https://hta-diffusion.github.io.
♻ ☆ Graphite: A GPU-Accelerated Mixed-Precision Graph Optimization Framework ICRA 2026
We present Graphite, a GPU-accelerated nonlinear least squares graph optimization framework. It provides a CUDA C++ interface to enable the sharing of code between a real-time application, such as a SLAM system, and its optimization tasks. The framework supports techniques to reduce memory usage, including in-place optimization, support for multiple floating point types and mixed-precision modes, and dynamically computed Jacobians. We evaluate Graphite on well-known bundle adjustment problems and find that it achieves similar performance to MegBA, a solver specialized for bundle adjustment, while maintaining generality and using less memory. We also apply Graphite to global visual-inertial bundle adjustment on maps generated from stereo-inertial SLAM datasets, and observe speed-ups of up to 59x compared to a CPU baseline. Our results indicate that our framework enables faster large-scale optimization on both desktop and resource-constrained devices.
comment: Accepted to ICRA 2026
♻ ☆ Optimal Modified Feedback Strategies in LQ Games under Control Imperfections
Game-theoretic approaches and Nash equilibrium have been widely applied across various engineering domains. However, practical challenges such as disturbances, delays, and actuator limitations can hinder the precise execution of Nash equilibrium strategies. This work investigates the impact of such implementation imperfections on game trajectories and players' costs in the context of a two-player finite-horizon linear quadratic (LQ) nonzero-sum game. Specifically, we analyze how small deviations by one player, measured or estimated at each stage affect the state trajectory and the other player's cost. To mitigate these effects, we construct a compensation law for the influenced player by augmenting the nominal game with the measurable deviation dynamics. The resulting policy is shown to be optimal within a causal affine policy class, and, for sufficiently small deviations, it locally outperforms the uncompensated equilibrium-derived feedback. Rigorous analysis and proofs are provided, and the effectiveness of the proposed approach is demonstrated through a representative numerical example.
comment: 8 pages, 2 figures, Manuscript accepted to ACC 2026
Robotics 76
☆ PhysMoDPO: Physically-Plausible Humanoid Motion with Preference Optimization
Recent progress in text-conditioned human motion generation has been largely driven by diffusion models trained on large-scale human motion data. Building on this progress, recent methods attempt to transfer such models for character animation and real robot control by applying a Whole-Body Controller (WBC) that converts diffusion-generated motions into executable trajectories. While WBC trajectories become compliant with physics, they may expose substantial deviations from original motion. To address this issue, we here propose PhysMoDPO, a Direct Preference Optimization framework. Unlike prior work that relies on hand-crafted physics-aware heuristics such as foot-sliding penalties, we integrate WBC into our training pipeline and optimize diffusion model such that the output of WBC becomes compliant both with physics and original text instructions. To train PhysMoDPO we deploy physics-based and task-specific rewards and use them to assign preference to synthesized trajectories. Our extensive experiments on text-to-motion and spatial control tasks demonstrate consistent improvements of PhysMoDPO in both physical realism and task-related metrics on simulated robots. Moreover, we demonstrate that PhysMoDPO results in significant improvements when applied to zero-shot motion transfer in simulation and for real-world deployment on a G1 humanoid robot.
☆ DecoVLN: Decoupling Observation, Reasoning, and Correction for Vision-and-Language Navigation CVPR2026
Vision-and-Language Navigation (VLN) requires agents to follow long-horizon instructions and navigate complex 3D environments. However, existing approaches face two major challenges: constructing an effective long-term memory bank and overcoming the compounding errors problem. To address these issues, we propose DecoVLN, an effective framework designed for robust streaming perception and closed-loop control in long-horizon navigation. First, we formulate long-term memory construction as an optimization problem and introduce adaptive refinement mechanism that selects frames from a historical candidate pool by iteratively optimizing a unified scoring function. This function jointly balances three key criteria: semantic relevance to the instruction, visual diversity from the selected memory, and temporal coverage of the historical trajectory. Second, to alleviate compounding errors, we introduce a state-action pair-level corrective finetuning strategy. By leveraging geodesic distance between states to precisely quantify deviation from the expert trajectory, the agent collects high-quality state-action pairs in the trusted region while filtering out the polluted data with low relevance. This improves both the efficiency and stability of error correction. Extensive experiments demonstrate the effectiveness of DecoVLN, and we have deployed it in real-world environments.
comment: 16 pages, 8 figures, CVPR2026
☆ Panoramic Multimodal Semantic Occupancy Prediction for Quadruped Robots
Panoramic imagery provides holistic 360° visual coverage for perception in quadruped robots. However, existing occupancy prediction methods are mainly designed for wheeled autonomous driving and rely heavily on RGB cues, limiting their robustness in complex environments. To bridge this gap, (1) we present PanoMMOcc, the first real-world panoramic multimodal occupancy dataset for quadruped robots, featuring four sensing modalities across diverse scenes. (2) We propose a panoramic multimodal occupancy perception framework, VoxelHound, tailored for legged mobility and spherical imaging. Specifically, we design (i) a Vertical Jitter Compensation (VJC) module to mitigate severe viewpoint perturbations caused by body pitch and roll during mobility, enabling more consistent spatial reasoning, and (ii) an effective Multimodal Information Prompt Fusion (MIPF) module that jointly leverages panoramic visual cues and auxiliary modalities to enhance volumetric occupancy prediction. (3) We establish a benchmark based on PanoMMOcc and provide detailed data analysis to enable systematic evaluation of perception methods under challenging embodied scenarios. Extensive experiments demonstrate that VoxelHound achieves state-of-the-art performance on PanoMMOcc (+4.16%} in mIoU). The dataset and code will be publicly released to facilitate future research on panoramic multimodal 3D perception for embodied robotic systems at https://github.com/SXDR/PanoMMOcc, along with the calibration tools released at https://github.com/losehu/CameraLiDAR-Calib.
comment: The dataset and code will be publicly released at https://github.com/SXDR/PanoMMOcc
☆ A Feasibility-Enhanced Control Barrier Function Method for Multi-UAV Collision Avoidance
This paper presents a feasibility-enhanced control barrier function (FECBF) framework for multi-UAV collision avoidance. In dense multi-UAV scenarios, the feasibility of the CBF quadratic program (CBF-QP) can be compromised due to internal incompatibility among multiple CBF constraints. To address this issue, we analyze the internal compatibility of CBF constraints and derive a sufficient condition for internal compatibility. Based on this condition, a sign-consistency constraint is introduced to mitigate internal incompatibility. The proposed constraint is incorporated into a decentralized CBF-QP formulation using worst-case estimates and slack variables. Simulation results demonstrate that the proposed method significantly reduces infeasibility and improves collision avoidance performance compared with existing baselines in dense scenarios. Additional simulations under varying time delays demonstrate the robustness of the proposed method. Real-world experiments validate the practical applicability of the proposed method.
☆ Evaluating VLMs' Spatial Reasoning Over Robot Motion: A Step Towards Robot Planning with Motion Preferences ICLR 2026
Understanding user instructions and object spatial relations in surrounding environments is crucial for intelligent robot systems to assist humans in various tasks. The natural language and spatial reasoning capabilities of Vision-Language Models (VLMs) have the potential to enhance the generalization of robot planners on new tasks, objects, and motion specifications. While foundation models have been applied to task planning, it is still unclear the degree to which they have the capability of spatial reasoning required to enforce user preferences or constraints on motion, such as desired distances from objects, topological properties, or motion style preferences. In this paper, we evaluate the capability of four state-of-the-art VLMs at spatial reasoning over robot motion, using four different querying methods. Our results show that, with the highest-performing querying method, Qwen2.5-VL achieves 71.4% accuracy zero-shot and 75% on a smaller model after fine-tuning, and GPT-4o leads to lower performance. We evaluate two types of motion preferences (object-proximity and path-style), and we also analyze the trade-off between accuracy and computation cost in number of tokens. This work shows some promise in the potential of VLM integration with robot motion planning pipelines.
comment: Accepted to the First Workshop on Efficient Spatial Reasoning at ICLR 2026
☆ SldprtNet: A Large-Scale Multimodal Dataset for CAD Generation in Language-Driven 3D Design ICRA 2026
We introduce SldprtNet, a large-scale dataset comprising over 242,000 industrial parts, designed for semantic-driven CAD modeling, geometric deep learning, and the training and fine-tuning of multimodal models for 3D design. The dataset provides 3D models in both .step and .sldprt formats to support diverse training and testing. To enable parametric modeling and facilitate dataset scalability, we developed supporting tools, an encoder and a decoder, which support 13 types of CAD commands and enable lossless transformation between 3D models and a structured text representation. Additionally, each sample is paired with a composite image created by merging seven rendered views from different viewpoints of the 3D model, effectively reducing input token length and accelerating inference. By combining this image with the parameterized text output from the encoder, we employ the lightweight multimodal language model Qwen2.5-VL-7B to generate a natural language description of each part's appearance and functionality. To ensure accuracy, we manually verified and aligned the generated descriptions, rendered images, and 3D models. These descriptions, along with the parameterized modeling scripts, rendered images, and 3D model files, are fully aligned to construct SldprtNet. To assess its effectiveness, we fine-tuned baseline models on a dataset subset, comparing image-plus-text inputs with text-only inputs. Results confirm the necessity and value of multimodal datasets for CAD generation. It features carefully selected real-world industrial parts, supporting tools for scalable dataset expansion, diverse modalities, and ensured diversity in model complexity and geometric features, making it a comprehensive multimodal dataset built for semantic-driven CAD modeling and cross-modal learning.
comment: Accept by ICRA 2026
☆ InterEdit: Navigating Text-Guided Multi-Human 3D Motion Editing
Text-guided 3D motion editing has seen success in single-person scenarios, but its extension to multi-person settings is less explored due to limited paired data and the complexity of inter-person interactions. We introduce the task of multi-person 3D motion editing, where a target motion is generated from a source and a text instruction. To support this, we propose InterEdit3D, a new dataset with manual two-person motion change annotations, and a Text-guided Multi-human Motion Editing (TMME) benchmark. We present InterEdit, a synchronized classifier-free conditional diffusion model for TMME. It introduces Semantic-Aware Plan Token Alignment with learnable tokens to capture high-level interaction cues and an Interaction-Aware Frequency Token Alignment strategy using DCT and energy pooling to model periodic motion dynamics. Experiments show that InterEdit improves text-to-motion consistency and edit fidelity, achieving state-of-the-art TMME performance. The dataset and code will be released at https://github.com/YNG916/InterEdit.
comment: The dataset and code will be released at https://github.com/YNG916/InterEdit
☆ ESPIRE: A Diagnostic Benchmark for Embodied Spatial Reasoning of Vision-Language Models
A recent trend in vision-language models (VLMs) has been to enhance their spatial cognition for embodied domains. Despite progress, existing evaluations have been limited both in paradigm and in coverage, hindering rapid, iterative model development. To address these limitations, we propose ESPIRE, a diagnostic benchmark for embodied spatial reasoning. ESPIRE offers a simulated world that physically grounds VLMs and evaluates them on spatial-reasoning-centric robotic tasks, thus narrowing the gap between evaluation and real-world deployment. To adapt VLMs to robotic tasks, we decompose each task into localization and execution, and frame both as generative problems, in stark contrast to predominant discriminative evaluations (e.g., via visual-question answering) that rely on distractors and discard execution. This decomposition further enables a fine-grained analysis beyond passive spatial reasoning toward reasoning to act. We systematically design ESPIRE both at the instruction level and at the environment level, ensuring broad coverage of spatial reasoning scenarios. We use ESPIRE to diagnose a range of frontier VLMs and provide in-depth analysis of their spatial reasoning behaviors.
☆ From Passive Monitoring to Active Defence: Resilient Control of Manipulators Under Cyberattacks
Cyber-physical robotic systems are vulnerable to false data injection attacks (FDIAs), in which an adversary corrupts sensor signals while evading residual-based passive anomaly detectors such as the chi-squared test. Such stealthy attacks can induce substantial end-effector deviations without triggering alarms. This paper studies the resilience of redundant manipulators to stealthy FDIAs and advances the architecture from passive monitoring to active defence. We formulate a closed-loop model comprising a feedback-linearized manipulator, a steady-state Kalman filter, and a chi-squared-based anomaly detector. Building on this passive monitoring layer, we propose an active control-level defence that attenuates the control input through a monotone function of an anomaly score generated by a novel actuation-projected, measurement-free state predictor. The proposed design provides probabilistic guarantees on nominal actuation loss and preserves closed-loop stability. From the attacker perspective, we derive a convex QCQP for computing one-step optimal stealthy attacks. Simulations on a 6-DOF planar manipulator show that the proposed defence significantly reduces attack-induced end-effector deviation while preserving nominal task performance in the absence of attacks.
☆ Route Fragmentation Based on Resource-centric Prioritisation for Efficient Multi-Robot Path Planning in Agricultural Environments
Agricultural environments present high proportions of spatially dense navigation bottlenecks for long-term navigation and operational planning of agricultural mobile robots. The existing agent-centric multi-robot path planning (MRPP) approaches resolve conflicts from the perspective of agents, rather than from the resources under contention. Further, the density of such contentions limits the capabilities of spatial interleaving, a concept that many planners rely on to achieve high throughput. In this work, two variants of the priority-based Fragment Planner (FP) are presented as resource-centric MRPP algorithms that leverage route fragmentation to enable partial route progression and limit the impact of binary-based waiting. These approaches are evaluated in lifelong simulation over a 3.6km topological map representing a commercial polytunnel environment. Their performances are contrasted against 5 baseline algorithms with varying robotic fleet sizes. The Fragment Planners achieved significant gains in throughput compared with Prioritised Planning (PP) and Priority-Based Search (PBS) algorithms. They further demonstrated a task throughput of 95% of the optimal task throughput over the same time period. This work shows that, for long-term deployment of agricultural robots in corridor-dominant agricultural environments, resource-centric MRPP approaches are a necessity for high-efficacy operational planning.
comment: This work has been submitted to the IEEE for possible publication
☆ Language-Grounded Decoupled Action Representation for Robotic Manipulation CVPR2026
The heterogeneity between high-level vision-language understanding and low-level action control remains a fundamental challenge in robotic manipulation. Although recent methods have advanced task-specific action alignment, they often struggle to generate robust and accurate actions for novel or semantically related tasks. To address this, we propose the Language-Grounded Decoupled Action Representation (LaDA) framework, which leverages natural language as a semantic bridge to connect perception and control. LaDA introduces a fine-grained intermediate layer of three interpretable action primitives--translation, rotation, and gripper control--providing explicit semantic structure for low-level actions. It further employs a semantic-guided soft-label contrastive learning objective to align similar action primitives across tasks, enhancing generalization and motion consistency. An adaptive weighting strategy, inspired by curriculum learning, dynamically balances contrastive and imitation objectives for stable and effective training. Extensive experiments on simulated benchmarks (LIBERO and MimicGen) and real-world demonstrations validate that LaDA achieves strong performance and generalizes effectively to unseen or related tasks.
comment: Accepted by CVPR2026
☆ Efficient Real-World Autonomous Racing via Attenuated Residual Policy Optimization
Residual policy learning (RPL), in which a learned policy refines a static base policy using deep reinforcement learning (DRL), has shown strong performance across various robotic applications. Its effectiveness is particularly evident in autonomous racing, a domain that serves as a challenging benchmark for real-world DRL. However, deploying RPL-based controllers introduces system complexity and increases inference latency. We address this by introducing an extension of RPL named attenuated residual policy optimization ($α$-RPO). Unlike standard RPL, $α$-RPO yields a standalone neural policy by progressively attenuating the base policy, which initially serves to bootstrap learning. Furthermore, this mechanism enables a form of privileged learning, where the base policy is permitted to use sensor modalities not required for final deployment. We design $α$-RPO to integrate seamlessly with PPO, ensuring that the attenuated influence of the base controller is dynamically compensated during policy optimization. We evaluate $α$-RPO by building a framework for 1:10-scaled autonomous racing around it. In both simulation and zero-shot real-world transfer to Roboracer cars, $α$-RPO not only reduces system complexity but also improves driving performance compared to baselines - demonstrating its practicality for robotic deployment. Our code is available at: https://github.com/raphajaner/arpo_racing.
☆ ReMem-VLA: Empowering Vision-Language-Action Model with Memory via Dual-Level Recurrent Queries
Vision-language-action (VLA) models for closed-loop robot control are typically cast under the Markov assumption, making them prone to errors on tasks requiring historical context. To incorporate memory, existing VLAs either retrieve from a memory bank, which can be misled by distractors, or extend the frame window, whose fixed horizon still limits long-term retention. In this paper, we introduce ReMem-VLA, a Recurrent Memory VLA model equipped with two sets of learnable queries: frame-level recurrent memory queries for propagating information across consecutive frames to support short-term memory, and chunk-level recurrent memory queries for carrying context across temporal chunks for long-term memory. These queries are trained end-to-end to aggregate and maintain relevant context over time, implicitly guiding the model's decisions without additional training or inference cost. Furthermore, to enhance visual memory, we introduce Past Observation Prediction as an auxiliary training objective. Through extensive memory-centric simulation and real-world robot experiments, we demonstrate that ReMem-VLA exhibits strong memory capabilities across multiple dimensions, including spatial, sequential, episodic, temporal, and visual memory. ReMem-VLA significantly outperforms memory-free VLA baselines $π$0.5 and OpenVLA-OFT and surpasses MemoryVLA on memory-dependent tasks by a large margin.
comment: 14 pages, 6 figures
☆ Coordinated Manipulation of Hybrid Deformable-Rigid Objects in Constrained Environments
Coordinated robotic manipulation of deformable linear objects (DLOs), such as ropes and cables, has been widely studied; however, handling hybrid assemblies composed of both deformable and rigid elements in constrained environments remains challenging. This work presents a quasi-static optimization-based manipulation planner that employs a strain-based Cosserat rod model, extending rigid-body formulations to hybrid deformable linear objects (hDLO). The proposed planner exploits the compliance of deformable links to maneuver through constraints while achieving task-space objectives for the object that are unreachable with rigid tools. By leveraging a differentiable model with analytically derived gradients, the method achieves up to a 33x speedup over finite-difference baselines for inverse kinetostatic(IKS) problems. Furthermore, the subsequent trajectory optimization problem, warm-started using the IKS solution, is only practically realizable via analytical derivatives. The proposed algorithm is validated in simulation on various hDLO systems and experimentally on a three-link hDLO manipulated in a constrained environment using a dual-arm robotic system. Experimental results confirm the planner's accuracy, yielding an average deformation error of approximately 3 cm (5% of the deformable link length) between the desired and measured marker positions. Finally, the proposed optimal planner is compared against a sampling-based feasibility planner adapted to the strain-based formulation. The results demonstrate the effectiveness and applicability of the proposed approach for robotic manipulation of hybrid assemblies in constrained environments.
comment: 15 pages, 10 figures
☆ RoboStream: Weaving Spatio-Temporal Reasoning with Memory in Vision-Language Models for Robotics
Enabling reliable long-horizon robotic manipulation is a crucial step toward open-world embodied intelligence. However, VLM-based planners treat each step as an isolated observation-to-action mapping, forcing them to reinfer scene geometry from raw pixels at every decision point while remaining unaware of how prior actions have reshaped the environment. Despite strong short-horizon performance, these systems lack the spatio-temporal reasoning required for persistent geometric anchoring and memory of action-triggered state transitions. Without persistent state tracking, perceptual errors accumulate across the execution horizon, temporarily occluded objects are catastrophically forgotten, and these compounding failures lead to precondition violations that cascade through subsequent steps. In contrast, humans maintain a persistent mental model that continuously tracks spatial relations and action consequences across interactions rather than reconstructing them at each instant. Inspired by this human capacity for causal spatio-temporal reasoning with persistent memory, we propose RoboStream, a training-free framework that achieves geometric anchoring through Spatio-Temporal Fusion Tokens (STF-Tokens), which bind visual evidence to 3D geometric attributes for persistent object grounding, and maintains causal continuity via a Causal Spatio-Temporal Graph (CSTG) that records action-triggered state transitions across steps. This design enables the planner to trace causal chains and preserve object permanence under occlusion without additional training or fine-tuning. RoboStream achieves 90.5% on long-horizon RLBench and 44.4% on challenging real-world block-building tasks, where both SoFar and VoxPoser score 11.1%, demonstrating that spatio-temporal reasoning and causal memory are critical missing components for reliable long-horizon manipulation.
☆ MotionAnymesh: Physics-Grounded Articulation for Simulation-Ready Digital Twins
Converting static 3D meshes into interactable articulated assets is crucial for embodied AI and robotic simulation. However, existing zero-shot pipelines struggle with complex assets due to a critical lack of physical grounding. Specifically, ungrounded Vision-Language Models (VLMs) frequently suffer from kinematic hallucinations, while unconstrained joint estimation inevitably leads to catastrophic mesh inter-penetration during physical simulation. To bridge this gap, we propose MotionAnymesh, an automated zero-shot framework that seamlessly transforms unstructured static meshes into simulation-ready digital twins. Our method features a kinematic-aware part segmentation module that grounds VLM reasoning with explicit SP4D physical priors, effectively eradicating kinematic hallucinations. Furthermore, we introduce a geometry-physics joint estimation pipeline that combines robust type-aware initialization with physics-constrained trajectory optimization to rigorously guarantee collision-free articulation. Extensive experiments demonstrate that MotionAnymesh significantly outperforms state-of-the-art baselines in both geometric precision and dynamic physical executability, providing highly reliable assets for downstream applications.
comment: 5 figures
☆ GoalSwarm: Multi-UAV Semantic Coordination for Open-Vocabulary Object Navigation
Cooperative visual semantic navigation is a foundational capability for aerial robot teams operating in unknown environments. However, achieving robust open-vocabulary object-goal navigation remains challenging due to the computational constraints of deploying heavy perception models onboard and the complexity of decentralized multi-agent coordination. We present GoalSwarm, a fully decentralized multi-UAV framework for zero-shot semantic object-goal navigation. Each UAV collaboratively constructs a shared, lightweight 2D top-down semantic occupancy map by projecting depth observations from aerial vantage points, eliminating the computational burden of full 3D representations while preserving essential geometric and semantic structure. The core contributions of GoalSwarm are threefold: (1) integration of zero-shot foundation model -- SAM3 for open vocabulary detection and pixel-level segmentation, enabling open-vocabulary target identification without task-specific training; (2) a Bayesian Value Map that fuses multi-viewpoint detection confidences into a per-pixel goal-relevance distribution, enabling informed frontier scoring via Upper Confidence Bound (UCB) exploration; and (3) a decentralized coordination strategy combining semantic frontier extraction, cost-utility bidding with geodesic path costs, and spatial separation penalties to minimize redundant exploration across the swarm.
comment: 6 pages, 2 figures
☆ Consistent and Efficient MSCKF-based LiDAR-Inertial Odometry with Inferred Cluster-to-Plane Constraints for UAVs
Robust and accurate navigation is critical for Unmanned Aerial Vehicles (UAVs) especially for those with stringent Size, Weight, and Power (SWaP) constraints. However, most state-of-the-art (SOTA) LiDAR-Inertial Odometry (LIO) systems still suffer from estimation inconsistency and computational bottlenecks when deployed on such platforms. To address these issues, this paper proposes a consistent and efficient tightly-coupled LIO framework tailored for UAVs. Within the efficient Multi-State Constraint Kalman Filter (MSCKF) framework, we build coplanar constraints inferred from planar features observed across a sliding window. By applying null-space projection to sliding-window coplanar constraints, we eliminate the direct dependency on feature parameters in the state vector, thereby mitigating overconfidence and improving consistency. More importantly, to further boost the efficiency, we introduce a parallel voxel-based data association and a novel compact cluster-to-plane measurement model. This compact measurement model losslessly reduces observation dimensionality and significantly accelerating the update process. Extensive evaluations demonstrate that our method outperforms most state-of-the-art (SOTA) approaches by providing a superior balance of consistency and efficiency. It exhibits improved robustness in degenerate scenarios, achieves the lowest memory usage via its map-free nature, and runs in real-time on resource-constrained embedded platforms (e.g., NVIDIA Jetson TX2).
☆ Beyond Imitation: Reinforcement Learning Fine-Tuning for Adaptive Diffusion Navigation Policies
Diffusion-based robot navigation policies trained on large-scale imitation learning datasets, can generate multi-modal trajectories directly from the robot's visual observations, bypassing the traditional localization-mapping-planning pipeline and achieving strong zero-shot generalization. However, their performance remains constrained by the coverage of offline datasets, and when deployed in unseen settings, distribution shift often leads to accumulated trajectory errors and safety-critical failures. Adapting diffusion policies with reinforcement learning is challenging because their iterative denoising structure hinders effective gradient backpropagation, while also making the training of an additional value network computationally expensive and less stable. To address these issues, we propose a reinforcement learning fine-tuning framework tailored for diffusion-based navigation. The method leverages the inherent multi-trajectory sampling mechanism of diffusion models and adopts Group Relative Policy Optimization (GRPO), which estimates relative advantages across sampled trajectories without requiring a separate value network. To preserve pretrained representations while enabling adaptation, we freeze the visual encoder and selectively update the higher decoder layers and action head, enhancing safety-aware behaviors through online environmental feedback. On the PointGoal task in Isaac Sim, our approach improves the Success Rate from 52.0% to 58.7% and SPL from 0.49 to 0.54 on unseen scenes, while reducing collision frequency. Additional experiments show that the fine-tuned policy transfers zero-shot to a real quadruped platform and maintains stable performance in geometrically out-of-distribution environments, suggesting improved adaptability and safe generalization to new domains.
☆ AoI-FusionNet: Age-Aware Tightly Coupled Fusion of UWB-IMU under Sparse Ranging Conditions
Accurate motion tracking of snow particles in avalanche events requires robust localization in global navigation satellite system (GNSS)-denied outdoor environments. This paper introduces AoI-FusionNet, a tightly coupled deep learning-based fusion framework that directly combines raw ultra-wideband (UWB) time-of-flight (ToF) measurements with inertial measurement unit (IMU) data for 3D trajectory estimation. Unlike loose-coupled pipelines based on intermediate trilateration, the proposed approach operates directly on heterogeneous sensor inputs, enabling localization even under insufficient ranging availability. The framework integrates an Age-of-Information (AoI)-aware decay module to reduce the influence of stale UWB ranging measurements and a learned attention gating mechanism that adaptively balances the contribution of UWB and IMU modalities based on measurement availability and temporal freshness. To evaluate robustness under limited data and measurement variability, we apply a diffusion-based residual augmentation strategy during training, producing an augmented variant termed AoI-FusionNet-DGAN. We assess the performance of the proposed model using offline post-processing of real-world measurement data collected in an alpine environment and benchmark it against UWB multilateration and loose-coupled fusion baselines. The results demonstrate that AoI-FusionNet substantially reduces mean and tail localization errors under intermittent and degraded sensing conditions.
☆ SmoothTurn: Learning to Turn Smoothly for Agile Navigation with Quadrupedal Robots
Quadrupedal robots show great potential for valuable real-world applications such as fire rescue and industrial inspection. Such applications often require urgency and the ability to navigate agilely, which in turn demands the capability to change directions smoothly while running in high speed. Existing approaches for agile navigation typically learn a single-goal reaching policy by encouraging the robot to stay at the target position after reaching there. As a result, when the policy is used to reach sequential goals that require changing directions, it cannot anticipate upcoming maneuvers or maintain momentum across the switch of goals, thereby preventing the robot from fully exploiting its agility potential. In this work, we formulate the task as sequential local navigation, extending the single-goal-conditioned local navigation formulation in prior work. We then introduce SmoothTurn, a learning-based control framework that learns to turn smoothly while running rapidly for agile sequential local navigation. The framework adopts a novel sequential goal-reaching reward, an expanded observation space with a lookahead window for future goals, and an automatic goal curriculum that progressively expands the difficulty of sampled goal sequences based on the goal-reaching performance. The trained policy can be directly deployed on real quadrupedal robots with onboard sensors and computation. Both simulation and real-world empirical results show that SmoothTurn learns an agile locomotion policy that performs smooth turning across goals, with emergent behaviors such as controlling momentum when switching goals, facing towards the future goal in advance, and planning efficient paths. We have provided video demos of the learned motions in the supplementary materials. The source code and trained policies will be made available upon acceptance.
☆ Reinforcement Learning for Elliptical Cylinder Motion Control Tasks
The control of devices with limited input always bring attention to solve by research due to its difficulty and non-trival solution. For instance, the inverted pendulum is benchmarking problem in control theory and machine learning. In this work, we are focused on the elliptical cylinder and its motion under limited torque. The inspiration of the problem is from untethered magnetic devices, which due to distance have to operate with limited input torque. In this work, the main goal is to define the control problem of elliptic cylinder with limited input torque and solve it by Reinforcement Learning. As a classical baseline, we evaluate a two-stage controller composed of an energy-shaping swing-up law and a local Linear Quadratic Regulator (LQR) stabilizer around the target equilibrium. The swing-up controller increases the system's mechanical energy to drive the state toward a neighborhood of the desired equilibrium, a linearization of the nonlinear model yields an LQR that regulates the angle and angular-rate states to the target orientation with bounded input. This swing-up + LQR policy is a strong, interpretable reference for underactuated system and serves a point of comparison to the learned policy under identical limits and parameters. The solution shows that the learning is possible however, the different cases like stabilization in upward position or rotating of half turn are very difficult for increasing mass or ellipses with a strongly unequal perimeter ratio.
☆ FLUX: Accelerating Cross-Embodiment Generative Navigation Policies via Rectified Flow and Static-to-Dynamic Learning
Autonomous navigation requires a broad spectrum of skills, from static goal-reaching to dynamic social traversal, yet evaluation remains fragmented across disparate protocols. We introduce DynBench, a dynamic navigation benchmark featuring physically valid crowd simulation. Combined with existing static protocols, it supports comprehensive evaluation across six fundamental navigation tasks. Within this framework, we propose FLUX, the first flow-based unified navigation policy. By linearizing probability flow, FLUX replaces iterative denoising with straight-line trajectories, improving per-step inference efficiency by 47% over prior flow-based methods and 29% over diffusion-based ones. Following a static-to-dynamic curriculum, FLUX initially establishes geometric priors and is subsequently refined through reinforcement learning in dynamic social environments. This regime not only strengthens socially-aware navigation but also enhances static task robustness by capturing recovery behaviors through stochastic action distributions. FLUX achieves state-of-the-art performance across all tasks and demonstrates zero-shot sim-to-real transfer on wheeled, quadrupedal, and humanoid platforms without any fine-tuning.
comment: Project Page at this [Website](https://zeying-gong.github.io/projects/flux/)
☆ Motion-Specific Battery Health Assessment for Quadrotors Using High-Fidelity Battery Models ICRA
Quadrotor endurance is ultimately limited by battery behavior, yet most energy aware planning treats the battery as a simple energy reservoir and overlooks how flight motions induce dynamic current loads that accelerate battery degradation. This work presents an end to end framework for motion aware battery health assessment in quadrotors. We first design a wide range current sensing module to capture motion specific current profiles during real flights, preserving transient features. In parallel, a high fidelity battery model is calibrated using reference performance tests and a metaheuristic based on a degradation coupled electrochemical model.By simulating measured flight loads in the calibrated model, we systematically resolve how different flight motions translate into degradation modes loss of lithium inventory and loss of active material as well as internal side reactions. The results demonstrate that even when two flight profiles consume the same average energy, their transient load structures can drive different degradation pathways, emphasizing the need for motion-aware battery management that balances efficiency with battery degradation.
comment: 8 pages. Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
☆ PVI: Plug-in Visual Injection for Vision-Language-Action Models
VLA architectures that pair a pretrained VLM with a flow-matching action expert have emerged as a strong paradigm for language-conditioned manipulation. Yet the VLM, optimized for semantic abstraction and typically conditioned on static visual observations, tends to attenuate fine-grained geometric cues and often lacks explicit temporal evidence for the action expert. Prior work mitigates this by injecting auxiliary visual features, but existing approaches either focus on static spatial representations or require substantial architectural modifications to accommodate temporal inputs, leaving temporal information underexplored. We propose Plug-in Visual Injection (PVI), a lightweight, encoder-agnostic module that attaches to a pretrained action expert and injects auxiliary visual representations via zero-initialized residual pathways, preserving pretrained behavior with only single-stage fine-tuning. Using PVI, we obtain consistent gains over the base policy and a range of competitive alternative injection strategies, and our controlled study shows that temporal video features (V-JEPA2) outperform strong static image features (DINOv2), with the largest gains on multi-phase tasks requiring state tracking and coordination. Real-robot experiments on long-horizon bimanual cloth folding further demonstrate the practicality of PVI beyond simulation.
☆ Easy-IIL: Reducing Human Operational Burden in Interactive Imitation Learning via Assistant Experts
Interactive Imitation Learning (IIL) typically relies on extensive human involvement for both offline demonstration and online interaction. Prior work primarily focuses on reducing human effort in passive monitoring rather than active operation. Interestingly, structured model-based imitation approaches achieve comparable performance with significantly fewer demonstrations than end-to-end imitation learning policies in the low-data regime. However, these methods are typically surpassed by end-to-end policies as the data increases. Leveraging this insight, we propose Easy-IIL, a framework that utilizes off-the-shelf model-based imitation methods as an assistant expert to replace active human operation for the majority of data collection. The human expert only provides a single demonstration to initialize the assistant expert and intervenes in critical states where the task is approaching failure. Furthermore, Easy-IIL can maintain IIL performance by preserving both offline and online data quality. Extensive simulation and real-world experiments demonstrate that Easy-IIL significantly reduces human operational burden while maintaining performance comparable to mainstream IIL baselines. User studies further confirm that Easy-IIL reduces subjective workload on the human expert. Project page: https://sites.google.com/view/easy-iil
☆ Show, Don't Tell: Detecting Novel Objects by Watching Human Videos
How can a robot quickly identify and recognize new objects shown to it during a human demonstration? Existing closed-set object detectors frequently fail at this because the objects are out-of-distribution. While open-set detectors (e.g., VLMs) sometimes succeed, they often require expensive and tedious human-in-the-loop prompt engineering to uniquely recognize novel object instances. In this paper, we present a self-supervised system that eliminates the need for tedious language descriptions and expensive prompt engineering by training a bespoke object detector on an automatically created dataset, supervised by the human demonstration itself. In our approach, "Show, Don't Tell," we show the detector the specific objects of interest during the demonstration, rather than telling the detector about these objects via complex language descriptions. By bypassing language altogether, this paradigm enables us to quickly train bespoke detectors tailored to the relevant objects observed in human task demonstrations. We develop an integrated on-robot system to deploy our "Show, Don't Tell" paradigm of automatic dataset creation and novel object-detection on a real-world robot. Empirical results demonstrate that our pipeline significantly outperforms state-of-the-art detection and recognition methods for manipulated objects, leading to improved task completion for the robot.
☆ Conflict Mitigation in Shared Environments using Flow-Aware Multi-Agent Path Finding ICRA 2026
Deploying multi-robot systems in environments shared with dynamic and uncontrollable agents presents significant challenges, especially for large robot fleets. In such environments, individual robot operations can be delayed due to unforeseen conflicts with uncontrollable agents. While existing research primarily focuses on preserving the completeness of Multi-Agent Path Finding (MAPF) solutions considering delays, there is limited emphasis on utilizing additional environmental information to enhance solution quality in the presence of other dynamic agents. To this end, we propose Flow-Aware Multi-Agent Path Finding (FA-MAPF), a novel framework that integrates learned motion patterns of uncontrollable agents into centralized MAPF algorithms. Our evaluation, conducted on a diverse set of benchmark maps with simulated uncontrollable agents and on a real-world map with recorded human trajectories, demonstrates the effectiveness of FA-MAPF compared to state-of-the-art baselines. The experimental results show that FA-MAPF can consistently reduce conflicts with uncontrollable agents, up to 55%, without compromising task efficiency.
comment: To be presented at ICRA 2026
☆ AnchorVLA4D: an Anchor-Based Spatial-Temporal Vision-Language-Action Model for Robotic Manipulation
Since current Vision-Language-Action (VLA) systems suffer from limited spatial perception and the absence of memory throughout manipulation, we investigate visual anchors as a means to enhance spatial and temporal reasoning within VLA policies for robotic manipulation. Conventional VLAs generate actions by conditioning on a single current frame together with a language instruction. However, since the frame is encoded as a 2D image, it does not contain detailed spatial information, and the VLA similarly lacks any means to incorporate past context. As a result, it frequently forgets objects under occlusion and becomes spatially disoriented during the manipulation process. Thus, we propose AnchorVLA4D, a simple spatial-temporal VLA that augments the visual input with an anchor image to preserve the initial scene context throughout execution, and adds a lightweight spatial encoder that jointly processes the anchor and current frames to expose geometric relationships within an episode. Built on a Qwen2.5-VL backbone with a diffusion-based action head, AnchorVLA4D requires no additional sensing modalities (e.g., depth or point clouds) and introduces negligible inference overhead. Combining anchoring with a frozen pretrained spatial encoder yields further gains, realizing a 13.6% improvement on the Simpler WidowX benchmark and confirming the approach on real-world tasks, where it achieved an average success rate of 80%.
☆ Altered Thoughts, Altered Actions: Probing Chain-of-Thought Vulnerabilities in VLA Robotic Manipulation
Recent Vision-Language-Action (VLA) models increasingly adopt chain-of-thought (CoT) reasoning, generating a natural-language plan before decoding motor commands. This internal text channel between the reasoning module and the action decoder has received no adversarial scrutiny. We ask: which properties of this intermediate plan does the action decoder actually rely on, and can targeted corruption of the reasoning trace alone -- with all inputs left intact -- degrade a robot's physical task performance? We design a taxonomy of seven text corruptions organized into three attacker tiers (blind noise, mechanical-semantic, and LLM-adaptive) and apply them to a state-of-the-art reasoning VLA across 40 LIBERO tabletop manipulation tasks. Our results reveal a striking asymmetry: substituting object names in the reasoning trace reduces overall success rate by 8.3~percentage points (pp) -- reaching $-$19.3~pp on goal-conditioned tasks and $-$45~pp on individual tasks -- whereas sentence reordering, spatial-direction reversal, token noise, and even a 70B-parameter LLM crafting plausible-but-wrong plans all have negligible impact (within $\pm$4~pp). This asymmetry indicates that the action decoder depends on entity-reference integrity rather than reasoning quality or sequential structure. Notably, a sophisticated LLM-based attacker underperforms simple mechanical object-name substitution, because preserving plausibility inadvertently retains the entity-grounding structure the decoder needs. A cross-architecture control using a non-reasoning VLA confirms the vulnerability is exclusive to reasoning-augmented models, while instruction-level attacks degrade both architectures -- establishing that the internal reasoning trace is a distinct and stealthy threat vector invisible to input-validation defenses.
☆ HaltNav: Reactive Visual Halting over Lightweight Topological Priors for Robust Vision-Language Navigation
Vision-and-Language Navigation (VLN) is shifting from rigid, step-by-step instruction following toward open-vocabulary, goal-oriented autonomy. Achieving this transition without exhaustive routing prompts requires agents to leverage structural priors. While prior work often assumes computationally heavy 2D/3D metric maps, we instead exploit a lightweight, text-based osmAG (OpenStreetMap Area Graph), a floorplan-level topological representation that is easy to obtain and maintain. However, global planning over a prior map alone is brittle in real-world deployments, where local connectivity can change (e.g., closed doors or crowded passages), leading to execution-time failures. To address this gap, we propose a hierarchical navigation framework HaltNav that couples the robust global planning of osmAG with the local exploration and instruction-grounding capability of VLN. Our approach features an MLLM-based brain module, which is capable of high-level task grounding and obstruction awareness. Conditioned on osmAG, the brain converts the global route into a sequence of localized execution snippets, providing the VLN executor with prior-grounded, goal-centric sub-instructions. Meanwhile, it detects local anomalies via a mechanism we term Reactive Visual Halting (RVH), which interrupts the local control loop, updates osmAG by invalidating the corresponding topology, and triggers replanning to orchestrate a viable detour. To train this halting capability efficiently, we introduce a data synthesis pipeline that leverages generative models to inject realistic obstacles into otherwise navigable scenes, substantially enriching hard negative samples. Extensive experiments demonstrate that our hierarchical framework outperforms several baseline methods without tedious language instructions, and significantly improves robustness for long-horizon vision-language navigation under environmental changes.
☆ Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data
Human athletes demonstrate versatile and highly-dynamic tennis skills to successfully conduct competitive rallies with a high-speed tennis ball. However, reproducing such behaviors on humanoid robots is difficult, partially due to the lack of perfect humanoid action data or human kinematic motion data in tennis scenarios as reference. In this work, we propose LATENT, a system that Learns Athletic humanoid TEnnis skills from imperfect human motioN daTa. The imperfect human motion data consist only of motion fragments that capture the primitive skills used when playing tennis rather than precise and complete human-tennis motion sequences from real-world tennis matches, thereby significantly reducing the difficulty of data collection. Our key insight is that, despite being imperfect, such quasi-realistic data still provide priors about human primitive skills in tennis scenarios. With further correction and composition, we learn a humanoid policy that can consistently strike incoming balls under a wide range of conditions and return them to target locations, while preserving natural motion styles. We also propose a series of designs for robust sim-to-real transfer and deploy our policy on the Unitree G1 humanoid robot. Our method achieves surprising results in the real world and can stably sustain multi-shot rallies with human players. Project page: https://zzk273.github.io/LATENT/
☆ TacVLA: Contact-Aware Tactile Fusion for Robust Vision-Language-Action Manipulation
Vision-Language-Action (VLA) models have demonstrated significant advantages in robotic manipulation. However, their reliance on vision and language often leads to suboptimal performance in tasks involving visual occlusion, fine-grained manipulation, and physical contact. To address these challenges, we propose TacVLA, a fine-tuned VLA model by incorporating tactile modalities into the transformer-based policy to enhance fine-grained manipulation capabilities. Specifically, we introduce a contact-aware gating mechanism that selectively activates tactile tokens only when contact is detected, enabling adaptive multimodal fusion while avoiding irrelevant tactile interference. The fused visual, language, and tactile tokens are jointly processed within the transformer architecture to strengthen cross-modal grounding during contact-rich interaction. Extensive experiments on constraint-locked disassembly, in-box picking and robustness evaluations demonstrate that our model outperforms baselines, improving the performance by averaging 20% success rate in disassembly, 60% in in-box picking and 2.1x improvement in scenarios with visual occlusion. Videos are available at https://sites.google.com/view/tacvla and code will be released.
comment: 9 pages, 7 figures
☆ Learning Geometric and Photometric Features from Panoramic LiDAR Scans for Outdoor Place Categorization
Semantic place categorization, which is one of the essential tasks for autonomous robots and vehicles, allows them to have capabilities of self-decision and navigation in unfamiliar environments. In particular, outdoor places are more difficult targets than indoor ones due to perceptual variations, such as dynamic illuminance over twenty-four hours and occlusions by cars and pedestrians. This paper presents a novel method of categorizing outdoor places using convolutional neural networks (CNNs), which take omnidirectional depth/reflectance images obtained by 3D LiDARs as the inputs. First, we construct a large-scale outdoor place dataset named Multi-modal Panoramic 3D Outdoor (MPO) comprising two types of point clouds captured by two different LiDARs. They are labeled with six outdoor place categories: coast, forest, indoor/outdoor parking, residential area, and urban area. Second, we provide CNNs for LiDAR-based outdoor place categorization and evaluate our approach with the MPO dataset. Our results on the MPO dataset outperform traditional approaches and show the effectiveness in which we use both depth and reflectance modalities. To analyze our trained deep networks we visualize the learned features.
comment: Published in Advanced Robotics on 31 Jul 2018
☆ Autonomous Integration and Improvement of Robotic Assembly using Skill Graph Representations
Robotic assembly systems traditionally require substantial manual engineering effort to integrate new tasks, adapt to new environments, and improve performance over time. This paper presents a framework for autonomous integration and continuous improvement of robotic assembly systems based on Skill Graph representations. A Skill Graph organizes robot capabilities as verb-based skills, explicitly linking semantic descriptions (verbs and nouns) with executable policies, pre-conditions, post-conditions, and evaluators. We show how Skill Graphs enable rapid system integration by supporting semantic-level planning over skills, while simultaneously grounding execution through well-defined interfaces to robot controllers and perception modules. After initial deployment, the same Skill Graph structure supports systematic data collection and closed-loop performance improvement, enabling iterative refinement of skills and their composition. We demonstrate how this approach unifies system configuration, execution, evaluation, and learning within a single representation, providing a scalable pathway toward adaptive and reusable robotic assembly systems. The code is at https://github.com/intelligent-control-lab/AIDF.
☆ CarPLAN: Context-Adaptive and Robust Planning with Dynamic Scene Awareness for Autonomous Driving
Imitation learning (IL) is widely used for motion planning in autonomous driving due to its data efficiency and access to real-world driving data. For safe and robust real-world driving, IL-based planning requires capturing the complex driving contexts inherent in real-world data and enabling context-adaptive decision-making, rather than relying solely on expert trajectory imitation. In this paper, we propose CarPLAN, a novel IL-based motion planning framework that explicitly enhances driving context understanding and enables adaptive planning across diverse traffic scenarios. Our contributions are twofold: We introduce Displacement-Aware Predictive Encoding (DPE) to improve the model's spatial awareness by predicting future displacement vectors between the Autonomous Vehicle (AV) and surrounding scene elements. This allows the planner to account for relational spacing when generating trajectories. In addition to the standard imitation loss, we incorporate an augmented loss term that captures displacement prediction errors, ensuring planning decisions consider relative distances from other agents. To improve the model's ability to handle diverse driving contexts, we propose Context-Adaptive Multi-Expert Decoder (CMD), which leverages the Mixture of Experts (MoE) framework. CMD dynamically selects the most suitable expert decoders based on scene structure at each Transformer layer, enabling adaptive and context-aware planning in dynamic environments. We evaluate CarPLAN on the nuPlan benchmark and demonstrate state-of-the-art performance across all closed-loop simulation metrics. In particular, CarPLAN exhibits robust performance on challenging scenarios such as Test14-Hard, validating its effectiveness in complex driving conditions. Additional experiments on the Waymax benchmark further demonstrate its generalization capability across different benchmark settings.
comment: 10 pages, 6 figures. Under review at IEEE Transactions on Intelligent Transportation Systems
☆ Early Pruning for Public Transport Routing
Routing algorithms for public transport, particularly the widely used RAPTOR and its variants, often face performance bottlenecks during the transfer relaxation phase, especially on dense transfer graphs, when supporting unlimited transfers. This inefficiency arises from iterating over many potential inter-stop connections (walks, bikes, e-scooters, etc.). To maintain acceptable performance, practitioners often limit transfer distances or exclude certain transfer options, which can reduce path optimality and restrict the multimodal options presented to travellers. This paper introduces Early Pruning, a low-overhead technique that accelerates routing algorithms without compromising optimality. By pre-sorting transfer connections by duration and applying a pruning rule within the transfer loop, the method discards longer transfers at a stop once they cannot yield an earlier arrival than the current best solution. Early Pruning can be integrated with minimal changes to existing codebases and requires only a one-time preprocessing step. Across multiple state-of-the-art RAPTOR-based solutions, including RAPTOR, ULTRA-RAPTOR, McRAPTOR, BM-RAPTOR, ULTRA-McRAPTOR, and UBM-RAPTOR and tested on the Switzerland and London transit networks, we achieved query time reductions of up to 57%. This approach provides a generalizable improvement to the efficiency of transit pathfinding algorithms. Beyond algorithmic performance, Early Pruning has practical implications for transport planning. By reducing computational costs, it enables transit agencies to expand transfer radii and incorporate additional mobility modes into journey planners without requiring extra server infrastructure. This is particularly relevant for passengers in areas with sparse direct transit coverage, such as outer suburbs and smaller towns, where richer multimodal routing can reveal viable alternatives to private car use.
☆ Skill-informed Data-driven Haptic Nudges for High-dimensional Human Motor Learning
In this work, we propose a data-driven skill-informed framework to design optimal haptic nudge feedback for high-dimensional novel motor learning tasks. We first model the stochastic dynamics of human motor learning using an Input-Output Hidden Markov Model (IOHMM), which explicitly decouples latent skill evolution from observable kinematic emissions. Leveraging this predictive model, we formulate the haptic nudge feedback design problem as a Partially Observable Markov Decision Process (POMDP). This allows us to derive an optimal nudging policy that minimizes long-term performance cost, implicitly guiding the learner toward robust regions of the skill space. We validated our approach through a human-subject study ($N=30$) using a high-dimensional hand-exoskeleton task. Results demonstrate that participants trained with the POMDP-derived policy exhibited significantly accelerated task performance compared to groups receiving heuristic-based feedback or no feedback. Furthermore, synergy analysis revealed that the POMDP group discovered efficient low-dimensional motor representations more rapidly.
☆ From Woofs to Words: Towards Intelligent Robotic Guide Dogs with Verbal Communication AAAI 2026
Assistive robotics is an important subarea of robotics that focuses on the well-being of people with disabilities. A robotic guide dog is an assistive quadruped robot that helps visually impaired people in obstacle avoidance and navigation. Enabling language capabilities for robotic guide dogs goes beyond naively adding an existing dialog system onto a mobile robot. The novel challenges include grounding language in the dynamically changing environment and improving spatial awareness for the human handler. To address those challenges, we develop a novel dialog system for robotic guide dogs that uses LLMs to verbalize both navigational plans and scenes. The goal is to enable verbal communication for collaborative decision-making within the handler-robot team. In experiments, we conducted a human study to evaluate different verbalization strategies and a simulation study to assess the efficiency and accuracy in navigation tasks.
comment: 10 pages, 6 figures, AAAI 2026
☆ Beyond Dense Futures: World Models as Structured Planners for Robotic Manipulation
Recent world-model-based Vision-Language-Action (VLA) architectures have improved robotic manipulation through predictive visual foresight. However, dense future prediction introduces visual redundancy and accumulates errors, causing long-horizon plan drift. Meanwhile, recent sparse methods typically represent visual foresight using high-level semantic subtasks or implicit latent states. These representations often lack explicit kinematic grounding, weakening the alignment between planning and low-level execution. To address this, we propose StructVLA, which reformulates a generative world model into an explicit structured planner for reliable control. Instead of dense rollouts or semantic goals, StructVLA predicts sparse, physically meaningful structured frames. Derived from intrinsic kinematic cues (e.g., gripper transitions and kinematic turning points), these frames capture spatiotemporal milestones closely aligned with task progress. We implement this approach through a two-stage training paradigm with a unified discrete token vocabulary: the world model is first trained to predict structured frames and subsequently optimized to map the structured foresight into low-level actions. This approach provides clear physical guidance and bridges visual planning and motion control. In our experiments, StructVLA achieves strong average success rates of 75.0% on SimplerEnv-WidowX and 94.8% on LIBERO. Real-world deployments further demonstrate reliable task completion and robust generalization across both basic pick-and-place and complex long-horizon tasks.
☆ Beyond Binary Success: Sample-Efficient and Statistically Rigorous Robot Policy Comparison
Generalist robot manipulation policies are becoming increasingly capable, but are limited in evaluation to a small number of hardware rollouts. This strong resource constraint in real-world testing necessitates both more informative performance measures and reliable and efficient evaluation procedures to properly assess model capabilities and benchmark progress in the field. This work presents a novel framework for robot policy comparison that is sample-efficient, statistically rigorous, and applicable to a broad set of evaluation metrics used in practice. Based on safe, anytime-valid inference (SAVI), our test procedure is sequential, allowing the evaluator to stop early when sufficient statistical evidence has accumulated to reach a decision at a pre-specified level of confidence. Unlike previous work developed for binary success, our unified approach addresses a wide range of informative metrics: from discrete partial credit task progress to continuous measures of episodic reward or trajectory smoothness, spanning both parametric and nonparametric comparison problems. Through extensive validation on simulated and real-world evaluation data, we demonstrate up to 70% reduction in evaluation burden compared to standard batch methods and up to 50% reduction compared to state-of-the-art sequential procedures designed for binary outcomes, with no loss of statistical rigor. Notably, our empirical results show that competing policies can be separated more quickly when using fine-grained task progress than binary success metrics.
comment: 12 + 9 pages, 2 + 5 figures,
☆ Egocentric World Model for Photorealistic Hand-Object Interaction Synthesis
To serve as a scalable data source for embodied AI, world models should act as true simulators that infer interaction dynamics strictly from user actions, rather than mere conditional video generators relying on privileged future object states. In this context, egocentric Human-Object Interaction (HOI) world models are critical for predicting physically grounded first-person rollouts. However, building such models is profoundly challenging due to rapid head motions, severe occlusions, and high-DoF hand articulations that abruptly alter contact topologies. Consequently, existing approaches often circumvent these physics challenges by resorting to conditional video generation with access to known future object trajectories. We introduce EgoHOI, an egocentric HOI world model that breaks away from this shortcut to simulate photorealistic, contact-consistent interactions from action signals alone. To ensure physical accuracy without future-state inputs, EgoHOI distills geometric and kinematic priors from 3D estimates into physics-informed embeddings. These embeddings regularize the egocentric rollouts toward physically valid dynamics. Experiments on the HOT3D dataset demonstrate consistent gains over strong baselines, and ablations validate the effectiveness of our physics-informed design.
☆ Sonar-MASt3R: Real-Time Opti-Acoustic Fusion in Turbid, Unstructured Environments ICRA 2026
Underwater intervention is an important capability in several marine domains, with numerous industrial, scientific, and defense applications. However, existing perception systems used during intervention operations rely on data from optical cameras, which limits capabilities in poor visibility or lighting conditions. Prior work has examined opti-acoustic fusion methods, which use sonar data to resolve the depth ambiguity of the camera data while using camera data to resolve the elevation angle ambiguity of the sonar data. However, existing methods cannot achieve dense 3D reconstructions in real-time, and few studies have reported results from applying these methods in a turbid environment. In this work, we propose the opti-acoustic fusion method Sonar-MASt3R, which uses MASt3R to extract dense correspondences from optical camera data in real-time and pairs it with geometric cues from an acoustic 3D reconstruction to ensure robustness in turbid conditions. Experimental results using data recorded from an opti-acoustic eye-in-hand configuration across turbidity values ranging from <0.5 to >12 NTU highlight this method's improved robustness to turbidity relative to baseline methods.
comment: This paper has been accepted for publication in ICRA 2026. Copyright IEEE
☆ Creating manufacturable blueprints for coarse-grained virtual robots
Over the past three decades, countless embodied yet virtual agents have freely evolved inside computer simulations, but vanishingly few were realized as physical robots. This is because evolution was conducted at a level of abstraction that was convenient for freeform body generation (creation, mutation, recombination) but swept away almost all of the physical details of functional body parts. The resulting designs were crude and underdetermined, requiring considerable effort and expertise to convert into a manufacturable format. Here, we automate this mapping from simplified design spaces that are readily evolvable to complete blueprints that can be directly followed by a builder. The pipeline incrementally resolves manufacturing constraints by embedding the structural and functional semantics of motors, electronics, batteries, and wiring into the abstract virtual design. In lieu of evolution, a user-defined or AI-generated ``sketch'' of a body plan can also be fed as input to the pipeline, providing a versatile framework for accelerating the design of novel robots.
☆ End-to-End O-RAN Testbed for Edge-AI-Enabled 5G/6G Connected Industrial Robotics
Connected robotics is one of the principal use cases driving the transition towards more intelligent and capable 6G mobile cellular networks. Replacing wired connections with highly reliable, high-throughput, and low-latency 5G/6G radio interfaces enables robotic system mobility and the offloading of compute-intensive artificial intelligence (AI) models for robotic perception and control to servers located at the network edge. The transition towards Edge AI as a Service (E-AIaaS) simplifies on-site maintenance of robotic systems and reduces operational costs in industrial environments, while supporting flexible AI model life-cycle management and seamless upgrades of robotic functionalities over time. In this paper, we present a 5G/6G O-RAN-based end-to-end testbed that integrates E-AIaaS for connected industrial robotic applications. The objective is to design and deploy a generic experimental platform based on open technologies and interfaces, demonstrated through an E-AIaaS-enabled autonomous welding scenario. Within this scenario, the testbed is used to investigate trade-offs among different data acquisition, edge processing, and real-time streaming approaches for robotic perception, while supporting emerging paradigms such as semantic and goal-oriented communications.
comment: Submitted to Global 6G Conference 2026
☆ Fabric Pneumatic Artificial Muscle-Based Head-Neck Exosuit: Design, Modeling, and Evaluation
Wearable exosuits assist human movement in tasks ranging from rehabilitation to daily activities; specifically, head-neck support is necessary for patients with certain neurological disorders. Rigid-link exoskeletons have shown to enable head-neck mobility compared to static braces, but their bulkiness and restrictive structure inspire designs using "soft" actuation methods. In this paper, we propose a fabric pneumatic artificial muscle-based exosuit design for head-neck support. We describe the design of our prototype and physics-based model, enabling us to derive actuator pressures required to compensate for gravitational load. Our modeled range of motion and workspace analysis indicate that the limited actuator lengths impose slight limitations (83% workspace coverage), and gravity compensation imposes a more significant limitation (43% workspace coverage). We introduce compression force along the neck as a novel, potentially comfort-related metric. We further apply our model to compare the torque output of various actuator placement configurations, allowing us to select a design with stability in lateral deviation and high axial rotation torques. The model correctly predicts trends in measured data where wrapping the actuators around the neck is not a significant factor. Our test dummy and human user demonstration confirm that the exosuit can provide functional head support and trajectory tracking, underscoring the potential of artificial muscle-based soft actuation for head-neck mobility assistance.
comment: Manuscript (8 pages, 5 tables, 7 figures) accepted to IEEE International Conference on Robotics and Automation 2026. Video attachment: https://youtu.be/iGuEbvCXgJ0?si=WqP2q-P_Mp1Brmfc
☆ Learning Actionable Manipulation Recovery via Counterfactual Failure Synthesis
While recent foundation models have significantly advanced robotic manipulation, these systems still struggle to autonomously recover from execution errors. Current failure-learning paradigms rely on either costly and unsafe real-world data collection or simulator-based perturbations, which introduce a severe sim-to-real gap. Furthermore, existing visual analyzers predominantly output coarse, binary diagnoses rather than the executable, trajectory-level corrections required for actual recovery. To bridge the gap between failure diagnosis and actionable recovery, we introduce Dream2Fix, a framework that synthesizes photorealistic, counterfactual failure rollouts directly from successful real-world demonstrations. By perturbing actions within a generative world model, Dream2Fix creates paired failure-correction data without relying on simulators. To ensure the generated data is physically viable for robot learning, we implement a structured verification mechanism that strictly filters rollouts for task validity, visual coherence, and kinematic safety. This engine produces a high-fidelity dataset of over 120k paired samples. Using this dataset, we fine-tune a vision-language model to jointly predict failure types and precise recovery trajectories, mapping visual anomalies directly to corrective actions. Extensive real-world robotic experiments show our approach achieves state-of-the-art correction accuracy, improving from 19.7% to 81.3% over prior baselines, and successfully enables zero-shot closed-loop failure recovery in physical deployments.
☆ Verification and Forward Invariance of Control Barrier Functions for Differential-Algebraic Systems
Differential-algebraic equations (DAEs) arise in power networks, chemical processes, and multibody systems, where algebraic constraints encode physical conservation laws. The safety of such systems is critical, yet safe control is challenging because algebraic constraints restrict allowable state trajectories. Control barrier functions (CBFs) provide computationally efficient safety filters for ordinary differential equation (ODE) systems. However, existing CBF methods are not directly applicable to DAEs due to potential conflicts between the CBF condition and the constraint manifold. This paper introduces DAE-aware CBFs that incorporate the differential-algebraic structure through projected vector fields. We derive conditions that ensure forward invariance of safe sets while preserving algebraic constraints and extend the framework to higher-index DAEs. A systematic verification framework is developed, establishing necessary and sufficient conditions for geometric correctness and feasibility of DAE-aware CBFs. For polynomial systems, sum-of-squares certificates are provided, while for nonpolynomial and neural network candidates, satisfiability modulo theories are used for falsification. The approach is validated on wind turbine and flexible-link manipulator systems.
☆ Safety-guaranteed and Goal-oriented Semantic Sensing, Communication, and Control for Robotics
Wirelessly-connected robotic system empowers robots with real-time intelligence by leveraging remote computing resources for decision-making. However, the data exchange between robots and base stations often overwhelms communication links, introducing latency that undermines real-time response. To tackle this, goal-oriented semantic communication (GSC) has been introduced into wirelessly-connected robotic systems to extract and transmit only goal-relevant semantic representations, enhancing communication efficiency and task effectiveness. However, existing GSC approaches focused primarily on optimizing effectiveness metrics while overlooking safety requirements, which should be treated as the top priority in real-world robotic systems. To bridge this gap, we propose safety-guaranteed and goal-oriented semantic communication for wirelessly-connected robotic system, aiming to maximize the robotic task effectiveness subject to practical operational safety requirements. We first summarize the general safety requirements and effectiveness metrics across typical robotic tasks, including robot arm grasping, unmanned aerial vehicle (UAV)-assisted tasks, and multi-robot exploration. We then systematically analyze the unique safety and effectiveness challenges faced by wirelessly-connected robotic system in sensing, communication, and control. Based on these, we further present potential safety-guaranteed and goal-oriented sensing, communication, and control solutions. Finally, a UAV target tracking case study validates that our proposed GSC solutions can significantly improve safety rate and tracking success rate by more than 2 times and 4.5 times, respectively.
comment: 7 pages. This paper has been submitted to the IEEE Communications Magazine
☆ Spatially Grounded Long-Horizon Task Planning in the Wild
Recent advances in robot manipulation increasingly leverage Vision-Language Models (VLMs) for high-level reasoning, such as decomposing task instructions into sequential action plans expressed in natural language that guide downstream low-level motor execution. However, current benchmarks do not assess whether these plans are spatially executable, particularly in specifying the exact spatial locations where the robot should interact to execute the plan, limiting evaluation of real-world manipulation capability. To bridge this gap, we define a novel task of grounded planning and introduce GroundedPlanBench, a newly curated benchmark for spatially grounded long-horizon action planning in the wild. GroundedPlanBench jointly evaluates hierarchical sub-action planning and spatial action grounding (where to act), enabling systematic assessment of whether generated sub-actions are spatially executable for robot manipulation. We further introduce Video-to-Spatially Grounded Planning (V2GP), an automated data generation framework that leverages real-world robot video demonstrations to improve spatially grounded long-horizon planning. Our evaluations reveal that spatially grounded long-horizon planning remains a major bottleneck for current VLMs. Our results demonstrate that V2GP provides a promising approach for improving both action planning and spatial grounding performance, validated on our benchmark as well as through real-world robot manipulation experiments, advancing progress toward spatially actionable planning.
comment: 9 pages, 7 figures
♻ ☆ Better Safe Than Sorry: Enhancing Arbitration Graphs for Safe and Robust Autonomous Decision-Making
This paper introduces an extension to the arbitration graph framework designed to enhance the safety and robustness of autonomous systems in complex, dynamic environments. Building on the flexibility and scalability of arbitration graphs, the proposed method incorporates a verification step and structured fallback layers in the decision-making process. This ensures that only verified and safe commands are executed while enabling graceful degradation in the presence of unexpected faults or bugs. The approach is demonstrated using a Pac-Man simulation and further validated in the context of autonomous driving, where it shows significant reductions in accident risk and improvements in overall system safety. The bottom-up design of arbitration graphs allows for an incremental integration of new behavior components. The extension presented in this work enables the integration of experimental or immature behavior components while maintaining system safety by clearly and precisely defining the conditions under which behaviors are considered safe. The proposed method is implemented as a ready to use header-only C++ library, published under the MIT License. Together with the Pac-Man demo, it is available at github.com/KIT-MRT/arbitration_graphs.
comment: 7 pages, 5 figures, Presented at 2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC), source code available at github.com/KIT-MRT/arbitration_graphs, v2: Added paragraph discussing the differences between arbitration graphs and behavior trees, v3: Updated version as presented at SMC
♻ ☆ RobotArena $\infty$: Scalable Robot Benchmarking via Real-to-Sim Translation
The pursuit of robot generalists, agents capable of performing diverse tasks across diverse environments, demands rigorous and scalable evaluation. Yet real-world testing of robot policies remains fundamentally constrained: it is labor-intensive, slow, unsafe at scale, and difficult to reproduce. As policies expand in scope and complexity, these barriers only intensify, since defining "success" in robotics often hinges on nuanced human judgments of execution quality. We introduce RobotArena Infinity, a new benchmarking framework that overcomes these challenges by shifting vision-language-action (VLA) evaluation into large-scale simulated environments augmented with online human feedback. Leveraging advances in vision-language models, 2D-to-3D generative modeling, and differentiable rendering, our approach automatically converts video demonstrations from widely used robot datasets into simulated counterparts. Within these digital twins, we assess VLA policies using both automated vision-language-model-guided scoring and scalable human preference judgments collected from crowdworkers, transforming human involvement from tedious scene setup, resetting, and safety supervision into lightweight preference comparisons. To measure robustness, we systematically perturb simulated environments along multiple axes, including textures and object placements, stress-testing policy generalization under controlled variation. The result is a continuously evolving, reproducible, and scalable benchmark for real-world-trained robot manipulation policies, addressing a critical missing capability in today's robotics landscape.
comment: Website: https://robotarenainf.github.io
♻ ☆ Accelerating Residual Reinforcement Learning with Uncertainty Estimation
Residual Reinforcement Learning (RL) is a popular approach for adapting pretrained policies by learning a lightweight residual policy that provides corrective actions. While Residual RL is more sample-efficient than finetuning the entire base policy, existing methods struggle with sparse rewards and are designed for deterministic base policies. We propose two improvements to Residual RL that further enhance its sample efficiency and make it suitable for stochastic base policies. First, we leverage uncertainty estimates of the base policy to focus exploration on regions in which the base policy is not confident. Second, we propose a simple modification to off-policy residual learning that allows it to observe base actions and better handle stochastic base policies. We evaluate our method with both Gaussian-based and Diffusion-based stochastic base policies on tasks from Robosuite and D4RL, and compare against state-of-the-art finetuning methods, demo-augmented RL methods, and other residual RL methods. Our algorithm significantly outperforms existing baselines in a variety of simulation benchmark environments. We also deploy our learned polices in the real world to demonstrate their robustness with zero-shot sim-to-real transfer. Paper homepage : lakshitadodeja.github.io/uncertainty-aware-residual-rl/
♻ ☆ SegDAC: Visual Generalization in Reinforcement Learning via Dynamic Object Tokens
Visual reinforcement learning policies trained on pixel observations often struggle to generalize when visual conditions change at test time. Object-centric representations are a promising alternative, but most approaches use fixed-size slot representations, require image reconstruction, or need auxiliary losses to learn object decompositions. As a result, it remains unclear how to learn RL policies directly from object-level inputs without these constraints. We propose SegDAC, a Segmentation-Driven Actor-Critic that operates on a variable-length set of object token embeddings. At each timestep, text-grounded segmentation produces object masks from which spatially aware token embeddings are extracted. A transformer-based actor-critic processes these dynamic tokens, using segment positional encoding to preserve spatial information across objects. We ablate these design choices and show that both segment positional encoding and variable-length processing are individually necessary for strong performance. We evaluate SegDAC on 8 ManiSkill3 manipulation tasks under 12 visual perturbation types across 3 difficulty levels. SegDAC improves over prior visual generalization methods by 15% on easy, 66% on medium, and 88% on the hardest settings. SegDAC matches the sample efficiency of the state-of-the-art visual RL methods while achieving improved generalization under visual changes. Project Page: https://segdac.github.io/
comment: 12 pages
♻ ☆ Dynamic Aware: Adaptive Multi-Mode Out-of-Distribution Detection for Trajectory Prediction in Autonomous Vehicles
Trajectory prediction is central to the safe and seamless operation of autonomous vehicles (AVs). In deployment, however, prediction models inevitably face distribution shifts between training data and real-world conditions, where rare or underrepresented traffic scenarios induce out-of-distribution (OOD) cases. While most prior OOD detection research in AVs has concentrated on computer vision tasks such as object detection and segmentation, trajectory-level OOD detection remains largely underexplored. A recent study formulated this problem as a quickest change detection (QCD) task, providing formal guarantees on the trade-off between detection delay and false alarms [1]. Building on this foundation, we propose a new framework that introduces adaptive mechanisms to achieve robust detection in complex driving environments. Empirical analysis across multiple real-world datasets reveals that prediction errors -- even on in-distribution samples -- exhibit mode-dependent distributions that evolve over time with dataset-specific dynamics. By explicitly modeling these error modes, our method achieves substantial improvements in both detection delay and false alarm rates. Comprehensive experiments on established trajectory prediction benchmarks show that our framework significantly outperforms prior UQ- and vision-based OOD approaches in both accuracy and computational efficiency, offering a practical path toward reliable, driving-aware autonomy.
comment: 8 pages, 7 figures
♻ ☆ DriveMind: A Dual Visual Language Model-based Reinforcement Learning Framework for Autonomous Driving
End-to-end autonomous driving systems map sensor data directly to control commands, but remain opaque, lack interpretability, and offer no formal safety guarantees. While recent vision-language-guided reinforcement learning (RL) methods introduce semantic feedback, they often rely on static prompts and fixed objectives, limiting adaptability to dynamic driving scenes. We present DriveMind, a unified semantic reward framework that integrates: (i) a contrastive Vision-Language Model (VLM) encoder for stepwise semantic anchoring; (ii) a novelty-triggered VLM encoder-decoder, fine-tuned via chain-of-thought (CoT) distillation, for dynamic prompt generation upon semantic drift; (iii) a hierarchical safety module enforcing kinematic constraints (e.g., speed, lane centering, stability); and (iv) a compact predictive world model to reward alignment with anticipated ideal states. DriveMind achieves 19.4 +/- 2.3 km/h average speed, 0.98 +/- 0.03 route completion, and near-zero collisions in CARLA Town 2, outperforming baselines by over 4% in success rate. Its semantic reward generalizes zero-shot to real dash-cam data with minimal distributional shift, demonstrating robust cross-domain alignment and potential for real-world deployment.
comment: Submitted to IEEE Transactions on Intelligent Vehicles (T-IV)
♻ ☆ Safe Interaction via Monte Carlo Linear-Quadratic Games
Safety is critical during human-robot interaction. But -- because people are inherently unpredictable -- it is often difficult for robots to plan safe behaviors. Instead of relying on our ability to anticipate humans, here we identify robot policies that are robust to unexpected human decisions. We achieve this by formulating human-robot interaction as a zero-sum game, where (in the worst case) the human's actions directly conflict with the robot's objective. Solving for the Nash Equilibrium of this game provides robot policies that maximize safety and performance across a wide range of human actions. Existing approaches attempt to find these optimal policies by leveraging Hamilton-Jacobi analysis (which is intractable) or linear-quadratic approximations (which are inexact). By contrast, in this work we propose a computationally efficient and theoretically justified method that converges towards the Nash Equilibrium policy. Our approach (which we call MCLQ) leverages linear-quadratic games to obtain an initial guess at safe robot behavior, and then iteratively refines that guess with a Monte Carlo search. Not only does MCLQ provide real-time safety adjustments, but it also enables the designer to tune how conservative the robot is -- preventing the system from focusing on unrealistic human behaviors. Our simulations and user study suggest that this approach advances safety in terms of both computation time and expected performance. See videos of our experiments here: https://youtu.be/KJuHeiWVuWY.
♻ ☆ Continuous Design and Reprogramming of Totimorphic Structures for Space Applications
Recently, a class of mechanical lattices with reconfigurable, zero-stiffness structures has been proposed, called Totimorphic lattices. In this work, we introduce a computational framework that enables continuous reprogramming of a Totimorphic lattice's effective properties, such as mechanical and optical behaviour, through geometric changes alone, demonstrated using computer simulations. Our approach is differentiable and guarantees valid Totimorphic configurations throughout the optimisation process, providing not only target states with desired properties but also continuous trajectories in configuration space that connect them. This enables reprogrammable structures in which actuators are controlled via automatic differentiation on an objective-dependent cost function, continuously adapting the lattice to achieve a given goal. We focus on deep space applications, where harsh and resource-constrained environments demand solutions that combine flexibility, efficiency, and autonomy. As proof of concept, we present two scenarios: a reprogrammable disordered lattice material and a space telescope mirror with adjustable focal length. The introduced framework is adaptable to a wide range of Totimorphic designs and objectives, providing a lightweight model for endowing physical systems with autonomous self-configuration and self-repair capabilities.
comment: Code: https://github.com/esa/LattyMorph/tree/main
♻ ☆ Beyond Static Instruction: A Multi-agent AI Framework for Adaptive Augmented Reality Robot Training
Augmented Reality (AR) offers powerful visualization capabilities for industrial robot training, yet current interfaces remain predominantly static, failing to account for learners' diverse cognitive profiles. In this paper, we present an AR application for robot training and propose a multi-agent AI framework for future integration that bridges the gap between static visualization and pedagogical intelligence. We report on the evaluation of the baseline AR interface with 36 participants performing a robotic pick-and-place task. While overall usability was high, notable disparities in task duration and learner characteristics highlighted the necessity for dynamic adaptation. To address this, we propose a multi-agent framework that orchestrates multiple components to perform complex preprocessing of multimodal inputs (e.g., voice, physiology, robot data) and adapt the AR application to the learner's needs. By utilizing autonomous Large Language Model (LLM) agents, the proposed system would dynamically adapt the learning environment based on advanced LLM reasoning in real-time.
♻ ☆ Reference-Free Sampling-Based Model Predictive Control
We present a sampling-based model predictive control (MPC) framework that enables emergent locomotion without relying on handcrafted gait patterns or predefined contact sequences. Our method discovers diverse motion patterns, ranging from trotting to galloping, robust standing policies, jumping, and handstand balancing, purely through the optimization of high-level objectives. Building on model predictive path integral (MPPI), we propose a cubic Hermite spline parameterization that operates on position and velocity control points. Our approach enables contact-making and contact-breaking strategies that adapt automatically to task requirements, requiring only a limited number of sampled trajectories. This sample efficiency enables real-time control on standard CPU hardware, eliminating the GPU acceleration typically required by other state-of-the-art MPPI methods. We validate our approach on the Go2 quadrupedal robot, demonstrating a range of emergent gaits and basic jumping capabilities. In simulation, we further showcase more complex behaviors, such as backflips, dynamic handstand balancing and locomotion on a Humanoid, all without requiring reference tracking or offline pre-training.
♻ ☆ IROSA: Interactive Robot Skill Adaptation using Natural Language
Foundation models have demonstrated impressive capabilities across diverse domains, while imitation learning provides principled methods for robot skill adaptation from limited data. Combining these approaches holds significant promise for direct application to robotics, yet this combination has received limited attention, particularly for industrial deployment. We present a novel framework that enables open-vocabulary skill adaptation through a tool-based architecture, maintaining a protective abstraction layer between the language model and robot hardware. Our approach leverages pre-trained LLMs to select and parameterize specific tools for adapting robot skills without requiring fine-tuning or direct model-to-robot interaction. We demonstrate the framework on a 7-DoF torque-controlled robot performing an industrial bearing ring insertion task, showing successful skill adaptation through natural language commands for speed adjustment, trajectory correction, and obstacle avoidance while maintaining safety, transparency, and interpretability.
comment: Accepted IEEE Robotics and Automation Letters (RA-L) journal, 8 pages, 5 figures, 3 tables, 1 listing
♻ ☆ Guided Policy Optimization under Partial Observability
Reinforcement Learning (RL) in partially observable environments poses significant challenges due to the complexity of learning under uncertainty. While additional information, such as that available in simulations, can enhance training, effectively leveraging it remains an open problem. To address this, we introduce Guided Policy Optimization (GPO), a framework that co-trains a guider and a learner. The guider takes advantage of privileged information while ensuring alignment with the learner's policy that is primarily trained via imitation learning. We theoretically demonstrate that this learning scheme achieves optimality comparable to direct RL, thereby overcoming key limitations inherent in existing approaches. Empirical evaluations show strong performance of GPO across various tasks, including continuous control with partial observability and noise, and memory-based challenges, significantly outperforming existing methods.
♻ ☆ NavForesee: A Unified Vision-Language World Model for Hierarchical Planning and Dual-Horizon Navigation Prediction
Embodied navigation for long-horizon tasks, guided by complex natural language instructions, remains a formidable challenge in artificial intelligence. Existing agents often struggle with robust long-term planning about unseen environments, leading to high failure rates. To address these limitations, we introduce NavForesee, a novel Vision-Language Model (VLM) that unifies high-level language planning and predictive world model imagination within a single, unified framework. Our approach empowers a single VLM to concurrently perform planning and predictive foresight. Conditioned on the full instruction and historical observations, the model is trained to understand the navigation instructions by decomposing the task, tracking its progress, and formulating the subsequent sub-goal. Simultaneously, it functions as a generative world model, providing crucial foresight by predicting short-term environmental dynamics and long-term navigation milestones. The VLM's structured plan guides its targeted prediction, while the imagined future provides rich context to inform the navigation actions, creating a powerful internal feedback loop of perception-planning/prediction-action. We demonstrate through extensive experiments on the R2R-CE and RxR-CE benchmark that NavForesee achieves highly competitive performance in complex scenarios. Our work highlights the immense potential of fusing explicit language planning with implicit spatiotemporal prediction, paving the way for more intelligent and capable embodied agents.
♻ ☆ CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions ICRA 2026
Reinforcement learning (RL), while powerful and expressive, can often prioritize performance at the expense of safety. Yet safety violations can lead to catastrophic outcomes in real-world deployments. Control Barrier Functions (CBFs) offer a principled method to enforce dynamic safety -- traditionally deployed online via safety filters. While the result is safe behavior, the fact that the RL policy does not have knowledge of the CBF can lead to conservative behaviors. This paper proposes CBF-RL, a framework for generating safe behaviors with RL by enforcing CBFs in training. CBF-RL has two key attributes: (1) minimally modifying a nominal RL policy to encode safety constraints via a CBF term, (2) and safety filtering of the policy rollouts in training. Theoretically, we prove that continuous-time safety filters can be deployed via closed-form expressions on discrete-time roll-outs. Practically, we demonstrate that CBF-RL internalizes the safety constraints in the learned policy -- both enforcing safer actions and biasing towards safer rewards -- enabling safe deployment without the need for an online safety filter. We validate our framework through ablation studies on navigation tasks and on the Unitree G1 humanoid robot, where CBF-RL enables safer exploration, faster convergence, and robust performance under uncertainty, enabling the humanoid robot to avoid obstacles and climb stairs safely in real-world settings without a runtime safety filter.
comment: To appear at ICRA 2026
♻ ☆ How Safe Will I Be Given What I Saw? Calibrated Prediction of Safety Chances for Image-Controlled Autonomy
Autonomous robots that rely on deep neural network controllers pose critical challenges for safety prediction, especially under partial observability and distribution shift. Traditional model-based verification techniques are limited in scalability and require access to low-dimensional state models, while model-free methods often lack reliability guarantees. This paper addresses these limitations by introducing a framework for calibrated safety prediction in end-to-end vision-controlled systems, where neither the state-transition model nor the observation model is accessible. Building on the foundation of world models, we leverage variational autoencoders and recurrent predictors to forecast future latent trajectories from raw image sequences and estimate the probability of satisfying safety properties. We distinguish between monolithic and composite prediction pipelines and introduce a calibration mechanism to quantify prediction confidence. In long-horizon predictions from high-dimensional observations, the forecasted inputs to the safety evaluator can deviate significantly from the training distribution due to compounding prediction errors and changing environmental conditions, leading to miscalibrated risk estimates. To address this, we incorporate unsupervised domain adaptation to ensure robustness of safety evaluation under distribution shift in predictions without requiring manual labels. Our formulation provides theoretical calibration guarantees and supports practical evaluation across long prediction horizons. Experimental results on three benchmarks show that our UDA-equipped evaluators maintain high accuracy and substantially lower false positive rates under distribution shift. Similarly, world model-based composite predictors outperform their monolithic counterparts on long-horizon tasks, and our conformal calibration provides reliable statistical bounds.
comment: arXiv admin note: text overlap with arXiv:2308.12252
♻ ☆ HumDex: Humanoid Dexterous Manipulation Made Easy
This paper investigates humanoid whole-body dexterous manipulation, where the efficient collection of high-quality demonstration data remains a central bottleneck. Existing teleoperation systems often suffer from limited portability, occlusion, or insufficient precision, which hinders their applicability to complex whole-body tasks. To address these challenges, we introduce HumDex, a portable teleoperation system designed for humanoid whole-body dexterous manipulation. Our system leverages IMU-based motion tracking to address the portability-precision trade-off, enabling accurate full-body tracking while remaining easy to deploy. For dexterous hand control, we further introduce a learning-based retargeting method that generates smooth and natural hand motions without manual parameter tuning. Beyond teleoperation, HumDex enables efficient collection of human motion data. Building on this capability, we propose a two-stage imitation learning framework that first pre-trains on diverse human motion data to learn generalizable priors, and then fine-tunes on robot data to bridge the embodiment gap for precise execution. We demonstrate that this approach significantly improves generalization to new configurations, objects, and backgrounds with minimal data acquisition costs. The entire system is fully reproducible and open-sourced at https://github.com/physical-superintelligence-lab/humdex.
♻ ☆ MIND-V: Hierarchical World Model for Long-Horizon Robotic Manipulation with RL-based Physical Alignment
Scalable embodied intelligence is constrained by the scarcity of diverse, long-horizon robotic manipulation data. Existing video world models in this domain are limited to synthesizing short clips of simple actions and often rely on manually defined trajectories. To this end, we introduce MIND-V, a cognitive hierarchical world model designed to synthesize physically plausible and logically coherent videos of long-horizon robotic manipulation. Inspired by cognitive science, MIND-V bridges high-level reasoning with pixel-level synthesis through three core components: a Semantic Reasoning Hub (SRH) that leverages a pre-trained vision-language model for task planning; a Behavioral Semantic Bridge (BSB) that translates abstract instructions into domain-invariant representations; and a Motor Video Generator (MVG) for conditional video rendering. MIND-V employs Staged Visual Future Rollouts, a test-time optimization strategy to enhance long-horizon robustness. To enforce adherence to physical laws, we introduce a GRPO reinforcement learning post-training phase guided by a novel Physical Foresight Coherence (PFC) reward. PFC leverages the V-JEPA2 world model as a physics referee to penalize implausible dynamics in the latent feature space. Experiments confirm MIND-V's SOTA performance in long-horizon simulation and its significant value for policy learning, introducing a scalable and fully autonomous framework for embodied data synthesis.
♻ ☆ AOMGen: Photoreal, Physics-Consistent Demonstration Generation for Articulated Object Manipulation CVPR
Recent advances in Vision-Language-Action (VLA) and world-model methods have improved generalization in tasks such as robotic manipulation and object interaction. However, Successful execution of such tasks depends on large, costly collections of real demonstrations, especially for fine-grained manipulation of articulated objects. To address this, we present AOMGen, a scalable data generation framework for articulated manipulation which is instantiated from a single real scan, demonstration and a library of readily available digital assets, yielding photoreal training data with verified physical states. The framework synthesizes synchronized multi-view RGB temporally aligned with action commands and state annotations for joints and contacts, and systematically varies camera viewpoints, object styles, and object poses to expand a single execution into a diverse corpus. Experimental results demonstrate that fine-tuning VLA policies on AOMGen data increases the success rate from 0% to 88.7%, and the policies are tested on unseen objects and layouts.
comment: Accepted by CVPR Findings2026
♻ ☆ DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Driving
We propose DynVLA, a driving VLA model that introduces a new CoT paradigm termed Dynamics CoT. DynVLA forecasts compact world dynamics before action generation, enabling more informed and physically grounded decision-making. To obtain compact dynamics representations, DynVLA introduces a Dynamics Tokenizer that compresses future evolution into a small set of dynamics tokens. Considering the rich environment dynamics in interaction-intensive driving scenarios, DynVLA decouples ego-centric and environment-centric dynamics, yielding more accurate world dynamics modeling. We then train DynVLA to generate dynamics tokens before actions through SFT and RFT, improving decision quality while maintaining latency-efficient inference. Compared to Textual CoT, which lacks fine-grained spatiotemporal understanding, and Visual CoT, which introduces substantial redundancy due to dense image prediction, Dynamics CoT captures the evolution of the world in a compact, interpretable, and efficient form. Extensive experiments on NAVSIM, Bench2Drive, and a large-scale in-house dataset demonstrate that DynVLA consistently outperforms Textual CoT and Visual CoT methods, validating the effectiveness and practical value of Dynamics CoT. Project Page: https://yaoyao-jpg.github.io/dynvla.
comment: 18 pages, 10 figures. Project Page: https://yaoyao-jpg.github.io/dynvla
♻ ☆ A Photorealistic Dataset and Vision-Based Algorithm for Anomaly Detection During Proximity Operations in Lunar Orbit ICRA'26
NASA's forthcoming Lunar Gateway space station, which will be uncrewed most of the time, will need to operate with an unprecedented level of autonomy. One key challenge is enabling the Canadarm3, the Gateway's external robotic system, to detect hazards in its environment using its onboard inspection cameras. This task is complicated by the extreme and variable lighting conditions in space. In this paper, we introduce the visual anomaly detection and localization task for the space domain and establish a benchmark based on a synthetic dataset called ALLO (Anomaly Localization in Lunar Orbit). We show that state-of-the-art visual anomaly detection methods often fail in the space domain, motivating the need for new approaches. To address this, we propose MRAD (Model Reference Anomaly Detection), a statistical algorithm that leverages the known pose of the Canadarm3 and a CAD model of the Gateway to generate reference images of the expected scene appearance. Anomalies are then identified as deviations from this model-generated reference. On the ALLO dataset, MRAD surpasses state-of-the-art anomaly detection algorithms, achieving an AP score of 62.9% at the pixel level and an AUROC score of 75.0% at the image level. Given the low tolerance for risk in space operations and the lack of domain-specific data, we emphasize the need for novel, robust, and accurate anomaly detection methods to handle the challenging visual conditions found in lunar orbit and beyond.
comment: In IEEE Robotics and Automation Letters (RA-L) and presented at the IEEE International Conference on Robotics and Automation (ICRA'26), 1-5 Jun. 2026, Vienna, Austria
♻ ☆ Real-time Rendering-based Surgical Instrument Tracking via Evolutionary Optimization
Accurate and efficient tracking of surgical instruments is fundamental for Robot-Assisted Minimally Invasive Surgery. Although vision-based robot pose estimation has enabled markerless calibration without tedious physical setups, reliable tool tracking for surgical robots still remains challenging due to partial visibility and specialized articulation design of surgical instruments. Previous works in the field are usually prone to unreliable feature detections under degraded visual quality and data scarcity, whereas rendering-based methods often struggle with computational costs and suboptimal convergence. In this work, we incorporate CMA-ES, an evolutionary optimization strategy, into a versatile tracking pipeline that jointly estimates surgical instrument pose and joint configurations. Using batch rendering to efficiently evaluate multiple pose candidates in parallel, the method significantly reduces inference time and improves convergence robustness. The proposed framework further generalizes to joint angle-free and bi-manual tracking settings, making it suitable for both vision feedback control and online surgery video calibration. Extensive experiments on synthetic and real-world datasets demonstrate that the proposed method significantly outperforms prior approaches in both accuracy and runtime.
♻ ☆ VIGS-SLAM: Visual Inertial Gaussian Splatting SLAM
We present VIGS-SLAM, a visual-inertial 3D Gaussian Splatting SLAM system that achieves robust real-time tracking and high-fidelity reconstruction. Although recent 3DGS-based SLAM methods achieve dense and photorealistic mapping, their purely visual design degrades under challenging conditions such as motion blur, low texture, and exposure variations. Our method tightly couples visual and inertial cues within a unified optimization framework, jointly optimizing camera poses, depths, and IMU states. It features robust IMU initialization, time-varying bias modeling, and loop closure with consistent Gaussian updates. Experiments on five challenging datasets demonstrate our superiority over state-of-the-art methods. Project page: https://vigs-slam.github.io
comment: Project page: https://vigs-slam.github.io
♻ ☆ From Ellipsoids to Midair Control of Dynamic Hitches
The ability to manipulate and interlace cables using aerial vehicles can greatly improve aerial transportation tasks. Such interlacing cables create hitches by winding two or more cables around each other, which can enclose payloads or can further develop into knots. Dynamic modeling and control of such hitches are key to mastering inter-cable interactions in the context of cable-suspended aerial manipulation. This paper introduces an ellipsoid-based kinematic model to connect the geometric nature of a hitch created by two cables and the dynamics of the hitch driven by four aerial vehicles, which reveals the control-affine form of the system. As the constraint for maintaining tension of a cable is also control-affine, we design a quadratic programming-based controller that combines Control Lyapunov and High-Order Control Barrier Functions (CLF-HOCBF-QP) to precisely track a desired hitch position and system shape while enforcing safety constraints like cable tautness. We convert desired geometric reference configurations into target robot positions and introduce a composite error into the Lyapunov function to ensure a relative degree of one to the input. Numerical simulations validate our approach, demonstrating stable, high-speed tracking of dynamic references.
♻ ☆ UMI-on-Air: Embodiment-Aware Guidance for Embodiment-Agnostic Visuomotor Policies
We introduce UMI-on-Air, a framework for embodiment-aware deployment of embodiment-agnostic manipulation policies. Our approach leverages diverse, unconstrained human demonstrations collected with a handheld gripper (UMI) to train generalizable visuomotor policies. A central challenge in transferring these policies to constrained robotic embodiments-such as aerial manipulators-is the mismatch in control and robot dynamics, which often leads to out-of-distribution behaviors and poor execution. To address this, we propose Embodiment-Aware Diffusion Policy (EADP), which couples a high-level UMI policy with a low-level embodiment-specific controller at inference time. By integrating gradient feedback from the controller's tracking cost into the diffusion sampling process, our method steers trajectory generation towards dynamically feasible modes tailored to the deployment embodiment. This enables plug-and-play, embodiment-aware trajectory adaptation at test time. We validate our approach on multiple long-horizon and high-precision aerial manipulation tasks, showing improved success rates, efficiency, and robustness under disturbances compared to unguided diffusion baselines. Finally, we demonstrate deployment in previously unseen environments, using UMI demonstrations collected in the wild, highlighting a practical pathway for scaling generalizable manipulation skills across diverse-and even highly constrained-embodiments. All code, data, checkpoints, and result videos can be found at umi-on-air.github.io.
comment: Result videos can be found at umi-on-air.github.io
♻ ☆ Concurrent Prehensile and Nonprehensile Manipulation: A Practical Approach to Multi-Stage Dexterous Tasks
Dexterous hands enable concurrent prehensile and nonprehensile manipulation, such as holding one object while interacting with another, a capability essential for everyday tasks yet underexplored in robotics. Learning such long-horizon, contact-rich multi-stage behaviors is challenging because demonstrations are expensive to collect and end-to-end policies require substantial data to generalize across varied object geometries and placements. We present DexMulti, a sample-efficient approach for real-world dexterous multi-task manipulation that decomposes demonstrations into object-centric skills with well-defined temporal boundaries. Rather than learning monolithic policies, our method retrieves demonstrated skills based on current object geometry, aligns them to the observed object state using an uncertainty-aware estimator that tracks centroid and yaw, and executes them via a retrieve-align-execute paradigm. We evaluate on three multi-stage tasks requiring concurrent manipulation (Grasp + Pull, Grasp + Open, and Grasp + Grasp) across two dexterous hands (Allegro and LEAP) in over 1,000 real-world trials. Our approach achieves an average success rate of 66% on training objects with only 3-4 demonstrations per object, outperforming diffusion policy baselines by 2-3x while requiring far fewer demonstrations. Results demonstrate robust generalization to held-out objects and spatial variations up to +/-25 cm.
comment: 12 pages, 6 figures
♻ ☆ A Human-in-the-Loop Confidence-Aware Failure Recovery Framework for Modular Robot Policies
Robots operating in unstructured human environments inevitably encounter failures, especially in robot caregiving scenarios. While humans can often help robots recover, excessive or poorly targeted queries impose unnecessary cognitive and physical workload on the human partner. We present a human-in-the-loop failure-recovery framework for modular robotic policies, where a policy is composed of distinct modules such as perception, planning, and control, any of which may fail and often require different forms of human feedback. Our framework integrates calibrated estimates of module-level uncertainty with models of human intervention cost to decide which module to query and when to query the human. It separates these two decisions: a module selector identifies the module most likely responsible for failure, and a querying algorithm determines whether to solicit human input or act autonomously. We evaluate several module-selection strategies and querying algorithms in controlled synthetic experiments, revealing trade-offs between recovery efficiency, robustness to system and user variables, and user workload. Finally, we deploy the framework on a robot-assisted bite acquisition system and demonstrate, in studies involving individuals with both emulated and real mobility limitations, that it improves recovery success while reducing the workload imposed on users. Our results highlight how explicitly reasoning about both robot uncertainty and human effort can enable more efficient and user-centered failure recovery in collaborative robots. Supplementary materials and videos can be found at: http://emprise.cs.cornell.edu/modularhil
comment: The second and third authors contributed equally. The last two authors advised equally
Robotics 76
☆ CRAFT: A Tendon-Driven Hand with Hybrid Hard-Soft Compliance
We introduce CRAFT hand, a tendon-driven anthropomorphic hand with hybrid hard-soft compliance for contact-rich manipulation. The design is based on a simple idea: contact is not uniform across the hand. Impacts concentrate at joints, while links carry most of the load. CRAFT places soft material at joints and keeps links rigid, and uses rollingcontact joint surfaces to keep flexion on repeatable motion paths. Fifteen motors mounted on the fingers drive the hand through tendons, keeping the form factor compact and the fingers light. In structural tests, CRAFT improves strength and endurance while maintaining comparable repeatability. In teleoperation, CRAFT improves handling of fragile and low-friction items, and the hand covers 33/33 grasps in the Feix taxonomy. The full design costs under $600 and will be released open-source with visionbased teleoperation and simulation integration. Project page: http://craft-hand.github.io/
☆ Towards Dynamic Model Identification and Gravity Compensation for the dVRK-Si Patient Side Manipulator
The da Vinci Research Kit (dVRK) is widely used for research in robot-assisted surgery, but most modeling and control methods target the first-generation dVRK Classic. The recently introduced dVRK-Si, built from da Vinci Si hardware, features a redesigned Patient Side Manipulator (PSM) with substantially larger gravity loading, which can degrade control if unmodeled. This paper presents the first complete kinematic and dynamic modeling framework for the dVRK-Si PSM. We derive a modified DH kinematic model that captures the closed-chain parallelogram mechanism, formulate dynamics via the Euler-Lagrange method, and express inverse dynamics in a linear-in-parameters regressor form. Dynamic parameters are identified from data collected on a periodic excitation trajectory optimized for numerical conditioning and estimated by convex optimization with physical feasibility constraints. Using the identified model, we implement real-time gravity compensation and computed-torque feedforward in the dVRK control stack. Experiments on a physical dVRK-Si show that the gravity compensation reduces steady-state joint errors by 68-84% and decreases end-effector tip drift during static holds from 4.2 mm to 0.7 mm. Computed-torque feedforward further improves transient and position tracking accuracy. For sinusoidal trajectory tracking, computed-torque feedforward reduces position errors by 35% versus gravity-only feedforward and by 40% versus PID-only. The proposed pipeline supports reliable control, high-fidelity simulation, and learning-based automation on the dVRK-Si.
comment: Submitted to IEEE Transactions on Medical Robotics and Bionics (T-MRB), under review. Open-source GitHub Repo: https://github.com/jhu-dvrk/dvrk_psm_dynamics_identification
☆ Towards Universal Computational Aberration Correction in Photographic Cameras: A Comprehensive Benchmark Analysis CVPR 2026
Prevalent Computational Aberration Correction (CAC) methods are typically tailored to specific optical systems, leading to poor generalization and labor-intensive re-training for new lenses. Developing CAC paradigms capable of generalizing across diverse photographic lenses offers a promising solution to these challenges. However, efforts to achieve such cross-lens universality within consumer photography are still in their early stages due to the lack of a comprehensive benchmark that encompasses a sufficiently wide range of optical aberrations. Furthermore, it remains unclear which specific factors influence existing CAC methods and how these factors affect their performance. In this paper, we present comprehensive experiments and evaluations involving 24 image restoration and CAC algorithms, utilizing our newly proposed UniCAC, a large-scale benchmark for photographic cameras constructed via automatic optical design. The Optical Degradation Evaluator (ODE) is introduced as a novel framework to objectively assess the difficulty of CAC tasks, offering credible quantification of optical aberrations and enabling reliable evaluation. Drawing on our comparative analysis, we identify three key factors -- prior utilization, network architecture, and training strategy -- that most significantly influence CAC performance, and further investigate their respective effects. We believe that our benchmark, dataset, and observations contribute foundational insights to related areas and lay the groundwork for future investigations. Benchmarks, codes, and Zemax files will be available at https://github.com/XiaolongQian/UniCAC.
comment: Accepted to CVPR 2026. Benchmarks, codes, and Zemax files will be available at https://github.com/XiaolongQian/UniCAC
☆ Decentralized Cooperative Localization for Multi-Robot Systems with Asynchronous Sensor Fusion
Decentralized cooperative localization (DCL) is a promising approach for nonholonomic mobile robots operating in GPS-denied environments with limited communication infrastructure. This paper presents a DCL framework in which each robot performs localization locally using an Extended Kalman Filter, while sharing measurement information during update stages only when communication links are available and companion robots are successfully detected by LiDAR. The framework preserves cross-correlation consistency among robot state estimates while handling asynchronous sensor data with heterogeneous sampling rates and accommodating accelerations during dynamic maneuvers. Unlike methods that require pre-aligned coordinate systems, the proposed approach allows robots to initialize with arbitrary reference-frame orientations and achieves automatic alignment through transformation matrices in both the prediction and update stages. To improve robustness in feature-sparse environments, we introduce a dual-landmark evaluation framework that exploits both static environmental features and mobile robots as dynamic landmarks. The proposed framework enables reliable detection and feature extraction during sharp turns, while prediction accuracy is improved through information sharing from mutual observations. Experimental results in both Gazebo simulation and real-world basement environments show that DCL outperforms centralized cooperative localization (CCL), achieving a 34% reduction in RMSE, while the dual-landmark variant yields an improvement of 56%. These results demonstrate the applicability of DCL to challenging domains such as enclosed spaces, underwater environments, and feature-sparse terrains where conventional localization methods are ineffective.
comment: Presented at the 13th RSI International Conference on Robotics and Mechatronics (ICRoM 2025)
☆ Flight through Narrow Gaps with Morphing-Wing Drones
The size of a narrow gap traversable by a fixed-wing drone is limited by its wingspan. Inspired by birds, here, we enable the traversal of a gap of sub-wingspan width and height using a morphing-wing drone capable of temporarily sweeping in its wings mid-flight. This maneuver poses control challenges due to sudden lift loss during gap-passage at low flight speeds and the need for precisely timed wing-sweep actuation ahead of the gap. To address these challenges, we first develop an aerodynamic model for general wing-sweep morphing drone flight including low flight speeds and post-stall angles of attack. We integrate longitudinal drone dynamics into an optimal reference trajectory generation and Nonlinear Model Predictive Control framework with runtime adaptive costs and constraints. Validated on a 130 g wing-sweep-morphing drone, our method achieves an average altitude error of 5 cm during narrow-gap passage at forward speeds between 5 and 7 m/s, whilst enforcing fully swept wings near the gap across variable threshold distances. Trajectory analysis shows that the drone can compensate for lift loss during gap-passage by accelerating and pitching upwards ahead of the gap to an extent that differs between reference trajectory optimization objectives. We show that our strategy also allows for accurate gap passage on hardware whilst maintaining a constant forward flight speed reference and near-constant altitude.
☆ Sim-to-reality adaptation for Deep Reinforcement Learning applied to an underwater docking application IROS 2026
Deep Reinforcement Learning (DRL) offers a robust alternative to traditional control methods for autonomous underwater docking, particularly in adapting to unpredictable environmental conditions. However, bridging the "sim-to-real" gap and managing high training latencies remain significant bottlenecks for practical deployment. This paper presents a systematic approach for autonomous docking using the Girona Autonomous Underwater Vehicle (AUV) by leveraging a high-fidelity digital twin environment. We adapted the Stonefish simulator into a multiprocessing RL framework to significantly accelerate the learning process while incorporating realistic AUV dynamics, collision models, and sensor noise. Using the Proximal Policy Optimization (PPO) algorithm, we developed a 6-DoF control policy trained in a headless environment with randomized starting positions to ensure generalized performance. Our reward structure accounts for distance, orientation, action smoothness, and adaptive collision penalties to facilitate soft docking. Experimental results demonstrate that the agent achieved a success rate of over 90% in simulation. Furthermore, successful validation in a physical test tank confirmed the efficacy of the sim-to-reality adaptation, with the DRL controller exhibiting emergent behaviors such as pitch-based braking and yaw oscillations to assist in mechanical alignment.
comment: Currently under review by IROS 2026
☆ Learning Visuomotor Policy for Multi-Robot Laser Tag Game
In this paper, we study multi robot laser tag, a simplified yet practical shooting-game-style task. Classic modular approaches on these tasks face challenges such as limited observability and reliance on depth mapping and inter robot communication. To overcome these issues, we present an end-to-end visuomotor policy that maps images directly to robot actions. We train a high performing teacher policy with multi agent reinforcement learning and distill its knowledge into a vision-based student policy. Technical designs, including a permutation-invariant feature extractor and depth heatmap input, improve performance over standard architectures. Our policy outperforms classic methods by 16.7% in hitting accuracy and 6% in collision avoidance, and is successfully deployed on real robots. Code will be released publicly.
☆ Energy Prediction on Sloping Ground for Quadruped Robots
Energy management is a fundamental challenge for legged robots in outdoor environments. Endurance directly constrains mission success, while efficient resource use reduces ecological impact. This paper investigates how terrain slope and heading orientation influence the energetic cost of quadruped locomotion. We introduce a simple energy model that relies solely on standard onboard sensors, avoids specialized instrumentation, and remains applicable in previously unexplored environments. The model is identified from field runs on a commercial quadruped and expressed as a compact function of slope angle and heading. Field validation on natural terrain shows near-linear trends of force-equivalent cost with slope angle, consistently higher lateral costs, and additive behavior across trajectory segments, supporting path-level energy prediction for planning-oriented evaluation.
comment: Presented at 3D-Advice (Advanced 3D Vision for Complex Environments) Workshop, ECMR 2025
☆ RADAR: Closed-Loop Robotic Data Generation via Semantic Planning and Autonomous Causal Environment Reset IROS
The acquisition of large-scale physical interaction data, a critical prerequisite for modern robot learning, is severely bottlenecked by the prohibitive cost and scalability limits of human-in-the-loop collection paradigms. To break this barrier, we introduce Robust Autonomous Data Acquisition for Robotics (RADAR), a fully autonomous, closed-loop data generation engine that completely removes human intervention from the collection cycle. RADAR elegantly divides the cognitive load into a four-module pipeline. Anchored by 2-5 3D human demonstrations as geometric priors, a Vision-Language Model first orchestrates scene-relevant task generation via precise semantic object grounding and skill retrieval. Next, a Graph Neural Network policy translates these subtasks into physical actions via in-context imitation learning. Following execution, the VLM performs automated success evaluation using a structured Visual Question Answering pipeline. Finally, to shatter the bottleneck of manual resets, a Finite State Machine orchestrates an autonomous environment reset and asymmetric data routing mechanism. Driven by simultaneous forward-reverse planning with a strict Last-In, First-Out causal sequence, the system seamlessly restores unstructured workspaces and robustly recovers from execution failures. This continuous brain-cerebellum synergy transforms data collection into a self-sustaining process. Extensive evaluations highlight RADAR's exceptional versatility. In simulation, our framework achieves up to 90% success rates on complex, long-horizon tasks, effortlessly solving challenges where traditional baselines plummet to near-zero performance. In real-world deployments, the system reliably executes diverse, contact-rich skills (e.g., deformable object manipulation) via few-shot adaptation without domain-specific fine-tuning, providing a highly scalable paradigm for robotic data acquisition.
comment: 8 pages, 4 figures. Submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
☆ HiSync: Spatio-Temporally Aligning Hand Motion from Wearable IMU and On-Robot Camera for Command Source Identification in Long-Range HRI
Long-range Human-Robot Interaction (HRI) remains underexplored. Within it, Command Source Identification (CSI) - determining who issued a command - is especially challenging due to multi-user and distance-induced sensor ambiguity. We introduce HiSync, an optical-inertial fusion framework that treats hand motion as binding cues by aligning robot-mounted camera optical flow with hand-worn IMU signals. We first elicit a user-defined (N=12) gesture set and collect a multimodal command gesture dataset (N=38) in long-range multi-user HRI scenarios. Next, HiSync extracts frequency-domain hand motion features from both camera and IMU data, and a learned CSINet denoises IMU readings, temporally aligns modalities, and performs distance-aware multi-window fusion to compute cross-modal similarity of subtle, natural gestures, enabling robust CSI. In three-person scenes up to 34m, HiSync achieves 92.32% CSI accuracy, outperforming the prior SOTA by 48.44%. HiSync is also validated on real-robot deployment. By making CSI reliable and natural, HiSync provides a practical primitive and design guidance for public-space HRI.
☆ Adapting Dijkstra for Buffers and Unlimited Transfers
In recent years, RAPTOR based algorithms have been considered the state-of-the-art for path-finding with unlimited transfers without preprocessing. However, this status largely stems from the evolution of routing research, where Dijkstra-based solutions were superseded by timetable-based algorithms without a systematic comparison. In this work, we revisit classical Dijkstra-based approaches for public transit routing with unlimited transfers and demonstrate that Time-Dependent Dijkstra (TD-Dijkstra) outperforms MR. However, efficient TD-Dijkstra implementations rely on filtering dominated connections during preprocessing, which assumes passengers can always switch to a faster connection. We show that this filtering is unsound when stops have buffer times, as it cannot distinguish between seated passengers who may continue without waiting and transferring passengers who must respect the buffer. To address this limitation, we introduce Transfer Aware Dijkstra (TAD), a modification that scans entire trip sequences rather than individual edges, correctly handling buffer times while maintaining performance advantages over MR. Our experiments on London and Switzerland networks show that we can achieve a greater than two time speed-up over MR while producing optimal results on both networks with and without buffer times.
☆ Coupling Tensor Trains with Graph of Convex Sets: Effective Compression, Exploration, and Planning in the C-Space ICRA2026
We present TANGO (Tensor ANd Graph Optimization), a novel motion planning framework that integrates tensor-based compression with structured graph optimization to enable efficient and scalable trajectory generation. While optimization-based planners such as the Graph of Convex Sets (GCS) offer powerful tools for generating smooth, optimal trajectories, they typically rely on a predefined convex characterization of the high-dimensional configuration space-a requirement that is often intractable for general robotic tasks. TANGO builds further by using Tensor Train decomposition to approximate the feasible configuration space in a compressed form, enabling rapid discovery and estimation of task-relevant regions. These regions are then embedded into a GCS-like structure, allowing for geometry-aware motion planning that respects both system constraints and environmental complexity. By coupling tensor-based compression with structured graph reasoning, TANGO enables efficient, geometry-aware motion planning and lays the groundwork for more expressive and scalable representations of configuration space in future robotic systems. Rigorous simulation studies on planar and real robots reinforce our claims of effective compression and higher quality trajectories.
comment: 8 pages, 10 figures, accepted paper for ICRA2026
☆ Concurrent Prehensile and Nonprehensile Manipulation: A Practical Approach to Multi-Stage Dexterous Tasks
Dexterous hands enable concurrent prehensile and nonprehensile manipulation, such as holding one object while interacting with another, a capability essential for everyday tasks yet underexplored in robotics. Learning such long-horizon, contact-rich multi-stage behaviors is challenging because demonstrations are expensive to collect and end-to-end policies require substantial data to generalize across varied object geometries and placements. We present DexMulti, a sample-efficient approach for real-world dexterous multi-task manipulation that decomposes demonstrations into object-centric skills with well-defined temporal boundaries. Rather than learning monolithic policies, our method retrieves demonstrated skills based on current object geometry, aligns them to the observed object state using an uncertainty-aware estimator that tracks centroid and yaw, and executes them via a retrieve-align-execute paradigm. We evaluate on three multi-stage tasks requiring concurrent manipulation (Grasp + Pull, Grasp + Open, and Grasp + Grasp) across two dexterous hands (Allegro and LEAP) in over 1,000 real-world trials. Our approach achieves an average success rate of 66% on training objects with only 3-4 demonstrations per object, outperforming diffusion policy baselines by 2-3x while requiring far fewer demonstrations. Results demonstrate robust generalization to held-out objects and spatial variations up to +/-25 cm.
☆ Simple Recipe Works: Vision-Language-Action Models are Natural Continual Learners with Reinforcement Learning
Continual Reinforcement Learning (CRL) for Vision-Language-Action (VLA) models is a promising direction toward self-improving embodied agents that can adapt in openended, evolving environments. However, conventional wisdom from continual learning suggests that naive Sequential Fine-Tuning (Seq. FT) leads to catastrophic forgetting, necessitating complex CRL strategies. In this work, we take a step back and conduct a systematic study of CRL for large pretrained VLAs across three models and five challenging lifelong RL benchmarks. We find that, contrary to established belief, simple Seq. FT with low-rank adaptation (LoRA) is remarkably strong: it achieves high plasticity, exhibits little to no forgetting, and retains strong zero-shot generalization, frequently outperforming more sophisticated CRL methods. Through detailed analysis, we show that this robustness arises from a synergy between the large pretrained model, parameter-efficient adaptation, and on-policy RL. Together, these components reshape the stability-plasticity trade-off, making continual adaptation both stable and scalable. Our results position Sequential Fine-Tuning as a powerful method for continual RL with VLAs and provide new insights into lifelong learning in the large model era. Code is available at github.com/UT-Austin-RobIn/continual-vla-rl.
☆ A Hybrid Neural-Assisted Unscented Kalman Filter for Unmanned Ground Vehicle Navigation
Modern autonomous navigation for unmanned ground vehicles relies on different estimators to fuse inertial sensors and GNSS measurements. However, the constant noise covariance matrices often struggle to account for dynamic real-world conditions. In this work we propose a hybrid estimation framework that bridges classical state estimation foundations with modern deep learning approaches. Instead of altering the fundamental unscented Kalman filter equations, a dedicated deep neural network is developed to predict the process and measurement noise uncertainty directly from raw inertial and GNSS measurements. We present a sim2real approach, with training performed only on simulative data. In this manner, we offer perfect ground truth data and relieves the burden of extensive data recordings. To evaluate our proposed approach and examine its generalization capabilities, we employed a 160-minutes test set from three datasets each with different types of vehicles (off-road vehicle, passenger car, and mobile robot), inertial sensors, road surface, and environmental conditions. We demonstrate across the three datasets a position improvement of $12.7\%$ compared to the adaptive model-based approach. Thus, offering a scalable and a more robust solution for unmanned ground vehicles navigation tasks.
☆ Chunk-Boundary Artifact in Action-Chunked Generative Policies: A Noise-Sensitive Failure Mechanism
Action chunking has become a central design choice for generative visuomotor policies, yet the execution discontinuities that arise at chunk boundaries remain poorly understood. In a frozen pretrained action-chunked policy, we identify chunk-boundary artifact as a noise-sensitive failure mechanism. First, artifact is strongly associated with task failure (p < 1e-4, permutation test) and emerges during the rollout rather than only as a post-hoc symptom. Second, under a fixed observation context, changing only latent noise systematically modulates artifact magnitude. Third, by identifying artifact-related directions in noise space and applying trajectory-level steering, we reliably alter artifact magnitude across all evaluated tasks. In hard-task settings with remaining outcome headroom, the success/failure distribution shifts accordingly; on near-ceiling tasks, positive gains are compressed by policy saturation, while the negative causal effect remains visible. Overall, we recast boundary discontinuity from an unavoidable execution nuisance into an analyzable, noise-dominated, and intervenable failure mechanism.
comment: 13 pages, 5 figures
☆ Learn Structure, Adapt on the Fly: Multi-Scale Residual Learning and Online Adaptation for Aerial Manipulators
Autonomous Aerial Manipulators (AAMs) are inherently coupled, nonlinear systems that exhibit nonstationary and multiscale residual dynamics, particularly during manipulator reconfiguration and abrupt payload variations. Conventional analytical dynamic models rely on fixed parametric structures, while static data-driven model assume stationary dynamics and degrade under configuration changes and payload variations. Moreover, existing learning architectures do not explicitly factorize cross-variable coupling and multi-scale temporal effects, conflating instantaneous inertial dynamics with long-horizon regime evolution. We propose a predictive-adaptive framework for real-time residual modeling and compensation in AAMs. The core of this framework is the Factorized Dynamics Transformer (FDT), which treats physical variables as independent tokens. This design enables explicit cross-variable attention while structurally separating short-horizon inertial dependencies from long-horizon aerodynamic effects. To address deployment-time distribution shifts, a Latent Residual Adapter (LRA) performs rapid linear adaptation in the latent space via Recursive Least Squares, preserving the offline nonlinear representation without prohibitive computational overhead. The adapted residual forecast is directly integrated into a residual-compensated adaptive controller. Real-world experiments on an aerial manipulator subjected to unseen payloads demonstrate higher prediction fidelity, accelerated disturbance attenuation, and superior closed-loop tracking precision compared to state-of-the-art learning baselines, all while maintaining strict real-time feasibility.
☆ Diversity You Can Actually Measure: A Fast, Model-Free Diversity Metric for Robotics Datasets
Robotics datasets for imitation learning typically consist of long-horizon trajectories of different lengths over states, actions, and high-dimensional observations (e.g., RGB video), making it non-trivial to quantify diversity in a way that respects the underlying trajectory structure and geometry. We extend Shannon and von Neumann entropy to this setting by defining signature transform-based entropy on the Gram matrix of a signature kernel over demonstrations, yielding entropy and diversity metrics that operate directly on the demonstration dataset. Building on these metrics, we study how dataset diversity affects generalization performance in robot imitation learning and propose a simple, model-free way to curate diverse demonstrations. We introduce FAKTUAL (FAst trajectory Kernel enTropy cUration for imitation Learning), a data curation algorithm that selects a subset of demonstrations maximizing entropy given a subset-size budget. FAKTUAL is fully model-free, requires no access to the imitation policy or rollouts, and adds negligible overhead relative to policy training. We evaluate our approach on image and state-based RoboMimic and MetaWorld benchmarks, as well as four real-world manipulation tasks. Across tasks and architectures, diversity-aware curation with FAKTUAL consistently improves downstream success rates over random selection, while being substantially more computationally efficient compared to recent robot data curation methods. Our results suggest that the entropy of demonstration datasets is a practical tool for understanding and improving dataset diversity in robot imitation learning.
☆ From Pets to Robots: MojiKit as a Data-Informed Toolkit for Affective HRI Design
Designing affective behaviors for animal-inspired social robots often relies on intuition and personal experience, leading to fragmented outcomes. To provide more systematic guidance, we first coded and analyzed human-pet interaction videos, validated insights through literature and interviews, and created structured reference cards that map the design space of pet-inspired affective interactions. Building on this, we developed MojiKit, a toolkit combining reference cards, a zoomorphic robot prototype (MomoBot), and a behavior control studio. We evaluated MojiKit in co-creation workshops with 18 participants, finding that MojiKit helped them design 35 affective interaction patterns beyond their own pet experiences, while the code-free studio lowered the technical barrier and enhanced creative agency. Our contributions include the data-informed structured resource for pet-inspired affective HRI design, an integrated toolkit that bridges reference materials with hands-on prototyping, and empirical evidence showing how MojiKit empowers users to systematically create richer, more diverse affective robot behaviors.
comment: 25 pages, 11 figures, Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI '26)
☆ Unsupervised LiDAR-Based Multi-UAV Detection and Tracking Under Extreme Sparsity ICMR
Non-repetitive solid-state LiDAR scanning leads to an extremely sparse measurement regime for detecting airborne UAVs: a small quadrotor at 10-25 m typically produces only 1-2 returns per scan, which is far below the point densities assumed by most existing detection approaches and inadequate for robust multi-target data association. We introduce an unsupervised, LiDAR-only pipeline that addresses both detection and tracking without the need for labeled training data. The detector integrates range-adaptive DBSCAN clustering with a three-stage temporal consistency check and is benchmarked on real-world air-to-air flight data under eight different parameter configurations. The best setup attains 0.891 precision, 0.804 recall, and 0.63 m RMSE, and a systematic minPts sweep verifies that most scans contain at most 1-2 target points, directly quantifying the sparsity regime. For multi-target tracking, we compare deterministic Hungarian assignment with joint probabilistic data association (JPDA), each coupled with Interacting Multiple Model filtering, in four simulated scenarios with increasing levels of ambiguity. JPDA cuts identity switches by 64% with negligible impact on MOTA, demonstrating that probabilistic association is advantageous when UAV trajectories approach one another closely. A two-environment evaluation strategy, combining real-world detection with RTK-GPS ground truth and simulation-based tracking with identity-annotated ground truth, overcomes the limitations of GNSS-only evaluation at inter-UAV distances below 2 m.
comment: Presented at the International Conference on Mechatronics and Robotics Engineering (ICMRE2026). To appear in IEEE conference proceedings
☆ SVLL: Staged Vision-Language Learning for Physically Grounded Embodied Task Planning
Embodied task planning demands vision-language models to generate action sequences that are both visually grounded and causally coherent over time. However, existing training paradigms face a critical trade-off: joint end-to-end training often leads to premature temporal binding, while standard reinforcement learning methods suffer from optimization instability. To bridge this gap, we present Staged Vision-Language Learning (SVLL), a unified three-stage framework for robust, physically-grounded embodied planning. In the first two stages, SVLL decouples spatial grounding from temporal reasoning, establishing robust visual dependency before introducing sequential action history. In the final stage, we identify a key limitation of standard Direct Preference Optimization (DPO), its purely relative nature -- optimizing only the preference gap between winning and losing trajectories while neglecting absolute likelihood constraints on optimal path, often yields unsafe or hallucinated behaviors. To address this, we further introduce Bias-DPO, a novel alignment objective that injects an inductive bias toward expert trajectories by explicitly maximizing likelihood on ground-truth actions while penalizing overconfident hallucinations. By anchoring the policy to the expert manifold and mitigating causal misalignment, SVLL, powered by Bias-DPO, ensures strict adherence to environmental affordances and effectively suppresses physically impossible shortcuts. Finally, extensive experiments on the interactive AI2-THOR benchmark and real-world robotic deployments demonstrate that SVLL outperforms both state-of-the-art open-source (e.g., Qwen2.5-VL-7B) and closed-source models (e.g., GPT-4o, Gemini-2.0-flash) in task success rate, while significantly reducing physical constraint violations.
☆ RoboClaw: An Agentic Framework for Scalable Long-Horizon Robotic Tasks
Vision-Language-Action (VLA) systems have shown strong potential for language-driven robotic manipulation. However, scaling them to long-horizon tasks remains challenging. Existing pipelines typically separate data collection, policy learning, and deployment, resulting in heavy reliance on manual environment resets and brittle multi-policy execution. We present RoboClaw, an agentic robotics framework that unifies data collection, policy learning, and task execution under a single VLM-driven controller. At the policy level, RoboClaw introduces Entangled Action Pairs (EAP), which couple forward manipulation behaviors with inverse recovery actions to form self-resetting loops for autonomous data collection. This mechanism enables continuous on-policy data acquisition and iterative policy refinement with minimal human intervention. During deployment, the same agent performs high-level reasoning and dynamically orchestrates learned policy primitives to accomplish long-horizon tasks. By maintaining consistent contextual semantics across collection and execution, RoboClaw reduces mismatch between the two phases and improves multi-policy robustness. Experiments in real-world manipulation tasks demonstrate improved stability and scalability compared to conventional open-loop pipelines, while significantly reducing human effort throughout the robot lifecycle, achieving a 25% improvement in success rate over baseline methods on long-horizon tasks and reducing human time investment by 53.7%.
☆ MANSION: Multi-floor lANguage-to-3D Scene generatIOn for loNg-horizon tasks
Real-world robotic tasks are long-horizon and often span multiple floors, demanding rich spatial reasoning. However, existing embodied benchmarks are largely confined to single-floor in-house environments, failing to reflect the complexity of real-world tasks. We introduce MANSION, the first language-driven framework for generating building-scale, multi-floor 3D environments. Being aware of vertical structural constraints, MANSION generates realistic, navigable whole-building structures with diverse, human-friendly scenes, enabling the development and evaluation of cross-floor long-horizon tasks. Building on this framework, we release MansionWorld, a dataset of over 1,000 diverse buildings ranging from hospitals to offices, alongside a Task-Semantic Scene Editing Agent that customizes these environments using open-vocabulary commands to meet specific user needs. Benchmarking reveals that state-of-the-art agents degrade sharply in our settings, establishing MANSION as a critical testbed for the next generation of spatial reasoning and planning.
☆ MiNI-Q: A Miniature, Wire-Free Quadruped with Unbounded, Independently Actuated Leg Joints
Physical joint limits are common in legged robots and can restrict workspace, constrain gait design, and increase the risk of hardware damage. This paper introduces MiNI-Q^2, a miniature, wire-free quadruped robot with independently actuated, mechanically unbounded 2-DOF leg joints. We present the mechanical design, kinematic analysis, and experimental validation of the proposed robot. The leg mechanism enables both oscillatory gaits and rotary locomotion while allowing the robot to fold to a minimum height of 2.5 cm. Experimentally, MiNI-Q achieves speeds up to 0.46 m/s and demonstrates low-clearance crawling, stair climbing, inverted locomotion, jumping, and backflipping. The wire-free architecture extends our previous Q8bot design, improving assembly reliability at miniature scale. All mechanical and electrical design files are released open source to support reproducibility and further research.
comment: 7 pages, 11 figures. Submitted to the IEEE RAS Conference on Ubiquitous Robots (UR 2026)
☆ SPARK: Skeleton-Parameter Aligned Retargeting on Humanoid Robots with Kinodynamic Trajectory Optimization
Human motion provides rich priors for training general-purpose humanoid control policies, but raw demonstrations are often incompatible with a robot's kinematics and dynamics, limiting their direct use. We present a two-stage pipeline for generating natural and dynamically feasible motion references from task-space human data. First, we convert human motion into a unified robot description format (URDF)-based skeleton representation and calibrate it to the target humanoid's dimensions. By aligning the underlying skeleton structure rather than heuristically modifying task-space targets, this step significantly reduces inverse kinematics error and tuning effort. Second, we refine the retargeted trajectories through progressive kinodynamic trajectory optimization (TO), solved in three stages: kinematic TO, inverse dynamics, and full kinodynamic TO, each warm-started from the previous solution. The final result yields dynamically consistent state trajectories and joint torque profiles, providing high-quality references for learning-based controllers. Together, skeleton calibration and kinodynamic TO enable the generation of natural, physically consistent motion references across diverse humanoid platforms.
☆ NFPO: Stabilized Policy Optimization of Normalizing Flow for Robotic Policy Learning
Deep Reinforcement Learning (DRL) has experienced significant advancements in recent years and has been widely used in many fields. In DRL-based robotic policy learning, however, current de facto policy parameterization is still multivariate Gaussian (with diagonal covariance matrix), which lacks the ability to model multi-modal distribution. In this work, we explore the adoption of a modern network architecture, i.e. Normalizing Flow (NF) as the policy parameterization for its ability of multi-modal modeling, closed form of log probability and low computation and memory overhead. However, naively training NF in online Reinforcement Learning (RL) usually leads to training instability. We provide a detailed analysis for this phenomenon and successfully address it via simple but effective technique. With extensive experiments in multiple simulation environments, we show our method, NFPO could obtain robust and strong performance in widely used robotic learning tasks and successfully transfer into real-world robots.
☆ CoViLLM: An Adaptive Human-Robot Collaborative Assembly Framework Using Large Language Models for Manufacturing
With increasing demand for mass customization, traditional manufacturing robots that rely on rule-based operations lack the flexibility to accommodate customized or new product variants. Human-Robot Collaboration (HRC) has demonstrated potential to improve system adaptability by leveraging human versatility and decision-making capabilities. However, existing HRC frame- works typically depend on predefined perception-manipulation pipelines, limiting their ability to autonomously generate task plans for new product assembly. In this work, we propose CoViLLM, an adaptive human-robot collaborative assembly frame- work that supports the assembly of customized and previously unseen products. CoViLLM combines depth-camera-based localization for object position estimation, human operator classification for identifying new components, and an Large Language Model (LLM) for assembly task planning based on natural language instructions. The framework is validated on the NIST Assembly Task Board for known, customized, and new product cases. Experimental results show that the proposed framework enables flexible collaborative assembly by extending HRC beyond predefined product and task settings.
comment: 7 pages, 7 figures. Accepted to ASME MSEC 2026
☆ Enhancing Lightweight Vision Language Models through Group Competitive Learning for Socially Compliant Navigation
Social robot navigation requires a sophisticated integration of scene semantics and human social norms. Scaling up Vision Language Models (VLMs) generally improves reasoning and decision-making capabilities for socially compliant navigation. However, increased model size incurs substantial computational overhead, limiting suitability for real-time robotic deployment. Conversely, lightweight VLMs enable efficient inference but often exhibit weaker reasoning and decision-making performance in socially complex environments. Achieving both strong reasoning ability and efficiency remains an open challenge. To bridge this gap, we propose Group Competitive Learning (GCL), a strategy designed to amplify the capabilities of lightweight VLMs. Our strategy introduces the Group Competitive Objective (GCO) to harmonize global semantics with distributional regularization, alongside Asymmetric Group Optimization (AGO) to explore the upper limits of model performance. Empirical evaluations on social navigation benchmarks demonstrate that GCL significantly elevates VLM performance. Specifically, GCL enables the Qwen2.5-VL-3B learner model and guide Qwen3-VL-4B to achieve an F1 score of 0.968 and 0.914, representing 40\% and 12\% improvement over vanilla supervised fine-tuning (SFT). Notably, under vanilla SFT, the 3B model initially trails the 8B model (F1: 0.692 vs. 0.755). However, through the GCL, the 3B model outperforms (28\%) the 8B baseline model. These results suggest that GCL provides an effective solution for achieving both high accuracy and computational efficiency in real-world deployment.
☆ A Generalized Theory of Load Distribution in Redundantly-actuated Robotic Systems
This paper presents a generalized theory which describes how applied loads are distributed within rigid bodies handled by redundantly-actuated robotic systems composed of multiple independent closed-loop kinematic chains. The theory fully characterizes the feasible set of manipulating wrench distributions for a given resultant wrench applied to the rigid body and has important implications for the force-control of multifingered grippers, legged robots, cooperating robots, and other overconstrained mechanisms. We also derive explicit solutions to the wrench synthesis and wrench analysis problems. These solutions are computationally efficient and scale linearly with the number of applied wrenches, requiring neither numerical methods nor the inversion of large matrices. Finally, we identify significant shortcomings in current state-of-the-art approaches and propose corrections. These are supported by illustrative examples that demonstrate the advantages of the improved methods.
comment: 20 pages, 11 figures. Submitted to The International Journal of Robotics Research
☆ Grounding Robot Generalization in Training Data via Retrieval-Augmented VLMs
Recent work on robot manipulation has advanced policy generalization to novel scenarios. However, it is often difficult to characterize how different evaluation settings actually represent generalization from the training distribution of a given policy. To work towards more precise evaluation of generalization in robotics, we propose RADAR, a scalable framework for directly comparing test-time evaluation tasks to policy training data, to determine what form of policy generalization is required. RADAR consists of a two-stage pipeline: first, retrieval using generalist policy embeddings identifies which training examples are relevant for a given evaluation task. Next, vision-language models (VLMs) analyze the evaluation task against the retrieved data, outputting interpretable analysis on how they compare along a variety of axes, and an overall classification of what type of policy generalization is required. Through controlled experiments, we demonstrate that VLMs are effective at analyzing data for generalization, and that our retrieval step effectively identifies examples needed to make accurate classifications with respect to the training data. Furthermore, we scale RADAR to large-scale datasets, where we observe agreement with human-defined benchmark conditions from prior work. We provide demonstrations at radar-analysis.github.io.
comment: 12 pages
☆ Real-time Rendering-based Surgical Instrument Tracking via Evolutionary Optimization
Accurate and efficient tracking of surgical instruments is fundamental for Robot-Assisted Minimally Invasive Surgery. Although vision-based robot pose estimation has enabled markerless calibration without tedious physical setups, reliable tool tracking for surgical robots still remains challenging due to partial visibility and specialized articulation design of surgical instruments. Previous works in the field are usually prone to unreliable feature detections under degraded visual quality and data scarcity, whereas rendering-based methods often struggle with computational costs and suboptimal convergence. In this work, we incorporate CMA-ES, an evolutionary optimization strategy, into a versatile tracking pipeline that jointly estimates surgical instrument pose and joint configurations. Using batch rendering to efficiently evaluate multiple pose candidates in parallel, the method significantly reduces inference time and improves convergence robustness. The proposed framework further generalizes to joint angle-free and bi-manual tracking settings, making it suitable for both vision feedback control and online surgery video calibration. Extensive experiments on synthetic and real-world datasets demonstrate that the proposed method significantly outperforms prior approaches in both accuracy and runtime.
☆ Deployment-Time Reliability of Learned Robot Policies
Recent advances in learning-based robot manipulation have produced policies with remarkable capabilities. Yet, reliability at deployment remains a fundamental barrier to real-world use, where distribution shift, compounding errors, and complex task dependencies collectively undermine system performance. This dissertation investigates how the reliability of learned robot policies can be improved at deployment time through mechanisms that operate around them. We develop three complementary classes of deployment-time mechanisms. First, we introduce runtime monitoring methods that detect impending failures by identifying inconsistencies in closed-loop policy behavior and deviations in task progress, without requiring failure data or task-specific supervision. Second, we propose a data-centric framework for policy interpretability that traces deployment-time successes and failures to influential training demonstrations using influence functions, enabling principled diagnosis and dataset curation. Third, we address reliable long-horizon task execution by formulating policy coordination as the problem of estimating and maximizing the success probability of behavior sequences, and we extend this formulation to open-ended, language-specified tasks through feasibility-aware task planning. By centering on core challenges of deployment, these contributions advance practical foundations for the reliable, real-world use of learned robot policies. Continued progress on these foundations will be essential for enabling trustworthy and scalable robot autonomy in the future.
comment: Stanford University PhD dissertation, 2026. 182 pages, 37 figures. Available from Stanford Digital Repository
☆ $Ψ_0$: An Open Foundation Model Towards Universal Humanoid Loco-Manipulation
We introduce $Ψ_0$ (Psi-Zero), an open foundation model to address challenging humanoid loco-manipulation tasks. While existing approaches often attempt to address this fundamental problem by co-training on large and diverse human and humanoid data, we argue that this strategy is suboptimal due to the fundamental kinematic and motion disparities between humans and humanoid robots. Therefore, data efficiency and model performance remain unsatisfactory despite the considerable data volume. To address this challenge, \ours\;decouples the learning process to maximize the utility of heterogeneous data sources. Specifically, we propose a staged training paradigm with different learning objectives: First, we autoregressively pre-train a VLM backbone on large-scale egocentric human videos to acquire generalizable visual-action representations. Then, we post-train a flow-based action expert on high-quality humanoid robot data to learn precise robot joint control. Our research further identifies a critical yet often overlooked data recipe: in contrast to approaches that scale with noisy Internet clips or heterogeneous cross-embodiment robot datasets, we demonstrate that pre-training on high-quality egocentric human manipulation data followed by post-training on domain-specific real-world humanoid trajectories yields superior performance. Extensive real-world experiments demonstrate that \ours\ achieves the best performance using only about 800 hours of human video data and 30 hours of real-world robot data, outperforming baselines pre-trained on more than 10$\times$ as much data by over 40\% in overall success rate across multiple tasks. We will open-source the entire ecosystem to the community, including a data processing and training pipeline, a humanoid foundation model, and a real-time action inference engine.
☆ HumDex:Humanoid Dexterous Manipulation Made Easy
This paper investigates humanoid whole-body dexterous manipulation, where the efficient collection of high-quality demonstration data remains a central bottleneck. Existing teleoperation systems often suffer from limited portability, occlusion, or insufficient precision, which hinders their applicability to complex whole-body tasks. To address these challenges, we introduce HumDex, a portable teleoperation system designed for humanoid whole-body dexterous manipulation. Our system leverages IMU-based motion tracking to address the portability-precision trade-off, enabling accurate full-body tracking while remaining easy to deploy. For dexterous hand control, we further introduce a learning-based retargeting method that generates smooth and natural hand motions without manual parameter tuning. Beyond teleoperation, HumDex enables efficient collection of human motion data. Building on this capability, we propose a two-stage imitation learning framework that first pre-trains on diverse human motion data to learn generalizable priors, and then fine-tunes on robot data to bridge the embodiment gap for precise execution. We demonstrate that this approach significantly improves generalization to new configurations, objects, and backgrounds with minimal data acquisition costs. The entire system is fully reproducible and open-sourced at https://github.com/physical-superintelligence-lab/HumDex.
☆ HandelBot: Real-World Piano Playing via Fast Adaptation of Dexterous Robot Policies
Mastering dexterous manipulation with multi-fingered hands has been a grand challenge in robotics for decades. Despite its potential, the difficulty of collecting high-quality data remains a primary bottleneck for high-precision tasks. While reinforcement learning and simulation-to-real-world transfer offer a promising alternative, the transferred policies often fail for tasks demanding millimeter-scale precision, such as bimanual piano playing. In this work, we introduce HandelBot, a framework that combines a simulation policy and rapid adaptation through a two-stage pipeline. Starting from a simulation-trained policy, we first apply a structured refinement stage to correct spatial alignments by adjusting lateral finger joints based on physical rollouts. Next, we use residual reinforcement learning to autonomously learn fine-grained corrective actions. Through extensive hardware experiments across five recognized songs, we demonstrate that HandelBot can successfully perform precise bimanual piano playing. Our system outperforms direct simulation deployment by a factor of 1.8x and requires only 30 minutes of physical interaction data.
comment: Website: https://amberxie88.github.io/handelbot
☆ SaPaVe: Towards Active Perception and Manipulation in Vision-Language-Action Models for Robotics CVPR 2026
Active perception and manipulation are crucial for robots to interact with complex scenes. Existing methods struggle to unify semantic-driven active perception with robust, viewpoint-invariant execution. We propose SaPaVe, an end-to-end framework that jointly learns these capabilities in a data-efficient manner. Our approach decouples camera and manipulation actions rather than placing them in a shared action space, and follows a bottom-up training strategy: we first train semantic camera control on a large-scale dataset, then jointly optimize both action types using hybrid data. To support this framework, we introduce ActiveViewPose-200K, a dataset of 200k image-language-camera movement pairs for semantic camera movement learning, and a 3D geometry-aware module that improves execution robustness under dynamic viewpoints. We also present ActiveManip-Bench, the first benchmark for evaluating active manipulation beyond fixed-view settings. Extensive experiments in both simulation and real-world environments show that SaPaVe outperforms recent vision-language-action models such as GR00T N1 and \(π_0\), achieving up to 31.25\% higher success rates in real-world tasks. These results show that tightly coupled perception and execution, when trained with decoupled yet coordinated strategies, enable efficient and generalizable active manipulation. Project page: https://lmzpai.github.io/SaPaVe
comment: Accepted to CVPR 2026. See project page at https://lmzpai.github.io/SaPaVe
☆ ComFree-Sim: A GPU-Parallelized Analytical Contact Physics Engine for Scalable Contact-Rich Robotics Simulation and Control
Physics simulation for contact-rich robotics is often bottlenecked by contact resolution: mainstream engines enforce non-penetration and Coulomb friction via complementarity constraints or constrained optimization, requiring per-step iterative solves whose cost grows superlinearly with contact density. We present ComFree-Sim, a GPU-parallelized analytical contact physics engine built on complementarity-free contact modeling. ComFree-Sim computes contact impulses in closed form via an impedance-style prediction--correction update in the dual cone of Coulomb friction. Contact computation decouples across contact pairs and becomes separable across cone facets, mapping naturally to GPU kernels and yielding near-linear runtime scaling with the number of contacts. We further extend the formulation to a unified 6D contact model capturing tangential, torsional, and rolling friction, and introduce a practical dual-cone impedance heuristic. ComFree-Sim is implemented in Warp and exposed through a MuJoCo-compatible interface as a drop-in backend alternative to MuJoCo Warp (MJWarp). Experiments benchmark penetration, friction behaviors, stability, and simulation runtime scaling against MJWarp, demonstrating near-linear scaling and 2--3 times higher throughput in dense contact scenes with comparable physical fidelity. We deploy ComFree-Sim in real-time MPC for in-hand dexterous manipulation on a real-world multi-fingered LEAP hand and in dynamics-aware motion retargeting, demonstrating that low-latency simulation yields higher closed-loop success rates and enables practical high-frequency control in contact-rich tasks.
comment: 9 pages
☆ O3N: Omnidirectional Open-Vocabulary Occupancy Prediction
Understanding and reconstructing the 3D world through omnidirectional perception is an inevitable trend in the development of autonomous agents and embodied intelligence. However, existing 3D occupancy prediction methods are constrained by limited perspective inputs and predefined training distribution, making them difficult to apply to embodied agents that require comprehensive and safe perception of scenes in open world exploration. To address this, we present O3N, the first purely visual, end-to-end Omnidirectional Open-vocabulary Occupancy predictioN framework. O3N embeds omnidirectional voxels in a polar-spiral topology via the Polar-spiral Mamba (PsM) module, enabling continuous spatial representation and long-range context modeling across 360°. The Occupancy Cost Aggregation (OCA) module introduces a principled mechanism for unifying geometric and semantic supervision within the voxel space, ensuring consistency between the reconstructed geometry and the underlying semantic structure. Moreover, Natural Modality Alignment (NMA) establishes a gradient-free alignment pathway that harmonizes visual features, voxel embeddings, and text semantics, forming a consistent "pixel-voxel-text" representation triad. Extensive experiments on multiple models demonstrate that our method not only achieves state-of-the-art performance on QuadOcc and Human360Occ benchmarks but also exhibits remarkable cross-scene generalization and semantic scalability, paving the way toward universal 3D world modeling. The source code will be made publicly available at https://github.com/MengfeiD/O3N.
comment: The source code will be made publicly available at https://github.com/MengfeiD/O3N
☆ Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies
Vision-Language-Action (VLA) models have significant potential to enable general-purpose robotic systems for a range of vision-language tasks. However, the performance of VLA-based robots is highly sensitive to the precise wording of language instructions, and it remains difficult to predict when such robots will fail. To improve the robustness of VLAs to different wordings, we present Q-DIG (Quality Diversity for Diverse Instruction Generation), which performs red-teaming by scalably identifying diverse natural language task descriptions that induce failures while remaining task-relevant. Q-DIG integrates Quality Diversity (QD) techniques with Vision-Language Models (VLMs) to generate a broad spectrum of adversarial instructions that expose meaningful vulnerabilities in VLA behavior. Our results across multiple simulation benchmarks show that Q-DIG finds more diverse and meaningful failure modes compared to baseline methods, and that fine-tuning VLAs on the generated instructions improves task success rates. Furthermore, results from a user study highlight that Q-DIG generates prompts judged to be more natural and human-like than those from baselines. Finally, real-world evaluations of Q-DIG prompts show results consistent with simulation, and fine-tuning VLAs on the generated prompts further success rates on unseen instructions. Together, these findings suggest that Q-DIG is a promising approach for identifying vulnerabilities and improving the robustness of VLA-based robots. Our anonymous project website is at qdigvla.github.io.
☆ Robots that redesign themselves through kinematic self-destruction
Every robot built to date was predesigned by an external process, prior to deployment. Here we show a robot that actively participates in its own design during its lifetime. Starting from a randomly assembled body, and using only proprioceptive feedback, the robot dynamically ``sculpts'' itself into a new design through kinematic self-destruction: identifying redundant links within its body that inhibit its locomotion, and then thrashing those links against the surface until they break at the joint and fall off the body. It does so using a single autoregressive sequence model, a universal controller that learns in simulation when and how to simplify a robot's body through self-destruction and then adaptively controls the reduced morphology. The optimized policy successfully transfers to reality and generalizes to previously unseen kinematic trees, generating forward locomotion that is more effective than otherwise equivalent policies that randomly remove links or cannot remove any. This suggests that self-designing robots may be more successful than predesigned robots in some cases, and that kinematic self-destruction, though reductive and irreversible, could provide a general adaptive strategy for a wide range of robots.
☆ COAD: Constant-Time Planning for Continuous Goal Manipulation with Compressed Library and Online Adaptation
In many robotic manipulation tasks, the robot repeatedly solves motion-planning problems that differ mainly in the location of the goal object and its associated obstacle, while the surrounding workspace remains fixed. Prior works have shown that leveraging experience and offline computation can accelerate repeated planning queries, but they lack guarantees of covering the continuous task space and require storing large libraries of solutions. In this work, we present COAD, a framework that provides constant-time planning over a continuous goal-parameterized task space. COAD discretizes the continuous task space into finitely many Task Coverage Regions. Instead of planning and storing solutions for every region offline, it constructs a compressed library by only solving representative root problems. Other problems are handled through fast adaptation from these root solutions. At query time, the system retrieves a root motion in constant time and adapts it to the desired goal using lightweight adaptation modules such as linear interpolation, Dynamic Movement Primitives, or simple trajectory optimization. We evaluate the framework on various manipulators and environments in simulation and the real world, showing that COAD achieves substantial compression of the motion library while maintaining high success rates and sub-millisecond-level queries, outperforming baseline methods in both efficiency and path quality. The source code is available at https://github.com/elpis-lab/CoAd.
comment: Adil Shiyas and Zhuoyun Zhong contributed equally to this work
☆ One-Step Flow Policy: Self-Distillation for Fast Visuomotor Policies
Generative flow and diffusion models provide the continuous, multimodal action distributions needed for high-precision robotic policies. However, their reliance on iterative sampling introduces severe inference latency, degrading control frequency and harming performance in time-sensitive manipulation. To address this problem, we propose the One-Step Flow Policy (OFP), a from-scratch self-distillation framework for high-fidelity, single-step action generation without a pre-trained teacher. OFP unifies a self-consistency loss to enforce coherent transport across time intervals, and a self-guided regularization to sharpen predictions toward high-density expert modes. In addition, a warm-start mechanism leverages temporal action correlations to minimize the generative transport distance. Evaluations across 56 diverse simulated manipulation tasks demonstrate that a one-step OFP achieves state-of-the-art results, outperforming 100-step diffusion and flow policies while accelerating action generation by over $100\times$. We further integrate OFP into the $π_{0.5}$ model on RoboTwin 2.0, where one-step OFP surpasses the original 10-step policy. These results establish OFP as a practical, scalable solution for highly accurate and low-latency robot control.
☆ Predictive and adaptive maps for long-term visual navigation in changing environments
In this paper, we compare different map management techniques for long-term visual navigation in changing environments. In this scenario, the navigation system needs to continuously update and refine its feature map in order to adapt to the environment appearance change. To achieve reliable long-term navigation, the map management techniques have to (i) select features useful for the current navigation task, (ii) remove features that are obsolete, (iii) and add new features from the current camera view to the map. We propose several map management strategies and evaluate their performance with regard to the robot localisation accuracy in long-term teach-and-repeat navigation. Our experiments, performed over three months, indicate that strategies which model cyclic changes of the environment appearance and predict which features are going to be visible at a particular time and location, outperform strategies which do not explicitly model the temporal evolution of the changes.
☆ Beyond Motion Imitation: Is Human Motion Data Alone Sufficient to Explain Gait Control and Biomechanics?
With the growing interest in motion imitation learning (IL) for human biomechanics and wearable robotics, this study investigates how additional foot-ground interaction measures, used as reward terms, affect human gait kinematics and kinetics estimation within a reinforcement learning-based IL framework. Results indicate that accurate reproduction of forward kinematics alone does not ensure biomechanically plausible joint kinetics. Adding foot-ground contacts and contact forces to the IL reward terms enables the prediction of joint moments in forward walking simulation, which are significantly closer to those computed by inverse dynamics. This finding highlights a fundamental limitation of motion-only IL approaches, which may prioritize kinematics matching over physical consistency. Incorporating kinetic constraints, particularly ground reaction force and center of pressure information, significantly enhances the realism of internal and external kinetics. These findings suggest that, when imitation learning is applied to human-related research domains such as biomechanics and wearable robot co-design, kinetics-based reward shaping is necessary to achieve physically consistent gait representations.
comment: 8 pages, 7 figures
☆ Push, Press, Slide: Mode-Aware Planar Contact Manipulation via Reduced-Order Models IROS 2026
Non-prehensile planar manipulation, including pushing and press-and-slide, is critical for diverse robotic tasks, but notoriously challenging due to hybrid contact mechanics, under-actuation, and asymmetric friction limits that traditionally necessitate computationally expensive iterative control. In this paper, we propose a mode-aware framework for planar manipulation with one or two robotic arms based on contact topology selection and reduced-order kinematic modeling. Our core insight is that complex wrench-twist limit surface mechanics can be abstracted into a discrete library of physically intuitive models. We systematically map various single-arm and bimanual contact topologies to simple non-holonomic formulations, e.g. unicycle for simplified press-and-slide motion. By anchoring trajectory generation to these reduced-order models, our framework computes the required object wrench and distributes feasible, friction-bounded contact forces via a direct algebraic allocator. We incorporate manipulator kinematics to ensure long-horizon feasibility and demonstrate our fast, optimization-free approach in simulation across diverse single-arm and bimanual manipulation tasks.
comment: 8 pages, 13 figures. Submitted to IEEE IROS 2026
☆ GNN-DIP: Neural Corridor Selection for Decomposition-Based Motion Planning
Motion planning through narrow passages remains a core challenge: sampling-based planners rarely place samples inside these narrow but critical regions, and even when samples land inside a passage, the straight-line connections between them run close to obstacle boundaries and are frequently rejected by collision checking. Decomposition-based planners resolve both issues by partitioning free space into convex cells -- every passage is captured exactly as a cell boundary, and any path within a cell is collision-free by construction. However, the number of candidate corridors through the cell graph grows combinatorially with environment complexity, creating a bottleneck in corridor selection. We present GNN-DIP, a framework that addresses this by integrating a Graph Neural Network (GNN) with a two-phase Decomposition-Informed Planner (DIP). The GNN predicts portal scores on the cell adjacency graph to bias corridor search toward near-optimal regions while preserving completeness. In 2D, Constrained Delaunay Triangulation (CDT) with the Funnel algorithm yields exact shortest paths within corridors; in 3D, Slab convex decomposition with portal-face sampling provides near-optimal path evaluation. Benchmarks on 2D narrow-passage scenarios, 3D bottleneck environments with up to 246 obstacles, and dynamic 2D settings show that GNN-DIP achieves 99--100% success rates with 2--280 times speedup over sampling-based baselines.
☆ A Learning-Based Approach for Contact Detection, Localization, and Force Estimation of Continuum Manipulators With Integrated OFDR Optical Fiber
Continuum manipulators (CMs) are widely used in minimally invasive procedures due to their compliant structure and ability to navigate deep and confined anatomical environments. However, their distributed deformation makes force sensing, contact detection, localization, and force estimation challenging, particularly when interactions occur at unknown arc-length locations along the robot. To address this problem, we propose a cascade learning-based framework (CLF) for CMs instrumented with a single distributed Optical Frequency Domain Reflectometry (OFDR) fiber embedded along one side of the robot. The OFDR sensor provides dense strain measurements along the manipulator backbone, capturing strain perturbations caused by external interactions. The proposed CLF first detects contact using a Gradient Boosting classifier and then estimates contact location and interaction force magnitude using a CNN--FiLM model that predicts a spatial force distribution along the manipulator. Experimental validation on a sensorized tendon-driven CM in an obstructed environment demonstrates that a single distributed OFDR fiber provides sufficient information to jointly infer contact occurrence, location, and force in continuum manipulators.
comment: 8 pages, 6 figures
♻ ☆ Whleaper: A 10-DOF Flexible Bipedal Wheeled Robot
Wheel-legged robots combine the advantages of both wheeled robots and legged robots, offering versatile locomotion capabilities with excellent stability on challenging terrains and high efficiency on flat surfaces. However, existing wheel-legged robots typically have limited hip joint mobility compared to humans, while hip joint plays a crucial role in locomotion. In this paper, we introduce Whleaper, a novel 10-degree-of-freedom (DOF) bipedal wheeled robot, with 3 DOFs at the hip of each leg. Its humanoid joint design enables adaptable motion in complex scenarios, ensuring stability and flexibility. This paper introduces the details of Whleaper, with a focus on innovative mechanical design, control algorithms and system implementation. Firstly, stability stems from the increased DOFs at the hip, which expand the range of possible postures and improve the robot's foot-ground contact. Secondly, the extra DOFs also augment its mobility. During walking or sliding, more complex movements can be adopted to execute obstacle avoidance tasks. Thirdly, we utilize two control algorithms to implement multimodal motion for walking and sliding. By controlling specific DOFs of the robot, we conducted a series of simulations and practical experiments, demonstrating that a high-DOF hip joint design can effectively enhance the stability and flexibility of wheel-legged robots. Whleaper shows its capability to perform actions such as squatting, obstacle avoidance sliding, and rapid turning in real-world scenarios.
♻ ☆ Robust Cooperative Localization in Featureless Environments: A Comparative Study of DCL, StCL, CCL, CI, and Standard-CL
Cooperative localization (CL) enables accurate position estimation in multi-robot systems operating in GPS-denied environments. This paper presents a comparative study of five CL approaches: Centralized Cooperative Localization (CCL), Decentralized Cooperative Localization (DCL), Sequential Cooperative Localization (StCL), Covariance Intersection (CI), and Standard Cooperative Localization (Standard-CL). All methods are implemented in ROS and evaluated through Monte Carlo simulations under two conditions: weak data association and robust detection. Our analysis reveals fundamental trade-offs among the methods. StCL and Standard-CL achieve the lowest position errors but exhibit severe filter inconsistency, making them unsuitable for safety-critical applications. DCL demonstrates remarkable stability under challenging conditions due to its measurement stride mechanism, which provides implicit regularization against outliers. CI emerges as the most balanced approach, achieving near-optimal consistency while maintaining competitive accuracy. CCL provides theoretically optimal estimation but shows sensitivity to measurement outliers. These findings offer practical guidance for selecting CL algorithms based on application requirements.
comment: Accepted and presented at the 2026 12th International Conference on Automation, Robotics and Applications (ICARA); to appear in IEEE conference proceedings
♻ ☆ Online Slip Detection and Friction Coefficient Estimation for Autonomous Racing
Accurate knowledge of the tire-road friction coefficient (TRFC) is essential for vehicle safety, stability, and performance, especially in autonomous racing, where vehicles often operate at the friction limit. However, TRFC cannot be directly measured with standard sensors, and existing estimation methods either depend on vehicle or tire models with uncertain parameters or require large training datasets. In this paper, we present a lightweight approach for online slip detection and TRFC estimation. Our approach relies solely on IMU and LiDAR measurements and the control actions, without special dynamical or tire models, parameter identification, or training data. Slip events are detected in real time by comparing commanded and measured motions, and the TRFC is then estimated directly from observed accelerations under no-slip conditions. Experiments with a 1:10-scale autonomous racing car across different friction levels demonstrate that the proposed approach achieves accurate and consistent slip detections and friction coefficients, with results closely matching ground-truth measurements. These findings highlight the potential of our simple, deployable, and computationally efficient approach for real-time slip monitoring and friction coefficient estimation in autonomous driving.
comment: Equal contribution by the first three authors
♻ ☆ Parallel-in-Time Nonlinear Optimal Control via GPU-native Sequential Convex Programming
Real-time trajectory optimization for nonlinear constrained autonomous systems is critical and typically performed by CPU-based sequential solvers. Specifically, reliance on global sparse linear algebra or the serial nature of dynamic programming algorithms restricts the utilization of massively parallel computing architectures like GPUs. To bridge this gap, we introduce a fully GPU-native trajectory optimization framework that combines sequential convex programming with a consensus-based alternating direction method of multipliers. By applying a temporal splitting strategy, our algorithm decouples the optimization horizon into independent, per-node subproblems that execute massively in parallel. The entire process runs fully on the GPU, eliminating costly memory transfers and large-scale sparse factorizations. This architecture naturally scales to multi-trajectory optimization. We validate the solver on a quadrotor agile flight task and a Mars powered descent problem using an on-board edge computing platform. Benchmarks reveal a sustained 4x throughput speedup and a 51% reduction in energy consumption over a heavily optimized 12-core CPU baseline. Crucially, the framework saturates the hardware, maintaining over 96% active GPU utilization to achieve planning rates exceeding 100 Hz. Furthermore, we demonstrate the solver's extensibility to robust Model Predictive Control by jointly optimizing dynamically coupled scenarios under stochastic disturbances, enabling scalable and safe autonomy.
♻ ☆ Safe and Stylized Trajectory Planning for Autonomous Driving via Diffusion Model
Achieving safe and stylized trajectory planning in complex real-world scenarios remains a critical challenge for autonomous driving systems. This paper proposes the SDD Planner, a diffusion-based framework designed to effectively reconcile safety constraints with driving styles in real time. The framework integrates two core modules: a Multi-Source Style-Aware Encoder, which employs distance-sensitive attention to fuse dynamic agent data and environmental contexts for heterogeneous safety-style perception; and a Style-Guided Dynamic Trajectory Generator, which adaptively modulates priority weights within the diffusion denoising process to generate user-preferred yet safe trajectories. Extensive experiments demonstrate that SDD Planner achieves state-of-the-art performance. On the StyleDrive benchmark, it improves the SM-PDMS metric by 3.9% over WoTE, the strongest baseline. Furthermore, on the NuPlan Test14 and Test14-hard benchmarks, SDD Planner ranks first with overall scores of 91.76 and 80.32, respectively, outperforming leading methods such as PLUTO. Real-vehicle closed-loop tests further confirm that SDD Planner maintains high safety standards while aligning with preset driving styles, validating its practical applicability for real-world deployment.
comment: 12 pages, 7 figures, submitted to IEEE Transactions on Intelligent Transportation Systems
♻ ☆ 4D Radar-Inertial Odometry based on Gaussian Modeling and Multi-Hypothesis Scan Matching
4D millimeter-wave (mmWave) radars are sensors that provide robustness against adverse weather conditions (rain, snow, fog, etc.), and as such they are increasingly used for odometry and SLAM (Simultaneous Location and Mapping). However, the noisy and sparse nature of the returned scan data proves to be a challenging obstacle for existing registration algorithms, especially those originally intended for more accurate sensors such as LiDAR. Following the success of 3D Gaussian Splatting for vision, in this paper we propose a summarized representation for radar scenes based on global simultaneous optimization of 3D Gaussians as opposed to voxel-based approaches, and leveraging its inherent Probability Density Function (PDF) for registration. Moreover, we propose optimizing multiple registration hypotheses for better protection against local optima of the PDF. We evaluate our modeling and registration system against state of the art techniques, finding that our system provides richer models and more accurate registration results. Finally, we evaluate the effectiveness of our system in a real Radar-Inertial Odometry task. Experiments using publicly available 4D radar datasets show that our Gaussian approach is comparable to existing registration algorithms, outperforming them in several sequences. Copyright 2026 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
comment: Our code and results can be publicly accessed at: https://github.com/robotics-upo/gaussian-rio-cpp Accepted for publication in IEEE Robotics and Automation Letters
♻ ☆ STONE Dataset: A Scalable Multi-Modal Surround-View 3D Traversability Dataset for Off-Road Robot Navigation ICRA 2026
Reliable off-road navigation requires accurate estimation of traversable regions and robust perception under diverse terrain and sensing conditions. However, existing datasets lack both scalability and multi-modality, which limits progress in 3D traversability prediction. In this work, we introduce STONE, a large-scale multi-modal dataset for off-road navigation. STONE provides (1) trajectory-guided 3D traversability maps generated by a fully automated, annotation-free pipeline, and (2) comprehensive surround-view sensing with synchronized 128-channel LiDAR, six RGB cameras, and three 4D imaging radars. The dataset covers a wide range of environments and conditions, including day and night, grasslands, farmlands, construction sites, and lakes. Our auto-labeling pipeline reconstructs dense terrain surfaces from LiDAR scans, extracts geometric attributes such as slope, elevation, and roughness, and assigns traversability labels beyond the robot's trajectory using a Mahalanobis-distance-based criterion. This design enables scalable, geometry-aware ground-truth construction without manual annotation. Finally, we establish a benchmark for voxel-level 3D traversability prediction and provide strong baselines under both single-modal and multi-modal settings. STONE is available at: https://konyul.github.io/STONE-dataset/
comment: ICRA 2026
♻ ☆ FSAG: Enhancing Human-to-Dexterous-Hand Finger-Specific Affordance Grounding via Diffusion Models
Dexterous grasp synthesis must jointly satisfy functional intent and physical feasibility, yet existing pipelines often decouple semantic grounding from refinement, yielding unstable or non-functional contacts under object and pose variations. This challenge is exacerbated by the high dimensionality and kinematic diversity of multi-fingered hands, which makes many methods rely on large, hardware-specific grasp datasets collected in simulation or through costly real-world trials. We propose a data-efficient framework that bypasses robot grasp data collection by exploiting object-centric semantic priors in pretrained generative diffusion models. Temporally aligned and fine-grained grasp affordances are extracted from raw human video demonstrations and fused with 3D scene geometry from depth images to infer semantically grounded contact targets. We further incorporate these affordance regions into the grasp refinement objective, explicitly guiding each fingertip toward its predicted region during optimization. The resulting system produces stable, human-intuitive multi-contact grasps across common objects and tools, while exhibiting strong generalization to previously unseen object instances within a category, pose variations, and multiple hand embodiments.This work (i) introduces a semantic affordance extraction pipeline leveraging vision--language generative priors for dexterous grasping, (ii) demonstrates cross-hand generalization without constructing hardware-specific grasp datasets, and (iii) establishes that a single depth modality suffices for high-performance grasp synthesis when coupled with foundation-model semantics. Our results highlight a path toward scalable, hardware-agnostic dexterous manipulation driven by human demonstrations and pretrained generative models.
♻ ☆ ManiVID-3D: Generalizable View-Invariant Reinforcement Learning for Robotic Manipulation via Disentangled 3D Representations
Deploying visual reinforcement learning (RL) policies in real-world manipulation is often hindered by camera viewpoint changes. A policy trained from a fixed front-facing camera may fail when the camera is shifted -- an unavoidable situation in real-world settings where sensor placement is hard to manage appropriately. Existing methods often rely on precise camera calibration or struggle with large perspective changes. To address these limitations, we propose ManiVID-3D, a novel 3D RL architecture designed for robotic manipulation, which learns view-invariant representations through self-supervised disentangled feature learning. The framework incorporates ViewNet, a lightweight yet effective module that automatically aligns point cloud observations from arbitrary viewpoints into a unified spatial coordinate system without the need for extrinsic calibration. Additionally, we develop an efficient GPU-accelerated batch rendering module capable of processing over 5000 frames per second, enabling large-scale training for 3D visual RL at unprecedented speeds. Extensive evaluation across 10 simulated and 5 real-world tasks demonstrates that our approach achieves a 40.6% higher success rate than state-of-the-art methods under viewpoint variations while using 80% fewer parameters. The system's robustness to severe perspective changes and strong sim-to-real performance highlight the effectiveness of learning geometrically consistent representations for scalable robotic manipulation in unstructured environments.
comment: Accepted to RA-L. Project website: https://zheng-joe-lee.github.io/manivid3d/
♻ ☆ Enhancing Heterogeneous Multi-Agent Cooperation in Decentralized MARL via GNN-driven Intrinsic Rewards AAMAS 2025
Multi-agent Reinforcement Learning (MARL) is emerging as a key framework for various sequential decision-making and control tasks. Unlike their single-agent counterparts, multi-agent systems necessitate successful cooperation among the agents. The deployment of these systems in real-world scenarios often requires decentralized training, a diverse set of agents, and learning from infrequent environmental reward signals. These challenges become more pronounced under partial observability and the lack of prior knowledge about agent heterogeneity. While notable studies use intrinsic motivation (IM) to address reward sparsity or cooperation in decentralized settings, those dealing with heterogeneity typically assume centralized training, parameter sharing, and agent indexing. To overcome these limitations, we propose the CoHet algorithm, which utilizes a novel Graph Neural Network (GNN) based intrinsic motivation to facilitate the learning of heterogeneous agent policies in decentralized settings, under the challenges of partial observability and reward sparsity. Evaluation of CoHet in the Multi-agent Particle Environment (MPE) and Vectorized Multi-Agent Simulator (VMAS) benchmarks demonstrates superior performance compared to the state-of-the-art in a range of cooperative multi-agent scenarios. Our research is supplemented by an analysis of the impact of the agent dynamics model on the intrinsic motivation module, insights into the performance of different CoHet variants, and its robustness to an increasing number of heterogeneous agents.
comment: Full paper version for AAMAS 2025 (https://ifaamas.org/Proceedings/aamas2025/pdfs/p2681.pdf), 9 pages, 5 figures
♻ ☆ XGrasp: Gripper-Aware Grasp Detection with Multi-Gripper Data Generation
Real-world robotic systems frequently require diverse end-effectors for different tasks, however most existing grasp detection methods are optimized for a single gripper type, demanding retraining or optimization for each novel gripper configuration. This gripper-specific retraining paradigm is neither scalable nor practical. We propose XGrasp, a real-time gripper-aware grasp detection framework that generalizes to novel gripper configurations without additional training or optimization. To resolve data scarcity, we augment existing single-gripper datasets with multi-gripper annotations by incorporating the physical characteristics and closing trajectories of diverse grippers. Each gripper is represented as a two-channel 2D image encoding its static shape (Gripper Mask) and dynamic closing trajectory (Gripper Path). XGrasp employs a hierarchical two-stage architecture consisting of a Grasp Point Predictor (GPP) and an Angle-Width Predictor (AWP). In the AWP, contrastive learning with a quality-aware anchor builds a gripper-agnostic embedding space, enabling generalization to novel grippers without additional training. Experimental results demonstrate that XGrasp outperforms existing gripper-aware methods in both grasp success rate and inference speed across diverse gripper types. Project page: https://sites.google.com/view/xgrasp
comment: 9 pages, 10 figures
♻ ☆ DRIFT: Dual-Representation Inter-Fusion Transformer for Automated Driving Perception with 4D Radar Point Clouds
4D radars, which provide 3D point cloud data along with Doppler velocity, are attractive components of modern automated driving systems due to their low cost and robustness under adverse weather conditions. However, they provide a significantly lower point cloud density than LiDAR sensors. This makes it important to exploit not only local but also global contextual scene information. This paper proposes DRIFT, a model that effectively captures and fuses both local and global contexts through a dual-path architecture. The model incorporates a point path to aggregate fine-grained local features and a pillar path to encode coarse-grained global features. These two parallel paths are intertwined via novel feature-sharing layers at multiple stages, enabling full utilization of both representations. DRIFT is evaluated on the widely used View-of-Delft (VoD) dataset and a proprietary internal dataset. It outperforms the baselines on the tasks of object detection and/or free road estimation. For example, DRIFT achieves a mean average precision (mAP) of 52.6% (compared to, say, 45.4% of CenterPoint) on the VoD dataset.
♻ ☆ Beyond Description: Cognitively Benchmarking Fine-Grained Action for Embodied Agents
Multimodal Large Language Models (MLLMs) show promising results as decision-making engines for embodied agents operating in complex, physical environments. However, existing benchmarks often prioritize high-level planning or spatial reasoning, leaving the fine-grained action intelligence required for embodied physical interaction underexplored. To address this gap, we introduce CFG-Bench, a new benchmark designed to systematically evaluate this crucial capability. CFG-Bench consists of 1,368 curated videos paired with 19,562 question-answer pairs spanning three evaluation paradigms targeting four cognitive abilities: 1) Physical Interaction, 2) Temporal-Causal Relation, 3) Intentional Understanding, and 4) Evaluative Judgment. Together, these dimensions provide a systematic framework for assessing a model's ability to translate visual observations into actionable knowledge, moving beyond mere surface-level recognition. Our comprehensive evaluation on CFG-Bench reveals that leading MLLMs struggle to produce detailed instructions for physical interactions and exhibit profound limitations in the higher-order reasoning of intention and evaluation. Moreover, supervised fine-tuning (SFT) on our data demonstrates that teaching an MLLMs to articulate fine-grained actions directly translates to significant performance gains on established embodied benchmarks. Our analysis highlights these limitations and offers insights for developing more capable and grounded embodied agents. Project page: https://cfg-bench.github.io/
♻ ☆ RoboRouter: Training-Free Policy Routing for Robotic Manipulation
Research on robotic manipulation has developed a diverse set of policy paradigms, including vision-language-action (VLA) models, vision-action (VA) policies, and code-based compositional approaches. Concrete policies typically attain high success rates on specific task distributions but lim-ited generalization beyond it. Rather than proposing an other monolithic policy, we propose to leverage the complementary strengths of existing approaches through intelligent policy routing. We introduce RoboRouter, a training-free framework that maintains a pool of heterogeneous policies and learns to select the best-performing policy for each task through accumulated execution experience. Given a new task, RoboRouter constructs a semantic task representation, retrieves historical records of similar tasks, predicts the optimal policy choice without requiring trial-and-error, and incorporates structured feedback to refine subsequent routing decisions. Integrating a new policy into the system requires only lightweight evaluation and incurs no training overhead. Across simulation benchmark and real-world evaluations, RoboRouter consistently outperforms than in-dividual policies, improving average success rate by more than 3% in simulation and over 13% in real-world settings, while preserving execution efficiency. Our results demonstrate that intelligent routing across heterogeneous, off-the-shelf policies provides a practical and scalable pathway toward building more capable robotic systems.
comment: We need to withdraw the paper as some of the reference papers are incorrect and need to be removed
♻ ☆ KnowVal: A Knowledge-Augmented and Value-Guided Autonomous Driving System CVPR 2026
Visual-language reasoning, driving knowledge, and value alignment are essential for advanced autonomous driving systems. However, existing approaches largely rely on data-driven learning, making it difficult to capture the complex logic underlying decision-making through imitation or limited reinforcement rewards. To address this, we propose KnowVal, a new autonomous driving system that enables visual-language reasoning through the synergistic integration of open-world perception and knowledge retrieval. Specifically, we construct a comprehensive driving knowledge graph that encodes traffic laws, defensive driving principles, and ethical norms, complemented by an efficient LLM-based retrieval mechanism tailored for driving scenarios. Furthermore, we develop a human-preference dataset and train a Value Model to guide interpretable, value-aligned trajectory assessment. Experimental results show that our method substantially improves planning performance while remaining compatible with existing architectures. Notably, KnowVal achieves the lowest collision rate on nuScenes and state-of-the-art results on Bench2Drive and NVISIM.
comment: Accepted to CVPR 2026
♻ ☆ Hyperbolic Multiview Pretraining for Robotic Manipulation CVPR 2026
3D-aware visual pretraining has proven effective in improving the performance of downstream robotic manipulation tasks. However, existing methods are constrained to Euclidean embedding spaces, whose flat geometry limits their ability to model structural relations among embeddings. As a result, they struggle to learn structured embeddings that are essential for robust spatial perception in robotic applications. To this end, we propose HyperMVP, a self-supervised framework for \underline{Hyper}bolic \underline{M}ulti\underline{V}iew \underline{P}retraining. Hyperbolic space offers geometric properties well suited for capturing structural relations. Methodologically, we extend the masked autoencoder paradigm and design a GeoLink encoder to learn multiview hyperbolic representations. The pretrained encoder is then finetuned with visuomotor policies on manipulation tasks. In addition, we introduce 3D-MOV, a large-scale dataset comprising multiple types of 3D point clouds to support pretraining. We evaluate HyperMVP on COLOSSEUM, RLBench, and real-world scenarios, where it consistently outperforms strong baselines across diverse tasks and perturbation settings. Our results highlight the potential of 3D-aware pretraining in a non-Euclidean space for learning robust and generalizable robotic manipulation policies.
comment: This paper was submitted to CVPR 2026 and was recommended for Findings, but the authors have withdrawn it and are currently adding more content to submit it elsewhere
♻ ☆ GUIDES: Guidance Using Instructor-Distilled Embeddings for Pre-trained Robot Policy Enhancement ICRA 2026
Pre-trained robot policies serve as the foundation of many validated robotic systems, which encapsulate extensive embodied knowledge. However, they often lack the semantic awareness characteristic of foundation models, and replacing them entirely is impractical in many situations due to high costs and the loss of accumulated knowledge. To address this gap, we introduce GUIDES, a lightweight framework that augments pre-trained policies with semantic guidance from foundation models without requiring architectural redesign. GUIDES employs a fine-tuned vision-language model (Instructor) to generate contextual instructions, which are encoded by an auxiliary module into guidance embeddings. These embeddings are injected into the policy's latent space, allowing the legacy model to adapt to this new semantic input through brief, targeted fine-tuning. For inference-time robustness, a large language model-based Reflector monitors the Instructor's confidence and, when confidence is low, initiates a reasoning loop that analyzes execution history, retrieves relevant examples, and augments the VLM's context to refine subsequent actions. Extensive validation in the RoboCasa simulation environment across diverse policy architectures shows consistent and substantial improvements in task success rates. Real-world deployment on a UR5 robot further demonstrates that GUIDES enhances motion precision for critical sub-tasks such as grasping. Overall, GUIDES offers a practical and resource-efficient pathway to upgrade, rather than replace, validated robot policies.
comment: IEEE International Conference on Robotics and Automation (ICRA 2026)
♻ ☆ Efficient Construction of Implicit Surface Models From a Single Image for Motion Generation ICRA
Implicit representations have been widely applied in robotics for obstacle avoidance and path planning. In this paper, we explore the problem of constructing an implicit distance representation from a single image. Past methods for implicit surface reconstruction, such as NeuS and its variants generally require a large set of multi-view images as input, and require long training times. In this work, we propose Fast Image-to-Neural Surface (FINS), a lightweight framework that can reconstruct high-fidelity surfaces and SDF fields based on a single or a small set of images. FINS integrates a multi-resolution hash grid encoder with lightweight geometry and color heads, making the training via an approximate second-order optimizer highly efficient and capable of converging within a few seconds. Additionally, we achieve the construction of a neural surface requiring only a single RGB image, by leveraging pre-trained foundation models to estimate the geometry inherent in the image. Our experiments demonstrate that under the same conditions, our method outperforms state-of-the-art baselines in both convergence speed and accuracy on surface reconstruction and SDF field estimation. Moreover, we demonstrate the applicability of FINS for robot surface following tasks and show its scalability to a variety of benchmark datasets. Code is publicly available at https://github.com/waynechu1109/FINS.
comment: 9 pages, 6 figures, 2026 IEEE International Conference on Robotics and Automation (ICRA)
♻ ☆ RAPID: Redundancy-Aware and Compatibility-Optimal Edge-Cloud Partitioned Inference for Diverse VLA Models
Vision Language Action (VLA) models are mainstream in embodied intelligence but face high inference costs. Edge-Cloud Collaborative (ECC) inference offers an effective fix by easing edge-device computing pressure to meet real-time needs. However, existing ECC frameworks are suboptimal for VLA models due to two challenges: (1) Mainstream environment-oriented edge-cloud partitioning methods are susceptible to interference from visual noise; (2) Existing edge-cloud partitioning methods overlook the step-wise redundancy unique to embodied tasks, thereby disrupting the physical continuity of motion. To address these issues, we propose a novel ECC inference framework, termed RAPID. Specifically, we developed an implementation tailored to the proposed framework. Experiments demonstrate this achieves a speedup of up to 1.73x with only 5%~7% overhead.
♻ ☆ When Semantics Connect the Swarm: LLM-Driven Fuzzy Control for Cooperative Multi-Robot Underwater Coverage
Underwater multi-robot cooperative coverage remains challenging due to partial observability, limited communication, environmental uncertainty, and the lack of access to global localization. To address these issues, this paper presents a semantics-guided fuzzy control framework that couples Large Language Models (LLMs) with interpretable control and lightweight coordination. Raw multimodal observations are compressed by the LLM into compact, human-interpretable semantic tokens that summarize obstacles, unexplored regions, and Objects Of Interest (OOIs) under uncertain perception. A fuzzy inference system with pre-defined membership functions then maps these tokens into smooth and stable steering and gait commands, enabling reliable navigation without relying on global positioning. Then, we further coordinate multiple robots by introducing semantic communication that shares intent and local context in linguistic form, enabling agreement on who explores where while avoiding redundant revisits. Extensive simulations in unknown reef-like environments show that, under limited sensing and communication, the proposed framework achieves robust OOI-oriented navigation and cooperative coverage with improved efficiency and adaptability, narrowing the gap between semantic cognition and distributed underwater control in GPS-denied, map-free conditions.
comment: Withdrawal for further improvement. The final version will be released in a few months
♻ ☆ ReViP: Mitigating False Completion in Vision-Language-Action Models with Vision-Proprioception Rebalance
Vision-Language-Action (VLA) models have advanced robotic manipulation by combining vision, language, and proprioception to predict actions. However, previous methods fuse proprioceptive signals directly with vision-language features, resulting in state-dominant bias and \textbf{false completions} despite visible execution failures. We systematically analyze this failure mode, attributing it to modality imbalance, where policies overly rely on internal state progression and underuse visual evidence. To address this, we introduce the first \textbf{False-Completion Benchmark Suite}, featuring eight tasks with three controlled perturbations (\emph{Object Drop}, \emph{Distractor Swap}, \emph{Relayout}) to comprehensively evaluate false completion. Moreover, we propose \textbf{ReViP}, a novel VLA framework with \textbf{Vi}sion-\textbf{P}roprioception \textbf{Re}balance to enhance visual grounding and robustness under perturbations. The key insight is to introduce auxiliary \emph{progress-aware visual cues} to adaptively modulate the coupling between semantic perception and proprioceptive dynamics. Specifically, progress-aware visual cues are extracted by an external Task-Stage Observer, which performs task-relevant reasoning on real-time observations to drive task-stage feature-wise linear modulation, enhancing environmental awareness and mitigating state-driven errors. Extensive experiments show that ReViP effectively mitigates false completion and improves success rates over strong VLA baselines, achieving a \textbf{26\%} gain over $π_0$ model on our suite, with gains extending to LIBERO, RoboTwin 2.0, and real-world evaluations.
♻ ☆ DriveCritic: Towards Context-Aware, Human-Aligned Evaluation for Autonomous Driving with Vision-Language Models ICRA 2026
Benchmarking autonomous driving planners to align with human judgment remains a critical challenge, as state-of-the-art metrics like the Extended Predictive Driver Model Score (EPDMS) lack context awareness in nuanced scenarios. To address this, we introduce DriveCritic, a novel framework featuring two key contributions: the DriveCritic dataset, a curated collection of challenging scenarios where context is critical for correct judgment and annotated with pairwise human preferences, and the DriveCritic model, a Vision-Language Model (VLM) based evaluator. Fine-tuned using a two-stage supervised and reinforcement learning pipeline, the DriveCritic model learns to adjudicate between trajectory pairs by integrating visual and symbolic context. Experiments show DriveCritic significantly outperforms existing metrics and baselines in matching human preferences and demonstrates strong context awareness. Overall, our work provides a more reliable, human-aligned foundation to evaluating autonomous driving systems. The project page for DriveCritic is https://song-jingyu.github.io/DriveCritic
comment: Accepted at ICRA 2026; 8 pages, 3 figures
♻ ☆ Decision-Aware Uncertainty Evaluation of Vision-Language Model-Based Early Action Anticipation for Human-Robot Interaction
Robots in shared workspaces must interpret human actions from partial, ambiguous observations, where overconfident early predictions can lead to unsafe or disruptive interaction. This challenge is amplified in egocentric views, where viewpoint changes and occlusions increase perceptual noise and ambiguity. As a result, downstream human-robot interaction modules require not only an action hypothesis but also a trustworthy estimate of confidence under partial observation. Recent vision-language model-based approaches have been proposed for short-term action recognition due to their open-vocabulary and context-aware reasoning, but their uncertainty reliability in the temporal-prefix regime is largely uncharacterized. We present the first systematic evaluation of uncertainty in vision-language model-based short-term action recognition for human-robot interaction. We introduce a temporal-prefix evaluation protocol and metrics for calibration and selective prediction. We also characterize miscalibration patterns and failure modes under partial observations. Our study provides the missing reliability evidence needed to use vision-language model predictions in confidence-gated human-robot interaction modules.
♻ ☆ Evaluating Zero-Shot and One-Shot Adaptation of Small Language Models in Leader-Follower Interaction
Leader-follower interaction is an important paradigm in human-robot interaction (HRI). Yet, assigning roles in real time remains challenging for resource-constrained mobile and assistive robots. While large language models (LLMs) have shown promise for natural communication, their size and latency limit on-device deployment. Small language models (SLMs) offer a potential alternative, but their effectiveness for role classification in HRI has not been systematically evaluated. In this paper, we present a benchmark of SLMs for leader-follower communication, introducing a novel dataset derived from a published database and augmented with synthetic samples to capture interaction-specific dynamics. We investigate two adaptation strategies: prompt engineering and fine-tuning, studied under zero-shot and one-shot interaction modes, compared with an untrained baseline. Experiments with Qwen2.5-0.5B reveal that zero-shot fine-tuning achieves robust classification performance (86.66% accuracy) while maintaining low latency (22.2 ms per sample), significantly outperforming baseline and prompt-engineered approaches. However, results also indicate a performance degradation in one-shot modes, where increased context length challenges the model's architectural capacity. These findings demonstrate that fine-tuned SLMs provide an effective solution for direct role assignment, while highlighting critical trade-offs between dialogue complexity and classification reliability on the edge.
♻ ☆ Scalable Surface-Based Manipulation Through Modularity and Inter-Module Object Transfer
Robotic Manipulation Surfaces (RMS) manipulate objects by deforming the surface on which they rest, offering safe, parallel handling of diverse and fragile items. However, existing designs face a fundamental tradeoff: achieving fine control typically demands dense actuator arrays that limit scalability. Modular architectures can extend the workspace, but transferring objects reliably across module boundaries on soft, continuously deformable surfaces remains an open challenge. We present a multi-modular soft manipulation platform that achieves coordinated inter-module object transfer and precise positioning across interconnected fabric-based modules. A hierarchical control framework, combining conflict-free Manhattan-based path planning with directional object passing and a geometric PID controller, achieves sub-centimeter positioning and consistent transfer of heterogeneous objects including fragile items. The platform employs shared-boundary actuation, where adjacent modules share edge actuators, reducing the required count from $4n^2$ to $(n + 1)^2$ for an $n \times n$ grid; a $2\times 2$ prototype covers $1\times 1$ m with only 9 actuators. This scaling comes at a cost: shared actuators mechanically couple neighbouring modules, creating interference during simultaneous manipulation. We systematically characterise this coupling across spatial configurations and propose compensation strategies that reduce passive-object displacement by 59--78\%. Together, these contributions establish a scalable foundation for soft manipulation surfaces in applications such as food processing and logistics.
comment: 8 pages
♻ ☆ Lifelong Imitation Learning with Multimodal Latent Replay and Incremental Adjustment CVPR 2026
We introduce a lifelong imitation learning framework that enables continual policy refinement across sequential tasks under realistic memory and data constraints. Our approach departs from conventional experience replay by operating entirely in a multimodal latent space, where compact representations of visual, linguistic, and robot's state information are stored and reused to support future learning. To further stabilize adaptation, we introduce an incremental feature adjustment mechanism that regularizes the evolution of task embeddings through an angular margin constraint, preserving inter-task distinctiveness. Our method establishes a new state of the art in the LIBERO benchmarks, achieving 10-17 point gains in AUC and up to 65% less forgetting compared to previous leading methods. Ablation studies confirm the effectiveness of each component, showing consistent gains over alternative strategies. The code is available at: https://github.com/yfqi/lifelong_mlr_ifa.
comment: Accepted to CVPR 2026
♻ ☆ Robust Attitude Control of Nonlinear UAV Dynamics with LFT Models and $\mathcal{H}_\infty$ Performance
Attitude stabilization of unmanned aerial vehicles (UAVs) in uncertain environments presents significant challenges due to nonlinear dynamics, parameter variations, and sensor limitations. This paper presents a comparative study of $\mathcal{H}_\infty$ and classical PID controllers for multi-rotor attitude regulation in the presence of wind disturbances and gyroscope noise. The flight dynamics are modeled using a linear parameter-varying (LPV) framework, where nonlinearities and parameter variations are systematically represented as structured uncertainties within a linear fractional transformation formulation. A robust controller based on $\mathcal{H}_\infty$ formulation is designed using only gyroscope measurements to ensure guaranteed performance bounds. Nonlinear simulation results demonstrate the effectiveness of the robust controllers compared to classical PID control, showing significant improvement in attitude regulation under severe wind disturbances.
comment: 6 pages, 6 figures, 3 tables, submitted to ACC 2026
♻ ☆ Warped Hypertime Representations for Long-term Autonomy of Mobile Robots
This paper presents a novel method for introducing time into discrete and continuous spatial representations used in mobile robotics, by modelling long-term, pseudo-periodic variations caused by human activities. Unlike previous approaches, the proposed method does not treat time and space separately, and its continuous nature respects both the temporal and spatial continuity of the modeled phenomena. The method extends the given spatial model with a set of wrapped dimensions that represent the periodicities of observed changes. By performing clustering over this extended representation, we obtain a model that allows us to predict future states of both discrete and continuous spatial representations. We apply the proposed algorithm to several long-term datasets and show that the method enables a robot to predict future states of representations with different dimensions. The experiments further show that the method achieves more accurate predictions than the previous state of the art.
♻ ☆ WHED: A Wearable Hand Exoskeleton for Natural, High-Quality Demonstration Collection
Scalable learning of dexterous manipulation remains bottlenecked by the difficulty of collecting natural, high-fidelity human demonstrations of multi-finger hands due to occlusion, complex hand kinematics, and contact-rich interactions. We present WHED, a wearable hand-exoskeleton system designed for in-the-wild demonstration capture, guided by two principles: wearability-first operation for extended use and a pose-tolerant, free-to-move thumb coupling that preserves natural thumb behaviors while maintaining a consistent mapping to the target robot thumb degrees of freedom. WHED integrates a linkage-driven finger interface with passive fit accommodation, a modified passive hand with robust proprioceptive sensing, and an onboard sensing/power module. We also provide an end-to-end data pipeline that synchronizes joint encoders, AR-based end-effector pose, and wrist-mounted visual observations, and supports post-processing for time alignment and replay. We demonstrate feasibility on representative grasping and manipulation sequences spanning precision pinch and full-hand enclosure grasps, and show qualitative consistency between collected demonstrations and replayed executions.
comment: This manuscript is withdrawn because the work is being substantially revised for submission to a peer-reviewed venue. The current version may be incomplete or misleading
Robotics 86
☆ A gripper for flap separation and opening of sealed bags ICRA2026
Separating thin, flexible layers that must be individually grasped is a common but challenging manipulation primitive for most off-the-shelf grippers. A prominent example arises in clinical settings: the opening of sterile flat pouches for the preparation of the operating room, where the first step is to separate and grasp the flaps. We present a novel gripper design and opening strategy that enables reliable flap separation and robust seal opening. This capability addresses a high-volume repetitive hospital procedure in which nurses manually open up to 240 bags per shift, a physically demanding task linked to musculoskeletal injuries. Our design combines an active dented-roller fingertip with compliant fingers that exploit environmental constraints to robustly grasp thin flexible flaps. Experiments demonstrate that the proposed gripper reliably grasps and separates sealed bag flaps and other thin-layered materials from the hospital, the most sensitive variable affecting performance being the normal force applied. When two copies of the gripper grasp both flaps, the system withstands the forces needed to open the seals robustly. To our knowledge, this is one of the first demonstrations of robotic assistance to automate this repetitive, low-value, but critical hospital task.
comment: 8 pages, Accepted at the 2026 IEEE International Conference on Robotics & Automation (ICRA2026)
☆ RL-Augmented MPC for Non-Gaited Legged and Hybrid Locomotion
We propose a contact-explicit hierarchical architecture coupling Reinforcement Learning (RL) and Model Predictive Control (MPC), where a high-level RL agent provides gait and navigation commands to a low-level locomotion MPC. This offloads the combinatorial burden of contact timing from the MPC by learning acyclic gaits through trial and error in simulation. We show that only a minimal set of rewards and limited tuning are required to obtain effective policies. We validate the architecture in simulation across robotic platforms spanning 50 kg to 120 kg and different MPC implementations, observing the emergence of acyclic gaits and timing adaptations in flat-terrain legged and hybrid locomotion, and further demonstrating extensibility to non-flat terrains. Across all platforms, we achieve zero-shot sim-to-sim transfer without domain randomization, and we further demonstrate zero-shot sim-to-real transfer without domain randomization on Centauro, our 120 kg wheeled-legged humanoid robot. We make our software framework and evaluation results publicly available at https://github.com/AndrePatri/AugMPC.
☆ FG-CLTP: Fine-Grained Contrastive Language Tactile Pretraining for Robotic Manipulation
Recent advancements in integrating tactile sensing into vision-language-action (VLA) models have demonstrated transformative potential for robotic perception. However, existing tactile representations predominantly rely on qualitative descriptors (e.g., texture), neglecting quantitative contact states such as force magnitude, contact geometry, and principal axis orientation, which are indispensable for fine-grained manipulation. To bridge this gap, we propose FG-CLTP, a fine-grained contrastive language tactile pretraining framework. We first introduce a novel dataset comprising over 100k tactile 3D point cloud-language pairs that explicitly capture multidimensional contact states from the sensor's perspective. We then implement a discretized numerical tokenization mechanism to achieve quantitative-semantic alignment, effectively injecting explicit physical metrics into the multimodal feature space. The proposed FG-CLTP model yields a 95.9% classification accuracy and reduces the regression error (MAE) by 52.6% compared to state-of-the-art methods. Furthermore, the integration of 3D point cloud representations establishes a sensor-agnostic foundation with a minimal sim-to-real gap of 3.5%. Building upon this fine-grained representation, we develop a 3D tactile-language-action (3D-TLA) architecture driven by a flow matching policy to enable multimodal reasoning and control. Extensive experiments demonstrate that our framework significantly outperforms strong baselines in contact-rich manipulation tasks, providing a robust and generalizable foundation for tactile-language-action models.
comment: 9 pages, 6 figures
☆ GRACE: A Unified 2D Multi-Robot Path Planning Simulator & Benchmark for Grid, Roadmap, And Continuous Environments ICRA 2026
Advancing Multi-Agent Pathfinding (MAPF) and Multi-Robot Motion Planning (MRMP) requires platforms that enable transparent, reproducible comparisons across modeling choices. Existing tools either scale under simplifying assumptions (grids, homogeneous agents) or offer higher fidelity with less comparable instrumentation. We present GRACE, a unified 2D simulator+benchmark that instantiates the same task at multiple abstraction levels (grid, roadmap, continuous) via explicit, reproducible operators and a common evaluation protocol. Our empirical results on public maps and representative planners enable commensurate comparisons on a shared instance set. Furthermore, we quantify the expected representation-fidelity trade-offs (MRMP solves instances at higher fidelity but lower speed, while grid/roadmap planners scale farther). By consolidating representation, execution, and evaluation, GRACE thereby aims to make cross-representation studies more comparable and provides a means to advance multi-robot planning research and its translation to practice.
comment: ICRA 2026, code will be released soon
☆ Semantic Landmark Particle Filter for Robot Localisation in Vineyards IROS 2026
Reliable localisation in vineyards is hindered by row-level perceptual aliasing: parallel crop rows produce nearly identical LiDAR observations, causing geometry-only and vision-based SLAM systems to converge towards incorrect corridors, particularly during headland transitions. We present a Semantic Landmark Particle Filter (SLPF) that integrates trunk and pole landmark detections with 2D LiDAR within a probabilistic localisation framework. Detected trunks are converted into semantic walls, forming structural row boundaries embedded in the measurement model to improve discrimination between adjacent rows. GNSS is incorporated as a lightweight prior that stabilises localisation when semantic observations are sparse. Field experiments in a 10-row vineyard demonstrate consistent improvements over geometry-only (AMCL), vision-based (RTAB-Map), and GNSS baselines. Compared to AMCL, SLPF reduces Absolute Pose Error by 22% and 65% across two traversal directions; relative to a NoisyGNSS baseline, APE decreases by 65% and 61%. Row correctness improves from 0.67 to 0.73, while mean cross-track error decreases from 1.40 m to 1.26 m. These results show that embedding row-level structural semantics within the measurement model enables robust localisation in highly repetitive outdoor agricultural environments.
comment: Submmitted to IROS 2026
☆ Sublinear-Time Reconfiguration of Programmable Matter with Joint Movements
We study centralized reconfiguration problems for geometric amoebot structures. A set of $n$ amoebots occupy nodes on the triangular grid and can reconfigure via expansion and contraction operations. We focus on the joint movement extension, where amoebots may expand and contract in parallel, enabling coordinated motion of larger substructures. Prior work introduced this extension and analyzed reconfiguration under additional assumptions such as metamodules. In contrast, we investigate the intrinsic dynamics of reconfiguration without such assumptions by restricting attention to centralized algorithms, leaving distributed solutions for future work. We study the reconfiguration problem between two classes of amoebot structures $A$ and $B$: For every structure $S\in A$, the goal is to compute a schedule that reconfigures $S$ into some structure $S'\in B$. Our focus is on sublinear-time algorithms. We affirmatively answer the open problem by Padalkin et al. (Auton. Robots, 2025) whether a within-the-model sublinear-time universal reconfiguration algorithm is possible, by proving that any structure can be reconfigured into a canonical line-segment structure in $O(\sqrt{n}\log n)$ rounds. Additionally, we give a constant-time algorithm for reconfiguring any spiral structure into a line segment. These results are enabled by new constant-time primitives that facilitate efficient parallel movement. Our findings demonstrate that the joint movement model supports sublinear reconfiguration without auxiliary assumptions. A central open question is whether universal reconfiguration within this model can be achieved in polylogarithmic or even constant time.
☆ ASTER: Attitude-aware Suspended-payload Quadrotor Traversal via Efficient Reinforcement Learning
Agile maneuvering of the quadrotor cable-suspended system is significantly hindered by its non-smooth hybrid dynamics. While model-free Reinforcement Learning (RL) circumvents explicit differentiation of complex models, achieving attitude-constrained or inverted flight remains an open challenge due to the extreme reward sparsity under strict orientation requirements. This paper presents ASTER, a robust RL framework that achieves, to our knowledge, the first successful autonomous inverted flight for the cable-suspended system. We propose hybrid-dynamics-informed state seeding (HDSS), an initialization strategy that back-propagates target configurations through physics-consistent kinematic inversions across both taut and slack cable phases. HDSS enables the policy to discover aggressive maneuvers that are unreachable via standard exploration. Extensive simulations and real-world experiments demonstrate remarkable agility, precise attitude alignment, and robust zero-shot sim-to-real transfer across complex trajectories.
☆ MAVEN: A Meta-Reinforcement Learning Framework for Varying-Dynamics Expertise in Agile Quadrotor Maneuvers
Reinforcement learning (RL) has emerged as a powerful paradigm for achieving online agile navigation with quadrotors. Despite this success, policies trained via standard RL typically fail to generalize across significant dynamic variations, exhibiting a critical lack of adaptability. This work introduces MAVEN, a meta-RL framework that enables a single policy to achieve robust end-to-end navigation across a wide range of quadrotor dynamics. Our approach features a novel predictive context encoder, which learns to infer a latent representation of the system dynamics from interaction history. We demonstrate our method in agile waypoint traversal tasks under two challenging scenarios: large variations in quadrotor mass and severe single-rotor thrust loss. We leverage a GPU-vectorized simulator to distribute tasks across thousands of parallel environments, overcoming the long training times of meta-RL to converge in less than an hour. Through extensive experiments in both simulation and the real world, we validate that MAVEN achieves superior adaptation and agility. The policy successfully executes zero-shot sim-to-real transfer, demonstrating robust online adaptation by performing high-speed maneuvers despite mass variations of up to 66.7% and single-rotor thrust losses as severe as 70%.
☆ FutureVLA: Joint Visuomotor Prediction for Vision-Language-Action Model
Predictive foresight is important to intelligent embodied agents. Since the motor execution of a robot is intrinsically constrained by its visual perception of environmental geometry, effectively anticipating the future requires capturing this tightly coupled visuomotor interplay. While recent vision-language-action models attempt to incorporate future guidance, they struggle with this joint modeling. Existing explicit methods divert capacity to task-irrelevant visual details, whereas implicit methods relying on sparse frame pairs disrupt temporal continuity. By heavily relying on visual reconstruction, these methods become visually dominated, entangling static scene context with dynamic action intent. We argue that effective joint visuomotor predictive modeling requires both temporal continuity and visually-conditioned supervision decoupling. To this end, we propose FutureVLA, featuring a novel Joint Visuomotor Predictive Architecture. FutureVLA is designed to extract joint visuomotor embeddings by first decoupling visual and motor information, and then jointly encoding generalized physical priors. Specifically, in the pretraining stage, we leverage heterogeneous manipulation datasets and introduce a Joint Visuomotor Gating mechanism to structurally separate visual state preservation from temporal action modeling. It allows the motor stream to focus on continuous physical dynamics while explicitly querying visual tokens for environmental constraints, yielding highly generalizable joint visuomotor embeddings. Subsequently, in the post-training stage, we employ a latent embeddings alignment strategy, enabling diverse downstream VLA models to internalize these temporal priors without modifying their inference architectures. Extensive experiments demonstrate that FutureVLA consistently improves VLA frameworks.
☆ Parallel-in-Time Nonlinear Optimal Control via GPU-native Sequential Convex Programming
Real-time trajectory optimization for nonlinear constrained autonomous systems is critical and typically performed by CPU-based sequential solvers. Specifically, reliance on global sparse linear algebra or the serial nature of dynamic programming algorithms restricts the utilization of massively parallel computing architectures like GPUs. To bridge this gap, we introduce a fully GPU-native trajectory optimization framework that combines sequential convex programming with a consensus-based alternating direction method of multipliers. By applying a temporal splitting strategy, our algorithm decouples the optimization horizon into independent, per-node subproblems that execute massively in parallel. The entire process runs fully on the GPU, eliminating costly memory transfers and large-scale sparse factorizations. This architecture naturally scales to multi-trajectory optimization. We validate the solver on a quadrotor agile flight task and a Mars powered descent problem using an on-board edge computing platform. Benchmarks reveal a sustained 4x throughput speedup and a 51% reduction in energy consumption over a heavily optimized 12-core CPU baseline. Crucially, the framework saturates the hardware, maintaining over 96% active GPU utilization to achieve planning rates exceeding 100 Hz. Furthermore, we demonstrate the solver's extensibility to robust Model Predictive Control by jointly optimizing dynamically coupled scenarios under stochastic disturbances, enabling scalable and safe autonomy.
☆ MapGCLR: Geospatial Contrastive Learning of Representations for Online Vectorized HD Map Construction
Autonomous vehicles rely on map information to understand the world around them. However, the creation and maintenance of offline high-definition (HD) maps remains costly. A more scalable alternative lies in online HD map construction, which only requires map annotations at training time. To further reduce the need for annotating vast training labels, self-supervised training provides an alternative. This work focuses on improving the latent birds-eye-view (BEV) feature grid representation within a vectorized online HD map construction model by enforcing geospatial consistency between overlapping BEV feature grids as part of a contrastive loss function. To ensure geospatial overlap for contrastive pairs, we introduce an approach to analyze the overlap between traversals within a given dataset and generate subsidiary dataset splits following adjustable multi-traversal requirements. We train the same model supervised using a reduced set of single-traversal labeled data and self-supervised on a broader unlabeled set of data following our multi-traversal requirements, effectively implementing a semi-supervised approach. Our approach outperforms the supervised baseline across the board, both quantitatively in terms of the downstream tasks vectorized map perception performance and qualitatively in terms of segmentation in the principal component analysis (PCA) visualization of the BEV feature space.
☆ OnFly: Onboard Zero-Shot Aerial Vision-Language Navigation toward Safety and Efficiency
Aerial vision-language navigation (AVLN) enables UAVs to follow natural-language instructions in complex 3D environments. However, existing zero-shot AVLN methods often suffer from unstable single-stream Vision-Language Model decision-making, unreliable long-horizon progress monitoring, and a trade-off between safety and efficiency. We propose OnFly, a fully onboard, real-time framework for zero-shot AVLN. OnFly adopts a shared-perception dual-agent architecture that decouples high-frequency target generation from low-frequency progress monitoring, thereby stabilizing decision-making. It further employs a hybrid keyframe-recent-frame memory to preserve global trajectory context while maintaining KV-cache prefix stability, enabling reliable long-horizon monitoring with termination and recovery signals. In addition, a semantic-geometric verifier refines VLM-predicted targets for instruction consistency and geometric safety using VLM features and depth cues, while a receding-horizon planner generates optimized collision-free trajectories under geometric safety constraints, improving both safety and efficiency. In simulation, OnFly improves task success from 26.4% to 67.8%, compared with the strongest state-of-the-art baseline, while fully onboard real-world flights validate its feasibility for real-time deployment. The code will be released at https://github.com/Robotics-STAR-Lab/OnFly
☆ Cybo-Waiter: A Physical Agentic Framework for Humanoid Whole-Body Locomotion-Manipulation
Robots are increasingly expected to execute open ended natural language requests in human environments, which demands reliable long horizon execution under partial observability. This is especially challenging for humanoids because locomotion and manipulation are tightly coupled through stance, reachability, and balance. We present a humanoid agent framework that turns VLM plans into verifiable task programs and closes the loop with multi object 3D geometric supervision. A VLM planner compiles each instruction into a typed JSON sequence of subtasks with explicit predicate based preconditions and success conditions. Using SAM3 and RGB-D, we ground all task relevant entities in 3D, estimate object centroids and extents, and evaluate predicates over stable frames to obtain condition level diagnostics. The supervisor uses these diagnostics to verify subtask completion and to provide condition-level feedback for progression and replanning. We execute each subtask by coordinating humanoid locomotion and whole-body manipulation, selecting feasible motion primitives under reachability and balance constraints. Experiments on tabletop manipulation and long horizon humanoid loco manipulation tasks show improved robustness from multi object grounding, temporal stability, and recovery driven replanning.
☆ Dynamic Modeling and Attitude Control of a Reaction-Wheel-Based Low-Gravity Bipedal Hopper
Planetary bodies characterized by low gravitational acceleration, such as the Moon and near-Earth asteroids, impose unique locomotion constraints due to diminished contact forces and extended airborne intervals. Among traversal strategies, hopping locomotion offers high energy efficiency but is prone to mid-flight attitude instability caused by asymmetric thrust generation and uneven terrain interactions. This paper presents an underactuated bipedal hopping robot that employs an internal reaction wheel to regulate body posture during the ballistic flight phase. The system is modeled as a gyrostat, enabling analysis of the dynamic coupling between torso rotation and reaction wheel momentum. The locomotion cycle comprises three phases: a leg-driven propulsive jump, mid-air attitude stabilization via an active momentum exchange controller, and a shock-absorbing landing. A reduced-order model is developed to capture the critical coupling between torso rotation and reaction wheel dynamics. The proposed framework is evaluated in MuJoCo-based simulations under lunar gravity conditions (g = 1.625 m/s^2). Results demonstrate that activation of the reaction wheel controller reduces peak mid-air angular deviation by more than 65% and constrains landing attitude error to within 3.5 degrees at touchdown. Additionally, actuator saturation per hop cycle is reduced, ensuring sufficient control authority. Overall, the approach significantly mitigates in-flight attitude excursions and enables consistent upright landings, providing a practical and control-efficient solution for locomotion on irregular extraterrestrial terrains.
comment: Preprint. Under review
☆ STM32-Based Smart Waste Bin for Hygienic Disposal Using Embedded Sensing and Automated Control
The increasing demand for hygienic and contactless solutions in public and private environments has encouraged the development of automated systems for everyday applications. This paper presents the design and implementation of a motion- sensing automatic waste bin using an STM32 microcontroller, ultrasonic sensors, and a servo motor. The system detects user presence through ultrasonic sensing and automatically opens the bin lid using a servo motor controlled by the microcontroller. An additional ultrasonic sensor is used to monitor the internal waste level of the bin, while an OLED display provides real- time feedback regarding system status. The proposed system offers a low-cost, reliable, and easily deployable solution for touch-free waste disposal. Experimental evaluation demonstrates fast response time, stable sensing performance, and smooth mechanical operation. The system can be effectively deployed in homes, educational institutions, hospitals, and public facilities to improve hygiene and user convenience.
comment: This paper consists of 6 pages, with 3 figures, 3 tables, and 1 algorithm
☆ Interleaving Scheduling and Motion Planning with Incremental Learning of Symbolic Space-Time Motion Abstractions
Task and Motion Planning combines high-level task sequencing (what to do) with low-level motion planning (how to do it) to generate feasible, collision-free execution plans. However, in many real-world domains, such as automated warehouses, tasks are predefined, shifting the challenge to if, when, and how to execute them safely and efficiently under resource, time and motion constraints. In this paper, we formalize this as the Scheduling and Motion Planning problem for multi-object navigation in shared workspaces. We propose a novel solution framework that interleaves off-the-shelf schedulers and motion planners in an incremental learning loop. The scheduler generates candidate plans, while the motion planner checks feasibility and returns symbolic feedback, i.e., spatial conflicts and timing adjustments, to guide the scheduler towards motion-feasible solutions. We validate our proposal on logistics and job-shop scheduling benchmarks augmented with motion tasks, using state-of-the-art schedulers and sampling-based motion planners. Our results show the effectiveness of our framework in generating valid plans under complex temporal and spatial constraints, where synchronized motion is critical.
☆ AdaClearGrasp: Learning Adaptive Clearing for Zero-Shot Robust Dexterous Grasping in Densely Cluttered Environments
In densely cluttered environments, physical interference, visual occlusions, and unstable contacts often cause direct dexterous grasping to fail, while aggressive singulation strategies may compromise safety. Enabling robots to adaptively decide whether to clear surrounding objects or directly grasp the target is therefore crucial for robust manipulation. We propose AdaClearGrasp, a closed-loop decision-execution framework for adaptive clearing and zero-shot dexterous grasping in densely cluttered environments. The framework formulates manipulation as a controllable high-level decision process that determines whether to directly grasp the target or first clear surrounding objects. A pretrained vision-language model (VLM) interprets visual observations and language task descriptions to reason about grasp interference and generate a high-level planning skeleton, which invokes structured atomic skills through a unified action interface. For dexterous grasping, we train a reinforcement learning policy with a relative hand-object distance representation, enabling zero-shot generalization across diverse object geometries and physical properties. During execution, visual feedback monitors outcomes and triggers replanning upon failures, forming a closed-loop correction mechanism. To evaluate language-conditioned dexterous grasping in clutter, we introduce Clutter-Bench, the first simulation benchmark with graded clutter complexity. It includes seven target objects across three clutter levels, yielding 210 task scenarios. We further perform sim-to-real experiments on three objects under three clutter levels (18 scenarios). Results demonstrate that AdaClearGrasp significantly improves grasp success rates in densely cluttered environments. For more videos and code, please visit our project website: https://chenzixuan99.github.io/adaclear-grasp.github.io/.
comment: 12 pages. Under review
☆ Learning Bimanual Cloth Manipulation with Vision-based Tactile Sensing via Single Robotic Arm
Robotic cloth manipulation remains challenging due to the high-dimensional state space of fabrics, their deformable nature, and frequent occlusions that limit vision-based sensing. Although dual-arm systems can mitigate some of these issues, they increase hardware and control complexity. This paper presents Touch G.O.G., a compact vision-based tactile gripper and perception/control framework for single-arm bimanual cloth manipulation. The proposed framework combines three key components: (1) a novel gripper design and control strategy for in-gripper cloth sliding with a single robot arm, (2) a Vision Foundation Model-backboned Vision Transformer pipeline for cloth part classification (PC-Net) and edge pose estimation (PE-Net) using real and synthetic tactile images, and (3) an encoder-decoder synthetic data generator (SD-Net) that reduces manual annotation by producing high-fidelity tactile images. Experiments show 96% accuracy in distinguishing edges, corners, interior regions, and grasp failures, together with sub-millimeter edge localization and 4.5° orientation error. Real-world results demonstrate reliable cloth unfolding, even for crumpled fabrics, using only a single robotic arm. These results highlight Touch G.O.G. as a compact and cost-effective solution for deformable object manipulation.
comment: 11 pages, 13 figures
☆ Recover to Predict: Progressive Retrospective Learning for Variable-Length Trajectory Prediction CVPR 2026
Trajectory prediction is critical for autonomous driving, enabling safe and efficient planning in dense, dynamic traffic. Most existing methods optimize prediction accuracy under fixed-length observations. However, real-world driving often yields variable-length, incomplete observations, posing a challenge to these methods. A common strategy is to directly map features from incomplete observations to those from complete ones. This one-shot mapping, however, struggles to learn accurate representations for short trajectories due to significant information gaps. To address this issue, we propose a Progressive Retrospective Framework (PRF), which gradually aligns features from incomplete observations with those from complete ones via a cascade of retrospective units. Each unit consists of a Retrospective Distillation Module (RDM) and a Retrospective Prediction Module (RPM), where RDM distills features and RPM recovers previous timesteps using the distilled features. Moreover, we propose a Rolling-Start Training Strategy (RSTS) that enhances data efficiency during PRF training. PRF is plug-and-play with existing methods. Extensive experiments on datasets Argoverse 2 and Argoverse 1 demonstrate the effectiveness of PRF. Code is available at https://github.com/zhouhao94/PRF.
comment: Paper is accepted by CVPR 2026
☆ Need for Speed: Zero-Shot Depth Completion with Single-Step Diffusion
We introduce Marigold-SSD, a single-step, late-fusion depth completion framework that leverages strong diffusion priors while eliminating the costly test-time optimization typically associated with diffusion-based methods. By shifting computational burden from inference to finetuning, our approach enables efficient and robust 3D perception under real-world latency constraints. Marigold-SSD achieves significantly faster inference with a training cost of only 4.5 GPU days. We evaluate our method across four indoor and two outdoor benchmarks, demonstrating strong cross-domain generalization and zero-shot performance compared to existing depth completion approaches. Our approach significantly narrows the efficiency gap between diffusion-based and discriminative models. Finally, we challenge common evaluation protocols by analyzing performance under varying input sparsity levels. Page: https://dtu-pas.github.io/marigold-ssd/
☆ Safety-critical Control Under Partial Observability: Reach-Avoid POMDP meets Belief Space Control
Partially Observable Markov Decision Processes (POMDPs) provide a principled framework for robot decision-making under uncertainty. Solving reach-avoid POMDPs, however, requires coordinating three distinct behaviors: goal reaching, safety, and active information gathering to reduce uncertainty. Existing online POMDP solvers attempt to address all three within a single belief tree search, but this unified approach struggles with the conflicting time scales inherent to these objectives. We propose a layered, certificate-based control architecture that operates directly in belief space, decoupling goal reaching, information gathering, and safety into modular components. We introduce Belief Control Lyapunov Functions (BCLFs) that formalize information gathering as a Lyapunov convergence problem in belief space, and show how they can be learned via reinforcement learning. For safety, we develop Belief Control Barrier Functions (BCBFs) that leverage conformal prediction to provide probabilistic safety guarantees over finite horizons. The resulting control synthesis reduces to lightweight quadratic programs solvable in real time, even for non-Gaussian belief representations with dimension $>10^4$. Experiments in simulation and on a space-robotics platform demonstrate real-time performance and improved safety and task success compared to state-of-the-art constrained POMDP solvers.
☆ TacLoc: Global Tactile Localization on Objects from a Registration Perspective
Pose estimation is essential for robotic manipulation, particularly when visual perception is occluded during gripper-object interactions. Existing tactile-based methods generally rely on tactile simulation or pre-trained models, which limits their generalizability and efficiency. In this study, we propose TacLoc, a novel tactile localization framework that formulates the problem as a one-shot point cloud registration task. TacLoc introduces a graph-theoretic partial-to-full registration method, leveraging dense point clouds and surface normals from tactile sensing for efficient and accurate pose estimation. Without requiring rendered data or pre-trained models, TacLoc achieves improved performance through normal-guided graph pruning and a hypothesis-and-verification pipeline. TacLoc is evaluated extensively on the YCB dataset. We further demonstrate TacLoc on real-world objects across two different visual-tactile sensors.
comment: 8 pages, 12 figures
☆ BinWalker: Development and Field Evaluation of a Quadruped Manipulator Platform for Sustainable Litter Collection
Litter pollution represents a growing environmental problem affecting natural and urban ecosystems worldwide. Waste discarded in public spaces often accumulates in areas that are difficult to access, such as uneven terrains, coastal environments, parks, and roadside vegetation. Over time, these materials degrade and release harmful substances, including toxic chemicals and microplastics, which can contaminate soil and water and pose serious threats to wildlife and human health. Despite increasing awareness of the problem, litter collection is still largely performed manually by human operators, making large-scale cleanup operations labor-intensive, time-consuming, and costly. Robotic solutions have the potential to support and partially automate environmental cleanup tasks. In this work, we present a quadruped robotic system designed for autonomous litter collection in challenging outdoor scenarios. The robot combines the mobility advantages of legged locomotion with a manipulation system consisting of a robotic arm and an onboard litter container. This configuration enables the robot to detect, grasp, and store litter items while navigating through uneven terrains. The proposed system aims to demonstrate the feasibility of integrating perception, locomotion, and manipulation on a legged robotic platform for environmental cleanup tasks. Experimental evaluations conducted in outdoor scenarios highlight the effectiveness of the approach and its potential for assisting large-scale litter removal operations in environments that are difficult to reach with traditional robotic platforms. The code associated with this work can be found at: https://github.com/iit-DLSLab/trash-collection-isaaclab.
☆ Muscle Synergy Priors Enhance Biomechanical Fidelity in Predictive Musculoskeletal Locomotion Simulation
Human locomotion emerges from high-dimensional neuromuscular control, making predictive musculoskeletal simulation challenging. We present a physiology-informed reinforcement-learning framework that constrains control using muscle synergies. We extracted a low-dimensional synergy basis from inverse musculoskeletal analyses of a small set of overground walking trials and used it as the action space for a muscle-driven three-dimensional model trained across variable speeds, slopes and uneven terrain. The resulting controller generated stable gait from 0.7-1.8 m/s and on $\pm$ 6$^{\circ}$ grades and reproduced condition-dependent modulation of joint angles, joint moments and ground reaction forces. Compared with an unconstrained controller, synergy-constrained control reduced non-physiological knee kinematics and kept knee moment profiles within the experimental envelope. Across conditions, simulated vertical ground reaction forces correlated strongly with human measurements, and muscle-activation timing largely fell within inter-subject variability. These results show that embedding neurophysiological structure into reinforcement learning can improve biomechanical fidelity and generalization in predictive human locomotion simulation with limited experimental data.
comment: 12 pages, 5 figures
☆ DepthCache: Depth-Guided Training-Free Visual Token Merging for Vision-Language-Action Model Inference
Vision-Language-Action (VLA) models enable generalist robotic manipulation but suffer from high inference latency. This bottleneck stems from the massive number of visual tokens processed by large language backbones. Existing methods either prune or merge tokens uniformly, degrading the spatial reasoning essential for robotic control. We present DepthCache, a training-free framework that leverages depth as a structural prior for visual token compression. It partitions observations into depth-based regions and applies spatially differentiated merge ratios, preserving the near-field workspace while compressing the distant background. To exploit temporal redundancy, DepthCache distributes the merging process across consecutive frames, ensuring consistent representations while reducing per-step computation. A motion-adaptive pipeline further optimizes auxiliary view compression based on end-effector dynamics. The framework requires no model modification, generalizing across diverse VLA architectures. On the LIBERO benchmark, DepthCache achieves up to 1.28x inference speedup with less than 1% average success rate degradation across three VLA models (pi_0.5, OpenVLA, GR00T), whereas pruning and merging baselines incur 4--24% degradation at comparable compression. Real-world experiments on a physical manipulator demonstrate that DepthCache enables faster task throughput and more responsive closed-loop control in latency-sensitive scenarios.
comment: 8 pages, 6 figures
☆ SUBTA: A Framework for Supported User-Guided Bimanual Teleoperation in Structured Assembly ICRA 2026
In human-robot collaboration, shared autonomy enhances human performance through precise, intuitive support. Effective robotic assistance requires accurately inferring human intentions and understanding task structures to determine optimal support timing and methods. In this paper, we present SUBTA, a supported teleoperation system for bimanual assembly that couples learned intention estimation, scene-graph task planning, and context-dependent motion assists. We validate our approach through a user study (N=12) comparing standard teleoperation, motion-support only, and SUBTA. Linear mixed-effects analysis revealed that SUBTA significantly outperformed standard teleoperation in position accuracy (p<0.001, d=1.18) and orientation accuracy (p<0.001, d=1.75), while reducing mental demand (p=0.002, d=1.34). Post-experiment ratings indicate clearer, more trustworthy visual feedback and predictable interventions in SUBTA. The results demonstrate that SUBTA greatly improves both effectiveness and user experience in teleoperation.
comment: 8 pages, 7 figures, accepted at ICRA 2026
☆ FAR-Dex: Few-shot Data Augmentation and Adaptive Residual Policy Refinement for Dexterous Manipulation ICRA
Achieving human-like dexterous manipulation through the collaboration of multi-fingered hands with robotic arms remains a longstanding challenge in robotics, primarily due to the scarcity of high-quality demonstrations and the complexity of high-dimensional action spaces. To address these challenges, we propose FAR-Dex, a hierarchical framework that integrates few-shot data augmentation with adaptive residual refinement to enable robust and precise arm-hand coordination in dexterous tasks. First, FAR-DexGen leverages the IsaacLab simulator to generate diverse and physically constrained trajectories from a few demonstrations, providing a data foundation for policy training. Second, FAR-DexRes introduces an adaptive residual module that refines policies by combining multi-step trajectory segments with observation features, thereby enhancing accuracy and robustness in manipulation scenarios. Experiments in both simulation and real-world demonstrate that FAR-Dex improves data quality by 13.4% and task success rates by 7% over state-of-the-art methods. It further achieves over 80% success in real-world tasks, enabling fine-grained dexterous manipulation with strong positional generalization.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
☆ DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control
Vision-Language-Action (VLA) models have emerged as a promising paradigm for robot learning, but their representations are still largely inherited from static image-text pretraining, leaving physical dynamics to be learned from comparatively limited action data. Generative video models, by contrast, encode rich spatiotemporal structure and implicit physics, making them a compelling foundation for robotic manipulation. But their potentials are not fully explored in the literature. To bridge the gap, we introduce DiT4DiT, an end-to-end Video-Action Model that couples a video Diffusion Transformer with an action Diffusion Transformer in a unified cascaded framework. Instead of relying on reconstructed future frames, DiT4DiT extracts intermediate denoising features from the video generation process and uses them as temporally grounded conditions for action prediction. We further propose a dual flow-matching objective with decoupled timesteps and noise scales for video prediction, hidden-state extraction, and action inference, enabling coherent joint training of both modules. Across simulation and real-world benchmarks, DiT4DiT achieves state-of-the-art results, reaching average success rates of 98.6% on LIBERO and 50.8% on RoboCasa GR1 while using substantially less training data. On the Unitree G1 robot, it also delivers superior real-world performance and strong zero-shot generalization. Importantly, DiT4DiT improves sample efficiency by over 10x and speeds up convergence by up to 7x, demonstrating that video generation can serve as an effective scaling proxy for robot policy learning. We release code and models at https://dit4dit.github.io/.
comment: https://dit4dit.github.io/
☆ KnowDiffuser: A Knowledge-Guided Diffusion Planner with LM Reasoning and Prior-Informed Trajectory Initialization
Recent advancements in Language Models (LMs) have demonstrated strong semantic reasoning capabilities, enabling their application in high-level decision-making for autonomous driving (AD). However, LMs operate over discrete token spaces and lack the ability to generate continuous, physically feasible trajectories required for motion planning. Meanwhile, diffusion models have proven effective at generating reliable and dynamically consistent trajectories, but often lack semantic interpretability and alignment with scene-level understanding. To address these limitations, we propose \textbf{KnowDiffuser}, a knowledge-guided motion planning framework that tightly integrates the semantic understanding of language models with the generative power of diffusion models. The framework employs a language model to infer context-aware meta-actions from structured scene representations, which are then mapped to prior trajectories that anchor the subsequent denoising process. A two-stage truncated denoising mechanism refines these trajectories efficiently, preserving both semantic alignment and physical feasibility. Experiments on the nuPlan benchmark demonstrate that KnowDiffuser significantly outperforms existing planners in both open-loop and closed-loop evaluations, establishing a robust and interpretable framework that effectively bridges the semantic-to-physical gap in AD systems.
comment: 10pages, 1 figure
☆ AsyncMDE: Real-Time Monocular Depth Estimation via Asynchronous Spatial Memory
Foundation-model-based monocular depth estimation offers a viable alternative to active sensors for robot perception, yet its computational cost often prohibits deployment on edge platforms. Existing methods perform independent per-frame inference, wasting the substantial computational redundancy between adjacent viewpoints in continuous robot operation. This paper presents AsyncMDE, an asynchronous depth perception system consisting of a foundation model and a lightweight model that amortizes the foundation model's computational cost over time. The foundation model produces high-quality spatial features in the background, while the lightweight model runs asynchronously in the foreground, fusing cached memory with current observations through complementary fusion, outputting depth estimates, and autoregressively updating the memory. This enables cross-frame feature reuse with bounded accuracy degradation. At a mere 3.83M parameters, it operates at 237 FPS on an RTX 4090, recovering 77% of the accuracy gap to the foundation model while achieving a 25X parameter reduction. Validated across indoor static, dynamic, and synthetic extreme-motion benchmarks, AsyncMDE degrades gracefully between refreshes and achieves 161FPS on a Jetson AGX Orin with TensorRT, clearly demonstrating its feasibility for real-time edge deployment.
comment: 8 pages, 5 figures, 5 tables
☆ COHORT: Hybrid RL for Collaborative Large DNN Inference on Multi-Robot Systems Under Real-Time Constraints
Large deep neural networks (DNNs), especially transformer-based and multimodal architectures, are computationally demanding and challenging to deploy on resource-constrained edge platforms like field robots. These challenges intensify in mission-critical scenarios (e.g., disaster response), where robots must collaborate under tight constraints on bandwidth, latency, and battery life, often without infrastructure or server support. To address these limitations, we present COHORT, a collaborative DNN inference and task-execution framework for multi-robot systems built on the Robotic Operating System (ROS). COHORT employs a hybrid offline-online reinforcement learning (RL) strategy to dynamically schedule and distribute DNN module execution across robots. Our key contributions are threefold: (a) Offline RL policy learning combined with Advantage-Weighted Regression (AWR), trained on auction-based task allocation data from heterogeneous DNN workloads across distributed robots, (b) Online policy adaptation via Multi-Agent PPO (MAPPO), initialized from the offline policy and fine-tuned in real time, and (c) comprehensive evaluation of COHORT on vision-language model (VLM) inference tasks such as CLIP and SAM, analyzing scalability with increasing robot/workload and robustness under . We benchmark COHORT against genetic algorithms and multiple RL baselines. Experimental results demonstrate that COHORT reduces battery consumption by 15.4% and increases GPU utilization by 51.67%, while satisfying frame-rate and deadline constraints 2.55 times of the time.
comment: Recently accepted at 27th IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks ( IEEE WoWMoM 2026)
☆ Rethinking Gaussian Trajectory Predictors: Calibrated Uncertainty for Safe Planning
Accurate trajectory prediction is critical for safe autonomous navigation in crowded environments. While many trajectory predictors output Gaussian distributions to represent the multi-modal distribution over future pedestrian positions, the reliability of their confidence levels often remains unaddressed. This limitation can lead to unsafe or overly conservative motion planning when the predictor is integrated with an uncertainty-aware planner. Existing Gaussian trajectory predictors primarily rely on the Negative Log-Likelihood loss, which is prone to predict over- or under-confident distributions, and may compromise downstream planner safety. This paper introduces a novel loss function for calibrating prediction uncertainty which leverages Kernel Density Estimation to estimate the empirical distribution of confidence levels. The proposed formulation enforces consistency with the properties of a Gaussian assumption by explicitly matching the estimated empirical distribution to the Chi-squared distribution. To ensure accurate mean prediction, a Mean Squared Error term is also incorporated in the final loss formulation. Experimental results on real-world trajectory datasets show that our method significantly improves the reliability of confidence levels predicted by different State-Of-The-Art Gaussian trajectory predictors. We also demonstrate the importance of providing planners with reliable probabilistic insights (i.e. calibrated confidence levels) for collision-free navigation in complex scenarios. For this purpose, we integrate Gaussian trajectory predictors trained with our loss function with an uncertainty-aware Model Predictive Control on scenarios extracted from real-world datasets, achieving improved planning performance through calibrated confidence levels.
☆ Shape Control of a Planar Hyper-Redundant Robot via Hybrid Kinematics-Informed and Learning-based Approach
Hyper-redundant robots offer high dexterity, making them good at operating in confined and unstructured environments. To extend the reachable workspace, we built a multi-segment flexible rack actuated planar robot. However, the compliance of the flexible mechanism introduces instability, rendering it sensitive to external and internal uncertainties. To address these limitations, we propose a hybrid kinematics-informed and learning-based shape control method, named SpatioCoupledNet. The neural network adopts a hierarchical design that explicitly captures bidirectional spatial coupling between segments while modeling local disturbance along the robot body. A confidence-gating mechanism integrates prior kinematic knowledge, allowing the controller to adaptively balance model-based and learned components for improved convergence and fidelity. The framework is validated on a five-segment planar hyper-redundant robot under three representative shape configurations. Experimental results demonstrate that the proposed method consistently outperforms both analytical and purely neural controllers. In complex scenarios, it reduces steady-state error by up to 75.5% against the analytical model, and accelerates convergence by up to 20.5% compared to the data-driven baseline. Furthermore, gating analysis reveals a state-dependent authority fusion, shifting toward data-driven predictions in unstable states, while relying on physical priors in the remaining cases. Finally, we demonstrate robust performance in a dynamic task where the robot maintains a fixed end-effector position while avoiding moving obstacles, achieving a precise tip-positioning accuracy with a mean error of 10.47 mm.
☆ Safe Probabilistic Planning for Human-Robot Interaction using Conformal Risk Control
In this paper, we present a novel probabilistic safe control framework for human-robot interaction that combines control barrier functions (CBFs) with conformal risk control to provide formal safety guarantees while considering complex human behavior. The approach uses conformal risk control to quantify and control the prediction errors in CBF safety values and establishes formal guarantees on the probability of constraint satisfaction during interaction. We introduce an algorithm that dynamically adjusts the safety margins produced by conformal risk control based on the current interaction context. Through experiments on human-robot navigation scenarios, we demonstrate that our approach significantly reduces collision rates and safety violations as compared to baseline methods while maintaining high success rates in goal-reaching tasks and efficient control. The code, simulations, and other supplementary material can be found on the project website: https://jakeagonzales.github.io/crc-cbf-website/.
☆ ScanDP: Generalizable 3D Scanning with Diffusion Policy
Learning-based 3D Scanning plays a crucial role in enabling efficient and accurate scanning of target objects. However, recent reinforcement learning-based methods often require large-scale training data and still struggle to generalize to unseen object categories.In this work, we propose a data-efficient 3D scanning framework that uses Diffusion Policy to imitate human-like scanning strategies. To enhance robustness and generalization, we adopt the Occupancy Grid Mapping instead of direct point cloud processing, offering improved noise resilience and handling of diverse object geometries. We also introduce a hybrid approach combining a sphere-based space representation with a path optimization procedure that ensures path safety and scanning efficiency. This approach addresses limitations in conventional imitation learning, such as redundant or unpredictable behavior. We evaluate our method on diverse unseen objects in both shape and scale. Ours achieves higher coverage and shorter paths than baselines, while remaining robust to sensor noise. We further confirm practical feasibility and stable operation in real-world execution.
comment: 8 pages, 7 figures, 5 tables. Project Page: https://treeitsuki.github.io/ScanDP/
☆ Few-Shot Adaptation to Non-Stationary Environments via Latent Trend Embedding for Robotics
Robotic systems operating in real-world environments often suffer from concept shift, where the input-output relationship changes due to latent environmental factors that are not directly observable. Conventional adaptation methods update model parameters, which may cause catastrophic forgetting and incur high computational cost. This paper proposes a latent Trend ID-based framework for few-shot adaptation in non-stationary environments. Instead of modifying model weights, a low-dimensional environmental state, referred to as the Trend ID, is estimated via backpropagation while the model parameters remain fixed. To prevent overfitting caused by per-sample latent variables, we introduce temporal regularization and a state transition model that enforces smooth evolution of the latent space. Experiments on a quantitative food grasping task demonstrate that the learned Trend IDs are distributed across distinct regions of the latent space with temporally consistent trajectories, and that few-shot adaptation to unseen environments is achieved without modifying model parameters. The proposed framework provides a scalable and interpretable solution for robotics applications operating across diverse and evolving environments.
☆ Adaptive Manipulation Potential and Haptic Estimation for Tool-Mediated Interaction
Achieving human-level dexterity in contact-rich, tool-mediated manipulation remains a significant challenge due to visual occlusion and the underdetermined nature of haptic sensing. This paper introduces a parameterized Equilibrium Manifold (EM) as a unified representation for tool-mediated interaction, and develops a closed-loop framework that integrates haptic estimation, online planning, and adaptive stiffness control. We establish a physical-geometric duality using an adaptive manipulation potential incorporating a differentiable contact model, which induces the manifold's geometric structure and ensures that complex physical interactions are encapsulated as continuous operations on the EM. Within this framework, we reformulate haptic estimation as a manifold parameter estimation problem. Specifically, a hybrid inference strategy (haptic SLAM) is employed in which discrete object shapes are classified via particle filtering, while the continuous object pose is estimated using analytical gradients for efficient optimization. By continuously updating the parameters of the manipulation potential, the framework dynamically reshapes the induced EM to guide online trajectory replanning and implement uncertainty-aware impedance control, thereby closing the perception-action loop. The system is validated through simulation and over 260 real-world screw-loosening trials. Experimental results demonstrate robust identification and manipulation success in standard scenarios while maintaining accurate tracking. Furthermore, ablation studies confirm that haptic SLAM and uncertainty-aware stiffness modulation outperform fixed impedance baselines, effectively preventing jamming during tight tolerance interactions.
☆ Overcoming Visual Clutter in Vision Language Action Models via Concept-Gated Visual Distillation
Vision-Language-Action (VLA) models demonstrate impressive zero-shot generalization but frequently suffer from a "Precision-Reasoning Gap" in cluttered environments. This failure is driven by background-induced feature dilution, where high-frequency semantic noise corrupts the geometric grounding required for precise manipulation. To bridge this gap, we propose Concept-Gated Visual Distillation (CGVD), a training-free, model-agnostic inference framework that stabilizes VLA policies. CGVD operates by parsing instructions into safe and distractor sets, utilizing a two-layer target refinement process--combining cross-validation and spatial disambiguation--to explicitly penalize false positives and isolate genuine manipulation targets. We then process the scene via Fourier-based inpainting, generating a clean observation that actively suppresses semantic distractors while preserving critical spatial geometry and visual proprioception. Extensive evaluations in highly cluttered manipulation tasks demonstrate that CGVD prevents performance collapse. In environments with dense semantic distractors, our method significantly outperforms state-of-the-art baselines, achieving a 77.5% success rate compared to the baseline's 43.0%. By enforcing strict attribute adherence, CGVD establishes inference-time visual distillation as a critical prerequisite for robust robotic manipulation in the clutter.
comment: 7 pages, 4 figures, 3 tables
☆ PC-Diffuser: Path-Consistent Capsule CBF Safety Filtering for Diffusion-Based Trajectory Planner
Autonomous driving in complex traffic requires planners that generalize beyond hand-crafted rules, motivating data-driven approaches that learn behavior from expert demonstrations. Diffusion-based trajectory planners have recently shown strong closed-loop performance by iteratively denoising a full-horizon plan, but they remain difficult to certify and can fail catastrophically in rare or out-of-distribution scenarios. To address this challenge, we present PC-Diffuser, a safety augmentation framework that embeds a certifiable, path-consistent barrier-function structure directly into the denoising loop of diffusion planning. The key idea is to make safety an intrinsic part of trajectory generation rather than a post-hoc fix: we enforce forward invariance along the rollout while preserving the diffusion model's intended path geometry. Specifically, PC-Diffuser (i) evaluates collision risk using a capsule-distance barrier function that better reflects vehicle geometry and reduces unnecessary conservativeness, (ii) converts denoised waypoints into dynamically feasible motion under a kinematic bicycle model, and (iii) applies a path-consistent safety filter that eliminates residual constraint violations without geometric distortion, so the corrected plan remains close to the learned distribution. By injecting these safety-consistent corrections at every denoising step and feeding the refined trajectory back into the diffusion process, PC-Diffuser enables iterative, context-aware safeguarding instead of post-hoc repair...
☆ SteadyTray: Learning Object Balancing Tasks in Humanoid Tray Transport via Residual Reinforcement Learning
Stabilizing unsecured payloads against the inherent oscillations of dynamic bipedal locomotion remains a critical engineering bottleneck for humanoids in unstructured environments. To solve this, we introduce ReST-RL, a hierarchical reinforcement learning architecture that explicitly decouples locomotion from payload stabilization, evaluated via the SteadyTray benchmark. Rather than relying on monolithic end-to-end learning, our framework integrates a robust base locomotion policy with a dynamic residual module engineered to actively cancel gait-induced perturbations at the end-effector. This architectural separation ensures steady tray transport without degrading the underlying bipedal stability. In simulation, the residual design significantly outperforms end-to-end baselines in gait smoothness and orientation accuracy, achieving a 96.9% success rate in variable velocity tracking and 74.5% robustness against external force disturbances. Successfully deployed on the Unitree G1 humanoid hardware, this modular approach demonstrates highly reliable zero-shot sim-to-real generalization across various objects and external force disturbances.
comment: Project website: https://steadytray.github.io/
☆ Vision-Based Hand Shadowing for Robotic Manipulation via Inverse Kinematics
Teleoperation of low-cost robotic manipulators remains challenging due to the complexity of mapping human hand articulations to robot joint commands. We present an offline hand-shadowing and retargeting pipeline from a single egocentric RGB-D camera mounted on 3D-printed glasses. The pipeline detects 21 hand landmarks per hand using MediaPipe Hands, deprojects them into 3D via depth sensing, transforms them into the robot coordinate frame, and solves a damped-least-squares inverse kinematics problem in PyBullet to produce joint commands for the 6-DOF SO-ARM101 robot. A gripper controller maps thumb-index finger geometry to grasp aperture with a four-level fallback hierarchy. Actions are first previewed in a physics simulation before replay on the physical robot through the LeRobot framework. We evaluate the IK retargeting pipeline on a structured pick-and-place benchmark (5-tile grid, 10 grasps per tile) achieving a 90% success rate, and compare it against four vision-language-action policies (ACT, SmolVLA, pi0.5, GR00T N1.5) trained on leader-follower teleoperation data. We also test the IK pipeline in unstructured real-world environments (grocery store, pharmacy), where hand occlusion by surrounding objects reduces success to 9.3% (N=75), highlighting both the promise and current limitations of marker-free analytical retargeting.
☆ D-SLAMSpoof: An Environment-Agnostic LiDAR Spoofing Attack using Dynamic Point Cloud Injection
In this work, we introduce Dynamic SLAMSpoof (D-SLAMSpoof), a novel attack that compromises LiDAR SLAM even in feature-rich environments. The attack leverages LiDAR spoofing, which injects spurious measurements into LiDAR scans through external laser interference. By designing both spatial injection shapes and temporally coordinated dynamic injection patterns guided by scan-matching principles, D-SLAMSpoof significantly improves attack success rates in real-world, feature-rich environments such as urban areas and indoor spaces, where conventional LiDAR spoofing methods often fail. Furthermore, we propose a practical defense method, ISD-SLAM, that relies solely on inertial dead reckoning signals commonly available in autonomous systems. We demonstrate that ISD-SLAM accurately detects LiDAR spoofing attacks, including D-SLAMSpoof, and effectively mitigates the resulting position drift. Our findings expose inherent vulnerabilities in LiDAR-based SLAM and introduce the first practical defense against LiDAR-based SLAM spoofing using only standard onboard sensors, providing critical insights for improving the security and reliability of autonomous systems.
☆ MirrorDrift: Actuated Mirror-Based Attacks on LiDAR SLAM
LiDAR SLAM provides high-accuracy localization but is fragile to point-cloud corruption because scan matching assumes geometric consistency. Prior physical attacks on LiDAR SLAM largely rely on LiDAR spoofing via external signal injection, which requires sensor-specific timing knowledge and is increasingly mitigated by modern defense mechanisms such as timing obfuscation and injection rejection. In this work, we show that specular reflection offers an injection-free alternative and demonstrate an attack, MirrorDrift, that uses an actuated planar mirror to cause ghost points in LiDAR scans and systematically bias scan-matching correspondences. MirrorDrift optimizes mirror placement, alignment, and actuation. In simulation, it increases the average pose error (APE) by 6.1x over random placement, degrading three SLAM systems to 2.29-3.31 m mean APE. In real-world experiments on a modern LiDAR with state-of-the-art interference mitigation, it induces localization errors of up to 6.03 m. To the best of our knowledge, this is the first successful SLAM-targeted attack against production-grade secure LiDARs.
☆ Novelty Adaptation Through Hybrid Large Language Model (LLM)-Symbolic Planning and LLM-guided Reinforcement Learning
In dynamic open-world environments, autonomous agents often encounter novelties that hinder their ability to find plans to achieve their goals. Specifically, traditional symbolic planners fail to generate plans when the robot's planning domain lacks the operators that enable it to interact appropriately with novel objects in the environment. We propose a neuro-symbolic architecture that integrates symbolic planning, reinforcement learning, and a large language model (LLM) to learn how to handle novel objects. In particular, we leverage the common sense reasoning capability of the LLM to identify missing operators, generate plans with the symbolic AI planner, and write reward functions to guide the reinforcement learning agent in learning control policies for newly identified operators. Our method outperforms the state-of-the-art methods in operator discovery as well as operator learning in continuous robotic domains.
☆ Learning to Assist: Physics-Grounded Human-Human Control via Multi-Agent Reinforcement Learning CVPR 2026
Humanoid robotics has strong potential to transform daily service and caregiving applications. Although recent advances in general motion tracking within physics engines (GMT) have enabled virtual characters and humanoid robots to reproduce a broad range of human motions, these behaviors are primarily limited to contact-less social interactions or isolated movements. Assistive scenarios, by contrast, require continuous awareness of a human partner and rapid adaptation to their evolving posture and dynamics. In this paper, we formulate the imitation of closely interacting, force-exchanging human-human motion sequences as a multi-agent reinforcement learning problem. We jointly train partner-aware policies for both the supporter (assistant) agent and the recipient agent in a physics simulator to track assistive motion references. To make this problem tractable, we introduce a partner policies initialization scheme that transfers priors from single-human motion-tracking controllers, greatly improving exploration. We further propose dynamic reference retargeting and contact-promoting reward, which adapt the assistant's reference motion to the recipient's real-time pose and encourage physically meaningful support. We show that AssistMimic is the first method capable of successfully tracking assistive interaction motions on established benchmarks, demonstrating the benefits of a multi-agent RL formulation for physically grounded and socially aware humanoid control.
comment: Accepted at CVPR 2026 (main). Project page: https://yutoshibata07.github.io/AssistMimic-projectpage/
☆ ADMM-based Continuous Trajectory Optimization in Graphs of Convex Sets
This paper presents a numerical solver for computing continuous trajectories in non-convex environments. Our approach relies on a customized implementation of the Alternating Direction Method of Multipliers (ADMM) built upon two key components: First, we parameterize trajectories as polynomials, allowing the primal update to be computed in closed form as a minimum-control-effort problem. Second, we introduce the concept of a spatio-temporal allocation graph based on a mixed-integer formulation and pose the slack update as a shortest-path search. The combination of these ingredients results in a solver with several distinct advantages over the state of the art. By jointly optimizing over both discrete spatial and continuous temporal domains, our method accesses a larger search space than existing decoupled approaches, enabling the discovery of superior trajectories. Additionally, the solver's structural robustness ensures reliable convergence from naive initializations, removing the bottleneck of complex warm starting in non-convex environments.
☆ Distributed Kalman--Consensus Filtering with Adaptive Uncertainty Weighting for Multi-Object Tracking in Mobile Robot Networks
This paper presents an implementation and evaluation of a Distributed Kalman--Consensus Filter (DKCF) for Multi-Object Tracking (MOT) in mobile robot networks operating under partial observability and heterogeneous localization uncertainty. A key challenge in such systems is the fusion of information from agents with differing localization quality, where frame misalignment can lead to inconsistent estimates, track duplication, and ghost tracks. To address this issue, we build upon the MOTLEE framework and retain its frame-alignment methodology, which uses consistently tracked dynamic objects as transient landmarks to improve relative pose estimates between robots. On top of this framework, we propose an uncertainty-aware adaptive consensus weighting mechanism that dynamically adjusts the influence of neighbor information based on the covariance of the transmitted estimates, thereby reducing the impact of unreliable data during distributed fusion. Local tracking is performed using a Kalman Filter (KF) with a Constant Velocity Model (CVM) and Global Nearest Neighbor (GNN) data association. simulation results demonstrate that adaptive weighting effectively protects local estimates from inconsistent data, yielding a MOTA improvement of 0.09 for agents suffering from localization drift, although system performance remains constrained by communication latency.
comment: Presented at ICARA 2026. To appear in the IEEE conference proceedings
☆ A Causal Approach to Predicting and Improving Human Perceptions of Social Navigation Robots
As mobile robots are increasingly deployed in human environments, enabling them to predict how people perceive them is critical for socially adaptable navigation. Predicting perceptions is challenging for two main reasons: (1) HRI prediction models must learn from limited data, and (2) the obtained models must be interpretable to enable safe and effective interactions. Interpretability is particularly important when a robot is perceived as incompetent (e.g., when the robot suddenly stops or rotates away from the goal), as it allows the robot to explain its reasoning and identify controllable factors to improve performance, requiring causal rather than associative reasoning. To address these challenges, we propose a Causal Bayesian Network designed to predict how people perceive a mobile robot's competence and how they interpret its intent during navigation. Additionally, we introduce a novel method to improve perceived robot competence employing a combinatorial search, guided by the proposed causal model, to identify better navigation behaviors. Our method enhances interpretability and generates counterfactual robot motions while achieving comparable or superior predictive performance to state-of-the-art methods, reaching an F1-score of 0.78 and 0.75 for competence and intention on a binary scale. To further assess our method's ability to improve the perceived robot competence, we conducted an online evaluation in which users rated robot behaviors on a 5-point Likert scale. Our method statistically significantly increased the perceived competence of low-competent robot behavior by 83%.
comment: 8 pages, to be submitted to RA-L
☆ Multi-Robot Multitask Gaussian Process Estimation and Coverage
Coverage control is essential for the optimal deployment of agents to monitor or cover areas with sensory demands. While traditional coverage involves single-task robots, increasing autonomy now enables multitask operations. This paper introduces a novel multitask coverage problem and addresses it for both the cases of known and unknown sensory demands. For known demands, we design a federated multitask coverage algorithm and establish its convergence properties. For unknown demands, we employ a multitask Gaussian Process (GP) framework to learn sensory demand functions and integrate it with the multitask coverage algorithm to develop an adaptive algorithm. We introduce a novel notion of multitask coverage regret that compares the performance of the adaptive algorithm against an oracle with prior knowledge of the demand functions. We establish that our algorithm achieves sublinear cumulative regret, and numerically illustrate its performance.
☆ DynVLA: Learning World Dynamics for Action Reasoning in Autonomous Driving
We propose DynVLA, a driving VLA model that introduces a new CoT paradigm termed Dynamics CoT. DynVLA forecasts compact world dynamics before action generation, enabling more informed and physically grounded decision-making. To obtain compact dynamics representations, DynVLA introduces a Dynamics Tokenizer that compresses future evolution into a small set of dynamics tokens. Considering the rich environment dynamics in interaction-intensive driving scenarios, DynVLA decouples ego-centric and environment-centric dynamics, yielding more accurate world dynamics modeling. We then train DynVLA to generate dynamics tokens before actions through SFT and RFT, improving decision quality while maintaining latency-efficient inference. Compared to Textual CoT, which lacks fine-grained spatiotemporal understanding, and Visual CoT, which introduces substantial redundancy due to dense image prediction, Dynamics CoT captures the evolution of the world in a compact, interpretable, and efficient form. Extensive experiments on NAVSIM, Bench2Drive, and a large-scale in-house dataset demonstrate that DynVLA consistently outperforms Textual CoT and Visual CoT methods, validating the effectiveness and practical value of Dynamics CoT.
comment: 18 pages, 10 figures
☆ PPGuide: Steering Diffusion Policies with Performance Predictive Guidance ICRA'26
Diffusion policies have shown to be very efficient at learning complex, multi-modal behaviors for robotic manipulation. However, errors in generated action sequences can compound over time which can potentially lead to failure. Some approaches mitigate this by augmenting datasets with expert demonstrations or learning predictive world models which might be computationally expensive. We introduce Performance Predictive Guidance (PPGuide), a lightweight, classifier-based framework that steers a pre-trained diffusion policy away from failure modes at inference time. PPGuide makes use of a novel self-supervised process: it uses attention-based multiple instance learning to automatically estimate which observation-action chunks from the policy's rollouts are relevant to success or failure. We then train a performance predictor on this self-labeled data. During inference, this predictor provides a real-time gradient to guide the policy toward more robust actions. We validated our proposed PPGuide across a diverse set of tasks from the Robomimic and MimicGen benchmarks, demonstrating consistent improvements in performance.
comment: Accepted by ICRA'26
☆ Learning Adaptive Force Control for Contact-Rich Sample Scraping with Heterogeneous Materials IROS
The increasing demand for accelerated scientific discovery, driven by global challenges, highlights the need for advanced AI-driven robotics. Deploying robotic chemists in human-centric labs is key for the next horizon of autonomous discovery, as complex tasks still demand the dexterity of human scientists. Robotic manipulation in this context is uniquely challenged by handling diverse chemicals (granular, powdery, or viscous liquids), under varying lab conditions. For example, humans use spatulas for scraping materials from vial walls. Automating this process is challenging because it goes beyond simple robotic insertion tasks and traditional lab automation, requiring the execution of fine-granular movements within a constrained environment (the sample vial). Our work proposes an adaptive control framework to address this, relying on a low-level Cartesian impedance controller for stable and compliant physical interaction and a high-level reinforcement learning agent that learns to dynamically adjust interaction forces at the end-effector. The agent is guided by perception feedback, which provides the material's location. We first created a task-representative simulation environment with a Franka Research 3 robot, a scraping tool, and a sample vial containing heterogeneous materials. To facilitate the learning of an adaptive policy and model diverse characteristics, the sample is modelled as a collection of spheres, where each sphere is assigned a unique dislodgement force threshold, which is procedurally generated using Perlin noise. We train an agent to autonomously learn and adapt the optimal contact wrench for a sample scraping task in simulation and then successfully transfer this policy to a real robotic setup. Our method was evaluated across five different material setups, outperforming a fixed-wrench baseline by an average of 10.9%.
comment: 8 pages, 6 figures, 4 tables; Submitted to IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2026
☆ Contact Coverage-Guided Exploration for General-Purpose Dexterous Manipulation
Deep Reinforcement learning (DRL) has achieved remarkable success in domains with well-defined reward structures, such as Atari games and locomotion. In contrast, dexterous manipulation lacks general-purpose reward formulations and typically depends on task-specific, handcrafted priors to guide hand-object interactions. We propose Contact Coverage-Guided Exploration (CCGE), a general exploration method designed for general-purpose dexterous manipulation tasks. CCGE represents contact state as the intersection between object surface points and predefined hand keypoints, encouraging dexterous hands to discover diverse and novel contact patterns, namely which fingers contact which object regions. It maintains a contact counter conditioned on discretized object states obtained via learned hash codes, capturing how frequently each finger interacts with different object regions. This counter is leveraged in two complementary ways: (1) to assign a count-based contact coverage reward that promotes exploration of novel contact patterns, and (2) an energy-based reaching reward that guides the agent toward under-explored contact regions. We evaluate CCGE on a diverse set of dexterous manipulation tasks, including cluttered object singulation, constrained object retrieval, in-hand reorientation, and bimanual manipulation. Experimental results show that CCGE substantially improves training efficiency and success rates over existing exploration methods, and that the contact patterns learned with CCGE transfer robustly to real-world robotic systems. Project page is https://contact-coverage-guided-exploration.github.io.
comment: 16 pages
☆ STADA: Specification-based Testing for Autonomous Driving Agents
Simulation-based testing has become a standard approach to validating autonomous driving agents prior to real-world deployment. A high-quality validation campaign will exercise an agent in diverse contexts comprised of varying static environments, e.g., lanes, intersections, signage, and dynamic elements, e.g., vehicles and pedestrians. To achieve this, existing test generation techniques rely on template-based, manually constructed, or random scenario generation. When applied to validate formally specified safety requirements, such methods either require significant human effort or run the risk of missing important behavior related to the requirement. To address this gap, we present STADA, a Specification-based Test generation framework for Autonomous Driving Agents that systematically generates the space of scenarios defined by a formal specification expressed in temporal logic (LTLf). Given a specification, STADA constructs all distinct initial scenes, a diverse space of continuations of those scenes, and simulations that reflect the behaviors of the specification. Evaluation of STADA on a variety of LTLf specifications formalized in SCENEFLOW using three complementary coverage criteria demonstrates that STADA yields more than 2x higher coverage than the best baseline on the finest criteria and a 75% increase for the coarsest criteria. Moreover, it matches the coverage of the best baseline with 6 times fewer simulations. While set in the context of autonomous driving, the approach is applicable to other domains with rich simulation environments.
☆ Lifelong Imitation Learning with Multimodal Latent Replay and Incremental Adjustment
We introduce a lifelong imitation learning framework that enables continual policy refinement across sequential tasks under realistic memory and data constraints. Our approach departs from conventional experience replay by operating entirely in a multimodal latent space, where compact representations of visual, linguistic, and robot's state information are stored and reused to support future learning. To further stabilize adaptation, we introduce an incremental feature adjustment mechanism that regularizes the evolution of task embeddings through an angular margin constraint, preserving inter-task distinctiveness. Our method establishes a new state of the art in the LIBERO benchmarks, achieving 10-17 point gains in AUC and up to 65% less forgetting compared to previous leading methods. Ablation studies confirm the effectiveness of each component, showing consistent gains over alternative strategies. The code is available at: https://github.com/yfqi/lifelong_mlr_ifa.
☆ Robust Co-design Optimisation for Agile Fixed-Wing UAVs
Co-design optimisation of autonomous systems has emerged as a powerful alternative to sequential approaches by jointly optimising physical design and control strategies. However, existing frameworks often neglect the robustness required for autonomous systems navigating unstructured, real-world environments. For agile Unmanned Aerial Vehicles (UAVs) operating at the edge of the flight envelope, this lack of robustness yields designs that are sensitive to perturbations and model mismatch. To address this, we propose a robust co-design framework for agile fixed-wing UAVs that integrates parametric uncertainty and wind disturbances directly into the concurrent optimisation process. Our bi-level approach optimises physical design in a high-level loop while discovering nominal solutions via a constrained trajectory planner and evaluating performance across a stochastic Monte Carlo ensemble using feedback LQR control. Validated across three agile flight missions, our strategy consistently outperforms deterministic baselines. The results demonstrate that our robust co-design strategy inherently tailors aerodynamic features, such as wing placement and aspect ratio, to achieve an optimal trade-off between mission performance and disturbance rejection.
☆ ResWM: Residual-Action World Model for Visual RL KDD2026
Learning predictive world models from raw visual observations is a central challenge in reinforcement learning (RL), especially for robotics and continuous control. Conventional model-based RL frameworks directly condition future predictions on absolute actions, which makes optimization unstable: the optimal action distributions are task-dependent, unknown a priori, and often lead to oscillatory or inefficient control. To address this, we introduce the Residual-Action World Model (ResWM), a new framework that reformulates the control variable from absolute actions to residual actions -- incremental adjustments relative to the previous step. This design aligns with the inherent smoothness of real-world control, reduces the effective search space, and stabilizes long-horizon planning. To further strengthen the representation, we propose an Observation Difference Encoder that explicitly models the changes between adjacent frames, yielding compact latent dynamics that are naturally coupled with residual actions. ResWM is integrated into a Dreamer-style latent dynamics model with minimal modifications and no extra hyperparameters. Both imagination rollouts and policy optimization are conducted in the residual-action space, enabling smoother exploration, lower control variance, and more reliable planning. Empirical results on the DeepMind Control Suite demonstrate that ResWM achieves consistent improvements in sample efficiency, asymptotic returns, and control smoothness, significantly surpassing strong baselines such as Dreamer and TD-MPC. Beyond performance, ResWM produces more stable and energy-efficient action trajectories, a property critical for robotic systems deployed in real-world environments. These findings suggest that residual action modeling provides a simple yet powerful principle for bridging algorithmic advances in RL with the practical requirements of robotics.
comment: Submit KDD2026
☆ RC-NF: Robot-Conditioned Normalizing Flow for Real-Time Anomaly Detection in Robotic Manipulation CVPR
Recent advances in Vision-Language-Action (VLA) models have enabled robots to execute increasingly complex tasks. However, VLA models trained through imitation learning struggle to operate reliably in dynamic environments and often fail under Out-of-Distribution (OOD) conditions. To address this issue, we propose Robot-Conditioned Normalizing Flow (RC-NF), a real-time monitoring model for robotic anomaly detection and intervention that ensures the robot's state and the object's motion trajectory align with the task. RC-NF decouples the processing of task-aware robot and object states within the normalizing flow. It requires only positive samples for unsupervised training and calculates accurate robotic anomaly scores during inference through the probability density function. We further present LIBERO-Anomaly-10, a benchmark comprising three categories of robotic anomalies for simulation evaluation. RC-NF achieves state-of-the-art performance across all anomaly types compared to previous methods in monitoring robotic tasks. Real-world experiments demonstrate that RC-NF operates as a plug-and-play module for VLA models (e.g., pi0), providing a real-time OOD signal that enables state-level rollback or task-level replanning when necessary, with a response latency under 100 ms. These results demonstrate that RC-NF noticeably enhances the robustness and adaptability of VLA-based robotic systems in dynamic environments.
comment: Accepted to the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026
☆ Thousand-GPU Large-Scale Training and Optimization Recipe for AI-Native Cloud Embodied Intelligence Infrastructure
Embodied intelligence is a key step towards Artificial General Intelligence (AGI), yet its development faces multiple challenges including data, frameworks, infrastructure, and evaluation systems. To address these issues, we have, for the first time in the industry, launched a cloud-based, thousand-GPU distributed training platform for embodied intelligence, built upon the widely adopted LeRobot framework, and have systematically overcome bottlenecks across the entire pipeline. At the data layer, we have restructured the data pipeline to optimize the flow of embodied training data. In terms of training, for the GR00T-N1.5 model, utilizing thousand-GPU clusters and data at the scale of hundreds of millions, the single-round training time has been reduced from 15 hours to just 22 minutes, achieving a 40-fold speedup. At the model layer, by combining variable-length FlashAttention and Data Packing, we have moved from sample redundancy to sequence integration, resulting in a 188% speed increase; π-0.5 attention optimization has accelerated training by 165%; and FP8 quantization has delivered a 140% speedup. On the infrastructure side, relying on high-performance storage, a 3.2T RDMA network, and a Ray-driven elastic AI data lake, we have achieved deep synergy among data, storage, communication, and computation. We have also built an end-to-end evaluation system, creating a closed loop from training to simulation to assessment. This framework has already been fully validated on thousand-GPU clusters, laying a crucial technical foundation for the development and application of next-generation autonomous intelligent robots, and is expected to accelerate the arrival of the era of human-machine integration.
☆ A Survey of Reasoning in Autonomous Driving Systems: Open Challenges and Emerging Paradigms
The development of high-level autonomous driving (AD) is shifting from perception-centric limitations to a more fundamental bottleneck, namely, a deficit in robust and generalizable reasoning. Although current AD systems manage structured environments, they consistently falter in long-tail scenarios and complex social interactions that require human-like judgment. Meanwhile, the advent of large language and multimodal models (LLMs and MLLMs) presents a transformative opportunity to integrate a powerful cognitive engine into AD systems, moving beyond pattern matching toward genuine comprehension. However, a systematic framework to guide this integration is critically lacking. To bridge this gap, we provide a comprehensive review of this emerging field and argue that reasoning should be elevated from a modular component to the system's cognitive core. Specifically, we first propose a novel Cognitive Hierarchy to decompose the monolithic driving task according to its cognitive and interactive complexity. Building on this, we further derive and systematize seven core reasoning challenges, such as the responsiveness-reasoning trade-off and social-game reasoning. Furthermore, we conduct a dual-perspective review of the state-of-the-art, analyzing both system-centric approaches to architecting intelligent agents and evaluation-centric practices for their validation. Our analysis reveals a clear trend toward holistic and interpretable "glass-box" agents. In conclusion, we identify a fundamental and unresolved tension between the high-latency, deliberative nature of LLM-based reasoning and the millisecond-scale, safety-critical demands of vehicle control. For future work, a primary objective is to bridge the symbolic-to-physical gap by developing verifiable neuro-symbolic architectures, robust reasoning under uncertainty, and scalable models for implicit social negotiation.
comment: Published in TMLR (March 2026) | OpenReview: https://openreview.net/forum?id=XwQ7dc4bqn
☆ Edge-Assisted Multi-Robot Visual-Inertial SLAM with Efficient Communication
The integration of cloud computing and edge computing is an effective way to achieve global consistent and real-time multi-robot Simultaneous Localization and Mapping (SLAM). Cloud computing effectively solves the problem of limited computing, communication and storage capacity of terminal equipment. However, limited bandwidth and extremely long communication links between terminal devices and the cloud result in serious performance degradation of multi-robot SLAM systems. To reduce the computational cost of feature tracking and improve the real-time performance of the robot, a lightweight SLAM method of optical flow tracking based on pyramid IMU prediction is proposed. On this basis, a centralized multi-robot SLAM system based on a robot-edge-cloud layered architecture is proposed to realize real-time collaborative SLAM. It avoids the problems of limited on-board computing resources and low execution efficiency of single robot. In this framework, only the feature points and keyframe descriptors are transmitted and lossless encoding and compression are carried out to realize real-time remote information transmission with limited bandwidth resources. This design reduces the actual bandwidth occupied in the process of data transmission, and does not cause the loss of SLAM accuracy caused by data compression. Through experimental verification on the EuRoC dataset, compared with the current most advanced local feature compression method, our method can achieve lower data volume feature transmission, and compared with the current advanced centralized multi-robot SLAM scheme, it can achieve the same or better positioning accuracy under low computational load.
comment: 13 pages, 18 figures
♻ ☆ Partially Equivariant Reinforcement Learning in Symmetry-Breaking Environments ICLR 2026
Group symmetries provide a powerful inductive bias for reinforcement learning (RL), enabling efficient generalization across symmetric states and actions via group-invariant Markov Decision Processes (MDPs). However, real-world environments almost never realize fully group-invariant MDPs; dynamics, actuation limits, and reward design usually break symmetries, often only locally. Under group-invariant Bellman backups for such cases, local symmetry-breaking introduces errors that propagate across the entire state-action space, resulting in global value estimation errors. To address this, we introduce Partially group-Invariant MDP (PI-MDP), which selectively applies group-invariant or standard Bellman backups depending on where symmetry holds. This framework mitigates error propagation from locally broken symmetries while maintaining the benefits of equivariance, thereby enhancing sample efficiency and generalizability. Building on this framework, we present practical RL algorithms -- Partially Equivariant (PE)-DQN for discrete control and PE-SAC for continuous control -- that combine the benefits of equivariance with robustness to symmetry-breaking. Experiments across Grid-World, locomotion, and manipulation benchmarks demonstrate that PE-DQN and PE-SAC significantly outperform baselines, highlighting the importance of selective symmetry exploitation for robust and sample-efficient RL. Project page: https://pranaboy72.github.io/perl_page/
comment: ICLR 2026
♻ ☆ RACAS: Controlling Diverse Robots With a Single Agentic System
Many robotic platforms expose an API through which external software can command their actuators and read their sensors. However, transitioning from these low-level interfaces to high-level autonomous behaviour requires a complicated pipeline, whose components demand distinct areas of expertise. Existing approaches to bridging this gap either require retraining for every new embodiment or have only been validated across structurally similar platforms. We introduce RACAS (Robot-Agnostic Control via Agentic Systems), a cooperative agentic architecture in which three LLM/VLM-based modules (Monitors, a Controller, and a Memory Curator) communicate exclusively through natural language to provide closed-loop robot control. RACAS requires only a natural language description of the robot, a definition of available actions, and a task specification; no source code, model weights, or reward functions need to be modified to move between platforms. We evaluate RACAS on several tasks using a wheeled ground robot, a recently published novel multi-jointed robotic limb, and an underwater vehicle. RACAS consistently solved all assigned tasks across these radically different platforms, demonstrating the potential of agentic AI to substantially reduce the barrier to prototyping robotic solutions.
comment: 7 pages in main text + 1 page of appendices + 1 page of references, 5 figures in main text + 1 figure in appendices, 2 tables in main text; source code available at https://github.com/janprz11/robot-agnostic-control
♻ ☆ vS-Graphs: Tightly Coupling Visual SLAM and 3D Scene Graphs Exploiting Hierarchical Scene Understanding
Current Visual Simultaneous Localization and Mapping (VSLAM) systems often struggle to create maps that are both semantically rich and easily interpretable. While incorporating semantic scene knowledge aids in building richer maps with contextual associations among mapped objects, representing them in structured formats, such as scene graphs, has not been widely addressed, resulting in complex map comprehension and limited scalability. This paper introduces vS-Graphs, a novel real-time VSLAM framework that integrates vision-based scene understanding with map reconstruction and comprehensible graph-based representation. The framework infers structural elements (i.e., rooms and floors) from detected building components (i.e., walls and ground surfaces) and incorporates them into optimizable 3D scene graphs. This solution enhances the reconstructed map's semantic richness, comprehensibility, and localization accuracy. Extensive experiments on standard benchmarks and real-world datasets demonstrate that vS-Graphs achieves an average of 15.22% accuracy gain across all tested datasets compared to state-of-the-art VSLAM methods. Furthermore, the proposed framework achieves environment-driven semantic entity detection accuracy comparable to that of precise LiDAR-based frameworks, using only visual features. The code is publicly available at https://github.com/snt-arg/visual_sgraphs and is actively being improved. Moreover, a web page containing more media and evaluation outcomes is available on https://snt-arg.github.io/vsgraphs-results/.
comment: 20 pages, 10 figures, 5 tables
♻ ☆ A Chain-Driven, Sandwich-Legged Quadruped Robot: Design and Experimental Analysis
This paper introduces a chain-driven, sandwich-legged mid-size quadruped robot designed as an accessible research platform. The design prioritizes enhanced locomotion, improved actuation reliability and safety, and simplified, cost-effective manufacturing. Locomotion performance is improved through a sandwiched leg architecture and dual-motor configuration, reducing leg inertia for agile motion. Reliability and safety are enhanced using robust cable strain reliefs, motor heat sinks for thermal management, and mechanical limits to restrict leg motion. The design incorporates quasi-direct-drive (QDD) actuators and low-cost fabrication methods such as laser cutting and 3D printing for rapid prototyping. The $25\,\mathrm{kg}$ robot is built under \$8000, providing an affordable quadruped research platform. Experiments demonstrate trot and crawl gaits on flat terrain and slopes. We also open-source the mechanical designs. VIDEO: https://youtu.be/ygSMCPcFnP8?feature=shared CADs: https://github.com/singhaman1750/stoch3-design.git
comment: 6 pages, 9 figures
♻ ☆ CostNav: A Navigation Benchmark for Real-World Economic-Cost Evaluation of Physical AI Agents
While current navigation benchmarks prioritize task success in simplified settings, they neglect the multidimensional economic constraints essential for the real-world commercialization of autonomous delivery systems. We introduce CostNav, an Economic Navigation Benchmark that evaluates physical AI agents through comprehensive economic cost-revenue analysis aligned with real-world business operations. By integrating industry-standard data--such as Securities and Exchange Commission (SEC) filings and Abbreviated Injury Scale (AIS) injury reports--with Isaac Sim's detailed collision and cargo dynamics, CostNav transcends simple task completion to accurately evaluate business value in complex, real-world scenarios. To our knowledge, CostNav is the first physics-grounded economic benchmark that uses industry-standard regulatory and financial data to quantitatively expose the gap between navigation research metrics and commercial viability, revealing that optimizing for task success on a simplified task fundamentally differs from optimizing for real-world economic deployment. Evaluating seven baselines--two rule-based and five imitation learning--we find that no current method is economically viable, all yielding negative contribution margins. The best-performing method, CANVAS (-27.36\$/run), equipped with only an RGB camera and GPS, outperforms LiDAR-equipped Nav2 w/ GPS (-35.46\$/run). We challenge the community to develop navigation policies that achieve economic viability on CostNav. We remain method-agnostic, evaluating success solely on cost rather than the underlying architecture. All resources are available at https://github.com/worv-ai/CostNav.
♻ ☆ Cross-embodied Co-design for Dexterous Hands
Dexterous manipulation is limited by both control and design, without consensus as to what makes manipulators best for performing dexterous tasks. This raises a fundamental challenge: how should we design and control robot manipulators that are optimized for dexterity? We present a co-design framework that learns task-specific hand morphology and complementary dexterous control policies. The framework supports 1) an expansive morphology search space including joint, finger, and palm generation, 2) scalable evaluation across the wide design space via morphology-conditioned cross-embodied control, and 3) real-world fabrication with accessible components. We evaluate the approach across multiple dexterous tasks, including in-hand rotation with simulation and real deployment. Our framework enables an end-to-end pipeline that can design, train, fabricate, and deploy a new robotic hand in under 24 hours. The full framework will be open-sourced and available on our website: https://an-axolotl.github.io/HouseofDextra/ .
♻ ☆ CompassNav: Steering From Path Imitation To Decision Understanding In Navigation
The dominant paradigm for training Large Vision-Language Models (LVLMs) in navigation relies on imitating expert trajectories. This approach reduces the complex navigation task to a sequence-to-sequence replication of a single correct path, fundamentally limiting the agent's ability to explore and generalize. In this work, we argue for and introduce a new paradigm: a shift from Path Imitation to Decision Understanding. The goal of this paradigm is to build agents that do not just follow, but truly understand how to navigate. We materialize this through two core contributions: first, we introduce Compass-Data-22k, a novel 22k-trajectory dataset. Its Reinforcement Fine-Tuning (RFT) subset provides a panoramic view of the decision landscape by annotating all feasible actions with A* geodesic distances. Second, we design a novel gap-aware hybrid reward function that dynamically adapts its feedback to decision certainty, shifting between decisive signals for optimal actions and nuanced scores to encourage exploration. Integrated into an SFT-then-RFT recipe, our CompassNav agent is trained not to memorize static routes, but to develop an internal compass that constantly intuits the direction to the goal by evaluating the relative quality of all possible moves. This approach enables our 7B agent to set a new state-of-the-art on Goal navigation benchmarks, outperforming even larger proprietary models, and achieve robust real-world goal navigation on a physical robot.
♻ ☆ SPAARS: Safer RL Policy Alignment through Abstract Exploration and Refined Exploitation of Action Space
Offline-to-online reinforcement learning (RL) offers a promising paradigm for robotics by pre-training policies on safe, offline demonstrations and fine-tuning them via online interaction. However, a fundamental challenge remains: how to safely explore online without deviating from the behavioral support of the offline data? While recent methods leverage conditional variational autoencoders (CVAEs) to bound exploration within a latent space, they inherently suffer from an exploitation gap -- a performance ceiling imposed by the decoder's reconstruction loss. We introduce SPAARS, a curriculum learning framework that initially constrains exploration to the low-dimensional latent manifold for sample-efficient, safe behavioral improvement, then seamlessly transfers control to the raw action space, bypassing the decoder bottleneck. SPAARS has two instantiations: the CVAE-based variant requires only unordered (s,a) pairs and no trajectory segmentation; SPAARS-SUPE pairs SPAARS with OPAL temporal skill pretraining for stronger exploration structure at the cost of requiring trajectory chunks. We prove an upper bound on the exploitation gap using the Performance Difference Lemma, establish that latent-space policy gradients achieve provable variance reduction over raw-space exploration, and show that concurrent behavioral cloning during the latent phase directly controls curriculum transition stability. Empirically, SPAARS-SUPE achieves 0.825 normalized return on kitchen-mixed-v0 versus 0.75 for SUPE, with 5x better sample efficiency; standalone SPAARS achieves 92.7 and 102.9 normalized return on hopper-medium-v2 and walker2d-medium-v2 respectively, surpassing IQL baselines of 66.3 and 78.3 respectively, confirming the utility of the unordered-pair CVAE instantiation.
comment: 9 pages
♻ ☆ Context-Nav: Context-Driven Exploration and Viewpoint-Aware 3D Spatial Reasoning for Instance Navigation CVPR 2026
Text-goal instance navigation (TGIN) asks an agent to resolve a single, free-form description into actions that reach the correct object instance among same-category distractors. We present \textit{Context-Nav} that elevates long, contextual captions from a local matching cue to a global exploration prior and verifies candidates through 3D spatial reasoning. First, we compute dense text-image alignments for a value map that ranks frontiers -- guiding exploration toward regions consistent with the entire description rather than early detections. Second, upon observing a candidate, we perform a viewpoint-aware relation check: the agent samples plausible observer poses, aligns local frames, and accepts a target only if the spatial relations can be satisfied from at least one viewpoint. The pipeline requires no task-specific training or fine-tuning; we attain state-of-the-art performance on InstanceNav and CoIN-Bench. Ablations show that (i) encoding full captions into the value map avoids wasted motion and (ii) explicit, viewpoint-aware 3D verification prevents semantically plausible but incorrect stops. This suggests that geometry-grounded spatial reasoning is a scalable alternative to heavy policy training or human-in-the-loop interaction for fine-grained instance disambiguation in cluttered 3D scenes.
comment: Camera-ready version. Accepted to CVPR 2026
♻ ☆ Scalable Multi-Task Learning through Spiking Neural Networks with Adaptive Task-Switching Policy for Intelligent Autonomous Agents
Training resource-constrained autonomous agents on multiple tasks simultaneously is crucial for adapting to diverse real-world environments. Recent works employ reinforcement learning (RL) approach, but they still suffer from sub-optimal multi-task performance due to task interference. State-of-the-art works employ Spiking Neural Networks (SNNs) to improve RL-based multi-task learning and enable low-power/energy operations through network enhancements and spike-driven data stream processing. However, they rely on fixed task-switching intervals during its training, thus limiting its performance and scalability. To address this, we propose SwitchMT, a novel methodology that employs adaptive task-switching for effective, scalable, and simultaneous multi-task learning. SwitchMT employs the following key ideas: (1) leveraging a Deep Spiking Q-Network with active dendrites and dueling structure, that utilizes task-specific context signals to create specialized sub-networks; and (2) devising an adaptive task-switching policy that leverages both rewards and internal dynamics of the network parameters. Experimental results demonstrate that SwitchMT achieves competitive scores in multiple Atari games (i.e., Pong: -8.8, Breakout: 5.6, and Enduro: 355.2) and longer game episodes as compared to the state-of-the-art. These results also highlight the effectiveness of SwitchMT methodology in addressing task interference without increasing the network complexity, enabling intelligent autonomous agents with scalable multi-task learning capabilities.
comment: Accepted at the 63rd ACM/IEEE Design Automation Conference (DAC), July 26-29, 2026 in Long Beach, CA, USA Codes: https://github.com/rachmadvwp/SwitchMT
♻ ☆ Robust Cooperative Localization in Featureless Environments: A Comparative Study of DCL, StCL, CCL, CI, and Standard-CL
Cooperative localization (CL) enables accurate position estimation in multi-robot systems operating in GPS-denied environments. This paper presents a comparative study of five CL approaches: Centralized Cooperative Localization (CCL), Decentralized Cooperative Localization (DCL), Sequential Cooperative Localization (StCL), Covariance Intersection (CI), and Standard Cooperative Localization (Standard-CL). All methods are implemented in ROS and evaluated through Monte Carlo simulations under two conditions: weak data association and robust detection. Our analysis reveals fundamental trade-offs among the methods. StCL and Standard-CL achieve the lowest position errors but exhibit severe filter inconsistency, making them unsuitable for safety-critical applications. DCL demonstrates remarkable stability under challenging conditions due to its measurement stride mechanism, which provides implicit regularization against outliers. CI emerges as the most balanced approach, achieving near-optimal consistency while maintaining competitive accuracy. CCL provides theoretically optimal estimation but shows sensitivity to measurement outliers. These findings offer practical guidance for selecting CL algorithms based on application requirements.
comment: Presented at the 2026 12th International Conference on Automation, Robotics and Applications (ICARA); to be published in IEEE conference proceedings
♻ ☆ Cosmos-H-Surgical: Learning Surgical Robot Policies from Videos via World Modeling
Data scarcity remains a fundamental barrier to achieving fully autonomous surgical robots. While large scale vision language action (VLA) models have shown impressive generalization in household and industrial manipulation by leveraging paired video action data from diverse domains, surgical robotics suffers from the paucity of datasets that include both visual observations and accurate robot kinematics. In contrast, vast corpora of surgical videos exist, but they lack corresponding action labels, preventing direct application of imitation learning or VLA training. In this work, we aim to alleviate this problem by learning policy models from Cosmos-H-Surgical, a world model designed for surgical physical AI. We curated the Surgical Action Text Alignment (SATA) dataset with detailed action description specifically for surgical robots. Then we built Cosmos-H-Surgical based on the most advanced physical AI world model and SATA. It's able to generate diverse, generalizable and realistic surgery videos. We are also the first to use an inverse dynamics model to infer pseudokinematics from synthetic surgical videos, producing synthetic paired video action data. We demonstrate that a surgical VLA policy trained with these augmented data significantly outperforms models trained only on real demonstrations on a real surgical robot platform. Our approach offers a scalable path toward autonomous surgical skill acquisition by leveraging the abundance of unlabeled surgical video and generative world modeling, thus opening the door to generalizable and data efficient surgical robot policies.
♻ ☆ Symskill: Symbol and Skill Co-Invention for Data-Efficient and Reactive Long-Horizon Manipulation ICRA 2026
Multi-step manipulation in dynamic environments remains challenging. Imitation learning (IL) is reactive but lacks compositional generalization, since monolithic policies do not decide which skill to reuse when scenes change. Classical task-and-motion planning (TAMP) offers compositionality, but its high planning latency prevents real-time failure recovery. We introduce SymSkill, a unified framework that jointly learns predicates, operators, and skills from unlabeled, unsegmented demonstrations, combining compositional generalization with real-time recovery. Offline, SymSkill learns symbolic abstractions and goal-oriented skills directly from demonstrations. Online, given a conjunction of learned predicates, it uses a symbolic planner to compose and reorder skills to achieve symbolic goals while recovering from failures at both the motion and symbolic levels in real time. Coupled with a compliant controller, SymSkill supports safe execution under human and environmental disturbances. In RoboCasa simulation, SymSkill executes 12 single-step tasks with 85% success and composes them into multi-step plans without additional data. On a real Franka robot, it learns from 5 minutes of play data and performs 12-step tasks from goal specifications. Code and additional analysis are available at https://sites.google.com/view/symskill.
comment: ICRA 2026; CoRL 2025 Learning Effective Abstractions for Planning (LEAP) Workshop Best Paper Award (https://sites.google.com/view/symskill)
♻ ☆ Open-World Task and Motion Planning via Vision-Language Model Generated Constraints
Foundation models like Vision-Language Models (VLMs) excel at common sense vision and language tasks such as visual question answering. However, they cannot yet directly solve complex, long-horizon robot manipulation problems requiring precise continuous reasoning. Task and Motion Planning (TAMP) systems can handle long-horizon reasoning through discrete-continuous hybrid search over parameterized skills, but rely on detailed environment models and cannot interpret novel human objectives, such as arbitrary natural language goals. We propose integrating VLMs into TAMP systems by having them generate discrete and continuous language-parameterized constraints that enable open-world reasoning. Specifically, we use VLMs to generate discrete action ordering constraints that constrain TAMP search over action sequences, and continuous constraints in the form of code that augments traditional TAMP manipulation constraints. Experiments show that our approach, OWL-TAMP, outperforms baselines relying solely on TAMP or VLMs across several long-horizon manipulation tasks specified directly in natural language. We additionally demonstrate that OWL-TAMP can be deployed with an off-the-shelf TAMP system to solve challenging manipulation tasks on real-world hardware.
comment: A version of this paper appears in IEEE Robotics and Automation Letters (RA-L) Volume 11, Issue 3
♻ ☆ PlayWorld: Learning Robot World Models from Autonomous Play
Action-conditioned video models offer a promising path to building general-purpose robot simulators that can improve directly from data. Yet, despite training on large-scale robot datasets, current state-of-the-art video models still struggle to predict physically consistent robot-object interactions that are crucial in robotic manipulation. To close this gap, we present PlayWorld, a simple, scalable, and fully autonomous pipeline for training high-fidelity video world simulators from interaction experience. In contrast to prior approaches that rely on success-biased human demonstrations, PlayWorld is the first system capable of learning entirely from unsupervised robot self-play, enabling naturally scalable data collection while capturing complex, long-tailed physical interactions essential for modeling realistic object dynamics. Experiments across diverse manipulation tasks show that PlayWorld generates high-quality, physically consistent predictions for contact-rich interactions that are not captured by world models trained on human-collected data. We further demonstrate the versatility of PlayWorld in enabling fine-grained failure prediction and policy evaluation, with up to 40% improvements over human-collected data. Finally, we demonstrate how PlayWorld enables reinforcement learning in the world model, improving policy performance by 65% in success rates when deployed in the real world.
comment: https://robot-playworld.github.io/
♻ ☆ PvP: Data-Efficient Humanoid Robot Learning with Proprioceptive-Privileged Contrastive Representations
Achieving efficient and robust whole-body control (WBC) is essential for enabling humanoid robots to perform complex tasks in dynamic environments. Despite the success of reinforcement learning (RL) in this domain, its sample inefficiency remains a significant challenge due to the intricate dynamics and partial observability of humanoid robots. To address this limitation, we propose PvP, a Proprioceptive-Privileged contrastive learning framework that leverages the intrinsic complementarity between proprioceptive and privileged states. PvP learns compact and task-relevant latent representations without requiring hand-crafted data augmentations, enabling faster and more stable policy learning. To support systematic evaluation, we develop SRL4Humanoid, the first unified and modular framework that provides high-quality implementations of representative state representation learning (SRL) methods for humanoid robot learning. Extensive experiments on the LimX Oli robot across velocity tracking and motion imitation tasks demonstrate that PvP significantly improves sample efficiency and final performance compared to baseline SRL methods. Our study further provides practical insights into integrating SRL with RL for humanoid WBC, offering valuable guidance for data-efficient humanoid robot learning.
comment: 15 pages, 17 figures
♻ ☆ Self-Improving Loops for Visual Robotic Planning ICLR 2026
Video generative models trained on expert demonstrations have been utilized as performant text-conditioned visual planners for solving robotic tasks. However, generalization to unseen tasks remains a challenge. Whereas improved generalization may be facilitated by leveraging learned prior knowledge from additional pre-collected offline data sources, such as web-scale video datasets, in the era of experience we aim to design agents that can continuously improve in an online manner from self-collected behaviors. In this work we thus propose the Self-Improving Loops for Visual Robotic Planning (SILVR), where an in-domain video model iteratively updates itself on self-produced trajectories, and steadily improves its performance for a specified task of interest. We apply SILVR to a diverse suite of MetaWorld tasks, as well as two manipulation tasks on a real robot arm, and find that performance improvements continuously emerge over multiple iterations for novel tasks unseen during initial in-domain video model training. We demonstrate that SILVR is robust in the absence of human-provided ground-truth reward functions or expert-quality demonstrations, and is preferable to alternate approaches that utilize online experience in terms of performance and sample efficiency.
comment: ICLR 2026. Project Page: https://diffusion-supervision.github.io/silvr/
♻ ☆ MergeVLA: Cross-Skill Model Merging Toward a Generalist Vision-Language-Action Agent CVPR 2026
Recent Vision-Language-Action (VLA) models reformulate vision-language models by tuning them with millions of robotic demonstrations. While they perform well when fine-tuned for a single embodiment or task family, extending them to multi-skill settings remains challenging: directly merging VLA experts trained on different tasks results in near-zero success rates. This raises a fundamental question: what prevents VLAs from mastering multiple skills within one model? With an empirical decomposition of learnable parameters during VLA fine-tuning, we identify two key sources of non-mergeability: (1) Finetuning drives LoRA adapters in the VLM backbone toward divergent, task-specific directions beyond the capacity of existing merging methods to unify. (2) Action experts develop inter-block dependencies through self-attention feedback, causing task information to spread across layers and preventing modular recombination. To address these challenges, we present MergeVLA, a merging-oriented VLA architecture that preserves mergeability by design. MergeVLA introduces sparsely activated LoRA adapters via task masks to retain consistent parameters and reduce irreconcilable conflicts in the VLM. Its action expert replaces self-attention with cross-attention-only blocks to keep specialization localized and composable. When the task is unknown, it uses a test-time task router to adaptively select the appropriate task mask and expert head from the initial observation, enabling unsupervised task inference. Across LIBERO, LIBERO-Plus, RoboTwin, and multi-task experiments on the real SO101 robotic arm, MergeVLA achieves performance comparable to or even exceeding individually finetuned experts, demonstrating robust generalization across tasks, embodiments, and environments. Project page: https://mergevla.github.io/
comment: Accepted to CVPR 2026
♻ ☆ UniFField: A Generalizable Unified Neural Feature Field for Visual, Semantic, and Spatial Uncertainties in Any Scene ICRA 2026
Comprehensive visual, geometric, and semantic understanding of a 3D scene is crucial for successful execution of robotic tasks, especially in unstructured and complex environments. Additionally, to make robust decisions, it is necessary for the robot to evaluate the reliability of perceived information. While recent advances in 3D neural feature fields have enabled robots to leverage features from pretrained foundation models for tasks such as language-guided manipulation and navigation, existing methods suffer from two critical limitations: (i) they are typically scene-specific, and (ii) they lack the ability to model uncertainty in their predictions. We present UniFField, a unified uncertainty-aware neural feature field that combines visual, semantic, and geometric features in a single generalizable representation while also predicting uncertainty in each modality. Our approach, which can be applied zero shot to any new environment, incrementally integrates RGB-D images into our voxel-based feature representation as the robot explores the scene, simultaneously updating uncertainty estimation. We evaluate our uncertainty estimations to accurately describe the model prediction errors in scene reconstruction and semantic feature prediction. Furthermore, we successfully leverage our feature predictions and their respective uncertainty for an active object search task using a mobile manipulator robot, demonstrating the capability for robust decision-making.
comment: ICRA 2026 Project website: https://sites.google.com/view/uniffield
♻ ☆ Time as a Control Dimension in Robot Learning
Temporal awareness plays a central role in intelligent behavior by shaping how actions are paced, coordinated, and adapted to changing goals and environments. In contrast, most robot learning algorithms treat time only as a fixed episode horizon or scheduling constraint. Here we introduce time-aware policy learning, a reinforcement learning framework that treats time as a control dimension of robot behavior. The approach augments policies with two temporal signals, the remaining time and a time ratio that modulates the policy's internal progression of time, allowing a single policy to regulate its execution strategy across temporal regimes. Across diverse manipulation tasks including long-horizon manipulation, granular-media pouring, articulated-object interaction, and multi-agent coordination, the resulting policies adapt their behavior continuously from dynamic execution under tight schedules to stable and deliberate interaction when more time is available. This temporal awareness improves efficiency, robustness under sim-to-real mismatch and disturbances, and controllability through human input without retraining. Treating time as a controllable variable provides a new framework for adaptive and human-aligned robot autonomy.
♻ ☆ POrTAL: Plan-Orchestrated Tree Assembly for Lookahead IROS 26
When tasking robots in partially observable environments, these robots must efficiently and robustly plan to achieve task goals under uncertainty. Although many probabilistic planning algorithms exist for this purpose, these algorithms can be inefficient if executed with the robot's limited computational resources, or may produce policies that take more steps than expected to achieve the goal. We therefore created a new, lightweight, probabilistic planning algorithm, Plan-Orchestrated Tree Assembly for Lookahead (POrTAL), that combines the strengths of two baseline planning algorithms, FF-Replan and POMCP. We demonstrate that POrTAL is an anytime algorithm that generally outperforms these baselines in terms of the final executed plan length given bounded computation time, especially for problems with only moderate levels of uncertainty.
comment: Submitted to IROS 26
♻ ☆ Inference-Time Enhancement of Generative Robot Policies via Predictive World Modeling
We present Generative Predictive Control (GPC), an inference-time method for improving pretrained behavior-cloning policies without retraining. GPC augments a frozen diffusion policy at deployment with an action-conditioned world model trained on expert demonstrations and random exploration rollouts. The world model predicts the consequences of action proposals generated by the diffusion policy and enables lightweight online planning that ranks and refines these proposals through model-based look-ahead. By combining a generative prior with predictive foresight, GPC enables test-time adaptation while keeping the original policy fixed. Across diverse robotic manipulation tasks, including state- and vision-based settings in both simulation and real-world experiments, GPC consistently outperforms standard behavior cloning and compares favorably with other inference-time adaptation baselines.
comment: Acceptance to IEEE Robotics and Automation Letters. Website: https://computationalrobotics.seas.harvard.edu/GPC
♻ ☆ Human-Aware Robot Behaviour in Self-Driving Labs
Self-driving laboratories (SDLs) are rapidly transforming research in chemistry and materials science to accelerate new discoveries. Mobile robot chemists (MRCs) play a pivotal role by autonomously navigating the lab to transport samples, effectively connecting synthesis, analysis, and characterisation equipment. The instruments within an SDL are typically designed or retrofitted to be accessed by both human and robotic chemists, ensuring operational flexibility and integration between manual and automated workflows. In many scenarios, human and robotic chemists may need to use the same equipment simultaneously. Currently, MRCs rely on simple LiDAR-based obstruction detection, which forces the robot to passively wait if a human is present. This lack of situational awareness leads to unnecessary delays and inefficient coordination in time-critical automated workflows in human-robot shared labs. To address this, we present an initial study of an embodied, AI-driven perception method that facilitates proactive human-robot interaction in shared-access scenarios. Our method features a hierarchical human intention prediction model that allows the robot to distinguish between preparatory actions (waiting) and transient interactions (accessing the instrument). Our results demonstrate that the proposed approach enhances efficiency by enabling proactive human-robot interaction, streamlining coordination, and potentially increasing the efficiency of autonomous scientific labs.
♻ ☆ Moving On, Even When You're Broken: Fail-Active Trajectory Generation via Diffusion Policies Conditioned on Embodiment and Task
Robot failure is detrimental and disruptive, often requiring human intervention to recover. Our vision is 'fail-active' operation, allowing robots to safely complete their tasks even when damaged. Focusing on 'actuation failures', we introduce DEFT, a diffusion-based trajectory generator conditioned on the robot's current embodiment and task constraints. DEFT generalizes across failure types, supports constrained and unconstrained motions, and enables task completion under arbitrary failure. We evaluate DEFT in both simulation and real-world scenarios using a 7-DoF robotic arm. DEFT outperforms its baselines over thousands of failure conditions, achieving a 99.5% success rate for unconstrained motions versus RRT's 42.4%, and 46.4% for constrained motions versus differential IK's 30.9%. Furthermore, DEFT demonstrates robust zero-shot generalization by maintaining performance on failure conditions unseen during training. Finally, we perform real-world evaluations on two multi-step tasks, drawer manipulation and whiteboard erasing. These experiments demonstrate DEFT succeeding on tasks where classical methods fail. Our results show that DEFT achieves fail-active manipulation across arbitrary failure configurations and real-world deployments.
comment: To be published in the 2026 IEEE International Conference on Robotics & Automation
♻ ☆ Pixel Motion Diffusion is What We Need for Robot Control CVPR 2026
We present DAWN (Diffusion is All We Need for robot control), a unified diffusion-based framework for language-conditioned robotic manipulation that bridges high-level motion intent and low-level robot action via structured pixel motion representation. In DAWN, both the high-level and low-level controllers are modeled as diffusion processes, yielding a fully trainable, end-to-end system with interpretable intermediate motion abstractions. DAWN achieves state-of-the-art results on the challenging CALVIN benchmark, demonstrating strong multi-task performance, and further validates its effectiveness on MetaWorld. Despite the substantial domain gap between simulation and reality and limited real-world data, we demonstrate reliable real-world transfer with only minimal finetuning, illustrating the practical viability of diffusion-based motion abstractions for robotic control. Our results show the effectiveness of combining diffusion modeling with motion-centric representations as a strong baseline for scalable and robust robot learning. Project page: https://eronguyen.github.io/DAWN/
comment: Accepted to CVPR 2026. Project page: https://eronguyen.github.io/DAWN
Robotics 115
☆ OTPL-VIO: Robust Visual-Inertial Odometry with Optimal Transport Line Association and Adaptive Uncertainty
Robust stereo visual-inertial odometry (VIO) remains challenging in low-texture scenes and under abrupt illumination changes, where point features become sparse and unstable, leading to ambiguous association and under-constrained estimation. Line structures offer complementary geometric cues, yet many efficient point-line systems still rely on point-guided line association, which can break down when point support is weak and may lead to biased constraints. We present a stereo point-line VIO system in which line segments are equipped with dedicated deep descriptors and matched using an entropy-regularized optimal transport formulation, enabling globally consistent correspondences under ambiguity, outliers, and partial observations. The proposed descriptor is training-free and is computed by sampling and pooling network feature maps. To improve estimation stability, we analyze the impact of line measurement noise and introduce reliability-adaptive weighting to regulate the influence of line constraints during optimization. Experiments on EuRoC and UMA-VI, together with real-world deployments in low-texture and illumination-challenging environments, demonstrate improved accuracy and robustness over representative baselines while maintaining real-time performance.
☆ A Generalized Voronoi Graph based Coverage Control Approach for Non-Convex Environment
To address the challenge of efficient coverage by multi-robot systems in non-convex regions with multiple obstacles, this paper proposes a coverage control method based on the Generalized Voronoi Graph (GVG), which has two phases: Load-Balancing Algorithm phase and Collaborative Coverage phase. In Load-Balancing Algorithm phase, the non-convex region is partitioned into multiple sub-regions based on GVG. Besides, a weighted load-balancing algorithm is developed, which considers the quality differences among sub-regions. By iteratively optimizing the robot allocation ratio, the number of robots in each sub-region is matched with the sub-region quality to achieve load balance. In Collaborative Coverage phase, each robot is controlled by a new controller to effectively coverage the region. The convergence of the method is proved and its performance is evaluated through simulations.
comment: 8 pages, 7 figures, published to ACC 2026
☆ Towards Terrain-Aware Safe Locomotion for Quadrupedal Robots Using Proprioceptive Sensing
Achieving safe quadrupedal locomotion in real-world environments has attracted much attention in recent years. When walking over uneven terrain, achieving reliable estimation and realising safety-critical control based on the obtained information is still an open question. To address this challenge, especially for low-cost robots equipped solely with proprioceptive sensors (e.g., IMUs, joint encoders, and contact force sensors), this work first presents an estimation framework that generates a 2.5-D terrain map and extracts support plane parameters, which are then integrated into contact and state estimation. Then, we integrate this estimation framework into a safety-critical control pipeline by formulating control barrier functions that provide rigorous safety guarantees. Experiments demonstrate that the proposed terrain estimation method provides smooth terrain representations. Moreover, the coupled estimation framework of terrain, state, and contact reduces the mean absolute error of base position estimation by 64.8%, decreases the estimation variance by 47.2%, and improves the robustness of contact estimation compared to a decoupled framework. The terrain-informed CBFs integrate historical terrain information and current proprioceptive measurements to ensure global safety by keeping the robot out of hazardous areas and local safety by preventing body-terrain collision, relying solely on proprioceptive sensing.
comment: 8 pages, 10 figures
☆ SCDP: Learning Humanoid Locomotion from Partial Observations via Mixed-Observation Distillation
Distilling humanoid locomotion control from offline datasets into deployable policies remains a challenge, as existing methods rely on privileged full-body states that require complex and often unreliable state estimation. We present Sensor-Conditioned Diffusion Policies (SCDP) that enables humanoid locomotion using only onboard sensors, eliminating the need for explicit state estimation. SCDP decouples sensing from supervision through mixed-observation training: diffusion model conditions on sensor histories while being supervised to predict privileged future state-action trajectories, enforcing the model to infer the motion dynamics under partial observability. We further develop restricted denoising, context distribution alignment, and context-aware attention masking to encourage implicit state estimation within the model and to prevent train-deploy mismatch. We validate SCDP on velocity-commanded locomotion and motion reference tracking tasks. In simulation, SCDP achieves near-perfect success on velocity control (99-100%) and 93% tracking success in AMASS test set, performing comparable to privileged baselines while using only onboard sensors. Finally, we deploy the trained policy on a real G1 humanoid at 50 Hz, demonstrating robust real robot locomotion without external sensing or state estimation.
comment: 6 pages, 8 figures, 5 tables, iRos
☆ ReTac-ACT: A State-Gated Vision-Tactile Fusion Transformer for Precision Assembly
Precision assembly requires sub-millimeter corrections in contact-rich "last-millimeter" regions where visual feedback fails due to occlusion from the end-effector and workpiece. We present ReTac-ACT (Reconstruction-enhanced Tactile ACT), a vision-tactile imitation learning policy that addresses this challenge through three synergistic mechanisms: (i) bidirectional cross-attention enabling reciprocal visuo-tactile feature enhancement before fusion, (ii) a proprioception-conditioned gating network that dynamically elevates tactile reliance when visual occlusion occurs, and (iii) a tactile reconstruction objective enforcing learning of manipulation-relevant contact information rather than generic visual textures. Evaluated on the standardized NIST Assembly Task Board M1 benchmark, ReTac-ACT achieves 90% peg-in-hole success, substantially outperforming vision-only and generalist baseline methods, and maintains 80% success at industrial-grade 0.1mm clearance. Ablation studies validate that each architectural component is indispensable. The ReTac-ACT codebase and a vision-tactile demonstration dataset covering various clearance levels with both visual and tactile features will be released to support reproducible research.
☆ Trajectory Optimization for Self-Wrap-Aware Cable-Towed Planar Object Manipulation under Implicit Tension Constraints
Cable/rope elements are pervasive in deformable-object manipulation, often serving as a deformable force-transmission medium whose routing and contact determine how wrenches are delivered. In cable-towed manipulation, transmission is unilateral and hybrid: the tether can pull only when taut and becomes force-free when slack; in practice, the tether may also contact the object boundary and self-wrap around edges, which is not merely collision avoidance but a change of the wrench transmission channel by shifting the effective application point and moment arm, thereby coupling routing geometry with rigid-body motion and tensioning. We formulate self-wrap towing as a routing-aware, tensioning-implicit trajectory optimization (TITO) problem that couples (i) a tensioning-implicit taut/slack constraint and (ii) routing-conditioned transmission maps for effective length and wrench, and we build a relaxation hierarchy from a strict mode-conditioned reference to three tractable relaxations: Full-Mode Relaxation (FMR), Binary-Mode Relaxation (BMR), and Implicit-Mode Relaxation (IMR). Across planar towing tasks, we find that making routing an explicit decision often yields conservative solutions that stay near switching boundaries, whereas IMR induces self-wrap through state evolution and exploits the redirected torque channel whenever turning requires it.
☆ On the Cost of Evolving Task Specialization in Multi-Robot Systems
Task specialization can lead to simpler robot behaviors and higher efficiency in multi-robot systems. Previous works have shown the emergence of task specialization during evolutionary optimization, focusing on feasibility rather than costs. In this study, we take first steps toward a cost-benefit analysis of task specialization in robot swarms using a foraging scenario. We evolve artificial neural networks as generalist behaviors for the entire task and as task-specialist behaviors for subtasks within a limited evaluation budget. We show that generalist behaviors can be successfully optimized while the evolved task-specialist controllers fail to cooperate efficiently, resulting in worse performance than the generalists. Consequently, task specialization does not necessarily improve efficiency when optimization budget is limited.
comment: Accepted for publication in the proceeding of ANTS 2026 - 15th International Conference on Swarm Intelligence
☆ NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models
Vision-Language-Action (VLA) models are formulated to ground instructions in visual context and generate action sequences for robotic manipulation. Despite recent progress, VLA models still face challenges in learning related and reusable primitives, reducing reliance on large-scale data and complex architectures, and enabling exploration beyond demonstrations. To address these challenges, we propose a novel Neuro-Symbolic Vision-Language-Action (NS-VLA) framework via online reinforcement learning (RL). It introduces a symbolic encoder to embedding vision and language features and extract structured primitives, utilizes a symbolic solver for data-efficient action sequencing, and leverages online RL to optimize generation via expansive exploration. Experiments on robotic manipulation benchmarks demonstrate that NS-VLA outperforms previous methods in both one-shot training and data-perturbed settings, while simultaneously exhibiting superior zero-shot generalizability, high data efficiency and expanded exploration space. Our code is available.
☆ Beyond Short-Horizon: VQ-Memory for Robust Long-Horizon Manipulation in Non-Markovian Simulation Benchmarks
The high cost of collecting real-robot data has made robotic simulation a scalable platform for both evaluation and data generation. Yet most existing benchmarks concentrate on simple manipulation tasks such as pick-and-place, failing to capture the non-Markovian characteristics of real-world tasks and the complexity of articulated object interactions. To address this limitation, we present RuleSafe, a new articulated manipulation benchmark built upon a scalable LLM-aided simulation framework. RuleSafe features safes with diverse unlocking mechanisms, such as key locks, password locks, and logic locks, which require different multi-stage reasoning and manipulation strategies. These LLM-generated rules produce non-Markovian and long-horizon tasks that require temporal modeling and memory-based reasoning. We further propose VQ-Memory, a compact and structured temporal representation that uses vector-quantized variational autoencoders (VQ-VAEs) to encode past proprioceptive states into discrete latent tokens. This representation filters low-level noise while preserving high-level task-phase context, providing lightweight yet robust temporal cues that are compatible with existing Vision-Language-Action models (VLA). Extensive experiments on state-of-the-art VLA models and diffusion policies show that VQ-Memory consistently improves long-horizon planning, enhances generalization to unseen configurations, and enables more efficient manipulation with reduced computational cost. Project page: vqmemory.github.io
comment: 9 pages
☆ Context-Nav: Context-Driven Exploration and Viewpoint-Aware 3D Spatial Reasoning for Instance Navigation CVPR 2026
Text-goal instance navigation (TGIN) asks an agent to resolve a single, free-form description into actions that reach the correct object instance among same-category distractors. We present \textit{Context-Nav} that elevates long, contextual captions from a local matching cue to a global exploration prior and verifies candidates through 3D spatial reasoning. First, we compute dense text-image alignments for a value map that ranks frontiers -- guiding exploration toward regions consistent with the entire description rather than early detections. Second, upon observing a candidate, we perform a viewpoint-aware relation check: the agent samples plausible observer poses, aligns local frames, and accepts a target only if the spatial relations can be satisfied from at least one viewpoint. The pipeline requires no task-specific training or fine-tuning; we attain state-of-the-art performance on InstanceNav and CoIN-Bench. Ablations show that (i) encoding full captions into the value map avoids wasted motion and (ii) explicit, viewpoint-aware 3D verification prevents semantically plausible but incorrect stops. This suggests that geometry-grounded spatial reasoning is a scalable alternative to heavy policy training or human-in-the-loop interaction for fine-grained instance disambiguation in cluttered 3D scenes.
comment: Camera-ready version. Accepted to CVPR 2026
☆ StyleVLA: Driving Style-Aware Vision Language Action Model for Autonomous Driving
Vision Language Models (VLMs) bridge visual perception and linguistic reasoning. In Autonomous Driving (AD), this synergy has enabled Vision Language Action (VLA) models, which translate high-level multimodal understanding into driving behaviors, typically represented as future trajectories. However, existing VLA models mainly generate generic collision-free trajectories. Beyond collision avoidance, adapting to diverse driving styles (e.g., sporty, comfortable) is essential for personalized driving. Moreover, many methods treat trajectory generation as naive token prediction, which can produce kinematically infeasible actions. To address these limitations, we present StyleVLA, a physics-informed VLA framework for generating diverse and physically plausible driving behaviors. We introduce a hybrid loss that combines a kinematic consistency constraint with a continuous regression head to improve trajectory feasibility. To train StyleVLA, built on Qwen3-VL-4B, we construct a large-scale instruction dataset with over 1.2k scenarios, 76k Bird's Eye View (BEV) samples, and 42k First Person View (FPV) samples, with ground-truth trajectories for five driving styles and natural-language instructions. Experiments show that our 4B-parameter StyleVLA significantly outperforms proprietary models (e.g., Gemini-3-Pro) and state-of-the-art VLA models. Using a composite driving score measuring success rate, physical feasibility, and style adherence, StyleVLA achieves 0.55 on BEV and 0.51 on FPV, versus 0.32 and 0.35 for Gemini-3-Pro. These results show that a specialized, physics-informed, lightweight model can surpass closed-source models on domain-specific tasks.
comment: 8 pages
☆ Receptogenesis in a Vascularized Robotic Embodiment
Equipping robotic systems with the capacity to generate $\textit{ex novo}$ hardware during operation extends control of physical adaptability. Unlike modular systems that rely on discrete component integration pre- or post-deployment, we envision the possibility that physical adaptation and development emerge from dynamic material restructuring to shape the body's intrinsic functions. Drawing inspiration from circulatory systems that redistribute mass and function in biological organisms, we utilize fluidics to restructure the material interface, a capability currently unpaired in robotics. Here, we realize this synthetic growth capability through a vascularized robotic composite designed for programmable material synthesis, demonstrated via receptogenesis - the on-demand construction of sensors from internal fluid reserves based on environmental cues. By coordinating the fluidic transport of precursors with external localized UV irradiation, we drive an $\textit{in situ}$ photopolymerization that chemically reconstructs the vasculature from the inside out. This reaction converts precursors with photolatent initiator into a solid dispersion of UV-sensitive polypyrrole, establishing a sensing modality validated by a characteristic decrease in electrical impedance. The newly synthesized sensor closed a control loop to regulate wing flapping in a moth-inspired robotic demonstrator. This physical update increased the robot's capability in real time. This work establishes a materials-based framework for constitutive evolution, enabling robots to physically grow the hardware needed to support emerging behaviors in a complex environment; for example, suggesting a pathway toward autonomous systems capable of generating specialized features, such as neurovascular systems in situated robotics.
comment: Supplementary Files currently unavailable online. Please contact the First Author to request any Supplementary Files
☆ SEA-Nav: Efficient Policy Learning for Safe and Agile Quadruped Navigation in Cluttered Environments
Efficiently training quadruped robot navigation in densely cluttered environments remains a significant challenge. Existing methods are either limited by a lack of safety and agility in simple obstacle distributions or suffer from slow locomotion in complex environments, often requiring excessively long training phases. To this end, we propose SEA-Nav (Safe, Efficient, and Agile Navigation), a reinforcement learning framework for quadruped navigation. Within diverse and dense obstacle environments, a differentiable control barrier function (CBF)-based shield constraints the navigation policy to output safe velocity commands. An adaptive collision replay mechanism and hazardous exploration rewards are introduced to increase the probability of learning from critical experiences, guiding efficient exploration and exploitation. Finally, kinematic action constraints are incorporated to ensure safe velocity commands, facilitating successful physical deployment. To the best of our knowledge, this is the first approach that achieves highly challenging quadruped navigation in the real world with minute-level training time.
comment: Project website: https://11chens.github.io/sea-nav/
☆ Stein Variational Ergodic Surface Coverage with SE(3) Constraints
Surface manipulation tasks require robots to generate trajectories that comprehensively cover complex 3D surfaces while maintaining precise end-effector poses. Existing ergodic trajectory optimization (TO) methods demonstrate success in coverage tasks, while struggling with point-cloud targets due to the nonconvex optimization landscapes and the inadequate handling of SE(3) constraints in sampling-as-optimization (SAO) techniques. In this work, we introduce a preconditioned SE(3) Stein Variational Gradient Descent (SVGD) approach for SAO ergodic trajectory generation. Our proposed approach comprises multiple innovations. First, we reformulate point-cloud ergodic coverage as a manifold-aware sampling problem. Second, we derive SE(3)-specific SVGD particle updates, and, third, we develop a preconditioner to accelerate TO convergence. Our sampling-based framework consistently identifies superior local optima compared to strong optimization-based and SAO baselines while preserving the SE(3) geometric structure. Experiments on a 3D point-cloud surface coverage benchmark and robotic surface drawing tasks demonstrate that our method achieves superior coverage quality with tractable computation in our setting relative to existing TO and SAO approaches, and is validated in real-world robot experiments.
☆ Open-World Motion Forecasting
Motion forecasting aims to predict the future trajectories of dynamic agents in the scene, enabling autonomous vehicles to effectively reason about scene evolution. Existing approaches operate under the closed-world regime and assume fixed object taxonomy as well as access to high-quality perception. Therefore, they struggle in real-world settings where perception is imperfect and object taxonomy evolves over time. In this work, we bridge this fundamental gap by introducing open-world motion forecasting, a novel setting in which new object classes are sequentially introduced over time and future object trajectories are estimated directly from camera images. We tackle this setting by proposing the first end-to-end class-incremental motion forecasting framework to mitigate catastrophic forgetting while simultaneously learning to forecast newly introduced classes. When a new class is introduced, our framework employs a pseudo-labeling strategy to first generate motion forecasting pseudo-labels for all known classes which are then processed by a vision-language model to filter inconsistent and over-confident predictions. Parallelly, our approach further mitigates catastrophic forgetting by using a novel replay sampling strategy that leverages query feature variance to sample previous sequences with informative motion patterns. Extensive evaluation on the nuScenes and Argoverse 2 datasets demonstrates that our approach successfully resists catastrophic forgetting and maintains performance on previously learned classes while improving adaptation to novel ones. Further, we demonstrate that our approach supports zero-shot transfer to real-world driving and naturally extends to end-to-end class-incremental planning, enabling continual adaptation of the full autonomous driving system. We provide the code at https://omen.cs.uni-freiburg.de .
☆ From Flow to One Step: Real-Time Multi-Modal Trajectory Policies via Implicit Maximum Likelihood Estimation-based Distribution Distillation
Generative policies based on diffusion and flow matching achieve strong performance in robotic manipulation by modeling multi-modal human demonstrations. However, their reliance on iterative Ordinary Differential Equation (ODE) integration introduces substantial latency, limiting high-frequency closed-loop control. Recent single-step acceleration methods alleviate this overhead but often exhibit distributional collapse, producing averaged trajectories that fail to execute coherent manipulation strategies. We propose a framework that distills a Conditional Flow Matching (CFM) expert into a fast single-step student via Implicit Maximum Likelihood Estimation (IMLE). A bi-directional Chamfer distance provides a set-level objective that promotes both mode coverage and fidelity, enabling preservation of the teacher multi-modal action distribution in a single forward pass. A unified perception encoder further integrates multi-view RGB, depth, point clouds, and proprioception into a geometry-aware representation. The resulting high-frequency control supports real-time receding-horizon re-planning and improved robustness under dynamic disturbances.
comment: https://sites.google.com/view/flow2one, 8 pages
☆ Vision-Augmented On-Track System Identification for Autonomous Racing via Attention-Based Priors and Iterative Neural Correction
Operating autonomous vehicles at the absolute limits of handling requires precise, real-time identification of highly non-linear tire dynamics. However, traditional online optimization methods suffer from "cold-start" initialization failures and struggle to model high-frequency transient dynamics. To address these bottlenecks, this paper proposes a novel vision-augmented, iterative system identification framework. First, a lightweight CNN (MobileNetV3) translates visual road textures into a continuous heuristic friction prior, providing a robust "warm-start" for parameter optimization. Next, a S4 model captures complex temporal dynamic residuals, circumventing the memory and latency limitations of traditional MLPs and RNNs. Finally, a derivative-free Nelder-Mead algorithm iteratively extracts physically interpretable Pacejka tire parameters via a hybrid virtual simulation. Co-simulation in CarSim demonstrates that the lightweight vision backbone reduces friction estimation error by 76.1 using 85 fewer FLOPs, accelerating cold-start convergence by 71.4. Furthermore, the S4-augmented framework improves parameter extraction accuracy and decreases lateral force RMSE by over 60 by effectively capturing complex vehicle dynamics, demonstrating superior performance compared to conventional neural architectures.
☆ SPAARS: Safer RL Policy Alignment through Abstract Exploration and Refined Exploitation of Action Space
Offline-to-online reinforcement learning (RL) offers a promising paradigm for robotics by pre-training policies on safe, offline demonstrations and fine-tuning them via online interaction. However, a fundamental challenge remains: how to safely explore online without deviating from the behavioral support of the offline data? While recent methods leverage conditional variational autoencoders (CVAEs) to bound exploration within a latent space, they inherently suffer from an exploitation gap -- a performance ceiling imposed by the decoder's reconstruction loss. We introduce SPAARS, a curriculum learning framework that initially constrains exploration to the low-dimensional latent manifold for sample-efficient, safe behavioral improvement, then seamlessly transfers control to the raw action space, bypassing the decoder bottleneck. SPAARS has two instantiations: the CVAE-based variant requires only unordered (s,a) pairs and no trajectory segmentation; SPAARS-SUPE pairs SPAARS with OPAL temporal skill pretraining for stronger exploration structure at the cost of requiring trajectory chunks. We prove an upper bound on the exploitation gap using the Performance Difference Lemma, establish that latent-space policy gradients achieve provable variance reduction over raw-space exploration, and show that concurrent behavioral cloning during the latent phase directly controls curriculum transition stability. Empirically, SPAARS-SUPE achieves 0.825 normalized return on kitchen-mixed-v0 versus 0.75 for SUPE, with 5x better sample efficiency; standalone SPAARS achieves 92.7 and 102.9 normalized return on hopper-medium-v2 and walker2d-medium-v2 respectively, surpassing IQL baselines of 66.3 and 78.3 respectively, confirming the utility of the unordered-pair CVAE instantiation.
comment: 9 pages
☆ NLiPsCalib: An Efficient Calibration Framework for High-Fidelity 3D Reconstruction of Curved Visuotactile Sensors ICRA 2026
Recent advances in visuotactile sensors increasingly employ biomimetic curved surfaces to enhance sensorimotor capabilities. Although such curved visuotactile sensors enable more conformal object contact, their perceptual quality is often degraded by non-uniform illumination, which reduces reconstruction accuracy and typically necessitates calibration. Existing calibration methods commonly rely on customized indenters and specialized devices to collect large-scale photometric data, but these processes are expensive and labor-intensive. To overcome these calibration challenges, we present NLiPsCalib, a physics-consistent and efficient calibration framework for curved visuotactile sensors. NLiPsCalib integrates controllable near-field light sources and leverages Near-Light Photometric Stereo (NLiPs) to estimate contact geometry, simplifying calibration to just a few simple contacts with everyday objects. We further introduce NLiPsTac, a controllable-light-source tactile sensor developed to validate our framework. Experimental results demonstrate that our approach enables high-fidelity 3D reconstruction across diverse curved form factors with a simple calibration procedure. We emphasize that our approach lowers the barrier to developing customized visuotactile sensors of diverse geometries, thereby making visuotactile sensing more accessible to the broader community.
comment: 8 pages, 8 figures, accepted to 2026 IEEE International Conference on Robotics & Automation (ICRA 2026)
☆ CORAL: Scalable Multi-Task Robot Learning via LoRA Experts
Deploying Vision-Language-Action (VLA) models in real-world robotics exposes a core multi-task learning challenge: reconciling task interference in multi-task robotic learning. When multiple tasks are jointly fine-tuned in a single stage, gradients from different tasks can conflict, causing negative transfer and reducing per-task performance. Yet maintaining a separate full checkpoint per task is often storage- and deployment-prohibitive. To address this dilemma, we present CORAL, a backbone- and embodiment-agnostic framework designed primarily to mitigate multi-task interference while remaining naturally extensible to a continuous stream of new tasks. CORAL freezes a single pre-trained VLA backbone and attaches one lightweight Low-Rank Adaptation (LoRA) expert per task; at runtime, a dynamic inference engine (the CORAL Manager) routes language instructions to the appropriate expert and swaps experts on the fly with zero inference overhead. This strict parameter isolation avoids complex gating networks and prevents parameter-level cross-task interference by construction; as an added capability, it also enables sequentially introducing new tasks without parameter overwriting caused by catastrophic forgetting. We validate CORAL on a real-world Galaxea R1 dual-arm mobile manipulator and three simulation benchmarks (LIBERO, WidowX, Google Robot), where CORAL overcomes fine-grained instructional ambiguity and substantially outperforms joint training, yielding a practical and scalable system for lifelong multi-task robot learning. Website: https://frontierrobo.github.io/CORAL
☆ See, Plan, Rewind: Progress-Aware Vision-Language-Action Models for Robust Robotic Manipulation CVPR
Measurement of task progress through explicit, actionable milestones is critical for robust robotic manipulation. This progress awareness enables a model to ground its current task status, anticipate verifiable intermediate states, and detect and recover from failures when progress stalls. To embody this capability, we introduce See, Plan, Rewind (SPR), a progress-aware vision-language-action framework that dynamically grounds language instructions into a sequence of spatial subgoals. SPR operates through a continuous core cycle, Seeing the current state and upcoming milestone, Planning a trajectory towards the next 2D waypoint, and Rewinding to a recoverable state upon failure by monitoring progress against the expected sequence. This closed-loop approach enables robust error correction without requiring additional training data or auxiliary models. Extensive experiments demonstrate the framework's effectiveness, generalization and robustness: SPR outperforms the MolmoAct baseline by 5\% on the LIBERO benchmark. On the challenging LIBERO-Plus benchmark with unseen instructions and initial states, SPR achieves state-of-the-art robustness with the smallest performance drop, surpassing OpenVLA-OFT and UniVLA, demonstrating superior out-of-distribution robustness.
comment: Suggested to CVPR Findings. https://tingjundai.github.io/SPRVLA/
☆ Implicit Geometry Representations for Vision-and-Language Navigation from Web Videos CVPR 2025
Vision-and-Language Navigation (VLN) has long been constrained by the limited diversity and scalability of simulator-curated datasets, which fail to capture the complexity of real-world environments. To overcome this limitation, we introduce a large-scale video-instruction framework derived from web-based room tour videos, enabling agents to learn from natural human walking demonstrations in diverse, realistic indoor settings. Unlike existing datasets, our framework integrates both open-ended description-enriched trajectories and action-enriched trajectories reconstructed in 3D, providing richer spatial and semantic supervision. A key extension in this work is the incorporation of implicit geometry representations, which extract spatial cues directly from RGB frames without requiring fragile 3D reconstruction. This approach substantially improves data utilization, alleviates reconstruction failures, and unlocks large portions of previously unusable video data. Comprehensive experiments across multiple VLN benchmarks (CVDN, SOON, R2R, and REVERIE) demonstrate that our method not only sets new state-of-the-art performance but also enables the development of robust zero-shot navigation agents. By bridging large-scale web videos with implicit spatial reasoning, this work advances embodied navigation towards more scalable, generalizable, and real-world applicable solutions.
comment: Extension of CVPR 2025 RoomTour3D with implicit geometric representations
☆ RAE-NWM: Navigation World Model in Dense Visual Representation Space
Visual navigation requires agents to reach goals in complex environments through perception and planning. World models address this task by simulating action-conditioned state transitions to predict future observations. Current navigation world models typically learn state evolution under actions within the compressed latent space of a Variational Autoencoder, where spatial compression often discards fine-grained structural information and hinders precise control. To better understand the propagation characteristics of different representations, we conduct a linear dynamics probe and observe that dense DINOv2 features exhibit stronger linear predictability for action-conditioned transitions. Motivated by this observation, we propose the Representation Autoencoder-based Navigation World Model (RAE-NWM), which models navigation dynamics in a dense visual representation space. We employ a Conditional Diffusion Transformer with Decoupled Diffusion Transformer head (CDiT-DH) to model continuous transitions, and introduce a separate time-driven gating module for dynamics conditioning to regulate action injection strength during generation. Extensive evaluations show that modeling sequential rollouts in this space improves structural stability and action accuracy, benefiting downstream planning and navigation.
comment: Code is available at: https://github.com/20robo/raenwm
☆ MO-Playground: Massively Parallelized Multi-Objective Reinforcement Learning for Robotics
Multi-objective reinforcement learning (MORL) is a powerful tool to learn Pareto-optimal policy families across conflicting objectives. However, unlike traditional RL algorithms, existing MORL algorithms do not effectively leverage large-scale parallelization to concurrently simulate thousands of environments, resulting in vastly increased computation time. Ultimately, this has limited MORL's application towards complex multi-objective robotics problems. To address these challenges, we present 1) MORLAX, a new GPU-native, fast MORL algorithm, and 2) MO-Playground, a pip-installable playground of GPU-accelerated multi-objective environments. Together, MORLAX and MO-Playground approximate Pareto sets within minutes, offering 25-270x speed-ups compared to legacy CPU-based approaches whilst achieving superior Pareto front hypervolumes. We demonstrate the versatility of our approach by implementing a custom BRUCE humanoid robot environment using MO-Playground and learning Pareto-optimal locomotion policies across 6 realistic objectives for BRUCE, such as smoothness, efficiency and arm swinging.
comment: 8 pages, 4 figures, 3 tables
☆ TRIP-Bag: A Portable Teleoperation System for Plug-and-Play Robotic Arms and Leaders
Large scale, diverse demonstration data for manipulation tasks remains a major challenge in learning-based robot policies. Existing in-the-wild data collection approaches often rely on vision-based pose estimation of hand-held grippers or gloves, which introduces an embodiment gap between the collection platform and the target robot. Teleoperation systems eliminate the embodiment gap, but are typically impractical to deploy outside the laboratory environment. We propose TRIP-Bag (Teleoperation, Recording, Intelligence in a Portable Bag), a portable, puppeteer-style teleoperation system fully contained within a commercial suitcase, as a practical solution for collecting high-fidelity manipulation data across varied settings. With a setup time of under five minutes and direct joint-to-joint teleoperation, TRIP-Bag enables rapid and reliable data collection in any environment. We validated TRIP-Bag's usability through experiments with non-expert users, showing that the system is intuitive and easy to operate. Furthermore, we confirmed the quality of the collected data by training benchmark manipulation policies, demonstrating its value as a practical resource for robot learning.
☆ Embodied Human Simulation for Quantitative Design and Analysis of Interactive Robotics
Physical interactive robotics, ranging from wearable devices to collaborative humanoid robots, require close coordination between mechanical design and control. However, evaluating interactive dynamics is challenging due to complex human biomechanics and motor responses. Traditional experiments rely on indirect metrics without measuring human internal states, such as muscle forces or joint loads. To address this issue, we develop a scalable simulation-based framework for the quantitative analysis of physical human-robot interaction. At its core is a full-body musculoskeletal model serving as a predictive surrogate for the human dynamical system. Driven by a reinforcement learning controller, it generates adaptive, physiologically grounded motor behaviors. We employ a sequential training pipeline where the pre-trained human motion control policy acts as a consistent evaluator, making large-scale design space exploration computationally tractable. By simulating the coupled human-robot system, the framework provides access to internal biomechanical metrics, offering a systematic way to concurrently co-optimize a robot's structural parameters and control policy. We demonstrate its capability in optimizing human-exoskeleton interactions, showing improved joint alignment and reduced contact forces. This work establishes embodied human simulation as a scalable paradigm for interactive robotics design.
☆ WESPR: Wind-adaptive Energy-Efficient Safe Perception & Planning for Robust Flight with Quadrotors
Local wind conditions strongly influence drone performance: headwinds increase flight time, crosswinds and wind shear hinder agility in cluttered spaces, while tailwinds reduce travel time. Although adaptive controllers can mitigate turbulence, they remain unaware of the surrounding geometry that generates it, preventing proactive avoidance. Existing methods that model how wind interacts with the environment typically rely on computationally expensive fluid dynamics simulations, limiting real-time adaptation to new environments and conditions. To bridge this gap, we present WESPR, a fast framework that predicts how environmental geometry affects local wind conditions, enabling proactive path planning and control adaptation. Our lightweight pipeline integrates geometric perception and local weather data to estimate wind fields, compute cost-efficient paths, and adjust control strategies-all within 10 seconds. We validate WESPR on a Crazyflie drone navigating turbulent obstacle courses. Our results show a 12.5-58.7% reduction in maximum trajectory deviation and a 24.6% improvement in stability compared to a wind-agnostic adaptive controller.
comment: 8 pages, 9 Figures
☆ Robust Spatiotemporal Motion Planning for Multi-Agent Autonomous Racing via Topological Gap Identification and Accelerated MPC
High-speed multi-agent autonomous racing demands robust spatiotemporal planning and precise control under strict computational limits. Current methods often oversimplify interactions or abandon strict kinematic constraints. We resolve this by proposing a Topological Gap Identification and Accelerated MPC framework. By predicting opponent behaviors via SGPs, our method constructs dynamic occupancy corridors to robustly select optimal overtaking gaps. We ensure strict kinematic feasibility using a Linear Time-Varying MPC powered by a customized Pseudo-Transient Continuation (PTC) solver for high-frequency execution. Experimental results on the F1TENTH platform show that our method significantly outperforms state-of-the-art baselines: it reduces total maneuver time by 51.6% in sequential scenarios, consistently maintains an overtaking success rate exceeding 81% in dense bottlenecks, and lowers average computational latency by 20.3%, pushing the boundaries of safe and high-speed autonomous racing.
☆ STONE Dataset: A Scalable Multi-Modal Surround-View 3D Traversability Dataset for Off-Road Robot Navigation
Reliable off-road navigation requires accurate estimation of traversable regions and robust perception under diverse terrain and sensing conditions. However, existing datasets lack both scalability and multi-modality, which limits progress in 3D traversability prediction. In this work, we introduce STONE, a large-scale multi-modal dataset for off-road navigation. STONE provides (1) trajectory-guided 3D traversability maps generated by a fully automated, annotation-free pipeline, and (2) comprehensive surround-view sensing with synchronized 128-channel LiDAR, six RGB cameras, and three 4D imaging radars. The dataset covers a wide range of environments and conditions, including day and night, grasslands, farmlands, construction sites, and lakes. Our auto-labeling pipeline reconstructs dense terrain surfaces from LiDAR scans, extracts geometric attributes such as slope, elevation, and roughness, and assigns traversability labels beyond the robot's trajectory using a Mahalanobis-distance-based criterion. This design enables scalable, geometry-aware ground-truth construction without manual annotation. Finally, we establish a benchmark for voxel-level 3D traversability prediction and provide strong baselines under both single-modal and multi-modal settings. STONE is available at: https://konyul.github.io/STONE-dataset/
☆ ZeroWBC: Learning Natural Visuomotor Humanoid Control Directly from Human Egocentric Video
Achieving versatile and naturalistic whole-body control for humanoid robot scene-interaction remains a significant challenge. While some recent works have demonstrated autonomous humanoid interactive control, they are constrained to rigid locomotion patterns and expensive teleoperation data collection, lacking the versatility to execute more human-like natural behaviors such as sitting or kicking. Furthermore, acquiring the necessary real robot teleoperation data is prohibitively expensive and time-consuming. To address these limitations, we introduce ZeroWBC, a novel framework that learns a natural humanoid visuomotor control policy directly from human egocentric videos, eliminating the need for large-scale robot teleoperation data and enabling natural humanoid robot scene-interaction control. Specifically, our approach first fine-tunes a Vision-Language Model (VLM) to predict future whole-body human motions based on text instructions and egocentric visual context, then these generated motions are retargeted to real robot joints and executed via our robust general motion tracking policy for humanoid whole-body control. Extensive experiments on the Unitree G1 humanoid robot demonstrate that our method outperforms baseline approaches in motion naturalness and versatility, successfully establishing a pipeline that eliminates teleoperation data collection overhead for whole-body humanoid control, offering a scalable and efficient paradigm for general humanoid whole-body control.
☆ SPAN-Nav: Generalized Spatial Awareness for Versatile Vision-Language Navigation
Recent embodied navigation approaches leveraging Vision-Language Models (VLMs) demonstrate strong generalization in versatile Vision-Language Navigation (VLN). However, reliable path planning in complex environments remains challenging due to insufficient spatial awareness. In this work, we introduce SPAN-Nav, an end-to-end foundation model designed to infuse embodied navigation with universal 3D spatial awareness using RGB video streams. SPAN-Nav extracts spatial priors across diverse scenes through an occupancy prediction task on extensive indoor and outdoor environments. To mitigate the computational burden, we introduce a compact representation for spatial priors, finding that a single token is sufficient to encapsulate the coarse-grained cues essential for navigation tasks. Furthermore, inspired by the Chain-of-Thought (CoT) mechanism, SPAN-Nav utilizes this single spatial token to explicitly inject spatial cues into action reasoning through an end-to end framework. Leveraging multi-task co-training, SPAN-Nav captures task-adaptive cues from generalized spatial priors, enabling robust spatial awareness to generalize even to the task lacking explicit spatial supervision. To support comprehensive spatial learning, we present a massive dataset of 4.2 million occupancy annotations that covers both indoor and outdoor scenes across multi-type navigation tasks. SPAN-Nav achieves state-of-the-art performance across three benchmarks spanning diverse scenarios and varied navigation tasks. Finally, real-world experiments validate the robust generalization and practical reliability of our approach across complex physical scenarios.
☆ Walking on Rough Terrain with Any Number of Legs
Robotics would gain by replicating the remarkable agility of arthropods in navigating complex environments. Here we consider the control of multi-legged systems which have 6 or more legs. Current multi-legged control strategies in robots include large black-box machine learning models, Central Pattern Generator (CPG) networks, and open-loop feed-forward control with stability arising from mechanics. Here we present a multi-legged control architecture for rough terrain using a segmental robot with 3 actuators for every 2 legs, which we validated in simulation for robots with 6 to 16 legs. Segments have identical state machines, and each segment also receives input from the segment in front of it. Our design bridges the gap between WalkNet-like event cascade controllers and CPG-based controllers: it tightly couples to the ground when contact is present, but produces fictive locomotion when ground contact is missing. The approach may be useful as an adaptive and computationally lightweight controller for multi-legged robots, and as a baseline capability for scaffolding the learning of machine learning controllers.
comment: 10 pages, 6 figures
☆ DexHiL: A Human-in-the-Loop Framework for Vision-Language-Action Model Post-Training in Dexterous Manipulation
While Vision-Language-Action (VLA) models have demonstrated promising generalization capabilities in robotic manipulation, deploying them on specific and complex downstream tasks still demands effective post-training. In parallel, Human-in-the-Loop (HiL) learning has proven to be a powerful mechanism for refining robot policies. However, extending this paradigm to dexterous manipulation remains challenging: multi-finger control is high-dimensional, contact-intensive, and exhibits execution distributions that differ markedly from standard arm motions, leaving existing dexterous VLA systems limited in reliability and adaptability. We present DexHiL, the first integrated arm-hand human-in-the-loop framework for dexterous VLA models, enabling coordinated interventions over the arm and the dexterous hand within a single system. DexHiL introduces an intervention-aware data sampling strategy that prioritizes corrective segments for post-training, alongside a lightweight teleoperation interface that supports instantaneous human corrections during execution. Real-robot experiments demonstrate that DexHiL serves as an effective post-training framework, yielding a substantial performance leap, outperforming standard offline-only fine-tuning baselines by an average of 25% in success rates across distinct tasks. Project page: https://chenzhongxi-sjtu.github.io/dexhil/
comment: 9 pages, 5 figures
☆ PM-Nav: Priori-Map Guided Embodied Navigation in Functional Buildings
Existing language-driven embodied navigation paradigms face challenges in functional buildings (FBs) with highly similar features, as they lack the ability to effectively utilize priori spatial knowledge. To tackle this issue, we propose a Priori-Map Guided Embodied Navigation (PM-Nav), wherein environmental maps are transformed into navigation-friendly semantic priori-maps, a hierarchical chain-of-thought prompt template with an annotation priori-map is designed to enable precise path planning, and a multi-model collaborative action output mechanism is built to accomplish positioning decisions and execution control for navigation planning. Comprehensive tests using a home-made FB dataset show that the PM-Nav obtains average improvements of 511\% and 1175\%, and 650\% and 400\% over the SG-Nav and the InstructNav in simulation and real-world, respectively. These tremendous boosts elucidate the great potential of using the PM-Nav as a backbone navigation framework for FBs.
comment: 6 pages, 4 figures
☆ Latent World Models for Automated Driving: A Unified Taxonomy, Evaluation Framework, and Open Challenges
Emerging generative world models and vision-language-action (VLA) systems are rapidly reshaping automated driving by enabling scalable simulation, long-horizon forecasting, and capability-rich decision making. Across these directions, latent representations serve as the central computational substrate: they compress high-dimensional multi-sensor observations, enable temporally coherent rollouts, and provide interfaces for planning, reasoning, and controllable generation. This paper proposes a unifying latent-space framework that synthesizes recent progress in world models for automated driving. The framework organizes the design space by the target and form of latent representations (latent worlds, latent actions, latent generators; continuous states, discrete tokens, and hybrids) and by structural priors for geometry, topology, and semantics. Building on this taxonomy, the paper articulates five cross-cutting internal mechanics (i.e, structural isomorphism, long-horizon temporal stability, semantic and reasoning alignment, value-aligned objectives and post-training, as well as adaptive computation and deliberation) and connects these design choices to robustness, generalization, and deployability. The work also proposes concrete evaluation prescriptions, including a closed-loop metric suite and a resource-aware deliberation cost, designed to reduce the open-loop / closed-loop mismatch. Finally, the paper identifies actionable research directions toward advancing latent world model for decision-ready, verifiable, and resource-efficient automated driving.
comment: 17 pages, 6 figures, under review by IEEE Transactions on Intelligent Transportation Systems (IEEE-T-ITS)
☆ Provably Safe Trajectory Generation for Manipulators Under Motion and Environmental Uncertainties
Robot manipulators operating in uncertain and non-convex environments present significant challenges for safe and optimal motion planning. Existing methods often struggle to provide efficient and formally certified collision risk guarantees, particularly when dealing with complex geometries and non-Gaussian uncertainties. This article proposes a novel risk-bounded motion planning framework to address this unmet need. Our approach integrates a rigid manipulator deep stochastic Koopman operator (RM-DeSKO) model to robustly predict the robot's state distribution under motion uncertainty. We then introduce an efficient, hierarchical verification method that combines parallelizable physics simulations with sum-of-squares (SOS) programming as a filter for fine-grained, formal certification of collision risk. This method is embedded within a Model Predictive Path Integral (MPPI) controller that uniquely utilizes binary collision information from SOS decomposition to improve its policy. The effectiveness of the proposed framework is validated on two typical robot manipulators through extensive simulations and real-world experiments, including a challenging human-robot collaboration scenario, demonstrating sim-to-real transfer of the learned model and its ability to generate safe and efficient trajectories in complex, uncertain settings.
☆ GST-VLA: Structured Gaussian Spatial Tokens for 3D Depth-Aware Vision-Language-Action Models
VLA models encode visual observations as 2D patch tokens with no intrinsic geometric structure. We introduce GST-VLA with two contributions. First, the Gaussian Spatial Tokenizer (GST) converts frozen dense depth and frozen semantic patch features into $N_g{=}128$ anisotropic 3D Gaussian primitives, each parameterized by a metric residual mean $μ\in \mathbb{R}^3$, log-scale covariance $\log σ\in \mathbb{R}^3$, and learned opacity $α\in (0,1)$. The covariance eigenstructure encodes local surface orientation, and opacity provides per-primitive geometric confidence, both inaccessible from scalar depth. Spatial attention pooling with learned queries concentrates the fixed token budget on geometrically salient regions rather than distributing uniformly. Second, 3D Depth-Aware Chain-of-Thought (DA-CoT) reasoning supervises four structured intermediate spatial thoughts, covering 3D object grounding, grasp affordance contact geometry, pairwise metric distances, and coarse SE(3) waypoints, as explicit generation targets in the training loss. A cross-attention sublayer at every VLM transformer block provides direct access to the raw 256-primitive Gaussian field during DA-CoT generation. A 300M-parameter flow-matching action expert with mixture-of-experts feedforward sublayers decodes 7-DoF delta action chunks via conditional ODE integration, conditioned on both VLM hidden states and DA-CoT outputs through dual cross-attention. Trained with composite $\mathcal{L}_\mathrm{flow} + \mathcal{L}_\mathrm{CoT} + \mathcal{L}_\mathrm{depth}$ across three progressive stages, GST-VLA achieves 96.4% on LIBERO (+2.0%), and 80.2% on SimplerEnv (+5.4%). Ablations isolate the contribution of each GST component, each DA-CoT thought, and each training stage, confirming independent and synergistic gains concentrated on precision demanding tasks.
comment: The results presented in this paper are preliminary. Please note that the experiments are currently ongoing, and the final data is subject to change upon the completion of the study. All ideas, results, methods, and any content herein are the sole property of the authors
☆ High-Slip-Ratio Control for Peak Tire-Road Friction Estimation Using Automated Vehicles
Accurate estimation of the tire-road friction coefficient (TRFC) is critical for ensuring safe vehicle control, especially under adverse road conditions. However, most existing methods rely on naturalistic driving data from regular vehicles, which typically operate under mild acceleration and braking. As a result, the data provide insufficient slip excitation and offer limited observability of the peak TRFC. This paper presents a high-slip-ratio control framework that enables automated vehicles (AVs) to actively excite the peak friction region during empty-haul operations while maintaining operational safety. A simplified Magic Formula tire model is adopted to represent nonlinear slip-force dynamics and is locally fitted using repeated high-slip measurements. To support safe execution in car-following scenarios, we formulate a constrained optimal control strategy that balances slip excitation, trajectory tracking, and collision avoidance. In parallel, a binning-based statistical projection method is introduced to robustly estimate peak TRFC under noise and local sparsity. The framework is validated through both closed-loop simulations and real-vehicle experiments, demonstrating its accuracy, safety, and feasibility for scalable, cost-effective roadway friction screening.
☆ 3D UAV Trajectory Estimation and Classification from Internet Videos via Language Model
Reliable 3D trajectory estimation of unmanned aerial vehicles (UAVs) is a fundamental requirement for anti-UAV systems, yet the acquisition of large-scale and accurately annotated trajectory data remains prohibitively expensive. In this work, we present a novel framework that derives UAV 3D trajectories and category information directly from Internet-scale UAV videos, without relying on manual annotations. First, language-driven data acquisition is employed to autonomously discover and collect UAV-related videos, while vision-language reasoning progressively filters task-relevant segments. Second, a training-free cross-modal label generation module is introduced to infer 3D trajectory hypotheses and UAV type cues. Third, a physics-informed refinement process is designed to impose temporal smoothness and kinematic consistency on the estimated trajectories. The resulting video clips and trajectory annotations can be readily utilized for downstream anti-UAV tasks. To assess effectiveness and generalization, we conduct zero-shot transfer experiments on a public, well-annotated 3D UAV benchmark. Results reveal a clear data scaling behavior: as the amount of online video data increases, zero-shot transfer performance on the target dataset improves consistently, without any target-domain training. The proposed method closely approaches the current state-of-the-art, highlighting its robustness and applicability to real-world anti-UAV scenarios. Code and datasets will be released upon acceptance.
☆ Quality over Quantity: Demonstration Curation via Influence Functions for Data-Centric Robot Learning ICRA 2026
Learning from demonstrations has emerged as a promising paradigm for end-to-end robot control, particularly when scaled to diverse and large datasets. However, the quality of demonstration data, often collected through human teleoperation, remains a critical bottleneck for effective data-driven robot learning. Human errors, operational constraints, and teleoperator variability introduce noise and suboptimal behaviors, making data curation essential yet largely manual and heuristic-driven. In this work, we propose Quality over Quantity (QoQ), a grounded and systematic approach to identifying high-quality data by defining data quality as the contribution of each training sample to reducing loss on validation demonstrations. To efficiently estimate this contribution, we leverage influence functions, which quantify the impact of individual training samples on model performance. We further introduce two key techniques to adapt influence functions for robot demonstrations: (i) using maximum influence across validation samples to capture the most relevant state-action pairs, and (ii) aggregating influence scores of state-action pairs within the same trajectory to reduce noise and improve data coverage. Experiments in both simulated and real-world settings show that QoQ consistently improves policy performances over prior data selection methods.
comment: Accepted to ICRA 2026, 8 pages
☆ Cutting the Cord: System Architecture for Low-Cost, GPU-Accelerated Bimanual Mobile Manipulation
We present a bimanual mobile manipulator built on the open-source XLeRobot with integrated onboard compute for less than \$1300. Key contributions include: (1) optimized mechanical design maximizing stiffness-to-weight ratio, (2) a Tri-Bus power topology isolating compute from motor-induced voltage transients, and (3) embedded autonomy using NVIDIA Jetson Orin Nano for untethered operation. The platform enables teleoperation, autonomous SLAM navigation, and vision-based manipulation without external dependencies, providing a low-cost alternative for research and education in robotics and robot learning.
☆ Beyond Amplitude: Channel State Information Phase-Aware Deep Fusion for Robotic Activity Recognition ICASSP
Wi-Fi Channel State Information (CSI) has emerged as a promising non-line-of-sight sensing modality for human and robotic activity recognition. However, prior work has predominantly relied on CSI amplitude while underutilizing phase information, particularly in robotic arm activity recognition. In this paper, we present GateFusion-Bidirectional Long Short-Term Memory network (GF-BiLSTM) for WiFi sensing in robotic activity recognition. GF-BiLSTM is a two-stream gated fusion network that encodes amplitude and phase separately and adaptively integrates per-time features through a learned gating mechanism. We systematically evaluate state-of-the-art deep learning models under a Leave-One-Velocity-Out (LOVO) protocol across four input configurations: amplitude only, phase only, amplitude + unwrapped phase, and amplitude + sanitized phase. Experimental results demonstrate that incorporating phase alongside amplitude consistently improves recognition accuracy and cross-speed robustness, with GF-BiLSTM achieving the best performance. To the best of our knowledge, this work provides the first systematic exploration of CSI phase for robotic activity recognition, establishing its critical role in Wi-Fi-based sensing.
comment: Accepted at 2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), May 4--8, 2026, Barcelona, Spain
☆ ImpedanceDiffusion: Diffusion-Based Global Path Planning for UAV Swarm Navigation with Generative Impedance Control
Safe swarm navigation in cluttered indoor environment requires long-horizon planning, reactive obstacle avoidance, and adaptive compliance. We propose ImpedanceDiffusion, a hierarchical framework that leverages image-conditioned diffusion-based global path planning with Artificial Potential Field (APF) tracking and semantic-aware variable impedance control for aerial drone swarms. The diffusion model generates geometric global trajectories directly from RGB images without explicit map construction. These trajectories are tracked by an APF-based reactive layer, while a VLM-RAG module performs semantic obstacle classification with 90% retrieval accuracy to adapt impedance parameters for mixed obstacle environments during execution. Two diffusion planners are evaluated: (i) a top-view long-horizon planner using single-pass inference and (ii) a first-person-view (FPV) short-horizon planner deployed via a two-stage inference pipeline. Both planners achieve a 100% trajectory generation rate across twenty static and dynamic experimental configurations and are validated via zero-shot sim-to-real deployment on Crazyflie 2.1 drones through the hierarchical APF-impedance control stack. The top-view planner produces smoother trajectories that yield conservative tracking speeds of 1.0-1.2 m/s near hard obstacles and 0.6-1.0 m/s near soft obstacles. In contrast, the FPV planner generates trajectories with greater local clearance and typically higher speeds, reaching 1.4-2.0 m/s near hard obstacles and up to 1.6 m/s near soft obstacles. Across 20 experimental configurations (100 total runs), the framework achieved a 92% success rate while maintaining stable impedance-based formation control with bounded oscillations and no in-flight collisions, demonstrating reliable and adaptive swarm navigation in cluttered indoor environments.
comment: This is paper is under review
☆ Update-Free On-Policy Steering via Verifiers
In recent years, Behavior Cloning (BC) has become one of the most prevalent methods for enabling robots to mimic human demonstrations. However, despite their successes, BC policies are often brittle and struggle with precise manipulation. To overcome these issues, we propose UF-OPS, an Update-Free On-Policy Steering method that enables the robot to predict the success likelihood of its actions and adapt its strategy at execution time. We accomplish this by training verifier functions using policy rollout data obtained during an initial evaluation of the policy. These verifiers are subsequently used to steer the base policy toward actions with a higher likelihood of success. Our method improves the performance of black-box diffusion policy, without changing the base parameters, making it light-weight and flexible. We present results from both simulation and real-world data and achieve an average 49% improvement in success rate over the base policy across 5 real tasks.
comment: 9 pages, 6 figures
☆ Design of a Robot-Assisted Chemical Dialysis System
Scientists perform diverse manual procedures that are tedious and laborious. Such procedures are considered a bottleneck for modern experimental science, as they consume time and increase burdens in fields including material science and medicine. We employ a user-centered approach to designing a robot-assisted system for dialysis, a common multi-day purification method used in polymer and protein synthesis. Through two usability studies, we obtain participant feedback and revise design requirements to develop the final system that satisfies scientists' needs and has the potential for applications in other experimental workflows. We anticipate that integration of this system into real synthesis procedures in a chemical wet lab will decrease workload on scientists during long experimental procedures and provide an effective approach to designing more systems that have the potential to accelerate scientific discovery and liberate scientists from tedious labor.
comment: Accepted at ACM/IEEE International Conference on Human-Robot Interaction (HRI'26), Late Breaking Reports 5 pages, 2 figures
☆ From Prior to Pro: Efficient Skill Mastery via Distribution Contractive RL Finetuning
We introduce Distribution Contractive Reinforcement Learning (DICE-RL), a framework that uses reinforcement learning (RL) as a "distribution contraction" operator to refine pretrained generative robot policies. DICE-RL turns a pretrained behavior prior into a high-performing "pro" policy by amplifying high-success behaviors from online feedback. We pretrain a diffusion- or flow-based policy for broad behavioral coverage, then finetune it with a stable, sample-efficient residual off-policy RL framework that combines selective behavior regularization with value-guided action selection. Extensive experiments and analyses show that DICE-RL reliably improves performance with strong stability and sample efficiency. It enables mastery of complex long-horizon manipulation skills directly from high-dimensional pixel inputs, both in simulation and on a real robot. Project website: https://zhanyisun.github.io/dice.rl.2026/.
☆ Degeneracy-Resilient Teach and Repeat for Geometrically Challenging Environments Using FMCW Lidar
Teach and Repeat (T&R) topometric navigation enables robots to autonomously repeat previously traversed paths without relying on GPS, making it well suited for operations in GPS-denied environments such as underground mines and lunar navigation. State-of-the-art T&R systems typically rely on iterative closest point (ICP)-based estimation; however, in geometrically degenerate environments with sparsely structured terrain, ICP often becomes ill-conditioned, resulting in degraded localization and unreliable navigation performance. To address this challenge, we present a degeneracy-resilient Frequency-Modulated Continuous-Wave (FMCW) lidar T&R navigation system consisting of Doppler velocity-based odometry and degeneracy-aware scan-to-map localization. Leveraging FMCW lidar, which provides per-point radial velocity measurements via the Doppler effect, we extend a geometry-independent, correspondence-free motion estimation to include principled pose uncertainty estimation that remains stable in degenerate environments. We further propose a degeneracy-aware localization method that incorporates per-point curvature for improved data association, and unifies translational and rotational scales to enable consistent degeneracy detection. Closed-loop field experiments across three environments with varying structural richness demonstrate that the proposed system reliably completes autonomous navigation, including in a challenging flat airport test field where a conventional ICP-based system fails.
☆ Hierarchical Task Model Predictive Control for Sequential Mobile Manipulation Tasks
Mobile manipulators are envisioned to serve more complex roles in people's everyday lives. With recent breakthroughs in large language models, task planners have become better at translating human verbal instructions into a sequence of tasks. However, there is still a need for a decision-making algorithm that can seamlessly interface with the high-level task planner to carry out the sequence of tasks efficiently. In this work, building on the idea of nonlinear lexicographic optimization, we propose a novel Hierarchical-Task Model Predictive Control framework that is able to complete sequential tasks with improved performance and reactivity by effectively leveraging the robot's redundancy. Compared to the state-of-the-art task-prioritized inverse kinematic control method, our approach has improved hierarchical trajectory tracking performance by 42% on average when facing task changes, robot singularity and reference variations. Compared to a typical single-task architecture, our proposed hierarchical task control architecture enables the robot to traverse a shorter path in task space and achieves an execution time 2.3 times faster when executing a sequence of delivery tasks. We demonstrated the results with real-world experiments on a 9 degrees of freedom mobile manipulator.
comment: 8 pages, Published in IEEE Robotics and Automation Letters ( Volume: 9, Issue: 2, February 2024)
☆ Perceptive Hierarchical-Task MPC for Sequential Mobile Manipulation in Unstructured Semi-Static Environments
As compared to typical mobile manipulation tasks, sequential mobile manipulation poses a unique challenge -- as the robot operates over extended periods, successful task completion is not solely dependent on consistent motion generation but also on the robot's awareness and adaptivity to changes in the operating environment. While existing motion planners can generate whole-body trajectories to complete sequential tasks, they typically assume that the environment remains static and rely on precomputed maps. This assumption often breaks down during long-term operations, where semi-static changes such as object removal, introduction, or shifts are common. In this work, we propose a novel perceptive hierarchical-task model predictive control (HTMPC) framework for efficient sequential mobile manipulation in unstructured, changing environments. To tackle the challenge, we leverage a Bayesian inference framework to explicitly model object-level changes and thereby maintain a temporally accurate representation of the 3D environment; this up-to-date representation is embedded in a lexicographic optimization framework to enable efficient execution of sequential tasks. We validate our perceptive HTMPC approach through both simulated and real-robot experiments. In contrast to baseline methods, our approach systematically accounts for moved and phantom obstacles, successfully completing sequential tasks with higher efficiency and reactivity, without relying on prior maps or external infrastructure.
☆ Robotic Ultrasound Makes CBCT Alive
Intraoperative Cone Beam Computed Tomography (CBCT) provides a reliable 3D anatomical context essential for interventional planning. However, its static nature fails to provide continuous monitoring of soft-tissue deformations induced by respiration, probe pressure, and surgical manipulation, leading to navigation discrepancies. We propose a deformation-aware CBCT updating framework that leverages robotic ultrasound as a dynamic proxy to infer tissue motion and update static CBCT slices in real time. Starting from calibration-initialized alignment with linear correlation of linear combination (LC2)-based rigid refinement, our method establishes accurate multimodal correspondence. To capture intraoperative dynamics, we introduce the ultrasound correlation UNet (USCorUNet), a lightweight network trained with optical flow-guided supervision to learn deformation-aware correlation representations, enabling accurate, real-time dense deformation field estimation from ultrasound streams. The inferred deformation is spatially regularized and transferred to the CBCT reference to produce deformation-consistent visualizations without repeated radiation exposure. We validate the proposed approach through deformation estimation and ultrasound-guided CBCT updating experiments. Results demonstrate real-time end-to-end CBCT slice updating and physically plausible deformation estimation, enabling dynamic refinement of static CBCT guidance during robotic ultrasound-assisted interventions. The source code is publicly available at https://github.com/anonymous-codebase/us-cbct-demo.
comment: 10 pages, 4 figures
☆ Octopus-inspired Distributed Control for Soft Robotic Arms: A Graph Neural Network-Based Attention Policy with Environmental Interaction IROS 2026
This paper proposes SoftGM, an octopus-inspired distributed control architecture for segmented soft robotic arms that learn to reach targets in contact-rich environments using online obstacle discovery without relying on global obstacle geometry. SoftGM formulates each arm section as a cooperative agent and represents the arm-environment interaction as a graph. SoftGM uses a two-stage graph attention message passing scheme following a Centralised Training Decentralised Execution (CTDE) paradigm with a centralised critic and decentralised actor. We evaluate SoftGM in a Cosserat-rod simulator (PyElastica) across three tasks that increase the complexity of the environment: obstacle-free, structured obstacles, and a wall-with-hole scenario. Compared with six widely used MARL baselines (IDDPG, IPPO, ISAC, MADDPG, MAPPO, MASAC) under identical information content and training conditions, SoftGM matches strong CTDE methods in simpler settings and achieves the best performance in the wall-with-hole task. Robustness tests with observation noise, single-section actuation failure, and transient disturbances show that SoftGM preserves success while keeping control effort bounded, indicating resilient coordination driven by selective contact-relevant information routing.
comment: 9 pages, 6 figures, 2 tables, submitted for IROS 2026
☆ Autonomous Search for Sparsely Distributed Visual Phenomena through Environmental Context Modeling ICRA 2026
Autonomous underwater vehicles (AUVs) are increasingly used to survey coral reefs, yet efficiently locating specific coral species of interest remains difficult: target species are often sparsely distributed across the reef, and an AUV with limited battery life cannot afford to search everywhere. When detections of the target itself are too sparse to provide directional guidance, the robot benefits from an additional signal to decide where to look next. We propose using the visual environmental context -- the habitat features that tend to co-occur with a target species -- as that signal. Because context features are spatially denser and often vary more smoothly than target detections, we hypothesize that a reward function targeted at broader environmental context will enable adaptive planners to make better decisions on where to go next, even in regions where no target has yet been observed. Starting from a single labeled image, our method uses patch-level DINOv2 embeddings to perform one-shot detections of both the target species and its surrounding context online. We validate our approach using real imagery collected by an AUV at two reef sites in St. John, U.S. Virgin Islands, simulating the robot's motion offline. Our results demonstrate that one-shot detection combined with adaptive context modeling enables efficient autonomous surveying, sampling up to 75$\%$ of the target in roughly half the time required by exhaustive coverage when the target is sparsely distributed, and outperforming search strategies that only use target detections.
comment: Accepted to the 2026 IEEE International Conference on Robotics and Automation (ICRA 2026)
☆ Characterizing Healthy & Post-Stroke Neuromotor Behavior During 6D Upper-Limb Isometric Gaming: Implications for Design of End-Effector Rehabilitation Robot Interfaces
Successful robot-mediated rehabilitation requires designing games and robot interventions that promote healthy motor practice. However, the interplay between a given user's neuromotor behavior, the gaming interface, and the physical robot makes designing system elements -- and even characterizing what behaviors are "healthy" or pathological -- challenging. We leverage our OpenRobotRehab 1.0 open access data set to assess the characteristics of 13 healthy and 2 post-stroke users' force output, muscle activations, and game performance while executing isometric trajectory tracking tasks using an end-effector rehabilitation robot. We present an assessment of how subtle aspects of interface design impact user behavior; an analysis of how pathological neuromotor behaviors are reflected in end-effector force dynamics; and a novel hidden Markov model (HMM)-based neuromotor behavior classification method based on surface electromyography (sEMG) signals during cyclic motions. We demonstrate that task specification (including which axes are constrained and how users interpret tracking instructions) shapes user behavior; that pathology-related features are detectable in 6D end-effector force data during isometric task execution (with significant differences between healthy and post-stroke profiles in force error and average force production at $p=0.05$); and that healthy neuromotor strategies are heterogeneous and inherently difficult to characterize. We also show that our HMM-based models discriminate healthy and post-stroke neuromotor dynamics where synergy-based decompositions reflect no such differentiation. Lastly, we discuss these results' implications for the design of adaptive end-effector rehabilitation robots capable of promoting healthier movement strategies across diverse user populations.
comment: This work has been submitted to the IEEE for possible publication
☆ Dance2Hesitate: A Multi-Modal Dataset of Dancer-Taught Hesitancy for Understandable Robot Motion
In human-robot collaboration, a robot's expression of hesitancy is a critical factor that shapes human coordination strategies, attention allocation, and safety-related judgments. However, designing hesitant robot motion that generalizes is challenging because the observer's inference is highly dependent on embodiment and context. To address these challenges, we introduce and open-source a multi-modal, dancer-generated dataset of hesitant motion where we focus on specific context-embodiment pairs (i.e., manipulator/human upper-limb approaching a Jenga Tower, and anthropomorphic whole body motion in free space). The dataset includes (i) kinesthetic teaching demonstrations on a Franka Emika Panda reaching from a fixed start configuration to a fixed target (a Jenga tower) with three graded hesitancy levels (slight, significant, extreme) and (ii) synchronized RGB-D motion capture of dancers performing the same reaching behavior using their upper limb across three hesitancy levels, plus full human body sequences for extreme hesitancy. We further provide documentation to enable reproducible benchmarking across robot and human modalities. Across all dancers, we obtained 70 unique whole-body trajectories, 84 upper limb trajectories spanning over the three hesitancy levels, and 66 kinesthetic teaching trajectories spanning over the three hesitancy levels. The dataset can be accessed here: https://brsrikrishna.github.io/Dance2Hesitate/.
comment: Accepted to the Designing Transparent and Understandable Robots (D-TUR) Workshop at the ACM/IEEE International Conference on Human-Robot Interaction (HRI) 2026, Edinburgh, UK
☆ Cross-Hand Latent Representation for Vision-Language-Action Models
Dexterous manipulation is essential for real-world robot autonomy, mirroring the central role of human hand coordination in daily activity. Humans rely on rich multimodal perception--vision, sound, and language-guided intent--to perform dexterous actions, motivating vision-based, language-conditioned manipulation systems for robots. However, training reliable vision-language-action (VLA) models for dexterous manipulation requires large-scale demonstrations across many robotic hands. In addition, as new dexterous embodiments appear rapidly, collecting data for each becomes costly and impractical, creating a need for scalable cross-embodiment learning. We introduce XL-VLA, a vision-language-action framework integrated with a unified latent action space shared across diverse dexterous hands. This embodiment-invariant latent space is directly pluggable into standard VLA architectures, enabling seamless cross-embodiment training and efficient reuse of both existing and newly collected data. Experimental results demonstrate that XL-VLA consistently outperforms baseline VLA models operating in raw joint spaces, establishing it as an effective solution for scalable cross-embodiment dexterous manipulation.
comment: Website: https://xl-vla.github.io
☆ AR-VLA: True Autoregressive Action Expert for Vision-Language-Action Models
We propose a standalone autoregressive (AR) Action Expert that generates actions as a continuous causal sequence while conditioning on refreshable vision-language prefixes. In contrast to existing Vision-Language-Action (VLA) models and diffusion policies that reset temporal context with each new observation and predict actions reactively, our Action Expert maintains its own history through a long-lived memory and is inherently context-aware. This structure addresses the frequency mismatch between fast control and slow reasoning, enabling efficient independent pretraining of kinematic syntax and modular integration with heavy perception backbones, naturally ensuring spatio-temporally consistent action generation across frames. To synchronize these asynchronous hybrid V-L-A modalities, we utilize a re-anchoring mechanism that mathematically accounts for perception staleness during both training and inference. Experiments on simulated and real-robot manipulation tasks demonstrate that the proposed method can effectively replace traditional chunk-based action heads for both specialist and generalist policies. AR-VLA exhibits superior history awareness and substantially smoother action trajectories while maintaining or exceeding the task success rates of state-of-the-art reactive VLAs. Overall, our work introduces a scalable, context-aware action generation schema that provides a robust structural foundation for training effective robotic policies.
☆ TiPToP: A Modular Open-Vocabulary Planning System for Robotic Manipulation
We present TiPToP, an extensible modular system that combines pretrained vision foundation models with an existing Task and Motion Planner (TAMP) to solve multi-step manipulation tasks directly from input RGB images and natural-language instructions. Our system aims to be simple and easy-to-use: it can be installed and run on a standard DROID setup in under one hour and adapted to new embodiments with minimal effort. We evaluate TiPToP -- which requires zero robot data -- over 28 tabletop manipulation tasks in simulation and the real world and find it matches or outperforms $π_{0.5}\text{-DROID}$, a vision-language-action (VLA) model fine-tuned on 350 hours of embodiment-specific demonstrations. TiPToP's modular architecture enables us to analyze the system's failure modes at the component level. We analyze results from an evaluation of 173 trials and identify directions for improvement. We release TiPToP open-source to further research on modular manipulation systems and tighter integration between learning and planning. Project website and code: https://tiptop-robot.github.io
comment: Project website: https://tiptop-robot.github.io
☆ BEACON: Language-Conditioned Navigation Affordance Prediction under Occlusion
Language-conditioned local navigation requires a robot to infer a nearby traversable target location from its current observation and an open-vocabulary, relational instruction. Existing vision-language spatial grounding methods usually rely on vision-language models (VLMs) to reason in image space, producing 2D predictions tied to visible pixels. As a result, they struggle to infer target locations in occluded regions, typically caused by furniture or moving humans. To address this issue, we propose BEACON, which predicts an ego-centric Bird's-Eye View (BEV) affordance heatmap over a bounded local region including occluded areas. Given an instruction and surround-view RGB-D observations from four directions around the robot, BEACON predicts the BEV heatmap by injecting spatial cues into a VLM and fusing the VLM's output with depth-derived BEV features. Using an occlusion-aware dataset built in the Habitat simulator, we conduct detailed experimental analysis to validate both our BEV space formulation and the design choices of each module. Our method improves the accuracy averaged across geodesic thresholds by 22.74 percentage points over the state-of-the-art image-space baseline on the validation subset with occluded target locations. Our project page is: https://xin-yu-gao.github.io/beacon.
comment: 8 pages. Project page: https://xin-yu-gao.github.io/beacon
☆ Kinodynamic Motion Retargeting for Humanoid Locomotion via Multi-Contact Whole-Body Trajectory Optimization
We present the KinoDynamic Motion Retargeting (KDMR) framework, a novel approach for humanoid locomotion that models the retargeting process as a multi-contact, whole-body trajectory optimization problem. Conventional kinematics-based retargeting methods rely solely on spatial motion capture (MoCap) data, inevitably introducing physically inconsistent artifacts, such as foot sliding and ground penetration, that severely degrade the performance of downstream imitation learning policies. To bridge this gap, KDMR extends beyond pure kinematics by explicitly enforcing rigid-body dynamics and contact complementarity constraints. Further, by integrating ground reaction force (GRF) measurements alongside MoCap data, our method automatically detects heel-toe contact events to accurately replicate complex human-like contact patterns. We evaluate KDMR against the state-of-the-art baseline, GMR, across three key dimensions: 1) the dynamic feasibility and smoothness of the retargeted motions, 2) the accuracy of GRF tracking compared to raw source data, and 3) the training efficiency and final performance of downstream control policies trained via the BeyondMimic framework. Experimental results demonstrate that KDMR significantly outperforms purely kinematic methods, yielding dynamically viable reference trajectories that accelerate policy convergence and enhance overall locomotion stability. Our end-to-end pipeline will be open-sourced upon publication.
☆ NanoBench: A Multi-Task Benchmark Dataset for Nano-Quadrotor System Identification, Control, and State Estimation
Existing aerial-robotics benchmarks target vehicles from hundreds of grams to several kilograms and typically expose only high-level state data. They omit the actuator-level signals required to study nano-scale quadrotors, where low-Reynolds number aerodynamics, coreless DC motor nonlinearities, and severe computational constraints invalidate models and controllers developed for larger vehicles. We introduce NanoBench, an open-source multi-task benchmark collected on the commercially available Crazyflie 2.1 nano-quadrotor (takeoff weight 27 g) in a Vicon motion capture arena. The dataset contains over 170 flight trajectories spanning hover, multi-frequency excitation, standard tracking, and aggressive maneuvers across multiple speed regimes. Each trajectory provides synchronized Vicon ground truth, raw IMU data, onboard extended Kalman filter estimates, PID controller internals, and motor PWM commands at 100 Hz, alongside battery telemetry at 10 Hz, aligned with sub-0.5 ms consistency. NanoBench defines standardized evaluation protocols, train/test splits, and open-source baselines for three tasks: nonlinear system identification, closed-loop controller benchmarking, and onboard state estimation assessment. To our knowledge, it is the first public dataset to jointly provide actuator commands, controller internals, and estimator outputs with millimeter-accurate ground truth on a commercially available nano-scale aerial platform.
comment: 9 pages, 6 figures
☆ Emerging Extrinsic Dexterity in Cluttered Scenes via Dynamics-aware Policy Learning
Extrinsic dexterity leverages environmental contact to overcome the limitations of prehensile manipulation. However, achieving such dexterity in cluttered scenes remains challenging and underexplored, as it requires selectively exploiting contact among multiple interacting objects with inherently coupled dynamics. Existing approaches lack explicit modeling of such complex dynamics and therefore fall short in non-prehensile manipulation in cluttered environments, which in turn limits their practical applicability in real-world environments. In this paper, we introduce a Dynamics-Aware Policy Learning (DAPL) framework that can facilitate policy learning with a learned representation of contact-induced object dynamics in cluttered environments. This representation is learned through explicit world modeling and used to condition reinforcement learning, enabling extrinsic dexterity to emerge without hand-crafted contact heuristics or complex reward shaping. We evaluate our approach in both simulation and the real world. Our method outperforms prehensile manipulation, human teleoperation, and prior representation-based policies by over 25% in success rate on unseen simulated cluttered scenes with varying densities. The real-world success rate reaches around 50% across 10 cluttered scenes, while a practical grocery deployment further demonstrates robust sim-to-real transfer and applicability.
comment: Project Page: https://pku-epic.github.io/DAPL/
☆ Lightweight 3D LiDAR-Based UAV Tracking: An Adaptive Extended Kalman Filtering Approach
Accurate relative positioning is crucial for swarm aerial robotics, enabling coordinated flight and collision avoidance. Although vision-based tracking has been extensively studied, 3D LiDAR-based methods remain underutilized despite their robustness under varying lighting conditions. Existing systems often rely on bulky, power-intensive sensors, making them impractical for small UAVs with strict payload and energy constraints. This paper presents a lightweight LiDAR-based UAV tracking system incorporating an Adaptive Extended Kalman Filter (AEKF) framework. Our approach effectively addresses the challenges posed by sparse, noisy, and nonuniform point cloud data generated by non-repetitive scanning 3D LiDARs, ensuring reliable tracking while remaining suitable for small drones with strict payload constraints. Unlike conventional filtering techniques, the proposed method dynamically adjusts the noise covariance matrices using innovation and residual statistics, thereby enhancing tracking accuracy under real-world conditions. Additionally, a recovery mechanism ensures continuity of tracking during temporary detection failures caused by scattered LiDAR returns or occlusions. Experimental validation was performed using a Livox Mid-360 LiDAR mounted on a DJI F550 UAV in real-world flight scenarios. The proposed method demonstrated robust UAV tracking performance under sparse LiDAR returns and intermittent detections, consistently outperforming both standard Kalman filtering and particle filtering approaches during aggressive maneuvers. These results confirm that the framework enables reliable relative positioning in GPS-denied environments without the need for multi-sensor arrays or external infrastructure.
comment: Presented at the 19th International Conference on Intelligent Autonomous Systems, IAS-19, Genoa, Italy, June 30 to July 4, 2025. To appear in the Springer post-proceedings of the conference
☆ TIMID: Time-Dependent Mistake Detection in Videos of Robot Executions IROS
As robotic systems execute increasingly difficult task sequences, so does the number of ways in which they can fail. Video Anomaly Detection (VAD) frameworks typically focus on singular, low-level kinematic or action failures, struggling to identify more complex temporal or spatial task violations, because they do not necessarily manifest as low-level execution errors. To address this problem, the main contribution of this paper is a new VAD-inspired architecture, TIMID, which is able to detect robot time-dependent mistakes when executing high-level tasks. Our architecture receives as inputs a video and prompts of the task and the potential mistake, and returns a frame-level prediction in the video of whether the mistake is present or not. By adopting a VAD formulation, the model can be trained with weak supervision, requiring only a single label per video. Additionally, to alleviate the problem of data scarcity of incorrect executions, we introduce a multi-robot simulation dataset with controlled temporal errors and real executions for zero-shot sim-to-real evaluation. Our experiments demonstrate that out-of-the-box VLMs lack the explicit temporal reasoning required for this task, whereas our framework successfully detects different types of temporal errors. Project: https://ropertunizar.github.io/TIMID/
comment: 8 pages, 5 figures , IROS submission
☆ MuxGel: Simultaneous Dual-Modal Visuo-Tactile Sensing via Spatially Multiplexing and Deep Reconstruction IROS 2026
High-fidelity visuo-tactile sensing is important for precise robotic manipulation. However, most vision-based tactile sensors face a fundamental trade-off: opaque coatings enable tactile sensing but block pre-contact vision. To address this, we propose MuxGel, a spatially multiplexed sensor that captures both external visual information and contact-induced tactile signals through a single camera. By using a checkerboard coating pattern, MuxGel interleaves tactile-sensitive regions with transparent windows for external vision. This design maintains standard form factors, allowing for plug-and-play integration into GelSight-style sensors by simply replacing the gel pad. To recover full-resolution vision and tactile signals from the multiplexed inputs, we develop a U-Net-based reconstruction framework. Leveraging a sim-to-real pipeline, our model effectively decouples and restores high-fidelity tactile and visual fields simultaneously. Experiments on unseen objects demonstrate the framework's generalization and accuracy. Furthermore, we demonstrate MuxGel's utility in grasping tasks, where dual-modality feedback facilitates both pre-contact alignment and post-contact interaction. Results show that MuxGel enhances the perceptual capabilities of existing vision-based tactile sensors while maintaining compatibility with their hardware stacks. Project webpage: https://zhixianhu.github.io/muxgel/.
comment: Submitted to IROS 2026
☆ PanoAffordanceNet: Towards Holistic Affordance Grounding in 360° Indoor Environments
Global perception is essential for embodied agents in 360° spaces, yet current affordance grounding remains largely object-centric and restricted to perspective views. To bridge this gap, we introduce a novel task: Holistic Affordance Grounding in 360° Indoor Environments. This task faces unique challenges, including severe geometric distortions from Equirectangular Projection (ERP), semantic dispersion, and cross-scale alignment difficulties. We propose PanoAffordanceNet, an end-to-end framework featuring a Distortion-Aware Spectral Modulator (DASM) for latitude-dependent calibration and an Omni-Spherical Densification Head (OSDH) to restore topological continuity from sparse activations. By integrating multi-level constraints comprising pixel-wise, distributional, and region-text contrastive objectives, our framework effectively suppresses semantic drift under low supervision. Furthermore, we construct 360-AGD, the first high-quality panoramic affordance grounding dataset. Extensive experiments demonstrate that PanoAffordanceNet significantly outperforms existing methods, establishing a solid baseline for scene-level perception in embodied intelligence. The source code and benchmark dataset will be made publicly available at https://github.com/GL-ZHU925/PanoAffordanceNet.
comment: The source code and benchmark dataset will be made publicly available at https://github.com/GL-ZHU925/PanoAffordanceNet
☆ Caterpillar-Inspired Spring-Based Compressive Continuum Robot for Bristle-based Exploration
Exploration of confined spaces, such as pipelines and ducts, remains challenging for conventional rigid robots due to limited space, irregular geometry, and restricted access. Inspired by caterpillar locomotion and sensing, this paper presents a compact spring-based tendon-driven continuum robot that integrates with commercial robotic arms for confined-space inspection. The system combines a mechanically compliant continuum body with a tendon actuation module, enabling coupled bending and axial length change, and uses a constant-curvature kinematic model for positional control. Experiments show a mean position error of 4.32 mm under the proposed model and control pipeline. To extend the system from motion to inspection, we integrate an artificial bristle contact sensor and demonstrate surface perception and confined-space exploration through contact interactions. This compact and compliant design offers a cost-effective upgrade for commercial robots and promises effective exploration in challenging environments.
comment: Accepted by RoboSoft 2026
☆ Let's Reward Step-by-Step: Step-Aware Contrastive Alignment for Vision-Language Navigation in Continuous Environments
Vision-Language Navigation in Continuous Environments (VLN-CE) requires agents to learn complex reasoning from long-horizon human interactions. While Multi-modal Large Language Models (MLLMs) have driven recent progress, current training paradigms struggle to balance generalization capability, error recovery and training stability. Specifically, (i) policies derived from SFT suffer from compounding errors, struggling to recover from out-of-distribution states, and (ii) Reinforcement Fine-Tuning (RFT) methods e.g. GRPO are bottlenecked by sparse outcome rewards. Their binary feedback fails to assign credit to individual steps, leading to gradient signal collapse in failure dominant batches. To address these challenges, we introduce Step-Aware Contrastive Alignment (SACA), a framework designed to extract dense supervision from imperfect trajectories. At its core, the Perception-Grounded Step-Aware auditor evaluates progress step-by-step, disentangling failed trajectories into valid prefixes and exact divergence points. Leveraging these signals, Scenario-Conditioned Group Construction mechanism dynamically routes batches to specialized resampling and optimization strategies. Extensive experiments on VLN-CE benchmarks demonstrate that SACA achieves state-of-the-art performance.
comment: 28 pages, 10 figures
☆ $M^2$-Occ: Resilient 3D Semantic Occupancy Prediction for Autonomous Driving with Incomplete Camera Inputs
Semantic occupancy prediction enables dense 3D geometric and semantic understanding for autonomous driving. However, existing camera-based approaches implicitly assume complete surround-view observations, an assumption that rarely holds in real-world deployment due to occlusion, hardware malfunction, or communication failures. We study semantic occupancy prediction under incomplete multi-camera inputs and introduce $M^2$-Occ, a framework designed to preserve geometric structure and semantic coherence when views are missing. $M^2$-Occ addresses two complementary challenges. First, a Multi-view Masked Reconstruction (MMR) module leverages the spatial overlap among neighboring cameras to recover missing-view representations directly in the feature space. Second, a Feature Memory Module (FMM) introduces a learnable memory bank that stores class-level semantic prototypes. By retrieving and integrating these global priors, the FMM refines ambiguous voxel features, ensuring semantic consistency even when observational evidence is incomplete. We introduce a systematic missing-view evaluation protocol on the nuScenes-based SurroundOcc benchmark, encompassing both deterministic single-view failures and stochastic multi-view dropout scenarios. Under the safety-critical missing back-view setting, $M^2$-Occ improves the IoU by 4.93%. As the number of missing cameras increases, the robustness gap further widens; for instance, under the setting with five missing views, our method boosts the IoU by 5.01%. These gains are achieved without compromising full-view performance. The source code will be publicly released at https://github.com/qixi7up/M2-Occ.
comment: The source code will be publicly released at https://github.com/qixi7up/M2-Occ
☆ Efficient and robust control with spikes that constrain free energy
Animal brains exhibit remarkable efficiency in perception and action, while being robust to both external and internal perturbations. The means by which brains accomplish this remains, for now, poorly understood, hindering our understanding of animal and human cognition, as well as our own implementation of efficient algorithms for control of dynamical systems.A potential candidate for a robust mechanism of state estimation and action computation is the free energy principle, but existing implementations of this principle have largely relied on conventional, biologically implausible approaches without spikes. We propose a novel, efficient, and robust spiking control framework with realistic biological characteristics. The resulting networks function as free energy constrainers, in which neurons only fire if they reduce the free energy of their internal representation. The networks offer efficient operation through highly sparse activity while matching performance with other similar spiking frameworks, and have high resilience against both external (e.g. sensory noise or collisions) and internal perturbations (e.g. synaptic noise and delays or neuron silencing) that such a network would be faced with when deployed by either an organism or an engineer. Overall, our work provides a novel mathematical account for spiking control through constraining free energy, providing both better insight into how brain networks might leverage their spiking substrate and a new route for implementing efficient control algorithms in neuromorphic hardware.
☆ Robotic Scene Cloning:Advancing Zero-Shot Robotic Scene Adaptation in Manipulation via Visual Prompt Editing
Modern robots can perform a wide range of simple tasks and adapt to diverse scenarios in the well-trained environment. However, deploying pre-trained robot models in real-world user scenarios remains challenging due to their limited zero-shot capabilities, often necessitating extensive on-site data collection. To address this issue, we propose Robotic Scene Cloning (RSC), a novel method designed for scene-specific adaptation by editing existing robot operation trajectories. RSC achieves accurate and scene-consistent sample generation by leveraging a visual prompting mechanism and a carefully tuned condition injection module. Not only transferring textures but also performing moderate shape adaptations in response to the visual prompts, RSC demonstrates reliable task performance across a variety of object types. Experiments across various simulated and real-world environments demonstrate that RSC significantly enhances policy generalization in target environments.
☆ DRIFT: Dual-Representation Inter-Fusion Transformer for Automated Driving Perception with 4D Radar Point Clouds
4D radars, which provide 3D point cloud data along with Doppler velocity, are attractive components of modern automated driving systems due to their low cost and robustness under adverse weather conditions. However, they provide a significantly lower point cloud density than LiDAR sensors. This makes it important to exploit not only local but also global contextual scene information. This paper proposes DRIFT, a model that effectively captures and fuses both local and global contexts through a dual-path architecture. The model incorporates a point path to aggregate fine-grained local features and a pillar path to encode coarse-grained global features. These two parallel paths are intertwined via novel feature-sharing layers at multiple stages, enabling full utilization of both representations. DRIFT is evaluated on the widely used View-of-Delft (VoD) dataset and a proprietary internal dataset. It outperforms the baselines on the tasks of object detection and/or free road estimation. For example, DRIFT achieves a mean average precision (mAP) of 52.6\% (compared to, say, 45.4\% of CenterPoint) on the VoD dataset.
☆ SELF-VLA: A Skill Enhanced Agentic Vision-Language-Action Framework for Contact-Rich Disassembly
Disassembly automation has long been pursued to address the growing demand for efficient and proper recovery of valuable components from the end-of-life (EoL) electronic products. Existing approaches have demonstrated promising and regimented performance by decomposing the disassembly process into different subtasks. However, each subtask typically requires extensive data preparation, model training, and system management. Moreover, these approaches are often task- and component-specific, making them poorly suited to handle the variability and uncertainty of EoL products and limiting their generalization capabilities. All these factors restrict the practical deployment of current robotic disassembly systems and leave them highly reliant on human labor. With the recent development of foundation models in robotics, vision-language-action (VLA) models have shown impressive performance on standard robotic manipulation tasks, but their applicability to complex, contact-rich, and long-horizon industrial practices like disassembly, which requires sequential and precise manipulation, remains limited. To address this challenge, we propose SELF-VLA, an agentic VLA framework that integrates explicit disassembly skills. Experimental studies demonstrate that our framework significantly outperforms current state-of-the-art end-to-end VLA models on two contact-rich disassembly tasks. The video illustration can be found via https://zh.engr.tamu.edu/wp-content/uploads/sites/310/2026/03/IROS-VLA-Video.mp4.
☆ TATIC: Task-Aware Temporal Learning for Human Intent Inference from Physical Corrections in Human-Robot Collaboration
In human-robot collaboration (HRC), robots must adapt online to dynamic task constraints and evolving human intent. While physical corrections provide a natural, low-latency channel for operators to convey motion-level adjustments, extracting task-level semantic intent from such brief interactions remains challenging. Existing foundation-model-based approaches primarily rely on vision and language inputs and lack mechanisms to interpret physical feedback. Meanwhile, traditional physical human-robot interaction (pHRI) methods leverage physical corrections for trajectory guidance but struggle to infer task-level semantics. To bridge this gap, we propose TATIC, a unified framework that utilizes torque-based contact force estimation and a task-aware Temporal Convolutional Network (TCN) to jointly infer discrete task-level intent and estimate continuous motion-level parameters from brief physical corrections. Task-aligned feature canonicalization ensures robust generalization across diverse layouts, while an intent-driven adaptation scheme translates inferred human intent into robot motion adaptations. Experiments achieve a 0.904 Macro-F1 score in intent recognition and demonstrate successful hardware validation in collaborative disassembly (see experimental video at https://youtu.be/xF8A52qwEc8).
♻ ☆ From Demonstrations to Safe Deployment: Path-Consistent Safety Filtering for Diffusion Policies ICRA 2026
Diffusion policies (DPs) achieve state-of-the-art performance on complex manipulation tasks by learning from large-scale demonstration datasets, often spanning multiple embodiments and environments. However, they cannot guarantee safe behavior, requiring external safety mechanisms. These, however, alter actions in ways unseen during training, causing unpredictable behavior and performance degradation. To address these problems, we propose path-consistent safety filtering (PACS) for DPs. Our approach performs path-consistent braking on a trajectory computed from the sequence of generated actions. In this way, we keep the execution consistent with the training distribution of the policy, maintaining the learned, task-completing behavior. To enable real-time deployment and handle uncertainties, we verify safety using set-based reachability analysis. Our experimental evaluation in simulation and on three challenging real-world human-robot interaction tasks shows that PACS (a) provides formal safety guarantees in dynamic environments, (b) preserves task success rates, and (c) outperforms reactive safety approaches, such as control barrier functions, by up to 68 % in terms of task success. Videos are available at our project website: https://tum-lsy.github.io/pacs.
comment: Accepted to IEEE ICRA 2026. Project page: https://tum-lsy.github.io/pacs/. 8 pages, 4 figures
♻ ☆ LLM-Advisor: An LLM Benchmark for Cost-efficient Path Planning across Multiple Terrains
Cost-efficient path planning across multiple terrains is a crucial task in robot navigation, requiring the identification of a path from the start to the goal that not only avoids obstacles but also minimizes the overall travel cost. This is especially crucial for real-world applications where robots need to navigate diverse terrains in outdoor environments with limited opportunities for recharging or refueling. Despite its practical importance, cost-efficient path planning across heterogeneous terrains has received relatively limited attention in prior work. In this paper, we propose LLM-Advisor, a prompt-based, planner-agnostic framework that leverages large language models (LLMs) as non-decisive post-processing advisors for cost refinement, without modifying the underlying planner. While we observe that LLMs may occasionally produce implausible suggestions, we introduce two effective hallucination-mitigation strategies. We further introduce two datasets, MultiTerraPath and RUGD_v2, for systematic evaluation of cost-efficient path planning. Extensive experiments reveal that state-of-the-art LLMs, including GPT-4o, GPT-4-turbo, Gemini-2.5-Flash, and Claude-Opus-4, perform poorly in zero-shot terrain-aware path planning, highlighting their limited spatial reasoning capability. In contrast, the proposed LLM-Advisor (with GPT-4o) improves cost efficiency for 72.37% of A*-planned paths, 69.47% of RRT*-planned paths, and 78.70% of LLM-A*-planned paths. On the MultiTerraPath dataset, LLM-Advisor demonstrates stronger performance on the hard subset, further validating its applicability to real-world scenarios.
♻ ☆ Exploring Single Domain Generalization of LiDAR-based Semantic Segmentation under Imperfect Labels
Accurate perception is critical for vehicle safety, with LiDAR as a key enabler in autonomous driving. To ensure robust performance across environments, sensor types, and weather conditions without costly re-annotation, domain generalization in LiDAR-based 3D semantic segmentation is essential. However, LiDAR annotations are often noisy due to sensor imperfections, occlusions, and human errors. Such noise degrades segmentation accuracy and is further amplified under domain shifts, threatening system reliability. While noisy-label learning is well-studied in images, its extension to 3D LiDAR segmentation under domain generalization remains largely unexplored, as the sparse and irregular structure of point clouds limits direct use of 2D methods. To address this gap, we introduce the novel task Domain Generalization for LiDAR Semantic Segmentation under Noisy Labels (DGLSS-NL) and establish the first benchmark by adapting three representative noisy-label learning strategies from image classification to 3D segmentation. However, we find that existing noisy-label learning approaches adapt poorly to LiDAR data. We therefore propose DuNe, a dual-view framework with strong and weak branches that enforce feature-level consistency and apply cross-entropy loss based on confidence-aware filtering of predictions. Our approach shows state-of-the-art performance by achieving 56.86% mIoU on SemanticKITTI, 42.28% on nuScenes, and 52.58% on SemanticPOSS under 10% symmetric label noise, with an overall Arithmetic Mean (AM) of 49.57% and Harmonic Mean (HM) of 48.50%, thereby demonstrating robust domain generalization in DGLSS-NL tasks. The code is available on our project page.
♻ ☆ VLN-Cache: Enabling Token Caching for VLN Models with Visual/Semantic Dynamics Awareness
Vision-and-Language Navigation (VLN) increasingly relies on large vision-language models, but their inference cost conflicts with real-time deployment. Token caching is a promising training-free strategy that avoids redundant computation by reusing stable visual tokens across frames. However, existing methods assume a static camera and fixed semantic focus, assumptions that VLN fundamentally violates. We identify two failure modes: (1) visual dynamics, where viewpoint shift displaces token positions across frames, causing position-wise matching to pair misaligned content; (2) semantic dynamics, where token relevance shifts across task stages as navigation progresses, making cached states stale. We propose VLN-Cache, a visual-dynamic-aware and semantic-dynamic-aware caching framework that introduces view-aligned remapping to recover geometric correspondences and a task-relevance saliency filter to veto reuse at semantic transitions. A layer-adaptive entropy policy further balances the per-layer reuse budget. Experiments on the R2R-CE simulation benchmark show up to 1.52x speedup while maintaining competitive navigation success rates.
♻ ☆ Bootstrap Dynamic-Aware 3D Visual Representation for Scalable Robot Learning CVPR 2026
Despite strong results on recognition and segmentation, current 3D visual pre-training methods often underperform on robotic manipulation. We attribute this gap to two factors: the lack of state-action-state dynamics modeling and the unnecessary redundancy of explicit geometric reconstruction. We introduce AFRO, a self-supervised framework that learns dynamics-aware 3D representations without action or reconstruction supervision. AFRO casts state prediction as a generative diffusion process and jointly models forward and inverse dynamics in a shared latent space to capture causal transition structure. To prevent feature leakage in action learning, we employ feature differencing and inverse-consistency supervision, improving the quality and stability of visual features. When combined with Diffusion Policy, AFRO substantially increases manipulation success rates across 16 simulated and 4 real-world tasks, outperforming existing pre-training approaches. The framework also scales favorably with data volume and task complexity. Qualitative visualizations indicate that AFRO learns semantically rich, discriminative features, offering an effective pre-training solution for 3D representation learning in robotics. Project page: https://kolakivy.github.io/AFRO/
comment: Project Page: https://kolakivy.github.io/AFRO/, accepted by CVPR 2026
♻ ☆ Compose Your Policies! Improving Diffusion-based or Flow-based Robot Policies via Test-time Distribution-level Composition ICLR 2026
Diffusion-based models for robotic control, including vision-language-action (VLA) and vision-action (VA) policies, have demonstrated significant capabilities. Yet their advancement is constrained by the high cost of acquiring large-scale interaction datasets. This work introduces an alternative paradigm for enhancing policy performance without additional model training. Perhaps surprisingly, we demonstrate that the composed policies can exceed the performance of either parent policy. Our contribution is threefold. First, we establish a theoretical foundation showing that the convex composition of distributional scores from multiple diffusion models can yield a superior one-step functional objective compared to any individual score. A Grönwall-type bound is then used to show that this single-step improvement propagates through entire generation trajectories, leading to systemic performance gains. Second, motivated by these results, we propose General Policy Composition (GPC), a training-free method that enhances performance by combining the distributional scores of multiple pre-trained policies via a convex combination and test-time search. GPC is versatile, allowing for the plug-and-play composition of heterogeneous policies, including VA and VLA models, as well as those based on diffusion or flow-matching, irrespective of their input visual modalities. Third, we provide extensive empirical validation. Experiments on Robomimic, PushT, and RoboTwin benchmarks, alongside real-world robotic evaluations, confirm that GPC consistently improves performance and adaptability across a diverse set of tasks. Further analysis of alternative composition operators and weighting strategies offers insights into the mechanisms underlying the success of GPC. These results establish GPC as a simple yet effective method for improving control performance by leveraging existing policies.
comment: Accepted to ICLR 2026. Project Page: https://sagecao1125.github.io/GPC-Site/
♻ ☆ SPARC: Spatial-Aware Path Planning via Attentive Robot Communication
Efficient communication is critical for decentralized Multi-Robot Path Planning (MRPP), yet existing learned communication methods treat all neighboring robots equally regardless of their spatial proximity, leading to diluted attention in congested regions where coordination matters most. We propose Relation enhanced Multi Head Attention (RMHA), a communication mechanism that explicitly embeds pairwise Manhattan distances into the attention weight computation, enabling each robot to dynamically prioritize messages from spatially relevant neighbors. Combined with a distance-constrained attention mask and GRU gated message fusion, RMHA integrates seamlessly with MAPPO for stable end-to-end training. In zero-shot generalization from 8 training robots to 128 test robots on 40x40 grids, RMHA achieves approximately 75 percent success rate at 30 percent obstacle density outperforming the best baseline by over 25 percentage points. Ablation studies confirm that distance-relation encoding is the key contributor to success rate improvement in high-density environments. Index Terms-Multi-robot path planning, graph attention mechanism, multi-head attention, communication optimization, cooperative decision-making
comment: The manuscript is being withdrawn at the request of the first author for the purpose of revising content and re-uploading a revised version with updated data/figures/text . The revised manuscript will be resubmitted to arXiv promptly with the same author list and research theme
♻ ☆ SynHLMA:Synthesizing Hand Language Manipulation for Articulated Object with Discrete Human Object Interaction Representation
Generating hand grasps with language instructions is a widely studied topic that benefits from embodied AI and VR/AR applications. While transferring into hand articulatied object interaction (HAOI), the hand grasps synthesis requires not only object functionality but also long-term manipulation sequence along the object deformation. This paper proposes a novel HAOI sequence generation framework SynHLMA, to synthesize hand language manipulation for articulated objects. Given a complete point cloud of an articulated object, we utilize a discrete HAOI representation to model each hand object interaction frame. Along with the natural language embeddings, the representations are trained by an HAOI manipulation language model to align the grasping process with its language description in a shared representation space. A joint-aware loss is employed to ensure hand grasps follow the dynamic variations of articulated object joints. In this way, our SynHLMA achieves three typical hand manipulation tasks for articulated objects of HAOI generation, HAOI prediction and HAOI interpolation. We evaluate SynHLMA on our built HAOI-lang dataset and experimental results demonstrate the superior hand grasp sequence generation performance comparing with state-of-the-art. We also show a robotics grasp application that enables dexterous grasps execution from imitation learning using the manipulation sequence provided by our SynHLMA. Our codes and datasets will be made publicly available.
♻ ☆ StructBiHOI: Structured Articulation Modeling for Long--Horizon Bimanual Hand--Object Interaction Generation
Recent progress in 3D hand--object interaction (HOI) generation has primarily focused on single--hand grasp synthesis, while bimanual manipulation remains significantly more challenging. Long--horizon planning instability, fine--grained joint articulation, and complex cross--hand coordination make coherent bimanual generation difficult, especially under multimodal conditions. Existing approaches often struggle to simultaneously ensure temporal consistency, physical plausibility, and semantic alignment over extended sequences. We propose StructBiHOI, a Structured articulation modeling framework for long-horizon Bimanual HOI generation. Our key insight is to structurally disentangle temporal joint planning from frame--level manipulation refinement. Specifically, a jointVAE models long-term joint evolution conditioned on object geometry and task semantics, while a maniVAE refines fine-grained hand poses at the single--frame level. To enable stable and efficient long--sequence generation, we incorporate a state--space--inspired diffusion denoiser based on Mamba, which models long--range dependencies with linear complexity. This hierarchical design facilitates coherent dual-hand coordination and articulated object interaction. Extensive experiments on bimanual manipulation and single-hand grasping benchmarks demonstrate that our method achieves superior long--horizon stability, motion realism, and computational efficiency compared to strong baselines.
♻ ☆ Morphological-Symmetry-Equivariant Heterogeneous Graph Neural Network for Robotic Dynamics Learning
We present a morphological-symmetry-equivariant heterogeneous graph neural network, namely MS-HGNN, for robotic dynamics learning, that integrates robotic kinematic structures and morphological symmetries into a single graph network. These structural priors are embedded into the learning architecture as constraints, ensuring high generalizability, sample and model efficiency. The proposed MS-HGNN is a versatile and general architecture that is applicable to various multi-body dynamic systems and a wide range of dynamics learning problems. We formally prove the morphological-symmetry-equivariant property of our MS-HGNN and validate its effectiveness across multiple quadruped robot learning problems using both real-world and simulated data. Our code is made publicly available at https://github.com/lunarlab-gatech/MorphSym-HGNN/.
♻ ☆ From Spatial to Actions: Grounding Vision-Language-Action Model in Spatial Foundation Priors ICLR 2026
Existing vision-language-action (VLA) models act in 3D real-world but are typically built on 2D encoders, leaving a spatial reasoning gap that limits generalization and adaptability. Recent 3D integration techniques for VLAs either require specialized sensors and transfer poorly across modalities, or inject weak cues that lack geometry and degrade vision-language alignment. In this work, we introduce FALCON (From Spatial to Action), a novel paradigm that injects rich 3D spatial tokens into the action head. FALCON leverages spatial foundation models to deliver strong geometric priors from RGB alone, and includes an Embodied Spatial Model that can optionally fuse depth, or pose for higher fidelity when available, without retraining or architectural changes. To preserve language reasoning, spatial tokens are consumed by a Spatial-Enhanced Action Head rather than being concatenated into the vision-language backbone. These designs enable FALCON to address limitations in spatial representation, modality transferability, and alignment. In comprehensive evaluations across three simulation benchmarks and eleven real-world tasks, our proposed FALCON achieves state-of-the-art performance, consistently surpasses competitive baselines, and remains robust under clutter, spatial-prompt conditioning, and variations in object scale and height.
comment: Accepted at ICLR 2026. Project page: https://falcon-vla.github.io/
♻ ☆ Automated Coral Spawn Monitoring for Reef Restoration: The Coral Spawn and Larvae Imaging Camera System (CSLICS)
Coral aquaculture for reef restoration requires accurate and continuous spawn counting for resource distribution and larval health monitoring, but current methods are labor-intensive and represent a critical bottleneck in the coral production pipeline. We propose the Coral Spawn and Larvae Imaging Camera System (CSLICS), which uses low cost modular cameras and object detectors trained using human-in-the-loop labeling approaches for automated spawn counting in larval rearing tanks. This paper details the system engineering, dataset collection, and computer vision techniques to detect, classify and count coral spawn. Experimental results from mass spawning events demonstrate an F1 score of 82.4% for surface spawn detection at different embryogenesis stages, 65.3% F1 score for sub-surface spawn detection, and a saving of 5,720 hours of labor per spawning event compared to manual sampling methods at the same frequency. Comparison of manual counts with CSLICS monitoring during a mass coral spawning event on the Great Barrier Reef demonstrates CSLICS' accurate measurement of fertilization success and sub-surface spawn counts. These findings enhance the coral aquaculture process and enable upscaling of coral reef restoration efforts to address climate change threats facing ecosystems like the Great Barrier Reef.
comment: 8 pages, 7 figures, accepted for presentation at the IEEE International Conference on Robotics and Automation, 2026
♻ ☆ A 26-Gram Butterfly-Inspired Robot Achieving Autonomous Tailless Flight
The flight of biological butterflies represents a unique aerodynamic regime where high-amplitude, low-frequency wingstrokes induce significant body undulations and inertial fluctuations. While existing tailless flapping-wing micro air vehicles typically employ high-frequency kinematics to minimize such perturbations, the lepidopteran flight envelope remains a challenging and underexplored frontier for autonomous robotics. Here, we present \textit{AirPulse}, a 26-gram butterfly-inspired robot that achieves the first onboard, closed-loop controlled flight for a tailless two-winged platform at this scale. It replicates key biomechanical traits of butterfly flight, utilizing low-aspect-ratio, compliant carbon-fiber-reinforced wings and low-frequency flapping that reproduces characteristic biological body undulations. Leveraging a quantitative mapping of control effectiveness, we introduce a hierarchical control architecture featuring state estimator, attitude controller, and central pattern generator with Stroke Timing Asymmetry Rhythm (STAR), which translates attitude control demands into smooth and stable wingstroke timing and angle-offset modulations. Free-flight experiments demonstrate stable climbing and directed turning maneuvers, proving that autonomous locomotion is achievable even within oscillatory dynamical regimes. By bridging biological morphology with a minimalist control architecture, \textit{AirPulse} serves as both a hardware-validated model for decoding butterfly flight dynamics and a prototype for a new class of collision-resilient aerial robots. Its lightweight and compliant structure offers a non-invasive solution for a wide range of applications, such as ecological monitoring and confined-space inspection, where traditional drones may fall short.
♻ ☆ Enhancing Heterogeneous Multi-Agent Cooperation in Decentralized MARL via GNN-driven Intrinsic Rewards AAMAS 2025
Multi-agent Reinforcement Learning (MARL) is emerging as a key framework for various sequential decision-making and control tasks. Unlike their single-agent counterparts, multi-agent systems necessitate successful cooperation among the agents. The deployment of these systems in real-world scenarios often requires decentralized training, a diverse set of agents, and learning from infrequent environmental reward signals. These challenges become more pronounced under partial observability and the lack of prior knowledge about agent heterogeneity. While notable studies use intrinsic motivation (IM) to address reward sparsity or cooperation in decentralized settings, those dealing with heterogeneity typically assume centralized training, parameter sharing, and agent indexing. To overcome these limitations, we propose the CoHet algorithm, which utilizes a novel Graph Neural Network (GNN) based intrinsic motivation to facilitate the learning of heterogeneous agent policies in decentralized settings, under the challenges of partial observability and reward sparsity. Evaluation of CoHet in the Multi-agent Particle Environment (MPE) and Vectorized Multi-Agent Simulator (VMAS) benchmarks demonstrates superior performance compared to the state-of-the-art in a range of cooperative multi-agent scenarios. Our research is supplemented by an analysis of the impact of the agent dynamics model on the intrinsic motivation module, insights into the performance of different CoHet variants, and its robustness to an increasing number of heterogeneous agents.
comment: Full paper version for AAMAS 2025, 9 pages, 5 figures
♻ ☆ You Only Pose Once: A Minimalist's Detection Transformer for Monocular RGB Category-level 9D Multi-Object Pose Estimation ICRA 2026
Accurately recovering the full 9-DoF pose of unseen instances within specific categories from a single RGB image remains a core challenge for robotics and automation. Most existing solutions still rely on pseudo-depth, CAD models, or multi-stage cascades that separate 2D detection from pose estimation. Motivated by the need for a simpler, RGB-only alternative that learns directly at the category level, we revisit a longstanding question: Can object detection and 9-DoF pose estimation be unified with high performance, without any additional data? We show that they can with our method, YOPO, a single-stage, query-based framework that treats category-level 9-DoF estimation as a natural extension of 2D detection. YOPO augments a transformer detector with a lightweight pose head, a bounding-box-conditioned translation module, and a 6D-aware Hungarian matching cost. The model is trained end-to-end only with RGB images and category-level pose labels. Despite its minimalist design, YOPO sets a new state of the art on three benchmarks. On the REAL275 dataset, it achieves 79.6% $\rm{IoU}_{50}$ and 54.1% under the $10^\circ$$10{\rm{cm}}$ metric, surpassing prior RGB-only methods and closing much of the gap to RGB-D systems. The code, models, and additional qualitative results can be found on https://mikigom.github.io/YOPO-project-page.
comment: This paper has been accepted by IEEE ICRA 2026
♻ ☆ Image Compression Using Novel View Synthesis Priors
Real-time visual feedback is essential for tetherless control of remotely operated vehicles, particularly during inspection and manipulation tasks. Though acoustic communication is the preferred choice for medium-range communication underwater, its limited bandwidth renders it impractical to transmit images or videos in real-time. To address this, we propose a model-based image compression technique that leverages prior mission information. Our approach employs trained machine-learning based novel view synthesis models, and uses gradient descent optimization to refine latent representations to help generate compressible differences between camera images and rendered images. We evaluate the proposed compression technique using a dataset from an artificial ocean basin, demonstrating superior compression ratios and image quality over existing techniques. Moreover, our method exhibits robustness to introduction of new objects within the scene, highlighting its potential for advancing tetherless remotely operated vehicle operations.
comment: Preprint submitted to IEEE Journal of Oceanic Engineering (v2.0)
♻ ☆ RL-100: Performant Robotic Manipulation with Real-World Reinforcement Learning
Real-world robotic manipulation in homes and factories demands reliability, efficiency, and robustness that approach or surpass those of skilled human operators. We present RL-100, a real-world reinforcement learning framework built on diffusion visuomotor policies. RL-100 unifies imitation and reinforcement learning under a single clipped PPO surrogate objective applied within the denoising process, yielding conservative and stable improvements across offline and online stages. To meet deployment latency requirements, a lightweight consistency distillation method compresses multi-step diffusion into a one-step controller for high-frequency control. The framework is task-, embodiment-, and representation-agnostic, and supports both single-action and action-chunking control. We evaluate RL-100 on eight diverse real-robot tasks, from dynamic pushing and agile bowling to pouring, cloth folding, unscrewing, multi-stage juicing, and long-horizon box folding. RL-100 attains 100 percent success across evaluated trials, for a total of 1000 out of 1000 episodes, including up to 250 out of 250 consecutive trials on one task. It matches or surpasses expert teleoperators in time to completion. Without retraining, a single policy attains approximately 90 percent zero-shot success under environmental and dynamics shifts, adapts in a few-shot regime to significant task variations (86.7 percent), and remains robust to aggressive human perturbations (about 96 percent). Notably, our juicing robot served random customers continuously for about seven hours without failure when deployed zero-shot in a shopping mall. These results suggest a practical path to deployment-ready robot learning by starting from human priors, aligning training objectives with human-grounded metrics, and reliably extending performance beyond human demonstrations.
comment: https://lei-kun.github.io/RL-100/
♻ ☆ Revisiting Replanning from Scratch: Real-Time Incremental Planning with Fast Almost-Surely Asymptotically Optimal Planners ICRA
Robots operating in changing environments either predict obstacle changes and/or plan quickly enough to react to them. Predictive approaches require a strong prior about the position and motion of obstacles. Reactive approaches require no assumptions about their environment but must replan quickly and find high-quality paths to navigate effectively. Reactive approaches often reuse information between queries to reduce planning cost. These techniques are conceptually sound but updating dense planning graphs when information changes can be computationally prohibitive. It can also require significant effort to detect the changes in some applications. This paper revisits the long-held assumption that reactive replanning requires updating existing plans. It shows that the incremental planning problem can alternatively be solved more efficiently as a series of independent problems using fast almost-surely asymptotically optimal (ASAO) planning algorithms. These ASAO algorithms quickly find an initial solution and converge towards an optimal solution which allows them to find consistent global plans in the presence of changing obstacles without requiring explicit plan reuse. This is demonstrated with simulated experiments where Effort Informed Trees (EIT*) finds shorter median solution paths than the tested reactive planning algorithms and is further validated using Asymptotically Optimal RRT-Connect (AORRTC) on a real-world planning problem on a robot arm.
comment: IEEE International Conference on Robotics and Automation (ICRA) 2026, 8 pages, 5 figures, 1 table. A video of this work can be found at https://www.youtube.com/watch?v=XaZrFy8wGZs
♻ ☆ RoboRouter: Training-Free Policy Routing for Robotic Manipulation
Research on robotic manipulation has developed a diverse set of policy paradigms, including vision-language-action (VLA) models, vision-action (VA) policies, and code-based compositional approaches. Concrete policies typically attain high success rates on specific task distributions but lim-ited generalization beyond it. Rather than proposing an other monolithic policy, we propose to leverage the complementary strengths of existing approaches through intelligent policy routing. We introduce RoboRouter, a training-free framework that maintains a pool of heterogeneous policies and learns to select the best-performing policy for each task through accumulated execution experience. Given a new task, RoboRouter constructs a semantic task representation, retrieves historical records of similar tasks, predicts the optimal policy choice without requiring trial-and-error, and incorporates structured feedback to refine subsequent routing decisions. Integrating a new policy into the system requires only lightweight evaluation and incurs no training overhead. Across simulation benchmark and real-world evaluations, RoboRouter consistently outperforms than in-dividual policies, improving average success rate by more than 3% in simulation and over 13% in real-world settings, while preserving execution efficiency. Our results demonstrate that intelligent routing across heterogeneous, off-the-shelf policies provides a practical and scalable pathway toward building more capable robotic systems.
comment: We need to withdraw the paper as some of the reference papers are incorrect and need to be removed
♻ ☆ Multimodal Adversarial Quality Policy for Safe Grasping
Vision-guided robot grasping based on Deep Neural Networks (DNNs) generalizes well but poses safety risks in the Human-Robot Interaction (HRI). Recent works solved it by designing benign adversarial attacks and patches with RGB modality, yet depth-independent characteristics limit their effectiveness on RGBD modality. In this work, we propose the Multimodal Adversarial Quality Policy (MAQP) to realize multimodal safe grasping. Our framework introduces two key components. First, the Heterogeneous Dual-Patch Optimization Scheme (HDPOS) mitigates the distribution discrepancy between RGB and depth modalities in patch generation by adopting modality-specific initialization strategies, employing a Gaussian distribution for depth patches and a uniform distribution for RGB patches, while jointly optimizing both modalities under a unified objective function. Second, the Gradient-Level Modality Balancing Strategy (GLMBS) is designed to resolve the optimization imbalance from RGB and Depth patches in patch shape adaptation by reweighting gradient contributions based on per-channel sensitivity analysis and applying distance-adaptive perturbation bounds. We conduct extensive experiments on the benchmark datasets and a cobot, showing the effectiveness of MAQP.
comment: submitted
♻ ☆ Pri4R: Learning World Dynamics for Vision-Language-Action Models with Privileged 4D Representation
Humans learn not only how their bodies move, but also how the surrounding world responds to their actions. In contrast, while recent Vision-Language-Action (VLA) models exhibit impressive semantic understanding, they often fail to capture the spatiotemporal dynamics governing physical interaction. In this paper, we introduce Pri4R, a simple yet effective approach that endows VLA models with an implicit understanding of world dynamics by leveraging privileged 4D information during training. Specifically, Pri4R augments VLAs with a lightweight point track head that predicts 3D point tracks. By injecting VLA features into this head to jointly predict future 3D trajectories, the model learns to incorporate evolving scene geometry within its shared representation space, enabling more physically aware context for precise control. Due to its architectural simplicity, Pri4R is compatible with dominant VLA design patterns with minimal changes. During inference, we run the model using the original VLA architecture unchanged; Pri4R adds no extra inputs, outputs, or computational overhead. Across simulation and real-world evaluations, Pri4R significantly improves performance on challenging manipulation tasks, including a +10% gain on LIBERO-Long and a +40% gain on RoboCasa. We further show that 3D point track prediction is an effective supervision target for learning action-world dynamics, and validate our design choices through extensive ablations. Project page: https://jiiiisoo.github.io/Pri4R/
♻ ☆ Vectorized Online POMDP Planning ICRA 2026
Planning under partial observability is an essential capability of autonomous robots. The Partially Observable Markov Decision Process (POMDP) provides a powerful framework for planning under partial observability problems, capturing the stochastic effects of actions and the limited information available through noisy observations. POMDP solving could benefit tremendously from massive parallelization on today's hardware, but parallelizing POMDP solvers has been challenging. Most solvers rely on interleaving numerical optimization over actions with the estimation of their values, which creates dependencies and synchronization bottlenecks between parallel processes that can offset the benefits of parallelization. In this paper, we propose Vectorized Online POMDP Planner (VOPP), a novel parallel online solver that leverages a recent POMDP formulation which analytically solves part of the optimization component, leaving numerical computations to consist of only estimation of expectations. VOPP represents all data structures related to planning as a collection of tensors, and implements all planning steps as fully vectorized computations over this representation. The result is a massively parallel online solver with no dependencies or synchronization bottlenecks between concurrent processes. Experimental results indicate that VOPP is at least $20\times$ more efficient in computing near-optimal solutions compared to an existing state-of-the-art parallel online solver. Moreover, VOPP outperforms state-of-the-art sequential online solvers, while using a planning budget that is $1000\times$ smaller.
comment: 8 pages, 3 figures. Accepted at ICRA 2026
♻ ☆ NavSpace: How Navigation Agents Follow Spatial Intelligence Instructions ICRA 2026
Instruction-following navigation is a key step toward embodied intelligence. Prior benchmarks mainly focus on semantic understanding but overlook systematically evaluating navigation agents' spatial perception and reasoning capabilities. In this work, we introduce the NavSpace benchmark, which contains six task categories and 1,228 trajectory-instruction pairs designed to probe the spatial intelligence of navigation agents. On this benchmark, we comprehensively evaluate 22 navigation agents, including state-of-the-art navigation models and multimodal large language models. The evaluation results lift the veil on spatial intelligence in embodied navigation. Furthermore, we propose SNav, a new spatially intelligent navigation model. SNav outperforms existing navigation agents on NavSpace and real robot tests, establishing a strong baseline for future work.
comment: ICRA 2026
♻ ☆ EgoMI: Learning Active Vision and Whole-Body Manipulation from Egocentric Human Demonstrations
Imitation learning from human demonstrations offers a promising approach for robot skill acquisition, but egocentric human data introduces fundamental challenges due to the embodiment gap. During manipulation, humans actively coordinate head and hand movements, continuously reposition their viewpoint and use pre-action visual fixation search strategies to locate relevant objects. These behaviors create dynamic, task-driven head motions that static robot sensing systems cannot replicate, leading to a significant distribution shift that degrades policy performance. We present EgoMI (Egocentric Manipulation Interface), a framework that captures synchronized end-effector and active head trajectories during manipulation tasks, resulting in data that can be retargeted to compatible semi-humanoid robot embodiments. To handle rapid and wide-spanning head viewpoint changes, we introduce a memory-augmented policy that selectively incorporates historical observations. We evaluate our approach on a bimanual robot equipped with an actuated camera head and find that policies with explicit head-motion modeling consistently outperform baseline methods. Results suggest that coordinated hand-eye learning with EgoMI effectively bridges the human-robot embodiment gap for robust imitation learning on semi-humanoid embodiments. Project page: https://egocentric-manipulation-interface.github.io
♻ ☆ Score Matching Diffusion Based Feedback Control and Planning of Nonlinear Systems
In this paper, we propose a deterministic diffusion-based framework for controlling the probability density of nonlinear control-affine systems, with theoretical guarantees for drift-free and linear time-invariant (LTI) dynamics. The central idea is to first excite the system with white noise so that a forward diffusion process explores the reachable regions of state space, and then to design a deterministic feedback law that acts as a denoising mechanism driving the system back toward a desired target distribution supported on the target set. This denoising phase provides a feedback controller that steers the control system to the target set. In this framework, control synthesis reduces to constructing a deterministic reverse process that reproduces the desired evolution of state densities. We derive existence conditions ensuring such deterministic realizations of time-reversals for controllable drift-free and LTI systems, and show that the resulting feedback laws provide a tractable alternative to nonlinear control by viewing density control as a relaxation of controlling a system to target sets. Numerical studies on a unicycle model with obstacles, a five-dimensional driftless system, and a four-dimensional LTI system demonstrate reliable diffusion-inspired density control.
♻ ☆ Automated Layout and Control Co-Design of Robust Multi-UAV Transportation Systems
The joint optimization of physical parameters and controllers in robotic systems is challenging. This is due to the difficulties of predicting the effect that changes in physical parameters have on final performances. At the same time, physical and morphological modifications can improve robot capabilities, perhaps completely unlocking new skills and tasks. We present a novel approach to co-optimize the physical layout and the control of a cooperative aerial transportation system. The goal is to achieve the most precise and robust flight when carrying a payload. We assume the agents are connected to the payload through rigid attachments, essentially transforming the whole system into a larger flying object with ``thrust modules" at the attachment locations of the quadcopters. We investigate the optimal arrangement of the thrust modules around the payload, so that the resulting system achieves the best disturbance rejection capabilities. We propose a novel metric of robustness inspired by H2 control, and propose an algorithm to optimize the layout of the vehicles around the object and their controller altogether. We experimentally validate the effectiveness of our approach using fleets of three and four quadcopters and payloads of diverse shapes.
comment: 7 pages, 7 figures, journal paper (IEEE RA-L)
♻ ☆ Global End-Effector Pose Control of an Underactuated Aerial Manipulator via Reinforcement Learning ICRA 2026
Aerial manipulators, which combine robotic arms with multi-rotor drones, face strict constraints on arm weight and mechanical complexity. In this work, we study a lightweight 2-degree-of-freedom (DoF) arm mounted on a quadrotor via a differential mechanism, capable of full six-DoF end-effector pose control. While the minimal design enables simplicity and reduced payload, it also introduces challenges such as underactuation and sensitivity to external disturbances. To address these, we employ reinforcement learning, training a Proximal Policy Optimization (PPO) agent in simulation to generate feedforward commands for quadrotor acceleration and body rates, along with joint angle targets. These commands are tracked by an incremental nonlinear dynamic inversion (INDI) attitude controller and a PID joint controller, respectively. Flight experiments demonstrate centimeter-level position accuracy and degree-level orientation precision, with robust performance under external force disturbances, including manipulation of heavy loads and pushing tasks. The results highlight the potential of learning-based control strategies for enabling contact-rich aerial manipulation using simple, lightweight platforms. Videos of the experiment and the method are summarized in https://youtu.be/bWLTPqKcCOA.
comment: 8 pages, 6 figures, accepted by IEEE ICRA 2026
♻ ☆ World Models That Know When They Don't Know - Controllable Video Generation with Calibrated Uncertainty
Recent advances in generative video models have led to significant breakthroughs in high-fidelity video synthesis, specifically in controllable video generation where the generated video is conditioned on text and action inputs, e.g., in instruction-guided video editing and world modeling in robotics. Despite these exceptional capabilities, controllable video models often hallucinate - generating future video frames that are misaligned with physical reality - which raises serious concerns in many tasks such as robot policy evaluation and planning. However, state-of-the-art video models lack the ability to assess and express their confidence, impeding hallucination mitigation. To rigorously address this challenge, we propose C3, an uncertainty quantification (UQ) method for training continuous-scale calibrated controllable video models for dense confidence estimation at the subpatch level, precisely localizing the uncertainty in each generated video frame. Our UQ method introduces three core innovations to empower video models to estimate their uncertainty. First, our method develops a novel framework that trains video models for correctness and calibration via strictly proper scoring rules. Second, we estimate the video model's uncertainty in latent space, avoiding training instability and prohibitive training costs associated with pixel-space approaches. Third, we map the dense latent-space uncertainty to interpretable pixel-level uncertainty in the RGB space for intuitive visualization, providing high-resolution uncertainty heatmaps that identify untrustworthy regions. Through extensive experiments on large-scale robot learning datasets (Bridge and DROID) and real-world evaluations, we demonstrate that our method not only provides calibrated uncertainty estimates within the training distribution, but also enables effective out-of-distribution detection.
♻ ☆ Dull, Dirty, Dangerous: Understanding the Past, Present, and Future of a Key Motivation for Robotics
In robotics, the concept of "dull, dirty, and dangerous" (DDD) work has been used to motivate where robots might be useful. In this paper, we conduct an empirical analysis of robotics publications between 1980 and 2024 that mention DDD, and find that only 2.7% of publications define DDD and 8.7% of publications provide concrete examples of tasks or jobs that are DDD. We then review the social science literature on "dull," "dirty," and "dangerous" work to provide definitions and guidance on how to conceptualize DDD for robotics. Finally, we propose a framework that helps the robotics community consider the job context for our technology, encouraging a more informed perspective on how robotics may impact human labor.
♻ ☆ REI-Bench: Can Embodied Agents Understand Vague Human Instructions in Task Planning? ICLR 2026
Robot task planning decomposes human instructions into executable action sequences that enable robots to complete a series of complex tasks. Although recent large language model (LLM)-based task planners achieve amazing performance, they assume that human instructions are clear and straightforward. However, real-world users are not experts, and their instructions to robots often contain significant vagueness. Linguists suggest that such vagueness frequently arises from referring expressions (REs), whose meanings depend heavily on dialogue context and environment. This vagueness is even more prevalent among the elderly and children, who are the groups that robots should serve more. This paper studies how such vagueness in REs within human instructions affects LLM-based robot task planning and how to overcome this issue. To this end, we propose the first robot task planning benchmark that systematically models vague REs grounded in pragmatic theory (REI-Bench), where we discover that the vagueness of REs can severely degrade robot planning performance, leading to success rate drops of up to 36.9%. We also observe that most failure cases stem from missing objects in planners. To mitigate the REs issue, we propose a simple yet effective approach: task-oriented context cognition, which generates clear instructions for robots, achieving state-of-the-art performance compared to aware prompts, chains of thought, and in-context learning. By tackling the overlooked issue of vagueness, this work contributes to the research community by advancing real-world task planning and making robots more accessible to non-expert users, e.g., the elderly and children.
comment: Accepted at ICLR 2026
♻ ☆ Safe and Optimal Learning from Preferences via Weighted Temporal Logic with Applications in Robotics and Formula 1
Autonomous systems increasingly rely on human feedback to align their behavior, expressed as pairwise comparisons, rankings, or demonstrations. While existing methods can adapt behaviors, they often fail to guarantee safety in safety-critical domains. We propose a safety-guaranteed, optimal, and efficient approach for solving the learning problem from preferences, rankings, or demonstrations using Weighted Signal Temporal Logic (WSTL). WSTL learning problems, when implemented naively, lead to multi-linear constraints in the weights to be learned. By introducing structural pruning and log-transform procedures, we reduce the problem size and recast it as a Mixed-Integer Linear Program while preserving safety guarantees. Experiments on robotic navigation and real-world Formula 1 data demonstrate that the method captures nuanced preferences and models complex task objectives.
comment: 8 pages, 2 figures
♻ ☆ NaviGait: Navigating Dynamically Feasible Gait Libraries using Deep Reinforcement Learning
Reinforcement learning (RL) has emerged as a powerful method to learn robust control policies for bipedal locomotion. Yet, it can be difficult to tune desired robot behaviors due to unintuitive and complex reward design. In comparison, trajectory optimization-based methods offer more tuneable, interpretable, and mathematically grounded motion plans for high-dimensional legged systems. However, these methods often remain brittle to real-world disturbances like external perturbations. In this work, we present NaviGait, a hierarchical framework that combines the structure of trajectory optimization with the adaptability of RL for robust and intuitive locomotion control. NaviGait leverages RL to synthesize new motions by selecting, minimally morphing, and stabilizing gaits taken from an offline-generated gait library. NaviGait results in walking policies that match the reference motion well while maintaining robustness comparable to other locomotion controllers. Additionally, the structure imposed by NaviGait drastically simplifies the RL reward composition. Our experimental results demonstrate that NaviGait enables faster training compared to conventional and imitation-based RL, and produces motions that remain closest to the original reference. Overall, by decoupling high-level motion generation from low-level correction, NaviGait offers a more scalable and generalizable approach for achieving dynamic and robust locomotion. Videos and the full framework are publicly available at https://dynamicmobility.github.io/navigait/
comment: Accepted to the International Conference on Robotics and Automation (2026). 8 pages, 9 figures
♻ ☆ A Distributional Treatment of Real2Sim2Real for Object-Centric Agent Adaptation in Vision-Driven Deformable Linear Object Manipulation
We present an integrated (or end-to-end) framework for the Real2Sim2Real problem of manipulating deformable linear objects (DLOs) based on visual perception. Working with a parameterised set of DLOs, we use likelihood-free inference (LFI) to compute the posterior distributions for the physical parameters using which we can approximately simulate the behaviour of each specific DLO. We use these posteriors for domain randomisation while training, in simulation, object-specific visuomotor policies (i.e. assuming only visual and proprioceptive sensory) for a DLO reaching task, using model-free reinforcement learning. We demonstrate the utility of this approach by deploying sim-trained DLO manipulation policies in the real world in a zero-shot manner, i.e. without any further fine-tuning. In this context, we evaluate the capacity of a prominent LFI method to perform fine classification over the parametric set of DLOs, using only visual and proprioceptive data obtained in a dynamic manipulation trajectory. We then study the implications of the resulting domain distributions in sim-based policy learning and real-world performance.
♻ ☆ Physics-Conditioned Grasping for Stable Tool Use
Tool use often fails not because robots misidentify tools, but because grasps cannot withstand task-induced wrench. Existing vision-language manipulation systems ground tools and contact regions from language yet select grasps under quasi-static or geometry-only assumptions. During interaction, inertial impulse and lever-arm amplification generate wrist torque and tangential loads that trigger slip and rotation. We introduce inverse Tool-use Planning (iTuP), which selects grasps by minimizing predicted interaction wrench along a task-conditioned trajectory. From rigid-body mechanics, we derive torque, slip, and alignment penalties, and train a Stable Dynamic Grasp Network (SDG-Net) to approximate these trajectory-conditioned costs for real-time scoring. Across hammering, sweeping, knocking, and reaching in simulation and on hardware, SDG-Net suppresses induced torque up to 17.6%, shifts grasps below empirically observed instability thresholds, and improves real-world success by 17.5% over a compositional baseline. Improvements concentrate where wrench amplification dominates, showing that robot tool use requires wrench-aware grasp selection, not perception alone.
comment: In submission and under review
♻ ☆ Asset-Centric Metric-Semantic Maps of Indoor Environments
Large Language Models (LLMs) can help robots reason about abstract task specifications. This requires augmenting classical representations of the environment used by robots, such as point-clouds and meshes, with natural language-based priors. There are a number of approaches to do so in the existing literature. While some navigation frameworks leverage scene-level semantics at the expense of object-level detail, others such as language-guided neural radiance fields (NeRFs) or segment-anything 3D (SAM3D) prioritize object accuracy over global scene context. This paper argues that we can get the best of both worlds. We use a Unitree Go2 quadruped with a RealSense stereo camera (RGB-D data) to build an explicit metric-semantic representation of indoor environments. This is a scene-scale representation with each object (e.g., chairs, couches, doors, of various shapes and sizes) represented by a detailed mesh, its category, and a pose. We show that this representation is more accurate than foundation-model-based maps such as those built by SAM3D, as well as state-of-the-art scene-level robotics mapping pipelines such as Clio (Maggio et al., 2024). Our implementation is about 25$\times$ faster than SAM3D and is about 10$\times$ slower than Clio. We can also adapt our approach to enable open-set scene-level mapping, i.e., when object meshes are not known a priori, by building upon SAM3D to further improve precision and recall. We show how this representation can be readily used with LLMs such as Google's Gemini to demonstrate scene understanding, complex inferences, and planning. We also display the utility of having these representations for semantic navigation in simulated warehouse and hospital settings using Nvidia's Issac Sim.
comment: 9 pages, 8 figures, 3 tables
♻ ☆ Robot Control Stack: A Lean Ecosystem for Robot Learning at Scale ICRA 2026
Vision-Language-Action models (VLAs) mark a major shift in robot learning. They replace specialized architectures and task-tailored components of expert policies with large-scale data collection and setup-specific fine-tuning. In this machine learning-focused workflow that is centered around models and scalable training, traditional robotics software frameworks become a bottleneck, while robot simulations offer only limited support for transitioning from and to real-world experiments. In this work, we close this gap by introducing Robot Control Stack (RCS), a lean ecosystem designed from the ground up to support research in robot learning with large-scale generalist policies. At its core, RCS features a modular and easily extensible layered architecture with a unified interface for simulated and physical robots, facilitating sim-to-real transfer. Despite its minimal footprint and dependencies, it offers a complete feature set, enabling both real-world experiments and large-scale training in simulation. Our contribution is twofold: First, we introduce the architecture of RCS and explain its design principles. Second, we evaluate its usability and performance along the development cycle of VLA and RL policies. Our experiments also provide an extensive evaluation of Octo, OpenVLA, and Pi Zero on multiple robots and shed light on how simulation data can improve real-world policy performance. Our code, datasets, weights, and videos are available at: https://robotcontrolstack.github.io/
comment: Accepted at ICRA 2026
♻ ☆ Relative Localization System Design for SnailBot: A Modular Self-reconfigurable Robot
This paper presents the design and implementation of a relative localization system for SnailBot, a modular self reconfigurable robot. The system integrates ArUco marker recognition, optical flow analysis, and IMU data processing into a unified fusion framework, enabling robust and accurate relative positioning for collaborative robotic tasks. Experimental validation demonstrates the effectiveness of the system in realtime operation, with a rule based fusion strategy ensuring reliability across dynamic scenarios. The results highlight the potential for scalable deployment in modular robotic systems.
comment: The paper contains factual error and logic flaws, which needs to be repaired before submitting
♻ ☆ Learning responsibility allocations for multi-agent interactions: A differentiable optimization approach with control barrier functions
From autonomous driving to package delivery, ensuring safe yet efficient multi-agent interaction is challenging as the interaction dynamics are influenced by hard-to-model factors such as social norms and contextual cues. Understanding these influences can aid in the design and evaluation of socially-aware autonomous agents whose behaviors are aligned with human values. In this work, we seek to codify factors governing safe multi-agent interactions via the lens of responsibility, i.e., an agent's willingness to deviate from their desired control to accommodate safe interaction with others. Specifically, we propose a data-driven modeling approach based on control barrier functions and differentiable optimization that efficiently learns agents' responsibility allocation from data. We demonstrate on synthetic and real-world datasets that we can obtain an interpretable and quantitative understanding of how much agents adjust their behavior to ensure the safety of others given their current environment.
comment: 8 pages, 7 figures
♻ ☆ UniBYD: A Unified Framework for Learning Robotic Manipulation Across Embodiments Beyond Imitation of Human Demonstrations
In embodied intelligence, the embodiment gap between robotic and human hands brings significant challenges for learning from human demonstrations. Although some studies have attempted to bridge this gap using reinforcement learning, they remain confined to merely reproducing human manipulation, resulting in limited task performance. Moreover, current methods struggle to support diverse robotic hand configurations. In this paper, we propose UniBYD, a unified framework that uses a dynamic reinforcement learning algorithm to discover manipulation policies aligned with the robot's physical characteristics. To enable consistent modeling across diverse robotic hand morphologies, UniBYD incorporates a unified morphological representation (UMR). Building on UMR, we design a dynamic PPO with an annealed reward schedule, enabling reinforcement learning to transition from offline-informed imitation of human demonstrations to online-adaptive exploration of policies better adapted to diverse robotic morphologies, thereby going beyond mere imitation of human hands. To address the severe state drift caused by the incapacity of early-stage policies, we design a hybrid Markov-based shadow engine that provides fine-grained guidance to anchor the imitation within the expert's manifold. To evaluate UniBYD, we propose UniManip, the first benchmark for cross-embodiment manipulation spanning diverse robotic morphologies. Experiments demonstrate a 44.08% average improvement in success rate over the current state-of-the-art. Upon acceptance, we will release our code and benchmark.
♻ ☆ Reactive Slip Control in Multifingered Grasping: Hybrid Tactile Sensing and Internal-Force Optimization ICRA
We present a hybrid learning and model-based approach for reactive internal-force adaptation to halt in-hand slip in a multifingered robotic gripper. A multimodal tactile stack combines piezoelectric (PzE) sensing for fast slip cues with piezoresistive (PzR) arrays for contact localization, enabling online construction of the grasp matrix. Upon slip detection, internal forces are updated in the null space of the grasp through a quadratic program that reinforces normal forces while preserving the object wrench. We demonstrate reactive stabilization of multifingered grasps under external perturbations. Augmenting analytic force control with learned tactile cues enables fast and reliable closed-loop stabilization in the evaluated grasp scenarios. The pipeline yields a theoretical sensing-to-command latency of 35-40 ms, including 5 ms for PzR-based grasp geometry updates and approximately 4 ms for solving the quadratic program. In controlled trials, slip onset is detected after ~ 20 ms. The analysis supports the feasibility of sub-50 ms integrated closed-loop stabilization.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA), 2026
♻ ☆ Unveiling the Potential of iMarkers: Invisible Fiducial Markers for Advanced Robotics
Fiducial markers are widely used in robotics for navigation, object recognition, and scene understanding. While offering significant advantages for robots and Augmented Reality (AR) applications, they often disrupt the visual aesthetics of environments, as they are visible to humans, making them unsuitable for many everyday use cases. To address this gap, this paper presents iMarkers, innovative, unobtrusive fiducial markers detectable exclusively by robots and AR devices equipped with adequate sensors and detection algorithms. These markers offer high flexibility in production, allowing customization of their visibility range and encoding algorithms to suit various demands. The paper also introduces the hardware designs and open-sourced software algorithms developed for detecting iMarkers, highlighting their adaptability and robustness in the detection and recognition stages. Numerous evaluations have demonstrated the effectiveness of iMarkers relative to conventional (printed) and blended fiducial markers and have confirmed their applicability across diverse robotics scenarios.
comment: 19 pages, 10 figures, 4 tables
♻ ☆ Latent Policy Steering with Embodiment-Agnostic Pretrained World Models
The performance of learned robot visuomotor policies is heavily dependent on the size and quality of the training dataset. Although large-scale robot and human datasets are increasingly available, embodiment gaps and mismatched action spaces make them difficult to leverage. Our main insight is that skills performed across different embodiments produce visual similarities in motions that can be captured using off-the-shelf action representations such as optical flow. Moreover, World Models (WMs) can leverage sub-optimal data since they focus on modeling dynamics. In this work, we aim to improve visuomotor policies in low-data regimes by first pretraining a WM using optical flow as an embodiment-agnostic action representation to leverage accessible or easily collected data from multiple embodiments (robots, humans). Given a small set of demonstrations on a target embodiment, we finetune the WM on this data to better align the WM predictions, train a base policy, and learn a robust value function. Using our finetuned WM and value function, our approach evaluates action candidates from the base policy and selects the best one to improve performance. Our approach, which we term Latent Policy Steering (LPS), improves behavior-cloned policies by 10.6% on average across four Robomimic tasks, even though most of the pretraining data comes from the real world. In the real-world experiments, LPS achieves larger gains: 70% relative improvement with 30-50 target-embodiment demonstrations, and 44% relative improvement with 60-100 demonstrations, compared to a behavior-cloned baseline.
Robotics 83
☆ PlayWorld: Learning Robot World Models from Autonomous Play
Action-conditioned video models offer a promising path to building general-purpose robot simulators that can improve directly from data. Yet, despite training on large-scale robot datasets, current state-of-the-art video models still struggle to predict physically consistent robot-object interactions that are crucial in robotic manipulation. To close this gap, we present PlayWorld, a simple, scalable, and fully autonomous pipeline for training high-fidelity video world simulators from interaction experience. In contrast to prior approaches that rely on success-biased human demonstrations, PlayWorld is the first system capable of learning entirely from unsupervised robot self-play, enabling naturally scalable data collection while capturing complex, long-tailed physical interactions essential for modeling realistic object dynamics. Experiments across diverse manipulation tasks show that PlayWorld generates high-quality, physically consistent predictions for contact-rich interactions that are not captured by world models trained on human-collected data.We further demonstrate the versatility of PlayWorld in enabling fine-grained failure prediction and policy evaluation, with up to 40% improvements over human-collected data. Finally, we demonstrate how PlayWorld enables reinforcement learning in the world model, improving policy performance by 65% in success rates when deployed in the real world.
comment: https://robot-playworld.github.io/
☆ Improving through Interaction: Searching Behavioral Representation Spaces with CMA-ES-IG
Robots that interact with humans must adapt to individual users' preferences to operate effectively in human-centered environments. An intuitive and effective technique to learn non-expert users' preferences is through rankings of robot behaviors, e.g., trajectories, gestures, or voices. Existing techniques primarily focus on generating queries that optimize preference learning outcomes, such as sample efficiency or final preference estimation accuracy. However, the focus on outcome overlooks key user expectations in the process of providing these rankings, which can negatively impact users' adoption of robotic systems. This work proposes the Covariance Matrix Adaptation Evolution Strategies with Information Gain (CMA-ES-IG) algorithm. CMA-ES-IG explicitly incorporates user experience considerations into the preference learning process by suggesting perceptually distinct and informative trajectories for users to rank. We demonstrate these benefits through both simulated studies and real-robot experiments. CMA-ES-IG, compared to state-of-the-art alternatives, (1) scales more effectively to higher-dimensional preference spaces, (2) maintains computational tractability for high-dimensional problems, (3) is robust to noisy or inconsistent user feedback, and (4) is preferred by non-expert users in identifying their preferred robot behaviors. This project's code is available at github.com/interaction-lab/CMA-ES-IG
comment: Under submission to IJRR
☆ Characterization, Analytical Planning, and Hybrid Force Control for the Inspire RH56DFX Hand
Commercially accessible dexterous robot hands are increasingly prevalent, but many remain difficult to use as scientific instruments. For example, the Inspire RH56DFX hand exposes only uncalibrated proprioceptive information and shows unreliable contact behavior at high speed (up to 1618% force limit overshoot). Furthermore, its underactuated, coupled finger linkages make antipodal grasps non-trivial. We contribute three improvements to the Inspire RH56DFX to transform it from a black-box device to a research tool: (1) hardware characterization (force calibration, latency, and overshoot), (2) a sim2real validated MuJoCo model for analytical width-to-grasp planning, and (3) a hybrid, closed-loop speed-force grasp controller. We validate these components on peg-in-hole insertion, achieving 65% success and outperforming a wrist-force-only baseline of 10% and on 300 grasps across 15 physically diverse objects, achieving 87% success and outperforming plan-free grasps and learned grasps. Our approach is modular, designed for compatibility with external object detectors and vision-language models for width & force estimation and high-level planning, and provides an interpretable and immediately deployable interface for dexterous manipulation with the Inspire RH56DFX hand, open-sourced at this website https://correlllab.github.io/rh56dfx.html.
☆ SurgCalib: Gaussian Splatting-Based Hand-Eye Calibration for Robot-Assisted Minimally Invasive Surgery
We present a Gaussian Splatting-based framework for hand-eye calibration of the da Vinci surgical robot. In a vision-guided robotic system, accurate estimation of the rigid transformation between the robot base and the camera frame is essential for reliable closed-loop control. For cable-driven surgical robots, this task faces unique challenges. The encoders of surgical instruments often produce inaccurate proprioceptive measurements due to cable stretch and backlash. Conventional hand-eye calibration approaches typically rely on known fiducial patterns and solve the AX = XB formulation. While effective, introducing additional markers into the operating room (OR) environment can violate sterility protocols and disrupt surgical workflows. In this study, we propose SurgCalib, an automatic, markerless framework that has the potential to be used in the OR. SurgCalib first initializes the pose of the surgical instrument using raw kinematic measurements and subsequently refines this pose through a two-phase optimization procedure under the RCM constraint within a Gaussian Splatting-based differentiable rendering pipeline. We evaluate the proposed method on the public dVRK benchmark, SurgPose. The results demonstrate average 2D tool-tip reprojection errors of 12.24 px (2.06 mm) and 11.33 px (1.9 mm), and 3D tool-tip Euclidean distance errors of 5.98 mm and 4.75 mm, for the left and right instruments, respectively.
comment: 9 pages, 7 figures
☆ FAME: Force-Adaptive RL for Expanding the Manipulation Envelope of a Full-Scale Humanoid
Maintaining balance under external hand forces is critical for humanoid bimanual manipulation, where interaction forces propagate through the kinematic chain and constrain the feasible manipulation envelope. We propose \textbf{FAME}, a force-adaptive reinforcement learning framework that conditions a standing policy on a learned latent context encoding upper-body joint configuration and bimanual interaction forces. During training, we apply diverse, spherically sampled 3D forces on each hand to inject disturbances in simulation together with an upper-body pose curriculum, exposing the policy to manipulation-induced perturbations across continuously varying arm configurations. At deployment, interaction forces are estimated from the robot dynamics and fed to the same encoder, enabling online adaptation without wrist force/torque sensors. In simulation across five fixed arm configurations with randomized hand forces and commanded base heights, FAME improves mean standing success to 73.84%, compared to 51.40% for the curriculum-only baseline and 29.44% for the base policy. We further deploy the learned policy on a full-scale Unitree H12 humanoid and evaluate robustness in representative load-interaction scenarios, including asymmetric single-arm load and symmetric bimanual load. Code and videos are available on https://fame10.github.io/Fame/
☆ Formation-Aware Adaptive Conformalized Perception for Safe Leader-Follower Multi-Robot Systems
This paper considers the perception safety problem in distributed vision-based leader-follower formations, where each robot uses onboard perception to estimate relative states, track desired setpoints, and keep the leader within its camera field of view (FOV). Safety is challenging due to heteroscedastic perception errors and the coupling between formation maneuvers and visibility constraints. We propose a distributed, formation-aware adaptive conformal prediction method based on Risk-Aware Mondrian CP to produce formation-conditioned uncertainty quantiles. The resulting bounds tighten in high-risk configurations (near FOV limits) and relax in safer regions. We integrate these bounds into a Formation-Aware Conformal CBF-QP with a smooth margin to enforce visibility while maintaining feasibility and tracking performance. Gazebo simulations show improved formation success rates and tracking accuracy over non-adaptive (global) CP baselines that ignore formation-dependent visibility risk, while preserving finite-sample probabilistic safety guarantees. The experimental videos are available on the \href{https://nail-uh.github.io/iros2026.github.io/}{project website}\footnote{Project Website: https://nail-uh.github.io/iros2026.github.io/}.
comment: 8 pages, 8 figures
☆ Fly, Track, Land: Infrastructure-less Magnetic Localization for Heterogeneous UAV-UGV Teaming
We present a complete infrastructure-less magneto-inductive (MI) localization system enabling a lightweight UAV to autonomously hover, track, and land with centimeter precision on a mobile quadruped robot acting as a dynamic docking pad. This work advances the vision of heterogeneous robot collaboration, where ultra-lightweight flying robots serve as mobile perception agents for ground-based Unmanned Ground Vehicles (UGVs). By extending the sensing horizon and providing complementary viewpoints, the UAVs enhance exploration efficiency and improve the quality of data collection in large-scale, unknown environments. The proposed system aims to complements traditional localization modalities with a compact, embedded, and infrastructure-less magnetic sensing approach, providing accurate short-range relative positioning to bridge the gap between coarse navigation and precise UAV docking. A single lightweight receive coil and a fully embedded estimation pipeline on the UAV deliver 20 Hz relative pose estimates in the UGV's frame, achieving a 3D position root-mean-square error (RMSE) of 5 cm. The system uses real-time estimation and a warm-started solver to estimate the 3D position, which is then fused with inertial and optical-flow measurements in the onboard extended Kalman filter. Real-world experiments validate the effectiveness of the framework, demonstrating significant improvements in UAV--UGV teaming in infrastructure-less scenarios compared to state-of-the-art methods, requiring no external anchors or global positioning. In dynamic scenarios, the UAV tracks and docks with a moving UGV while maintaining a 7.2 cm RMSE and achieving successful autonomous landings.
comment: Submitted to IEEE Transactions on Robotics (T-RO). Supplementary video available
☆ Proprioceptive Safe Active Navigation and Exploration for Planetary Environments
Deformable granular terrains introduce significant locomotion and immobilization risks in planetary exploration and are difficult to detect via remote sensing (e.g., vision). Legged robots can sense terrain properties through leg-terrain interactions during locomotion, offering a direct means to assess traversability in deformable environments. How to systematically exploit this interaction-derived information for navigation planning, however, remains underexplored. We address this gap by presenting PSANE, a Proprioceptive Safe Active Navigation and Exploration framework that leverages leg-terrain interaction measurements for safe navigation and exploration in unknown deformable environments. PSANE learns a traversability model via Gaussian Process regression to estimate and certify safe regions and identify exploration frontiers online, and integrates these estimates with a reactive controller for real-time navigation. Frontier selection is formulated as a multi-objective optimization that balances safe-set expansion probability and goal-directed cost, with subgoals selected via scalarization over the Pareto-optimal frontier set. PSANE safely explores unknown granular terrain and reaches specified goals using only proprioceptively estimated traversability, while achieving performance improvements over baseline methods.
comment: 9 pages, 7 figures
☆ Why Channel-Centric Models are not Enough to Predict End-to-End Performance in Private 5G: A Measurement Campaign and Case Study
Communication-aware robot planning requires accurate predictions of wireless network performance. Current approaches rely on channel-level metrics such as received signal strength and signal-to-noise ratio, assuming these translate reliably into end-to-end throughput. We challenge this assumption through a measurement campaign in a private 5G industrial environment. We evaluate throughput predictions from a commercial ray-tracing simulator as well as data-driven Gaussian process regression models against measurements collected using a mobile robot. The study uses off-the-shelf user equipment in an underground, radio-shielded facility with detailed 3D modeling, representing a best-case scenario for prediction accuracy. The ray-tracing simulator captures the spatial structure of indoor propagation and predicts channel-level metrics with reasonable fidelity. However, it systematically over-predicts throughput, even in line-of-sight regions. The dominant error source is shown to be over-estimation of sustainable MIMO spatial layers: the simulator assumes near-uniform four-layer transmission while measurements reveal substantial adaptation between one and three layers. This mismatch inflates predicted throughput even when channel metrics appear accurate. In contrast, a Gaussian process model with a rational quadratic kernel achieves approximately two-thirds reduction in prediction error with near-zero bias by learning end-to-end throughput directly from measurements. These findings demonstrate that favorable channel conditions do not guarantee high throughput; communication-aware planners relying solely on channel-centric predictions risk overly optimistic trajectories that violate reliability requirements. Accurate throughput prediction for 5G systems requires either extensive calibration of link-layer models or data-driven approaches that capture real system behavior.
☆ Adaptive SINDy: Residual Force System Identification Based UAV Disturbance Rejection
The stability and control of Unmanned Aerial Vehicles (UAVs) in a turbulent environment is a matter of great concern. Devising a robust control algorithm to reject disturbances is challenging due to the highly nonlinear nature of wind dynamics, and modeling the dynamics using analytical techniques is not straightforward. While traditional techniques using disturbance observers and classical adaptive control have shown some progress, they are mostly limited to relatively non-complex environments. On the other hand, learning based approaches are increasingly being used for modeling of residual forces and disturbance rejection; however, their generalization and interpretability is a factor of concern. To this end, we propose a novel integration of data-driven system identification using Sparse Identification of Non-Linear Dynamics (SINDy) with a Recursive Least Square (RLS) adaptive control to adapt and reject wind disturbances in a turbulent environment. We tested and validated our approach on Gazebo harmonic environment and on real flights with wind speeds of up to 2 m/s from four directions, creating a highly dynamic and turbulent environment. Adaptive SINDy outperformed the baseline PID and INDI controllers on several trajectory tracking error metrics without crashing. A root mean square error (RMSE) of up to 12.2 cm and 17.6 cm, and a mean absolute error (MAE) of 13.7 cm and 10.5 cm were achieved on circular and lemniscate trajectories, respectively. The validation was performed on a very lightweight Crazyflie drone under a highly dynamic environment for complex trajectory tracking.
☆ APPLV: Adaptive Planner Parameter Learning from Vision-Language-Action Model
Autonomous navigation in highly constrained environments remains challenging for mobile robots. Classical navigation approaches offer safety assurances but require environment-specific parameter tuning; end-to-end learning bypasses parameter tuning but struggles with precise control in constrained spaces. To this end, recent robot learning approaches automate parameter tuning while retaining classical systems' safety, yet still face challenges in generalizing to unseen environments. Recently, Vision-Language-Action (VLA) models have shown promise by leveraging foundation models' scene understanding capabilities, but still struggle with precise control and inference latency in navigation tasks. In this paper, we propose Adaptive Planner Parameter Learning from Vision-Language-Action Model (\textsc{applv}). Unlike traditional VLA models that directly output actions, \textsc{applv} leverages pre-trained vision-language models with a regression head to predict planner parameters that configure classical planners. We develop two training strategies: supervised learning fine-tuning from collected navigation trajectories and reinforcement learning fine-tuning to further optimize navigation performance. We evaluate \textsc{applv} across multiple motion planners on the simulated Benchmark Autonomous Robot Navigation (BARN) dataset and in physical robot experiments. Results demonstrate that \textsc{applv} outperforms existing methods in both navigation performance and generalization to unseen environments.
☆ SEP-NMPC: Safety Enhanced Passivity-Based Nonlinear Model Predictive Control for a UAV Slung Payload System ICRA 2026
Model Predictive Control (MPC) is widely adopted for agile multirotor vehicles, yet achieving both stability and obstacle-free flight is particularly challenging when a payload is suspended beneath the airframe. This paper introduces a Safety Enhanced Passivity-Based Nonlinear MPC (SEP-NMPC) that provides formal guarantees of stability and safety for a quadrotor transporting a slung payload through cluttered environments. Stability is enforced by embedding a strict passivity inequality, which is derived from a shaped energy storage function with adaptive damping, directly into the NMPC. This formulation dissipates excess energy and ensures asymptotic convergence despite payload swings. Safety is guaranteed through high-order control barrier functions (HOCBFs) that render user-defined clearance sets forward-invariant, obliging both the quadrotor and the swinging payload to maintain separation while interacting with static and dynamic obstacles. The optimization remains quadratic-program compatible and is solved online at each sampling time without gain scheduling or heuristic switching. Extensive simulations and real-world experiments confirm stable payload transport, collision-free trajectories, and real-time feasibility across all tested scenarios. The SEP-NMPC framework therefore unifies passivity-based closed-loop stability with HOCBF-based safety guarantees for UAV slung-payload transportation.
comment: Accepted at ICRA 2026
☆ Predictive Control with Indirect Adaptive Laws for Payload Transportation by Quadrupedal Robots
This paper formally develops a novel hierarchical planning and control framework for robust payload transportation by quadrupedal robots, integrating a model predictive control (MPC) algorithm with a gradient-descent-based adaptive updating law. At the framework's high level, an indirect adaptive law estimates the unknown parameters of the reduced-order (template) locomotion model under varying payloads. These estimated parameters feed into an MPC algorithm for real-time trajectory planning, incorporating a convex stability criterion within the MPC constraints to ensure the stability of the template model's estimation error. The optimal reduced-order trajectories generated by the high-level adaptive MPC (AMPC) are then passed to a low-level nonlinear whole-body controller (WBC) for tracking. Extensive numerical investigations validate the framework's capabilities, showcasing the robot's proficiency in transporting unmodeled, unknown static payloads up to 109% in experiments on flat terrains and 91% on rough experimental terrains. The robot also successfully manages dynamic payloads with 73% of its mass on rough terrains. Performance comparisons with a normal MPC and an L1 MPC indicate a significant improvement. Furthermore, comprehensive hardware experiments conducted in indoor and outdoor environments confirm the method's efficacy on rough terrains despite uncertainties such as payload variations, push disturbances, and obstacles.
comment: 8 pages, 6 figures. Published in IEEE Robotics and Automation Letters
☆ Impact of Different Failures on a Robot's Perceived Reliability ICRA 2026
Robots fail, potentially leading to a loss in the robot's perceived reliability (PR), a measure correlated with trustworthiness. In this study we examine how various kinds of failures affect the PR of the robot differently, and how this measure recovers without explicit social repair actions by the robot. In a preregistered and controlled online video study, participants were asked to predict a robot's success in a pick-and-place task. We examined manipulation failures (slips), freezing (lapses), and three types of incorrect picked objects or place goals (mistakes). Participants were shown one of 11 videos -- one of five types of failure, one of five types of failure followed by a successful execution in the same video, or a successful execution video. This was followed by two additional successful execution videos. Participants bet money either on the robot or on a coin toss after each video. People's betting patterns along with a qualitative analysis of their survey responses highlight that mistakes are less damaging to PR than slips or lapses, and some mistakes are even perceived as successes. We also see that successes immediately following a failure have the same effect on PR as successes without a preceding failure. Finally, we show that successful executions recover PR after a failure. Our findings highlight which robot failures are in higher need of repair in a human-robot interaction, and how trust could be recovered by robot successes.
comment: Accepted to ICRA 2026. 8 pages, 6 figures
☆ HMR-1: Hierarchical Massage Robot with Vision-Language-Model for Embodied Healthcare
The rapid advancement of Embodied Intelligence has opened transformative opportunities in healthcare, particularly in physical therapy and rehabilitation. However, critical challenges remain in developing robust embodied healthcare solutions, such as the lack of standardized evaluation benchmarks and the scarcity of open-source multimodal acupoint massage datasets. To address these gaps, we construct MedMassage-12K - a multimodal dataset containing 12,190 images with 174,177 QA pairs, covering diverse lighting conditions and backgrounds. Furthermore, we propose a hierarchical embodied massage framework, which includes a high-level acupoint grounding module and a low-level control module. The high-level acupoint grounding module uses multimodal large language models to understand human language and identify acupoint locations, while the low-level control module provides the planned trajectory. Based on this, we evaluate existing MLLMs and establish a benchmark for embodied massage tasks. Additionally, we fine-tune the Qwen-VL model, demonstrating the framework's effectiveness. Physical experiments further confirm the practical applicability of the framework.Our dataset and code are publicly available at https://github.com/Xiaofeng-Han-Res/HMR-1.
☆ Scale-Plan: Scalable Language-Enabled Task Planning for Heterogeneous Multi-Robot Teams
Long-horizon task planning for heterogeneous multi-robot systems is essential for deploying collaborative teams in real-world environments; yet, it remains challenging due to the large volume of perceptual information, much of which is irrelevant to task objectives and burdens planning. Traditional symbolic planners rely on manually constructed problem specifications, limiting scalability and adaptability, while recent large language model (LLM)-based approaches often suffer from hallucinations and weak grounding-i.e., poor alignment between generated plans and actual environmental objects and constraints-in object-rich settings. We present Scale-Plan, a scalable LLM-assisted framework that generates compact, task-relevant problem representations from natural language instructions. Given a PDDL domain specification, Scale-Plan constructs an action graph capturing domain structure and uses shallow LLM reasoning to guide a structured graph search that identifies a minimal subset of relevant actions and objects. By filtering irrelevant information prior to planning, Scale-Plan enables efficient decomposition, allocation, and long-horizon plan generation. We evaluate our approach on complex multi-agent tasks and introduce MAT2-THOR, a cleaned benchmark built on AI2-THOR for reliable evaluation of multi-robot planning systems. Scale-Plan outperforms pure LLM and hybrid LLM-PDDL baselines across all metrics, improving scalability and reliability.
☆ Age-Related Differences in the Perception of Eye-Gaze from a Social Robot
There is an increasing interest in social robots assisting older adults during daily life tasks. In this context, non-verbal cues such as deictic gaze are important in natural communication in human-robot interaction. However, the sensibility to deictic-gaze declines naturally with age and results in a reduction in social perception. Therefore, this work explores the benefits of deictic gaze from social robots assisting older adults during daily life tasks, and how age-related differences may influence their social perception in contrast to younger populations. This may help on the design of adaptive age-related non-verbal cues in the Human-Robot Interaction context.
comment: This is the pre-print version. Final publication available at https://doi.org/10.1007/978-3-030-90525-5_30
☆ Exp-Force: Experience-Conditioned Pre-Grasp Force Selection with Vision-Language Models
Accurate pre-contact grasp force selection is critical for safe and reliable robotic manipulation. Adaptive controllers regulate force after contact but still require a reasonable initial estimate. Starting a grasp with too little force requires reactive adjustment, while starting a grasp with too high a force risks damaging fragile objects. This trade-off is particularly challenging for compliant grippers, whose contact mechanics are difficult to model analytically. We propose Exp-Force, an experience-conditioned framework that predicts the minimum feasible grasping force from a single RGB image. The method retrieves a small set of relevant prior grasping experiences and conditions a vision-language model on these examples for in-context inference, without analytic contact models or manually designed heuristics. On 129 object instances, ExpForce achieves a best-case MAE of 0.43 N, reducing error by 72% over zero-shot inference. In real-world tests on 30 unseen objects, it improves appropriate force selection rate from 63% to 87%. These results demonstrate that Exp-Force enables reliable and generalizable pre-grasp force selection by leveraging prior interaction experiences. http://expforcesubmission.github.io/Exp-Force-Website/
☆ Embedding Classical Balance Control Principles in Reinforcement Learning for Humanoid Recovery
Humanoid robots remain vulnerable to falls and unrecoverable failure states, limiting their practical utility in unstructured environments. While reinforcement learning has demonstrated stand-up behaviors, existing approaches treat recovery as a pure task-reward problem without an explicit representation of the balance state. We present a unified RL policy that addresses this limitation by embedding classical balance metrics: capture point, center-of-mass state, and centroidal momentum, as privileged critic inputs and shaping rewards directly around these quantities during training, while the actor relies solely on proprioception for zero-shot hardware transfer. Without reference trajectories or scripted contacts, a single policy spans the full recovery spectrum: ankle and hip strategies for small disturbances, corrective stepping under large pushes, and compliant falling with multi-contact stand-up using the hands, elbows, and knees. Trained on the Unitree H1-2 in Isaac Lab, the policy achieves a 93.4% recovery rate across randomized initial poses and unscripted fall configurations. An ablation study shows that removing the balance-informed structure causes stand-up learning to fail entirely, confirming that these metrics provide a meaningful learning signal rather than incidental structure. Sim-to-sim transfer to MuJoCo and preliminary hardware experiments further demonstrate cross-environment generalization. These results show that embedding interpretable balance structure into the learning framework substantially reduces time spent in failure states and broadens the envelope of autonomous recovery.
☆ Diff-Muscle: Efficient Learning for Musculoskeletal Robotic Table Tennis
Musculoskeletal robots provide superior advantages in flexibility and dexterity, positioning them as a promising frontier towards embodied intelligence. However, current research is largely confined to relative simple tasks, restricting the exploration of their full potential in multi-segment coordination. Furthermore, efficient learning remains a challenge, primarily due to the high-dimensional action space and inherent overactuated structures. To address these challenges, we propose Diff-Muscle, a musculoskeletal robot control algorithm that leverages differential flatness to reformulate policy learning from the redundant muscle-activation space into a significantly lower-dimensional joint space. Furthermore, we utilize the highly dynamic robotic table tennis task to evaluate our algorithm. Specifically, we propose a hierarchical reinforcement learning framework that integrates a Kinematics-based Muscle Actuation Controller (K-MAC) with high-level trajectory planning, enabling a musculoskeletal robot to perform dexterous and precise rallies. Experimental results demonstrate that Diff-Muscle significantly outperforms state-of-the-art baselines in success rates while maintaining minimal muscle activation. Notably, the proposed framework successfully enables the musculoskeletal robots to achieve continuous rallies in a challenging dual-robot setting.
comment: 8 pages, 7 figures
☆ FOMO-3D: Using Vision Foundation Models for Long-Tailed 3D Object Detection
In order to navigate complex traffic environments, self-driving vehicles must recognize many semantic classes pertaining to vulnerable road users or traffic control devices. However, many safety-critical objects (e.g., construction worker) appear infrequently in nominal traffic conditions, leading to a severe shortage of training examples from driving data alone. Recent vision foundation models, which are trained on a large corpus of data, can serve as a good source of external prior knowledge to improve generalization. We propose FOMO-3D, the first multi-modal 3D detector to leverage vision foundation models for long-tailed 3D detection. Specifically, FOMO-3D exploits rich semantic and depth priors from OWLv2 and Metric3Dv2 within a two-stage detection paradigm that first generates proposals with a LiDAR-based branch and a novel camera-based branch, and refines them with attention especially to image features from OWL. Evaluations on real-world driving data show that using rich priors from vision foundation models with careful multi-modal fusion designs leads to large gains for long-tailed 3D detection. Project website is at https://waabi.ai/fomo3d/.
comment: Published at 9th Annual Conference on Robot Learning (CoRL 2025)
☆ Bilevel Planning with Learned Symbolic Abstractions from Interaction Data
Intelligent agents must reason over both continuous dynamics and discrete representations to generate effective plans in complex environments. Previous studies have shown that symbolic abstractions can emerge from neural effect predictors trained with a robot's unsupervised exploration. However, these methods rely on deterministic symbolic domains, lack mechanisms to verify the generated symbolic plans, and operate only at the abstract level, often failing to capture the continuous dynamics of the environment. To overcome these limitations, we propose a bilevel neuro-symbolic framework in which learned probabilistic symbolic rules generate candidate plans rapidly at the high level, and learned continuous effect models verify these plans and perform forward search when necessary at the low level. Our experiments on multi-object manipulation tasks demonstrate that the proposed bilevel method outperforms symbolic-only approaches, reliably identifying failing plans through verification, and achieves planning performance statistically comparable to continuous forward search while resolving most problems via efficient symbolic reasoning.
☆ MetaWorld-X: Hierarchical World Modeling via VLM-Orchestrated Experts for Humanoid Loco-Manipulation
Learning natural, stable, and compositionally generalizable whole-body control policies for humanoid robots performing simultaneous locomotion and manipulation (loco-manipulation) remains a fundamental challenge in robotics. Existing reinforcement learning approaches typically rely on a single monolithic policy to acquire multiple skills, which often leads to cross-skill gradient interference and motion pattern conflicts in high-degree-of-freedom systems. As a result, generated behaviors frequently exhibit unnatural movements, limited stability, and poor generalization to complex task compositions. To address these limitations, we propose MetaWorld-X, a hierarchical world model framework for humanoid control. Guided by a divide-and-conquer principle, our method decomposes complex control problems into a set of specialized expert policies (Specialized Expert Policies, SEP). Each expert is trained under human motion priors through imitation-constrained reinforcement learning, introducing biomechanically consistent inductive biases that ensure natural and physically plausible motion generation. Building upon this foundation, we further develop an Intelligent Routing Mechanism (IRM) supervised by a Vision-Language Model (VLM), enabling semantic-driven expert composition. The VLM-guided router dynamically integrates expert policies according to high-level task semantics, facilitating compositional generalization and adaptive execution in multi-stage loco-manipulation tasks.
comment: 8 figures, https://syt2004.github.io/metaworldX/
☆ CONTACT: CONtact-aware TACTile Learning for Robotic Disassembly IROS 2026
Robotic disassembly involves contact-rich interactions in which successful manipulation depends not only on geometric alignment but also on force-dependent state transitions. While vision-based policies perform well in structured settings, their reliability often degrades in tight-tolerance, contact-dominated, or deformable scenarios. In this work, we systematically investigate the role of tactile sensing in robotic disassembly through both simulation and real-world experiments. We construct five rigid-body disassembly tasks in simulation with increasing geometric constraints and extraction difficulty. We further design five real-world tasks, including three rigid and two deformable scenarios, to evaluate contact-dependent manipulation. Within a unified learning framework, we compare three sensing configurations: Vision Only, Vision + tactile RGB (TacRGB), and Vision + tactile force field (TacFF). Across both simulation and real-world experiments, TacFF-based policies consistently achieve the highest success rates, with particularly notable gains in contact-dependent and deformable settings. Notably, naive fusion of TacRGB and TacFF underperforms either modality alone, indicating that simple concatenation can dilute task-relevant force information. Our results show that tactile sensing plays a critical, task-dependent role in robotic disassembly, with structured force-field representations being particularly effective in contact-dominated scenarios.
comment: Submitted to IROS 2026, 8 pages, 6 figures
☆ Interactive World Simulator for Robot Policy Training and Evaluation
Action-conditioned video prediction models (often referred to as world models) have shown strong potential for robotics applications, but existing approaches are often slow and struggle to capture physically consistent interactions over long horizons, limiting their usefulness for scalable robot policy training and evaluation. We present Interactive World Simulator, a framework for building interactive world models from a moderate-sized robot interaction dataset. Our approach leverages consistency models for both image decoding and latent-space dynamics prediction, enabling fast and stable simulation of physical interactions. In our experiments, the learned world models produce interaction-consistent pixel-level predictions and support stable long-horizon interactions for more than 10 minutes at 15 FPS on a single RTX 4090 GPU. Our framework enables scalable demonstration collection solely within the world models to train state-of-the-art imitation policies. Through extensive real-world evaluation across diverse tasks involving rigid objects, deformable objects, object piles, and their interactions, we find that policies trained on world-model-generated data perform comparably to those trained on the same amount of real-world data. Additionally, we evaluate policies both within the world models and in the real world across diverse tasks, and observe a strong correlation between simulated and real-world performance. Together, these results establish the Interactive World Simulator as a stable and physically consistent surrogate for scalable robotic data generation and faithful, reproducible policy evaluation.
comment: Project Page: https://yixuanwang.me/interactive_world_sim
☆ The Neural Compass: Probabilistic Relative Feature Fields for Robotic Search IROS 2026
Object co-occurrences provide a key cue for finding objects successfully and efficiently in unfamiliar environments. Typically, one looks for cups in kitchens and views fridges as evidence of being in a kitchen. Such priors have also been exploited in artificial agents, but they are typically learned from explicitly labeled data or queried from language models. It is still unclear whether these relations can be learned implicitly from unlabeled observations alone. In this work, we address this problem and propose ProReFF, a feature field model trained to predict relative distributions of features obtained from pre-trained vision language models. In addition, we introduce a learning-based strategy that enables training from unlabeled and potentially contradictory data by aligning inconsistent observations into a coherent relative distribution. For the downstream object search task, we propose an agent that leverages predicted feature distributions as a semantic prior to guide exploration toward regions with a high likelihood of containing the object. We present extensive evaluations demonstrating that ProReFF captures meaningful relative feature distributions in natural scenes and provides insight into the impact of our proposed alignment step. We further evaluate the performance of our search agent in 100 challenges in the Matterport3D simulator, comparing with feature-based baselines and human participants. The proposed agent is 20% more efficient than the strongest baseline and achieves up to 80% of human performance.
comment: 9 pages, 7 figures, 2 tables, submitted to IROS 2026
☆ EquiBim: Learning Symmetry-Equivariant Policy for Bimanual Manipulation IROS 2026
Robotic imitation learning has achieved impressive success in learning complex manipulation behaviors from demonstrations. However, many existing robot learning methods do not explicitly account for the physical symmetries of robotic systems, often resulting in asymmetric or inconsistent behaviors under symmetric observations. This limitation is particularly pronounced in dual-arm manipulation, where bilateral symmetry is inherent to both the robot morphology and the structure of many tasks. In this paper, we introduce EquiBim, a symmetry-equivariant policy learning framework for bimanual manipulation that enforces bilateral equivariance between observations and actions during training. Our approach formulates physical symmetry as a group action on both observation and action spaces, and imposes an equivariance constraint on policy predictions under symmetric transformations. The framework is model-agnostic and can be seamlessly integrated into a wide range of imitation learning pipelines with diverse observation modalities and action representations, including point cloud-based and image-based policies, as well as both end-effector-space and joint-space parameterizations. We evaluate EquiBim on RoboTwin, a dual-arm robotic platform with symmetric kinematics, and evaluate it across diverse observation and action configurations in simulation. We further validate the approach on a real-world dual-arm system. Across both simulation and physical experiments, our method consistently improves performance and robustness under distribution shifts. These results suggest that explicitly enforcing physical symmetry provides a simple yet effective inductive bias for bimanual robot learning.
comment: Submitted to IROS 2026. 8 pages, 6 figures
☆ CRED: Counterfactual Reasoning and Environment Design for Active Preference Learning ICRA
As a robot's operational environment and tasks to perform within it grow in complexity, the explicit specification and balancing of optimization objectives to achieve a preferred behavior profile moves increasingly farther out of reach. These systems benefit strongly by being able to align their behavior to reflect human preferences and respond to corrections, but manually encoding this feedback is infeasible. Active preference learning (APL) learns human reward functions by presenting trajectories for ranking. However, existing methods sample from fixed trajectory sets or replay buffers that limit query diversity and often fail to identify informative comparisons. We propose CRED, a novel trajectory generation method for APL that improves reward inference by jointly optimizing environment design and trajectory selection to efficiently query and extract preferences from users. CRED "imagines" new scenarios through environment design and leverages counterfactual reasoning -- by sampling possible rewards from its current belief and asking "What if this were the true preference?" -- to generate trajectory pairs that expose differences between competing reward functions. Comprehensive experiments and a user study show that CRED significantly outperforms state-of-the-art methods in reward accuracy and sample efficiency and receives higher user ratings.
comment: IEEE International Conference on Robotics and Automation (ICRA) 2026
☆ OccTrack360: 4D Panoptic Occupancy Tracking from Surround-View Fisheye Cameras
Understanding dynamic 3D environments in a spatially continuous and temporally consistent manner is fundamental for robotics and autonomous driving. While recent advances in occupancy prediction provide a unified representation of scene geometry and semantics, progress in 4D panoptic occupancy tracking remains limited by the lack of benchmarks that support surround-view fisheye sensing, long temporal sequences, and instance-level voxel tracking. To address this gap, we present OccTrack360, a new benchmark for 4D panoptic occupancy tracking from surround-view fisheye cameras. OccTrack360 provides substantially longer and more diverse sequences (174~2234 frames) than prior benchmarks, together with principled voxel visibility annotations, including an all-direction occlusion mask and an MEI-based fisheye field-of-view mask. To establish a strong fisheye-oriented baseline, we further propose Focus on Sphere Occ (FoSOcc), a framework that addresses two core challenges in fisheye occupancy tracking: distorted spherical projection and inaccurate voxel-space localization. FoSOcc includes a Center Focusing Module (CFM) to enhance instance-aware spatial localization through supervised focus guidance, and a Spherical Lift Module (SLM) that extends perspective lifting to fisheye imaging under the Unified Projection Model. Extensive experiments on Occ3D-Waymo and OccTrack360 show that our method improves occupancy tracking quality with notable gains on geometrically regular categories, and establishes a strong baseline for future research on surround-view fisheye 4D occupancy tracking. The benchmark and source code will be made publicly available at https://github.com/YouthZest-Lin/OccTrack360.
comment: The benchmark and source code will be made publicly available at https://github.com/YouthZest-Lin/OccTrack360
☆ AtomVLA: Scalable Post-Training for Robotic Manipulation via Predictive Latent World Models
Vision-Language-Action (VLA) models demonstrate remarkable potential for generalizable robotic manipulation. The execution of complex multi-step behaviors in VLA models can be improved by robust instruction grounding, a critical component for effective control. However, current paradigms predominantly rely on coarse, high-level task instructions during supervised fine-tuning. This instruction grounding gap leaves models without explicit intermediate guidance, leading to severe compounding errors in long-horizon tasks. Therefore, bridging this instruction gap and providing scalable post-training for VLA models is urgent. To tackle this problem, we propose \method, the first subtask-aware VLA framework integrated with a scalable offline post-training pipeline. Our framework leverages a large language model to decompose high-level demonstrations into fine-grained atomic subtasks. This approach utilizes a pretrained predictive world model to score candidate action chunks against subtask goals in the latent space, mitigating error accumulation while significantly improving long-horizon robustness. Furthermore, this approach enables highly efficient Group Relative Policy Optimization without the prohibitive expenses associated with online rollouts on physical robots. Extensive simulations validate that our AtomVLA maintains strong robustness under perturbations. When evaluated against fundamental baseline models, it achieves an average success rate of 97.0\% on the LIBERO benchmark and 48.0\% on the LIBERO-PRO benchmark. Finally, experiments conducted in the real world using the Galaxea R1 Lite platform confirm its broad applicability across diverse tasks, especially long-horizon tasks. All datasets, checkpoints, and code will be released to the public domain following the acceptance of this work for future research.
☆ Rethinking the semantic classification of indoor places by mobile robots
A significant challenge in service robots is the semantic understanding of their surrounding areas. Traditional approaches addressed this problem by segmenting the floor plan into regions corresponding to full rooms that are assigned labels consistent with human perception, e.g. office or kitchen. However, different areas inside the same room can be used in different ways: Could the table and the chair in my kitchen become my office? What is the category of that area now? office or kitchen? To adapt to these circumstances we propose a new paradigm where we intentionally relax the resulting labeling of semantic classifiers by allowing confusions inside rooms. Our hypothesis is that those confusions can be beneficial to a service robot. We present a proof of concept in the task of searching for objects.
comment: Presented at the Workshop on Semantic Scene Understanding for Human Robot Interaction, in the ACM/IEEE International Conference on Human-Robot Interaction (HRI), Stockholm, Sweden, 2023
☆ Spherical-GOF: Geometry-Aware Panoramic Gaussian Opacity Fields for 3D Scene Reconstruction
Omnidirectional images are increasingly used in robotics and vision due to their wide field of view. However, extending 3D Gaussian Splatting (3DGS) to panoramic camera models remains challenging, as existing formulations are designed for perspective projections and naive adaptations often introduce distortion and geometric inconsistencies. We present Spherical-GOF, an omnidirectional Gaussian rendering framework built upon Gaussian Opacity Fields (GOF). Unlike projection-based rasterization, Spherical-GOF performs GOF ray sampling directly on the unit sphere in spherical ray space, enabling consistent ray-Gaussian interactions for panoramic rendering. To make the spherical ray casting efficient and robust, we derive a conservative spherical bounding rule for fast ray-Gaussian culling and introduce a spherical filtering scheme that adapts Gaussian footprints to distortion-varying panoramic pixel sampling. Extensive experiments on standard panoramic benchmarks (OmniBlender and OmniPhotos) demonstrate competitive photometric quality and substantially improved geometric consistency. Compared with the strongest baseline, Spherical-GOF reduces depth reprojection error by 57% and improves cycle inlier ratio by 21%. Qualitative results show cleaner depth and more coherent normal maps, with strong robustness to global panorama rotations. We further validate generalization on OmniRob, a real-world robotic omnidirectional dataset introduced in this work, featuring UAV and quadruped platforms. The source code and the OmniRob dataset will be released at https://github.com/1170632760/Spherical-GOF.
comment: The source code and dataset will be released at https://github.com/1170632760/Spherical-GOF
☆ An Open-Source Robotics Research Platform for Autonomous Laparoscopic Surgery
Autonomous robot-assisted surgery demands reliable, high-precision platforms that strictly adhere to the safety and kinematic constraints of minimally invasive procedures. Existing research platforms, primarily based on the da Vinci Research Kit, suffer from cable-driven mechanical limitations that degrade state-space consistency and hinder the downstream training of reliable autonomous policies. We present an open-source, robot-agnostic Remote Center of Motion (RCM) controller based on a closed-form analytical velocity solver that enforces the trocar constraint deterministically without iterative optimization. The controller operates in Cartesian space, enabling any industrial manipulator to function as a surgical robot. We provide implementations for the UR5e and Franka Emika Panda manipulators, and integrate stereoscopic 3D perception. We integrate the robot control into a full-stack ROS-based surgical robotics platform supporting teleoperation, demonstration recording, and deployment of learned policies via a decoupled server-client architecture. We validate the system on a bowel grasping and retraction task across phantom, ex vivo, and in vivo porcine laparoscopic procedures. RCM deviations remain sub-millimeter across all conditions, and trajectory smoothness metrics (SPARC, LDLJ) are comparable to expert demonstrations from the JIGSAWS benchmark recorded on the da Vinci system. These results demonstrate that the platform provides the precision and robustness required for teleoperation, data collection and autonomous policy deployment in realistic surgical scenarios.
comment: Submitted to iROS 2026
☆ 3PoinTr: 3D Point Tracks for Robot Manipulation Pretraining from Casual Videos
Data-efficient training of robust robot policies is the key to unlocking automation in a wide array of novel tasks. Current systems require large volumes of demonstrations to achieve robustness, which is impractical in many applications. Learning policies directly from human videos is a promising alternative that removes teleoperation costs, but it shifts the challenge toward overcoming the embodiment gap (differences in kinematics and strategies between robots and humans), often requiring restrictive and carefully choreographed human motions. We propose 3PoinTr, a method for pretraining robot policies from casual and unconstrained human videos, enabling learning from motions natural for humans. 3PoinTr uses a transformer architecture to predict 3D point tracks as an intermediate embodiment-agnostic representation. 3D point tracks encode goal specifications, scene geometry, and spatiotemporal relationships. We use a Perceiver IO architecture to extract a compact representation for sample-efficient behavior cloning, even when point tracks violate downstream embodiment-specific constraints. We conduct thorough evaluation on simulated and real-world tasks, and find that 3PoinTr achieves robust spatial generalization on diverse categories of manipulation tasks with only 20 action-labeled robot demonstrations. 3PoinTr outperforms the baselines, including behavior cloning methods, as well as prior methods for pretraining from human videos. We also provide evaluations of 3PoinTr's 3D point track predictions compared to an existing point track prediction baseline. We find that 3PoinTr produces more accurate and higher quality point tracks due to a lightweight yet expressive architecture built on a single transformer, in addition to a training formulation that preserves supervision of partially occluded points. Project page: https://adamhung60.github.io/3PoinTr/.
☆ STRIDE: Structured Lagrangian and Stochastic Residual Dynamics via Flow Matching
Robotic systems operating in unstructured environments must operate under significant uncertainty arising from intermittent contacts, frictional variability, and unmodeled compliance. While recent model-free approaches have demonstrated impressive performance, many deployment settings still require predictive models that support planning, constraint handling, and online adaptation. Analytical rigid-body models provide strong physical structure but often fail to capture complex interaction effects, whereas purely data-driven models may violate physical consistency, exhibit data bias, and accumulate long-horizon drift. In this work, we propose STRIDE, a dynamics learning framework that explicitly separates conservative rigid-body mechanics from uncertain, effectively stochastic non-conservative interaction effects. The structured component is modeled using a Lagrangian Neural Network (LNN) to preserve energy-consistent inertial dynamics, while residual interaction forces are represented using Conditional Flow Matching (CFM) to capture multi-modal interaction phenomena. The two components are trained jointly end-to-end, enabling the model to retain physical structure while representing complex stochastic behavior. We evaluate STRIDE on systems of increasing complexity, including a pendulum, the Unitree Go1 quadruped, and the Unitree G1 humanoid. Results show 20% reduction in long-horizon prediction error and 30% reduction in contact force prediction error compared to deterministic residual baselines, supporting more reliable model-based control in uncertain robotic environments.
comment: 9 pages, 7 figures
☆ LAR-MoE: Latent-Aligned Routing for Mixture of Experts in Robotic Imitation Learning
Imitation learning enables robots to acquire manipulation skills from demonstrations, yet deploying a policy across tasks with heterogeneous dynamics remains challenging, as models tend to average over distinct behavioral modes present in the demonstrations. Mixture-of-Experts (MoE) architectures address this by activating specialized subnetworks, but requires meaningful skill decompositions for expert routing. We introduce Latent-Aligned Routing for Mixture of Experts (LAR-MoE), a two-stage framework that decouples unsupervised skill discovery from policy learning. In pre-training, we learn a joint latent representation between observations and future actions through student-teacher co-training. In a post-training stage, the expert routing is regularized to follow the structure of the learned latent space, preventing expert collapse while maintaining parameter efficiency. We evaluate LAR-MoE in simulation and on hardware. On the LIBERO benchmark, our method achieves a 95.2% average success rate with 150M parameters. On a surgical bowel grasping and retraction task, LAR-MoE matches a supervised MoE baseline without requiring any phase annotations, and transfers zero-shot to ex vivo porcine tissue. Our findings suggest that latent-aligned routing provides a principled alternative to supervised skill decomposition, enabling structured expert specialization from unlabeled demonstrations.
comment: Submitted to iROS 2026
☆ R2F: Repurposing Ray Frontiers for LLM-free Object Navigation
Zero-shot open-vocabulary object navigation has progressed rapidly with the emergence of large Vision-Language Models (VLMs) and Large Language Models (LLMs), now widely used as high-level decision-makers instead of end-to-end policies. Although effective, such systems often rely on iterative large-model queries at inference time, introducing latency and computational overhead that limit real-time deployment. To address this problem, we repurpose ray frontiers (R2F), a recently proposed frontier-based exploration paradigm, to develop an LLM-free framework for indoor open-vocabulary object navigation. While ray frontiers were originally used to bias exploration using semantic cues carried along rays, we reinterpret frontier regions as explicit, direction-conditioned semantic hypotheses that serve as navigation goals. Language-aligned features accumulated along out-of-range rays are stored sparsely at frontiers, where each region maintains multiple directional embeddings encoding plausible unseen content. In this way, navigation then reduces to embedding-based frontier scoring and goal tracking within a classical mapping and planning pipeline, eliminating iterative large-model reasoning. We further introduce R2F-VLN, a lightweight extension for free-form language instructions using syntactic parsing and relational verification without additional VLM or LLM components. Experiments in Habitat-sim and on a real robotic platform demonstrate competitive state-of-the-art zero-shot performance with real-time execution, achieving up to 6 times faster runtime than VLM-based alternatives.
☆ Adaptive Entropy-Driven Sensor Selection in a Camera-LiDAR Particle Filter for Single-Vessel Tracking
Robust single-vessel tracking from fixed coastal platforms is hindered by modality-specific degradations: cameras suffer from illumination and visual clutter, while LiDAR performance drops with range and intermittent returns. We present a heterogeneous multi-sensor fusion particle-filter tracker that incorporates an information-gain (entropy-reduction) adaptive sensing policy to select the most informative configuration at each fusion time bin. The approach is validated in a real maritime deployment at the CMMI Smart Marina Testbed (Ayia Napa Marina, Cyprus), using a shore-mounted 3D LiDAR and an elevated fixed camera to track a rigid inflatable boat with onboard GNSS ground truth. We compare LiDAR-only, camera-only, all-sensors, and adaptive configurations. Results show LiDAR dominates near-field accuracy, the camera sustains longer-range coverage when LiDAR becomes unavailable, and the adaptive policy achieves a favorable accuracy-continuity trade-off by switching modalities based on information gain. By avoiding continuous multi-stream processing, the adaptive configuration provides a practical baseline for resilient and resource-aware maritime surveillance.
comment: 8 pages, 5 figures, submitted to FUSION 2026 conference proceedings
☆ FoMo: A Multi-Season Dataset for Robot Navigation in Forêt Montmorency
The Forêt Montmorency (FoMo) dataset is a comprehensive multi-season data collection, recorded over the span of one year in a boreal forest. Featuring a unique combination of on- and off-pavement environments with significant environmental changes, the dataset challenges established odometry and SLAM pipelines. Some highlights of the data include the accumulation of snow exceeding 1 m, significant vegetation growth in front of sensors, and operations at the traction limits of the platform. In total, the FoMo dataset includes over 64 km of six diverse trajectories, repeated during 12 deployments throughout the year. The dataset features data from one rotating and one hybrid solid-state lidar, a Frequency Modulated Continuous Wave (FMCW) radar, full-HD images from a stereo camera and a wide lens monocular camera, as well as data from two IMUs. Ground Truth is calculated by post-processing three GNSS receivers mounted on the Uncrewed Ground Vehicle (UGV) and a static GNSS base station. Additional metadata, such as one measurement per minute from an on-site weather station, camera calibration intrinsics, and vehicle power consumption, is available for all sequences. To highlight the relevance of the dataset, we performed a preliminary evaluation of the robustness of a lidar-inertial, radar-gyro, and a visual-inertial localization and mapping techniques to seasonal changes. We show that seasonal changes have serious effects on the re-localization capabilities of the state-of-the-art methods. The dataset and development kit are available at https://fomo.norlab.ulaval.ca.
☆ Tactile Recognition of Both Shapes and Materials with Automatic Feature Optimization-Enabled Meta Learning ICRA 2026
Tactile perception is indispensable for robots to implement various manipulations dexterously, especially in contact-rich scenarios. However, alongside the development of deep learning techniques, it meanwhile suffers from training data scarcity and a time-consuming learning process in practical applications since the collection of a large amount of tactile data is costly and sometimes even impossible. Hence, we propose an automatic feature optimization-enabled prototypical network to realize meta-learning, i.e., AFOP-ML framework. As a ``learn to learn" network, it not only adapts to new unseen classes rapidly with few-shot, but also learns how to determine the optimal feature space automatically. Based on the four-channel signals acquired from a tactile finger, both shapes and materials are recognized. On a 36-category benchmark, it outperforms several existing approaches by attaining an accuracy of 96.08% in 5-way-1-shot scenario, where only 1 example is available for training. It still remains 88.7% in the extreme 36-way-1-shot case. The generalization ability is further validated through three groups of experiment involving unseen shapes, materials and force/speed perturbations. More insights are additionally provided by this work for the interpretation of recognition tasks and improved design of tactile sensors.
comment: 7 pages, 7 figures, conference paper accepted by ICRA 2026
☆ Human-Aware Robot Behaviour in Self-Driving Labs
Self-driving laboratories (SDLs) are rapidly transforming research in chemistry and materials science to accelerate new discoveries. Mobile robot chemists (MRCs) play a pivotal role by autonomously navigating the lab to transport samples, effectively connecting synthesis, analysis, and characterisation equipment. The instruments within an SDL are typically designed or retrofitted to be accessed by both human and robotic chemists, ensuring operational flexibility and integration between manual and automated workflows. In many scenarios, human and robotic chemists may need to use the same equipment simultaneously. Currently, MRCs rely on simple LiDAR-based obstruction detection, which forces the robot to passively wait if a human is present. This lack of situational awareness leads to unnecessary delays and inefficient coordination in time-critical automated workflows in human-robot shared labs. To address this, we present an initial study of an embodied, AI-driven perception method that facilitates proactive human-robot interaction in shared-access scenarios. Our method features a hierarchical human intention prediction model that allows the robot to distinguish between preparatory actions (waiting) and transient interactions (accessing the instrument). Our results demonstrate that the proposed approach enhances efficiency by enabling proactive human-robot interaction, streamlining coordination, and potentially increasing the efficiency of autonomous scientific labs.
☆ A Recipe for Stable Offline Multi-agent Reinforcement Learning
Despite remarkable achievements in single-agent offline reinforcement learning (RL), multi-agent RL (MARL) has struggled to adopt this paradigm, largely persisting with on-policy training and self-play from scratch. One reason for this gap comes from the instability of non-linear value decomposition, leading prior works to avoid complex mixing networks in favor of linear value decomposition (e.g., VDN) with value regularization used in single-agent setups. In this work, we analyze the source of instability in non-linear value decomposition within the offline MARL setting. Our observations confirm that they induce value-scale amplification and unstable optimization. To alleviate this, we propose a simple technique, scale-invariant value normalization (SVN), that stabilizes actor-critic training without altering the Bellman fixed point. Empirically, we examine the interaction among key components of offline MARL (e.g., value decomposition, value learning, and policy extraction) and derive a practical recipe that unlocks its full potential.
comment: Preprint
☆ MoMaStage: Skill-State Graph Guided Planning and Closed-Loop Execution for Long-Horizon Indoor Mobile Manipulation
Indoor mobile manipulation (MoMA) enables robots to translate natural language instructions into physical actions, yet long-horizon execution remains challenging due to cascading errors and limited generalization across diverse environments. Learning-based approaches often fail to maintain logical consistency over extended horizons, while methods relying on explicit scene representations impose rigid structural assumptions that reduce adaptability in dynamic settings. To address these limitations, we propose MoMaStage, a structured vision-language framework for long-horizon MoMA that eliminates the need for explicit scene mapping. MoMaStage grounds a Vision-Language Model (VLM) within a Hierarchical Skill Library and a topology-aware Skill-State Graph, constraining task decomposition and skill composition within a feasible transition space. This structured grounding ensures that generated plans remain logically consistent and topologically valid with respect to the agent's evolving physical state. To enhance robustness, MoMaStage incorporates a closed-loop execution mechanism that monitors proprioceptive feedback and triggers graph-constrained semantic replanning when deviations are detected, maintaining alignment between planned skills and physical outcomes. Extensive experiments in physics-rich simulations and real-world environments demonstrate that MoMaStage outperforms state-of-the-art baselines, achieving substantially higher planning success, reducing token overhead, and significantly improving overall task success rates in long-horizon mobile manipulation. Video demonstrations are available on the project website: https://chenxuli-cxli.github.io/MoMaStage/.
comment: 8 pages
☆ Perception-Aware Communication-Free Multi-UAV Coordination in the Wild
We present a communication-free method for safe multi-robot coordination in complex environments such as forests with dense canopy cover, where GNSS is unavailable. Our approach relies on an onboard anisotropic 3D LiDAR sensor used for SLAM as well as for detecting obstacles and neighboring robots. We develop a novel perception-aware 3D navigation framework that enables robots to safely and effectively progress toward a goal region despite limited sensor field-of-view. The approach is evaluated through extensive simulations across diverse scenarios and validated in real-world field experiments, demonstrating its scalability, robustness, and reliability.
☆ PhaForce: Phase-Scheduled Visual-Force Policy Learning with Slow Planning and Fast Correction for Contact-Rich Manipulation
Contact-rich manipulation requires not only vision-dominant task semantics but also closed-loop reactions to force/torque (F/T) transients. Yet, generative visuomotor policies are typically constrained to low-frequency updates due to inference latency and action chunking, underutilizing F/T for control-rate feedback. Furthermore, existing force-aware methods often inject force continuously and indiscriminately, lacking an explicit mechanism to schedule when / how much / where to apply force across different task phases. We propose PhaForce, a phase-scheduled visual--force policy that coordinates low-rate chunk-level planning and high-rate residual correction via a unified contact/phase schedule. PhaForce comprises (i) a contact-aware phase predictor (CAP) that estimates contact probability and phase belief, (ii) a Slow diffusion planner that performs dual-gated visual--force fusion with orthogonal residual injection to preserve vision semantics while conditioning on force, and (iii) a Fast corrector that applies control-rate phase-routed residuals in interpretable corrective subspaces for within-chunk micro-adjustments. Across multiple real-robot contact-rich tasks, PhaForce achieves an average success rate of 86% (+40 pp over baselines), while also substantially improving contact quality by regulating interaction forces and exhibiting robust adaptability to OOD geometric shifts.
☆ Hierarchical Multi-Modal Planning for Fixed-Altitude Sparse Target Search and Sampling
Efficient monitoring of sparse benthic phenomena, such as coral colonies, presents a great challenge for Autonomous Underwater Vehicles. Traditional exhaustive coverage strategies are energy-inefficient, while recent adaptive sampling approaches rely on costly vertical maneuvers. To address these limitations, we propose HIMoS (Hierarchical Informative Multi-Modal Search), a fixed-altitude framework for sparse coral search-and-sample missions. The system integrates a heterogeneous sensor suite within a two-layer planning architecture. At the strategic level, a Global Planner optimizes topological routes to maximize potential discovery. At the tactical level, a receding-horizon Local Planner leverages differentiable belief propagation to generate kinematically feasible trajectories that balance acoustic substrate exploration, visual coral search, and close-range sampling. Validated in high-fidelity simulations derived from real-world coral reef benthic surveys, our approach demonstrates superior mission efficiency compared to state-of-the-art baselines.
comment: 8 pages, 9 figures, conference
☆ EndoSERV: A Vision-based Endoluminal Robot Navigation System
Robot-assisted endoluminal procedures are increasingly used for early cancer intervention. However, the intricate, narrow and tortuous pathways within the luminal anatomy pose substantial difficulties for robot navigation. Vision-based navigation offers a promising solution, but existing localization approaches are error-prone due to tissue deformation, in vivo artifacts and a lack of distinctive landmarks for consistent localization. This paper presents a novel EndoSERV localization method to address these challenges. It includes two main parts, \textit{i.e.}, \textbf{SE}gment-to-structure and \textbf{R}eal-to-\textbf{V}irtual mapping, and hence the name. For long-range and complex luminal structures, we divide them into smaller sub-segments and estimate the odometry independently. To cater for label insufficiency, an efficient transfer technique maps real image features to the virtual domain to use virtual pose ground truth. The training phases of EndoSERV include an offline pretraining to extract texture-agnostic features, and an online phase that adapts to real-world conditions. Extensive experiments based on both public and clinical datasets have been performed to demonstrate the effectiveness of the method even without any real pose labels.
☆ Less is More: Robust Zero-Communication 3D Pursuit-Evasion via Representational Parsimony
Asymmetric 3D pursuit-evasion in cluttered voxel environments is difficult under communication latency, partial observability, and nonholonomic maneuver limits. While many MARL methods rely on richer inter-agent coupling or centralized signals, these dependencies can become fragility sources when communication is delayed or noisy. Building on an inherited path-guided decentralized pursuit scaffold, we study a robustness-oriented question: can representational parsimony improve communication-free coordination? We instantiate this principle with (i) a parsimonious actor observation interface that removes team-coupled channels (83-D to 50-D), and (ii) Contribution-Gated Credit Assignment (CGCA), a locality-aware credit structure for communication-denied cooperation. In Stage-5 evaluation (4 pursuers vs. 1 evader), our configuration reaches 0.753 +/- 0.091 success and 0.223 +/- 0.066 collision, outperforming the 83-D FULL OBS counterpart (0.721 +/- 0.071, 0.253 +/- 0.089). It further shows graceful degradation under speed/yaw/noise/delay stress tests and resilient zero-shot transfer on urban-canyon maps (about 61% success at density 0.24). These results support a practical paradigm shift: explicitly severing redundant cross-agent channels can suppress compounding error cascades and improve robustness in latency-prone deployment.
comment: 7 pages, 10 figures. This work has been submitted to the IEEE for possible publication
☆ SAIL: Test-Time Scaling for In-Context Imitation Learning with VLM
In-context imitation learning allows robots to acquire skills from demonstrations, yet one-shot trajectory generation remains fragile under environmental variation. We propose SAIL, a framework that reframes robot imitation as an iterative refinement problem capable of scaling with test-time compute. SAIL utilizes Monte Carlo Tree Search, where each node is a complete trajectory and edges correspond to trajectory refinements. The process is guided by three core components: an automated archive of successful trajectories for contextually relevant retrieval, a vision language model-based scoring mechanism for trajectory evaluation, and a step-level feedback that provides trajectory-aligned scores for iterative refinement. Experiments across six diverse manipulation tasks in simulation and real-world validation clearly demonstrate that increasing test-time compute consistently improves success rates, achieving up to 95% on complex tasks. Our results suggest that trajectory-level test-time scaling is a robust path toward more generalizable robotic agents.
comment: 8 pages, 3 figures
☆ Seed2Scale: A Self-Evolving Data Engine for Embodied AI via Small to Large Model Synergy and Multimodal Evaluation
Existing data generation methods suffer from exploration limits, embodiment gaps, and low signal-to-noise ratios, leading to performance degradation during self-iteration. To address these challenges, we propose Seed2Scale, a self-evolving data engine that overcomes the data bottleneck through a heterogeneous synergy of "small-model collection, large-model evaluation, and target-model learning". Starting with as few as four seed demonstrations, the engine employs the lightweight Vision-Language-Action model, SuperTiny, as a dedicated collector, leveraging its strong inductive bias for robust exploration in parallel environments. Concurrently, a pre-trained Vision-Language Model is integrated as a Verifer to autonomously perform success/failure judgment and quality scoring for the massive generated trajectories. Seed2Scale effectively mitigates model collapse, ensuring the stability of the self-evolution process. Experimental results demonstrate that Seed2Scale exhibits signifcant scaling potential: as iterations progress, the success rate of the target model shows a robust upward trend, achieving a performance improvement of 131.2%. Furthermore, Seed2Scale signifcantly outperforms existing data augmentation methods, providing a scalable and cost-effective pathway for the large-scale development of Generalist Embodied AI. Project page: https://terminators2025.github.io/Seed2Scale.github.io
☆ FlowTouch: View-Invariant Visuo-Tactile Prediction
Tactile sensation is essential for contact-rich manipulation tasks. It provides direct feedback on object geometry, surface properties, and interaction forces, enhancing perception and enabling fine-grained control. An inherent limitation of tactile sensors is that readings are available only when an object is touched. This precludes their use during planning and the initial execution phase of a task. Predicting tactile information from visual information can bridge this gap. A common approach is to learn a direct mapping from camera images to the output of vision-based tactile sensors. However, the resulting model will depend strongly on the specific setup and on how well the camera can capture the area where an object is touched. In this work, we introduce FlowTouch, a novel model for view-invariant visuo-tactile prediction. Our key idea is to use an object's local 3D mesh to encode rich information for predicting tactile patterns while abstracting away from scene-dependent details. FlowTouch integrates scene reconstruction and Flow Matching-based models for image generation. Our results show that FlowTouch is able to bridge the sim-to-real gap and generalize to new sensor instances. We further show that the resulting tactile images can be used for downstream grasp stability prediction. Our code, datasets and videos are available at https://flowtouch.github.io/
☆ A General Lie-Group Framework for Continuum Soft Robot Modeling
This paper introduces a general Lie group framework for modeling continuum soft robots, employing Cosserat rod theory combined with cumulative parameterization on the Lie group SE(3). This novel approach addresses limitations present in current strain-based and configuration-based methods by providing geometric local control and eliminating unit quaternion constraints. The paper derives unified analytical expressions for kinematics, statics, and dynamics, including recursive Jacobian computations and an energy-conserving integrator suitable for real-time simulation and control. Additionally, the framework is extended to handle complex robotic structures, including segmented, branched, nested, and rigid-soft composite configurations, facilitating a modular and unified modeling strategy. The effectiveness, generality, and computational efficiency of the proposed methodology are demonstrated through various scenarios, including large-deformation rods, concentric tube robots, parallel robots, cable-driven robots, and articulated fingers. This work enhances modeling flexibility and numerical performance, providing an improved toolset for designing, simulating, and controlling soft robotic systems.
☆ Fusion-Poly: A Polyhedral Framework Based on Spatial-Temporal Fusion for 3D Multi-Object Tracking
LiDAR-camera 3D multi-object tracking (MOT) combines rich visual semantics with accurate depth cues to improve trajectory consistency and tracking reliability. In practice, however, LiDAR and cameras operate at different sampling rates. To maintain temporal alignment, existing data pipelines usually synchronize heterogeneous sensor streams and annotate them at a reduced shared frequency, forcing most prior methods to perform spatial fusion only at synchronized timestamps through projection-based or learnable cross-sensor association. As a result, abundant asynchronous observations remain underexploited, despite their potential to support more frequent association and more robust trajectory estimation over short temporal intervals. To address this limitation, we propose Fusion-Poly, a spatial-temporal fusion framework for 3D MOT that integrates asynchronous LiDAR and camera data. Fusion-Poly associates trajectories with multi-modal observations at synchronized timestamps and with single-modal observations at asynchronous timestamps, enabling higher-frequency updates of motion and existence states. The framework contains three key components: a frequency-aware cascade matching module that adapts to synchronized and asynchronous frames according to available detection modalities; a frequency-aware trajectory estimation module that maintains trajectories through high-frequency motion prediction, differential updates, and confidence-calibrated lifecycle management; and a full-state observation alignment module that improves cross-modal consistency at synchronized timestamps by optimizing image-projection errors. On the nuScenes test set, Fusion-Poly achieves 76.5% AMOTA, establishing a new state of the art among tracking-by-detection 3D MOT methods. Extensive ablation studies further validate the effectiveness of each component. Code will be released.
☆ Edged USLAM: Edge-Aware Event-Based SLAM with Learning-Based Depth Priors ICRA 2026
Conventional visual simultaneous localization and mapping (SLAM) algorithms often fail under rapid motion, low illumination, or abrupt lighting transitions due to motion blur and limited dynamic range. Event cameras mitigate these issues with high temporal resolution and high dynamic range (HDR), but their sparse, asynchronous outputs complicate feature extraction and integration with other sensors; e.g. inertial measurement units (IMUs) and standard cameras. We present Edged USLAM, a hybrid visual-inertial system that extends Ultimate SLAM (USLAM) with an edge-aware front-end and a lightweight depth module. The frontend enhances event frames for robust feature tracking and nonlinear motion compensation, while the depth module provides coarse, region-of-interest (ROI)-based scene depth to improve motion compensation and scale consistency. Evaluations across public benchmarks and real-world unmanned air vehicle (UAV) flights demonstrate that performance varies significantly by scenario. For instance, event-only methods like point-line event-based visual-inertial odometry (PL-EVIO) or learning-based pipelines such as deep event-based visual odometry (DEVO) excel in highly aggressive or extreme HDR conditions. In contrast, Edged USLAM provides superior stability and minimal drift in slow or structured trajectories, ensuring consistently accurate localization on real flights under challenging illumination. These findings highlight the complementary strengths of event-only, learning-based, and hybrid approaches, while positioning Edged USLAM as a robust solution for diverse aerial navigation tasks.
comment: 8 pages, 7 figures, 3 tables. Accepted to ICRA 2026. Project code and datasets available at https://github.com/sebnem-byte/Edged-USLAM
☆ Multifingered force-aware control for humanoid robots ICRA 2026
In this paper, we address force-aware control and force distribution in robotic platforms with multi-fingered hands. Given a target goal and force estimates from tactile sensors, we design a controller that adapts the motion of the torso, arm, wrist, and fingers, redistributing forces to maintain stable contact with objects of varying mass distribution or unstable contacts. To estimate forces, we collect a dataset of tactile signals and ground-truth force measurements using five Xela magnetic sensors interacting with indenters, and train force estimators. We then introduce a model-based control scheme that minimizes the distance between the Center of Pressure (CoP) and the centroid of the fingertips contact polygon. Since our method relies on estimated forces rather than raw tactile signals, it has the potential to be applied to any sensor capable of force estimation. We validate our framework on a balancing task with five objects, achieving a $82.7\%$ success rate, and further evaluate it in multi-object scenarios, achieving $80\%$ accuracy. Code and data can be found here https://github.com/hsp-iit/multifingered-force-aware-control.
comment: This work has been accepted for publication in ICRA 2026
☆ POIROT: Investigating Direct Tangible vs. Digitally Mediated Interaction and Attitude Moderation in Multi-party Murder Mystery Games
As social robots take on increasingly complex roles like game masters (GMs) in multi-party games, the expectation that physicality universally enhances user experience remains debated. This study challenges the "one-size-fits-all" view of tangible interaction by identifying a critical boundary condition: users' Negative Attitudes towards Robots (NARS). In a between-subjects experiment (N = 67), a custom-built robot GM facilitated a multi-party murder mystery game (MMG) by delivering clues either through direct tangible interaction or a digitally mediated interface. Baseline multivariate analysis (MANOVA) showed no significant main effect of delivery modality, confirming that tangibility alone does not guarantee superior engagement. However, primary analysis using multilevel linear models (MLM) revealed a reliable moderation: participants high in NARS experienced markedly lower narrative immersion under tangible delivery, whereas those with low NARS scores showed no such decrement. Qualitative findings further illuminate this divergence: tangibility provides novelty and engagement for some but imposes excessive proxemic friction for anxious users, for whom the digital interface acts as a protective social buffer. These results advance a conditional model of HRI and emphasize the necessity for adaptive systems that can tailor interaction modalities to user predispositions.
comment: 16 pages, 7 figures. Accepted to the 21st ACM/IEEE International Conference on Human-Robot Interaction (HRI 2026)
♻ ☆ Multi-Quadruped Cooperative Object Transport: Learning Decentralized Pinch-Lift-Move ICRA 2026
We study decentralized cooperative transport using teams of N-quadruped robots with arm that must pinch, lift, and move ungraspable objects through physical contact alone. Unlike prior work that relies on rigid mechanical coupling between robots and objects, we address the more challenging setting where mechanically independent robots must coordinate through contact forces alone without any communication or centralized control. To this end, we employ a hierarchical policy architecture that separates base locomotion from arm control, and propose a constellation reward formulation that unifies position and orientation tracking to enforce rigid contact behavior. The key insight is encouraging robots to behave as if rigidly connected to the object through careful reward design and training curriculum rather than explicit mechanical constraints. Our approach enables coordination through shared policy parameters and implicit synchronization cues - scaling to arbitrary team sizes without retraining. We show extensive simulation experiments to demonstrate robust transport across 2-10 robots on diverse object geometries and masses, along with sim2real transfer results on lightweight objects.
comment: Accepted to ICRA 2026. Project page: https://decplm.github.io
♻ ☆ Open-World Task and Motion Planning via Vision-Language Model Genereated Constraints
Foundation models like Vision-Language Models (VLMs) excel at common sense vision and language tasks such as visual question answering. However, they cannot yet directly solve complex, long-horizon robot manipulation problems requiring precise continuous reasoning. Task and Motion Planning (TAMP) systems can handle long-horizon reasoning through discrete-continuous hybrid search over parameterized skills, but rely on detailed environment models and cannot interpret novel human objectives, such as arbitrary natural language goals. We propose integrating VLMs into TAMP systems by having them generate discrete and continuous language-parameterized constraints that enable open-world reasoning. Specifically, we use VLMs to generate discrete action ordering constraints that constrain TAMP search over action sequences, and continuous constraints in the form of code that augments traditional TAMP manipulation constraints. Experiments show that our approach, OWL-TAMP, outperforms baselines relying solely on TAMP or VLMs across several long-horizon manipulation tasks specified directly in natural language. We additionally demonstrate that OWL-TAMP can be deployed with an off-the-shelf TAMP system to solve challenging manipulation tasks on real-world hardware.
comment: A version of this paper appears in IEEE Robotics and Automation Letters (RA-L) Volume 11, Issue 3
♻ ☆ Connectivity Maintenance and Recovery for Multi-Robot Motion Planning
Connectivity is crucial in many multi-robot applications, yet balancing between maintaining it and the fleet's traversability in obstacle-rich environments remains a challenge. Reactive controllers, such as control barrier functions, while providing connectivity guarantees, often struggle to traverse obstacle-rich environments due to deadlocks. We propose a real-time Bézier-based constrained motion planning algorithm, namely, MPC--CLF--CBF, that produces trajectory and control concurrently, under high-order control barrier functions and control Lyapunov functions conditions. Our motion planner significantly improves the navigation success rate of connected fleets in a cluttered workspace and recovers after inevitable connection loss by bypassing obstacles or from an initially disconnected fleet configuration. In addition, our predictive motion planner, owing to its Bézier curve solution, can easily obtain continuous-time arbitrary orders of derivatives, making it suitable for agile differentially flat systems, such as quadrotors. We validate the proposed algorithm through simulations and a physical experiment with $8$ Crazyflie nano-quadrotors.
♻ ☆ Magnetically Driven Elastic Microswimmers: Exploiting Hysteretic Collapse for Autonomous Propulsion and Independent Control
When swimming at low Reynolds numbers, inertial effects are negligible and reciprocal movements cannot induce net motion. Instead, symmetry breaking is necessary to achieve net propulsion. Directed swimming can be supported by magnetic fields, which simultaneously provide a versatile means of remote actuation. Thus, we analyze the motion of a straight microswimmer composed of three magnetizable beads connected by two elastic links. The swimming mechanism is based on oriented external magnetic fields that oscillate in magnitude. Through induced reversible hysteretic collapse of the two segments of the swimmer, the two pairs of beads jump into contact and separate nonreciprocally. Due to higher-order hydrodynamic interactions, net displacement results after each cycle. Different microswimmers can be tuned to different driving amplitudes and frequencies, allowing for simultaneous independent control by just one external magnetic field. The swimmer geometry and magnetic field shape are optimized for maximum swimming speed using an evolutionary optimization strategy. Thanks to the simple working principle, an experimental realization of such a microrobot seems feasible and may open new approaches for microinvasive medical interventions such as targeted drug delivery.
comment: 12 pages, 7 figures, submitted to ACS Nanoscience Au
♻ ☆ CuriousBot: Interactive Mobile Exploration via Actionable 3D Relational Object Graph
Mobile exploration is a longstanding challenge in robotics, yet current methods primarily focus on active perception instead of active interaction, limiting the robot's ability to interact with and fully explore its environment. Existing robotic exploration approaches via active interaction are often restricted to tabletop scenes, neglecting the unique challenges posed by mobile exploration, such as large exploration spaces, complex action spaces, and diverse object relations. In this work, we introduce a 3D relational object graph that encodes diverse object relations and enables exploration through active interaction. We develop a system based on this representation and evaluate it across diverse scenes. Our qualitative and quantitative results demonstrate the system's effectiveness and generalization across object instances, relations, and scenes, outperforming methods solely relying on vision-language models (VLMs).
comment: Accepted to IEEE Robotics and Automation Letters (RA-L). Project Page: https://curiousbot.theaiinstitute.com/
♻ ☆ DemoDiffusion: One-Shot Human Imitation using pre-trained Diffusion Policy ICRA 2026
We propose DemoDiffusion, a simple method for enabling robots to perform manipulation tasks by imitating a single human demonstration, without requiring task-specific training or paired human-robot data. Our approach is based on two insights. First, the hand motion in a human demonstration provides a useful prior for the robot's end-effector trajectory, which we can convert into a rough open-loop robot motion trajectory via kinematic retargeting. Second, while this retargeted motion captures the overall structure of the task, it may not align well with plausible robot actions in-context. To address this, we leverage a pre-trained generalist diffusion policy to modify the trajectory, ensuring it both follows the human motion and remains within the distribution of plausible robot actions. Unlike approaches based on online reinforcement learning or paired human-robot data, our method enables robust adaptation to new tasks and scenes with minimal effort. In real-world experiments across 8 diverse manipulation tasks, DemoDiffusion achieves 83.8\% average success rate, compared to 13.8\% for the pre-trained policy and 52.5\% for kinematic retargeting, succeeding even on tasks where the pre-trained generalist policy fails entirely. Project page: https://demodiffusion.github.io/
comment: 11 pages. Published at ICRA 2026
♻ ☆ From Pixels to Predicates: Learning Symbolic World Models via Pretrained Vision-Language Models
Our aim is to learn to solve long-horizon decision-making problems in complex robotics domains given low-level skills and a handful of short-horizon demonstrations containing sequences of images. To this end, we focus on learning abstract symbolic world models that facilitate zero-shot generalization to novel goals via planning. A critical component of such models is the set of symbolic predicates that define properties of and relationships between objects. In this work, we leverage pretrained vision-language models (VLMs) to propose a large set of visual predicates potentially relevant for decision-making, and to evaluate those predicates directly from camera images. At training time, we pass the proposed predicates and demonstrations into an optimization-based model-learning algorithm to obtain an abstract symbolic world model that is defined in terms of a compact subset of the proposed predicates. At test time, given a novel goal in a novel setting, we use the VLM to construct a symbolic description of the current world state, and then use a search-based planning algorithm to find a sequence of low-level skills that achieves the goal. We demonstrate empirically across experiments in both simulation and the real world that our method can generalize aggressively, applying its learned world model to solve problems with a wide variety of object types, arrangements, numbers of objects, and visual backgrounds, as well as novel goals and much longer horizons than those seen at training time.
comment: A version of this paper appears in the official proceedings of RA-L, Volume 11, Issue 4
♻ ☆ BEV-Patch-PF: Particle Filtering with BEV-Aerial Feature Matching for Off-Road Geo-Localization
We propose BEV-Patch-PF, a GPS-free sequential geo-localization system that integrates a particle filter with learned bird's-eye-view (BEV) and aerial feature maps. From onboard RGB and depth images, we construct a BEV feature map. For each 3-DoF particle pose hypothesis, we crop the corresponding patch from an aerial feature map computed from a local aerial image queried around the approximate location. BEV-Patch-PF computes a per-particle log-likelihood by matching the BEV feature to the aerial patch feature. On two real-world off-road datasets, our method achieves 9.7x lower absolute trajectory error (ATE) on seen routes and 6.6x lower ATE on unseen routes than a retrieval-based baseline, while maintaining accuracy under dense canopy and shadow. The system runs in real time at 10 Hz on an NVIDIA Tesla T4, enabling practical robot deployment.
♻ ☆ Task-Oriented Robot-Human Handovers on Legged Manipulators
Task-oriented handovers (TOH) are fundamental to effective human-robot collaboration, requiring robots to present objects in a way that supports the human's intended post-handover use. Existing approaches are typically based on object- or task-specific affordances, but their ability to generalize to novel scenarios is limited. To address this gap, we present AFT-Handover, a framework that integrates large language model (LLM)-driven affordance reasoning with efficient texture-based affordance transfer to achieve zero-shot, generalizable TOH. Given a novel object-task pair, the method retrieves a proxy exemplar from a database, establishes part-level correspondences via LLM reasoning, and texturizes affordances for feature-based point cloud transfer. We evaluate AFT-Handover across diverse task-object pairs, showing improved handover success rates and stronger generalization compared to baselines. In a comparative user study, our framework is significantly preferred over the current state-of-the-art, effectively reducing human regrasping before tool use. Finally, we demonstrate TOH on legged manipulators, highlighting the potential of our framework for real-world robot-human handovers.
comment: Accepted to 21st ACM/IEEE International Conference on Human-Robot Interaction (HRI) 2026
♻ ☆ LIVE-GS: Online LiDAR-Inertial-Visual State Estimation and Globally Consistent Mapping with 3D Gaussian Splatting
While 3D Gaussian Splatting (3DGS) enabled photorealistic mapping, its integration into SLAM has largely followed traditional camera-centric pipelines. As a result, they inherit well-known weaknesses such as high computational load, failure in texture-poor or illumination-varying environments, and limited operational range, particularly for RGB-D setups. On the other hand, LiDAR emerges as a robust alternative, but its integration with 3DGS introduces new challenges, such as the need for tighter global alignment for photorealistic quality and prolonged optimization times caused by sparse data. To address these challenges, we propose LIVE-GS, an online LiDAR-Inertial Visual SLAM framework that tightly couples 3D Gaussian Splatting with LiDAR-based surfels to ensure high-precision map consistency through global geometric optimization. Particularly, to handle sparse data, our system employs a depth-invariant Gaussian initialization strategy for efficient representation and a bounded sigmoid constraint to prevent uncontrolled Gaussian growth. Experiments on public and our datasets demonstrate competitive performance in rendering quality and map-building efficiency compared with representative 3DGS SLAM baselines.
♻ ☆ $π$-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs
Flow-based vision-language-action (VLA) models excel in embodied control but suffer from intractable likelihoods during multi-step sampling, hindering online reinforcement learning. We propose \textbf{\textit{$\boldsymbolπ$-StepNFT}} (Step-wise Negative-aware Fine-Tuning), a critic-and-likelihood-free framework that requires only a single forward pass per optimization step and eliminates auxiliary value networks. We identify that wider exploration spaces necessitate finer-grained, step-wise guidance for alignment. Empirically, $π$-StepNFT unlocks latent potential on LIBERO with competitive few-shot robustness. Moreover, it achieves superior generalization on ManiSkill, outperforming value-based baselines in OOD scenarios by preventing overfitting to multimodal features. This property offers a scalable solution promising for complex real-world applications.
♻ ☆ Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations
Generalizing skill policies to novel conditions remains a key challenge in robot learning. Imitation learning methods, while data-efficient, are largely confined to the training region and consistently fail on input data outside it, leading to unpredictable policy failures. Alternatively, transfer learning approaches offer methods for trajectory generation robust to both changes in environment or tasks, but they remain data-hungry and lack accuracy in zero-shot generalization. We address these challenges by framing the problem in the context of task inversion learning and proposing a novel joint learning approach to achieve accurate and efficient knowledge transfer. Our method constructs a common representation of the forward and inverse tasks, and leverages auxiliary forward demonstrations from novel configurations to successfully execute the corresponding inverse tasks, without any direct supervision. We show the extrapolation capabilities of our framework via ablation studies and experiments in simulated and real-world environments that require complex manipulation skills with a diverse set of objects and tools, where we outperform diffusion-based alternatives.
comment: Corrected author affiliation
♻ ☆ CroSTAta: Cross-State Transition Attention Transformer for Robotic Manipulation
Learning robotic manipulation policies through supervised learning from demonstrations remains challenging when policies encounter execution variations not explicitly covered during training. While incorporating historical context through attention mechanisms can improve robustness, standard approaches process all past states in a sequence without explicitly modeling the temporal structure that demonstrations may include, such as failure and recovery patterns. We propose a Cross-State Transition Attention Transformer that employs a novel State Transition Attention (STA) mechanism to modulate standard attention weights based on learned state evolution patterns, enabling policies to better adapt their behavior based on execution history. Our approach combines this structured attention with temporal masking during training, where visual information is randomly removed from recent timesteps to encourage temporal reasoning from historical context. Evaluation in simulation shows that STA consistently outperforms standard attention approach and temporal modeling methods like TCN and LSTM networks, achieving more than 2x improvement over cross-attention on precision-critical tasks. The source code and data can be accessed at https://github.com/iit-DLSLab/croSTAta
comment: Code and data available at https://github.com/iit-DLSLab/croSTAta
♻ ☆ EasyInsert: A Data-Efficient and Generalizable Insertion Policy
Robotic insertion is a highly challenging task that requires exceptional precision in cluttered environments. Existing methods often have poor generalization capabilities. They typically function in restricted and structured environments, and frequently fail when the plug and socket are far apart, when the scene is densely cluttered, or when handling novel objects. They also rely on strong assumptions such as access to CAD models or a digital twin in simulation. To address these limitations, we propose EasyInsert. Inspired by human intuition, it formulates insertion as a delta-pose regression problem, which unlocks an efficient, highly scalable data collection pipeline with minimal human labor to train an end-to-end visual policy. During execution, the visual policy predicts the relative pose between plug and socket to drive a multi-phase, coarse-to-fine insertion process. EasyInsert demonstrates strong zero-shot generalization capability for unseen objects in cluttered environments, robustly handling cases with significant initial pose deviations. In real-world experiments, by leveraging just 1 hour of human teleoperation data to bootstrap a large-scale automated data collection process, EasyInsert achieves an over 90% success rate in zero-shot insertion for 13 out of 15 unseen novel objects, including challenging objects like Type-C cables, HDMI cables, and Ethernet cables. Furthermore, requiring only a single manual reset, EasyInsert allows for fast adaptation to novel test objects through automated data collection and fine-tuning, achieving an over 90% success rate across all 15 objects.
♻ ☆ Scalable Aerial GNSS Localization for Marine Robots
Accurate localization is crucial for water robotics, yet traditional onboard Global Navigation Satellite System (GNSS) approaches are difficult or ineffective due to signal reflection on the water's surface and its high cost of aquatic GNSS receivers. Existing approaches, such as inertial navigation, Doppler Velocity Loggers (DVL), SLAM, and acoustic-based methods, face challenges like error accumulation and high computational complexity. Therefore, a more efficient and scalable solution remains necessary. This paper proposes an alternative approach that leverages an aerial drone equipped with GNSS localization to track and localize a marine robot once it is near the surface of the water. Our results show that this novel adaptation enables accurate single and multi-robot marine robot localization.
comment: International Conference on Robotics and Automation 2025 Workshop Robots in the Wild
♻ ☆ ViTaPEs: Visuotactile Position Encodings for Cross-Modal Alignment in Multimodal Transformers
Tactile sensing provides local essential information that is complementary to visual perception, such as texture, compliance, and force. Despite recent advances in visuotactile representation learning, challenges remain in fusing these modalities and generalizing across tasks and environments without heavy reliance on pre-trained vision-language models. Moreover, existing methods do not study positional encodings, thereby overlooking the multi-stage spatial reasoning needed to capture fine-grained visuotactile correlations. We introduce ViTaPEs, a transformer-based architecture for learning task-agnostic visuotactile representations from paired vision and tactile inputs. Our key idea is a two-stage positional injection: local (modality-specific) positional encodings are added within each stream, and a global positional encoding is added on the joint token sequence immediately before attention, providing a shared positional vocabulary at the stage where cross-modal interaction occurs. We make the positional injection points explicit and conduct controlled ablations that isolate their effect before a token-wise nonlinearity versus immediately before self-attention. Experiments on multiple large-scale real-world datasets show that ViTaPEs not only surpasses state-of-the-art baselines across various recognition tasks but also demonstrates zero-shot generalization to unseen, out-of-domain scenarios. We further demonstrate the transfer-learning strength of ViTaPEs in a robotic grasping task, where it outperforms state-of-the-art baselines in predicting grasp success. Project page: https://sites.google.com/view/vitapes
♻ ☆ RoboLayout: Differentiable 3D Scene Generation for Embodied Agents
Recent advances in vision language models (VLMs) have shown strong potential for spatial reasoning and 3D scene layout generation from open-ended language instructions. However, generating layouts that are not only semantically coherent but also feasible for interaction by embodied agents remains challenging, particularly in physically constrained indoor environments. In this paper, RoboLayout is introduced as an extension of LayoutVLM that augments the original framework with agent-aware reasoning and improved optimization stability. RoboLayout integrates explicit reachability constraints into a differentiable layout optimization process, enabling the generation of layouts that are navigable and actionable by embodied agents. Importantly, the agent abstraction is not limited to a specific robot platform and can represent diverse entities with distinct physical capabilities, such as service robots, warehouse robots, humans of different age groups, or animals, allowing environment design to be tailored to the intended agent. In addition, a local refinement stage is proposed that selectively reoptimizes problematic object placements while keeping the remainder of the scene fixed, improving convergence efficiency without increasing global optimization iterations. Overall, RoboLayout preserves the strong semantic alignment and physical plausibility of LayoutVLM while enhancing applicability to agent-centric indoor scene generation, as demonstrated by experimental results across diverse scene configurations.
♻ ☆ FreeTacMan: Robot-free Visuo-Tactile Data Collection System for Contact-rich Manipulation
Enabling robots with contact-rich manipulation remains a pivotal challenge in robot learning, which is substantially hindered by the data collection gap, including its inefficiency and limited sensor setup. While prior work has explored handheld paradigms, their rod-based mechanical structures remain rigid and unintuitive, providing limited tactile feedback and posing challenges for operators. Motivated by the dexterity and force feedback of human motion, we propose FreeTacMan, a human-centric and robot-free data collection system for accurate and efficient robot manipulation. Concretely, we design a wearable gripper with visuo-tactile sensors for data collection, which can be worn by human fingers for intuitive control. A high-precision optical tracking system is introduced to capture end-effector poses while synchronizing visual and tactile feedback simultaneously. We leverage FreeTacMan to collect a large-scale multimodal dataset, comprising over 3000k paired visuo-tactile images with end-effector poses, 10k demonstration trajectories across 50 diverse contact-rich manipulation tasks. FreeTacMan achieves multiple improvements in data collection performance over prior works and enables effective policy learning from self-collected datasets. By open-sourcing the hardware and the dataset, we aim to facilitate reproducibility and support research in visuo-tactile manipulation.
♻ ☆ Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid Control
Physics-based humanoid control relies on training with motion datasets that have diverse data distributions. However, the fixed difficulty distribution of datasets limits the performance ceiling of the trained control policies. Additionally, the method of acquiring high-quality data through professional motion capture systems is constrained by costs, making it difficult to achieve large-scale scalability. To address these issues, we propose a closed-loop automated motion data generation and iterative framework. It can generate high-quality motion data with rich action semantics, including martial arts, dance, combat, sports, gymnastics, and more. Furthermore, our framework enables difficulty iteration of policies and data through physical metrics and objective evaluations, allowing the trained tracker to break through its original difficulty limits. On the PHC single-primitive tracker, using only approximately 1/10 of the AMASS dataset size, the average failure rate on the test set (2201 clips) is reduced by 45% compared to the baseline. Finally, we conduct comprehensive ablation and comparative experiments to highlight the rationality and advantages of our framework.
♻ ☆ MetricNet: Recovering Metric Scale in Generative Navigation Policies ICRA'26
Generative navigation policies have made rapid progress in improving end-to-end learned navigation. Despite their promising results, this paradigm has two structural problems. First, the sampled trajectories exist in an abstract, unscaled space without metric grounding. Second, the control strategy discards the full path, instead moving directly towards a single waypoint. This leads to short-sighted and unsafe actions, moving the robot towards obstacles that a complete and correctly scaled path would circumvent. To address these issues, we propose MetricNet, an effective add-on for generative navigation that predicts the metric distance between waypoints, grounding policy outputs in metric coordinates. We evaluate our method in simulation with a new benchmarking framework and show that executing MetricNet-scaled waypoints significantly improves both navigation and exploration performance. Beyond simulation, we further validate our approach in real-world experiments. Finally, we propose MetricNav, which integrates MetricNet into a navigation policy to guide the robot away from obstacles while still moving towards the goal.
comment: Accepted to ICRA'26
♻ ☆ Synchronized Online Friction Estimation and Adaptive Grasp Control for Robust Gentle Grasp
We introduce a unified framework for gentle robotic grasping that synergistically couples real-time friction estimation with adaptive grasp control. We propose a new particle filter-based method for real-time estimation of the friction coefficient using vision-based tactile sensors. This estimate is seamlessly integrated into a reactive controller that dynamically modulates grasp force to maintain a stable grip. The two processes operate synchronously in a closed-loop: the controller uses the current best estimate to adjust the force, while new tactile feedback from this action continuously refines the estimation. This creates a highly responsive and robust sensorimotor cycle. The reliability and efficiency of the complete framework are validated through extensive robotic experiments.
♻ ☆ MCGS-SLAM: A Multi-Camera SLAM Framework Using Gaussian Splatting for High-Fidelity Mapping ICRA
Recent progress in dense SLAM has primarily targeted monocular setups, often at the expense of robustness and geometric coverage. We present MCGS-SLAM, the first purely RGB-based multi-camera SLAM system built on 3D Gaussian Splatting (3DGS). Unlike prior methods relying on sparse maps or inertial data, MCGS-SLAM fuses dense RGB inputs from multiple viewpoints into a unified, continuously optimized Gaussian map. A multi-camera bundle adjustment (MCBA) jointly refines poses and depths via dense photometric and geometric residuals, while a scale consistency module enforces metric alignment across views using low-rank priors. The system supports RGB input and maintains real-time performance at large scale. Experiments on synthetic and real-world datasets show that MCGS-SLAM consistently yields accurate trajectories and photorealistic reconstructions, usually outperforming monocular baselines. Notably, the wide field of view from multi-camera input enables reconstruction of side-view regions that monocular setups miss, critical for safe autonomous operation. These results highlight the promise of multi-camera Gaussian Splatting SLAM for high-fidelity mapping in robotics and autonomous driving.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
♻ ☆ Input-to-State Stable Coupled Oscillator Networks for Closed-form Model-based Control in Latent Space NeurIPS 2024
Even though a variety of methods have been proposed in the literature, efficient and effective latent-space control (i.e., control in a learned low-dimensional space) of physical systems remains an open challenge. We argue that a promising avenue is to leverage powerful and well-understood closed-form strategies from control theory literature in combination with learned dynamics, such as potential-energy shaping. We identify three fundamental shortcomings in existing latent-space models that have so far prevented this powerful combination: (i) they lack the mathematical structure of a physical system, (ii) they do not inherently conserve the stability properties of the real systems, (iii) these methods do not have an invertible mapping between input and latent-space forcing. This work proposes a novel Coupled Oscillator Network (CON) model that simultaneously tackles all these issues. More specifically, (i) we show analytically that CON is a Lagrangian system - i.e., it possesses well-defined potential and kinetic energy terms. Then, (ii) we provide formal proof of global Input-to-State stability using Lyapunov arguments. Moving to the experimental side, we demonstrate that CON reaches SoA performance when learning complex nonlinear dynamics of mechanical systems directly from images. An additional methodological innovation contributing to achieving this third goal is an approximated closed-form solution for efficient integration of network dynamics, which eases efficient training. We tackle (iii) by approximating the forcing-to-input mapping with a decoder that is trained to reconstruct the input based on the encoded latent space force. Finally, we show how these properties enable latent-space control. We use an integral-saturated PID with potential force compensation and demonstrate high-quality performance on a soft robot using raw pixels as the only feedback information.
comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024) spotlight, 50 pages
♻ ☆ M4Diffuser: Multi-View Diffusion Policy with Manipulability-Aware Control for Robust Mobile Manipulation
Mobile manipulation requires the coordinated control of a mobile base and a robotic arm while simultaneously perceiving both global scene context and fine-grained object details. Existing single-view approaches often fail in unstructured environments due to limited fields of view, exploration, and generalization abilities. Moreover, classical controllers, although stable, struggle with efficiency and manipulability near singularities. To address these challenges, we propose M4Diffuser, a hybrid framework that integrates a Multi-View Diffusion Policy with a novel Reduced and Manipulability-aware QP (ReM-QP) controller for mobile manipulation. The diffusion policy leverages proprioceptive states and complementary camera perspectives with both close-range object details and global scene context to generate task-relevant end-effector goals in the world frame. These high-level goals are then executed by the ReM-QP controller, which eliminates slack variables for computational efficiency and incorporates manipulability-aware preferences for robustness near singularities. Comprehensive experiments in simulation and real-world environments show that M4Diffuser achieves 7 to 56 percent higher success rates and reduces collisions by 3 to 31 percent over baselines. Our approach demonstrates robust performance for smooth whole-body coordination, and strong generalization to unseen tasks, paving the way for reliable mobile manipulation in unstructured environments. Details of the demo and supplemental material are available on our project website https://sites.google.com/view/m4diffuser.
comment: Project page: https://sites.google.com/view/m4diffuser, 10 pages, 9 figures
♻ ☆ Pretraining in Actor-Critic Reinforcement Learning for Robot Locomotion
The pretraining-finetuning paradigm has facilitated numerous transformative advancements in artificial intelligence research in recent years. However, in the domain of reinforcement learning (RL) for robot locomotion, individual skills are often learned from scratch despite the high likelihood that some generalizable knowledge is shared across all task-specific policies belonging to the same robot embodiment. This work aims to define a paradigm for pretraining neural network models that encapsulate such knowledge and can subsequently serve as a basis for warm-starting the RL process in classic actor-critic algorithms, such as Proximal Policy Optimization (PPO). We begin with a task-agnostic exploration-based data collection algorithm to gather diverse, dynamic transition data, which is then used to train a Proprioceptive Inverse Dynamics Model (PIDM) through supervised learning. The pretrained weights are then loaded into both the actor and critic networks to warm-start the policy optimization of actual tasks. We systematically validated our proposed method with 9 distinct robot locomotion RL environments comprising 3 different robot embodiments, showing significant benefits of this initialization strategy. Our proposed approach on average improves sample efficiency by 36.9% and task performance by 7.3% compared to random initialization. We further present key ablation studies and empirical analyses that shed light on the mechanisms behind the effectiveness of this method.
♻ ☆ ClearDepth: Enhanced Stereo Perception of Transparent Objects for Robotic Manipulation
Transparent object depth perception poses a challenge in everyday life and logistics, primarily due to the inability of standard 3D sensors to accurately capture depth on transparent or reflective surfaces. This limitation significantly affects depth map and point cloud-reliant applications, especially in robotic manipulation. We developed a vision transformer-based algorithm for stereo depth recovery of transparent objects. This approach is complemented by an innovative feature post-fusion module, which enhances the accuracy of depth recovery by structural features in images. To address the high costs associated with dataset collection for stereo camera-based perception of transparent objects, our method incorporates a parameter-aligned, domain-adaptive, and physically realistic Sim2Real simulation for efficient data generation, accelerated by AI algorithm. Our experimental results demonstrate the model's exceptional Sim2Real generalizability in real-world scenarios, enabling precise depth mapping of transparent objects to assist in robotic manipulation. Project details are available at https://sites.google.com/view/cleardepth/ .
comment: 9 pages
♻ ☆ NaviTrace: Evaluating Embodied Navigation of Vision-Language Models ICRA 2026
Vision-language models demonstrate unprecedented performance and generalization across a wide range of tasks and scenarios. Integrating these foundation models into robotic navigation systems opens pathways toward building general-purpose robots. Yet, evaluating these models' navigation capabilities remains constrained by costly real-world trials, overly simplified simulations, and limited benchmarks. We introduce NaviTrace, a high-quality Visual Question Answering benchmark where a model receives an instruction and embodiment type (human, legged robot, wheeled robot, bicycle) and must output a 2D navigation trace in image space. Across 1000 scenarios and more than 3000 expert traces, we systematically evaluate eight state-of-the-art VLMs using a newly introduced semantic-aware trace score. This metric combines Dynamic Time Warping distance, goal endpoint error, and embodiment-conditioned penalties derived from per-pixel semantics and correlates with human preferences. Our evaluation reveals consistent gap to human performance caused by poor spatial grounding and goal localization. NaviTrace establishes a scalable and reproducible benchmark for real-world robotic navigation. The benchmark and leaderboard can be found at https://leggedrobotics.github.io/navitrace_webpage/.
comment: 11 pages, 6 figures, with appendix, accepted to ICRA 2026
Robotics 4
☆ Perceptive Variable-Timing Footstep Planning for Humanoid Locomotion on Disconnected Footholds
Many real-world walking scenarios contain obstacles and unsafe ground patches (e.g., slippery or cluttered areas), leaving a disconnected set of admissible footholds that can be modeled as stepping-stone-like regions. We propose an onboard, perceptive mixed-integer model predictive control framework that jointly plans foot placement and step duration using step-to-step Divergent Component of Motion (DCM) dynamics. Ego-centric depth images are fused into a probabilistic local heightmap, from which we extract a union of convex steppable regions. Region membership is enforced with binary variables in a mixed-integer quadratic program (MIQP). To keep the optimization tractable while certifying safety, we embed capturability bounds in the DCM space: a lateral one-step condition (preventing leg crossing) and a sagittal infinite-step bound that limits unstable growth. We further re-plan within the step by back-propagating the measured instantaneous DCM to update the initial DCM, improving robustness to model mismatch and external disturbances. We evaluate the approach in simulation on Digit on randomized stepping-stone fields, including external pushes. The planner generates terrain-aware, dynamically consistent footstep sequences with adaptive timing and millisecond-level solve times.
comment: 8 pages, 5 figures, 1 table, 3 algorithms. Supplemental video at: https://youtu.be/5EeuBnSb66s
☆ Underwater Embodied Intelligence for Autonomous Robots: A Constraint-Coupled Perspective on Planning, Control, and Deployment
Autonomous underwater robots are increasingly deployed for environmental monitoring, infrastructure inspection, subsea resource exploration, and long-horizon exploration. Yet, despite rapid advances in learning-based planning and control, reliable autonomy in real ocean environments remains fundamentally constrained by tightly coupled physical limits. Hydrodynamic uncertainty, partial observability, bandwidth-limited communication, and energy scarcity are not independent challenges; they interact within the closed perception-planning-control loop and often amplify one another over time. This Review develops a constraint-coupled perspective on underwater embodied intelligence, arguing that planning and control must be understood within tightly coupled sensing, communication, coordination, and resource constraints in real ocean environments. We synthesize recent progress in reinforcement learning, belief-aware planning, hybrid control, multi-robot coordination, and foundation-model integration through this embodied perspective. Across representative application domains, we show how environmental monitoring, inspection, exploration, and cooperative missions expose distinct stress profiles of cross-layer coupling. To unify these observations, we introduce a cross-layer failure taxonomy spanning epistemic, dynamic, and coordination breakdowns, and analyze how errors cascade across autonomy layers under uncertainty. Building on this structure, we outline research directions toward physics-grounded world models, certifiable learning-enabled control, communication-aware coordination, and deployment-aware system design. By internalizing constraint coupling rather than treating it as an external disturbance, underwater embodied intelligence may evolve from performance-driven adaptation toward resilient, scalable, and verifiable autonomy under real ocean conditions.
comment: This article is currently under review
♻ ☆ Let's Think in Two Steps: Mitigating Agreement Bias in MLLMs with Self-Grounded Verification ICLR 2026
Verifiers--functions assigning rewards to agent behavior--have been key to AI progress in math, code, and games. However, extending gains to domains without clear-cut success criteria remains a challenge: while humans can recognize desired outcomes, translating this intuition into scalable rules is nontrivial. Multimodal LLMs (MLLMs) offer a promising solution, given their world knowledge, human-preference alignment, and reasoning capabilities. We evaluate MLLM verifiers across web navigation, computer use, and robotics, spanning 13+ models, 28+ designs, and thousands of trajectories from diverse agents. We identify a critical limitation: a strong tendency for MLLMs to over-validate agent behavior--a phenomenon we term agreement bias. This bias is pervasive, resilient to test-time scaling, and can harm applications relying on MLLM judgments/rewards (e.g., self-improvement, steering, online supervision). We discuss several considerations for evaluating and designing MLLM verifiers, and introduce SGV, a lightweight method that better leverages their capabilities by modulating (un)conditional generation. First, an MLLM is elicited to generate broad priors about desired behavior, independent of the data under evaluation. Then, conditioned on self-generated priors, it reasons over and evaluates a candidate trajectory. Our methods yield more human-aligned verifiers, improving failure detection by 25pp and accuracy by 14pp. In self-improvement and online supervision, they boost task completion of a GUI specialist in OSWorld, a diffusion policy in robomimic, and a ReAct agent in VisualWebArena--surpassing the previous state of the art by 20pp. As a byproduct, we release an update of VisualWebArena featuring strong agent baselines, more human-aligned oracles, container parallelism with high fidelity and proper resets, >10x speedups, and VWA-Lite, a 1/3 subset with comparable evaluation fidelity.
comment: ICLR 2026. Code, models, and data publicly available at https://mshalimay.github.io/agreement-bias-sgv/
♻ ☆ MEM: Multi-Scale Embodied Memory for Vision Language Action Models
Conventionally, memory in end-to-end robotic learning involves inputting a sequence of past observations into the learned policy. However, in complex multi-stage real-world tasks, the robot's memory must represent past events at multiple levels of granularity: from long-term memory that captures abstracted semantic concepts (e.g., a robot cooking dinner should remember which stages of the recipe are already done) to short-term memory that captures recent events and compensates for occlusions (e.g., a robot remembering the object it wants to pick up once its arm occludes it). In this work, our main insight is that an effective memory architecture for long-horizon robotic control should combine multiple modalities to capture these different levels of abstraction. We introduce Multi-Scale Embodied Memory (MEM), an approach for mixed-modal long-horizon memory in robot policies. MEM combines video-based short-horizon memory, compressed via a video encoder, with text-based long-horizon memory. Together, they enable robot policies to perform tasks that span up to fifteen minutes, like cleaning up a kitchen, or preparing a grilled cheese sandwich. Additionally, we find that memory enables MEM policies to intelligently adapt manipulation strategies in-context.
comment: Website: https://pi.website/research/memory
Robotics 47
☆ A Distributed Gaussian Process Model for Multi-Robot Mapping ICRA 2026
We propose DistGP: a multi-robot learning method for collaborative learning of a global function using only local experience and computation. We utilise a sparse Gaussian process (GP) model with a factorisation that mirrors the multi-robot structure of the task, and admits distributed training via Gaussian belief propagation (GBP). Our loopy model outperforms Tree-Structured GPs \cite{bui2014tree} and can be trained online and in settings with dynamic connectivity. We show that such distributed, asynchronous training can reach the same performance as a centralised, batch-trained model, albeit with slower convergence. Last, we compare to DiNNO \cite{yu2022dinno}, a distributed neural network (NN) optimiser, and find DistGP achieves superior accuracy, is more robust to sparse communication and is better able to learn continually.
comment: ICRA 2026, 8 pages
☆ A Lightweight Digital-Twin-Based Framework for Edge-Assisted Vehicle Tracking and Collision Prediction
Vehicle tracking, motion estimation, and collision prediction are fundamental components of traffic safety and management in Intelligent Transportation Systems (ITS). Many recent approaches rely on computationally intensive prediction models, which limits their practical deployment on resource-constrained edge devices. This paper presents a lightweight digital-twin-based framework for vehicle tracking and spatiotemporal collision prediction that relies solely on object detection, without requiring complex trajectory prediction networks. The framework is implemented and evaluated in Quanser Interactive Labs (QLabs), a high-fidelity digital twin of an urban traffic environment that enables controlled and repeatable scenario generation. A YOLO-based detector is deployed on simulated edge cameras to localize vehicles and extract frame-level centroid trajectories. Offline path maps are constructed from multiple traversals and indexed using K-D trees to support efficient online association between detected vehicles and road segments. During runtime, consistent vehicle identifiers are maintained, vehicle speed and direction are estimated from the temporal evolution of path indices, and future positions are predicted accordingly. Potential collisions are identified by analyzing both spatial proximity and temporal overlap of predicted future trajectories. Our experimental results across diverse simulated urban scenarios show that the proposed framework predicts approximately 88% of collision events prior to occurrence while maintaining low computational overhead suitable for edge deployment. Rather than introducing a computationally intensive prediction model, this work introduces a lightweight digital-twin-based solution for vehicle tracking and collision prediction, tailored for real-time edge deployment in ITS.
comment: 6 pages, 2 figures, IEEE ICC 2026 Workshops (under submission)
☆ Faster-HEAL: An Efficient and Privacy-Preserving Collaborative Perception Framework for Heterogeneous Autonomous Vehicles
Collaborative perception (CP) is a promising paradigm for improving situational awareness in autonomous vehicles by overcoming the limitations of single-agent perception. However, most existing approaches assume homogeneous agents, which restricts their applicability in real-world scenarios where vehicles use diverse sensors and perception models. This heterogeneity introduces a feature domain gap that degrades detection performance. Prior works address this issue by retraining entire models/major components, or using feature interpreters for each new agent type, which is computationally expensive, compromises privacy, and may reduce single-agent accuracy. We propose Faster-HEAL, a lightweight and privacy-preserving CP framework that fine-tunes a low-rank visual prompt to align heterogeneous features with a unified feature space while leveraging pyramid fusion for robust feature aggregation. This approach reduces the trainable parameters by 94%, enabling efficient adaptation to new agents without retraining large models. Experiments on the OPV2V-H dataset show that Faster-HEAL improves detection performance by 2% over state-of-the-art methods with significantly lower computational overhead, offering a practical solution for scalable heterogeneous CP.
comment: Accepted to appear in the 2026 IEEE Intelligent Vehicles Symposium (IV 2026), Detroit, MI, USA, June 22-25, 2026. 6 pages, 1 figure, 4 tables
☆ Soft Rigid Hybrid Gripper with Inflatable Silicone Pockets for Tunable Frictional Grasping
Grasping objects with diverse mechanical properties, such as heavy, slippery, or fragile items, remains a significant challenge in robotics. Conventional rigid grippers typically rely on increasing the normal forces to secure an object, however, this can cause damage to fragile objects due to excessive force. To address this limitation, we propose a soft rigid hybrid gripper finger that combines rigid structural shells with soft, inflatable silicone pockets, which could be integrated into a conventional gripper. The hybrid gripper can actively modulate its surface friction by varying the internal air pressure of the silicone pockets, enabling the gripper to securely grasp objects without increasing the gripping force. This is demonstrated by fundamental experimental results, in which an increase in internal pressure leads to a proportional increase in the effective coefficient of friction. The gripping experiments also show that the integrated gripper can stably lift heavy and slippery objects or fragile, deformable objects, such as eggs, tofu, fruits, and paper cups, with minimal damage by increasing friction rather than applying high force.
☆ Kinematics-Aware Latent World Models for Data-Efficient Autonomous Driving SC
Data-efficient learning remains a central challenge in autonomous driving due to the high cost and safety risks of large-scale real-world interaction. Although world-model-based reinforcement learning enables policy optimization through latent imagination, existing approaches often lack explicit mechanisms to encode spatial and kinematic structure essential for driving tasks. In this work, we build upon the Recurrent State-Space Model (RSSM) and propose a kinematics-aware latent world model framework for autonomous driving. Vehicle kinematic information is incorporated into the observation encoder to ground latent transitions in physically meaningful motion dynamics, while geometry-aware supervision regularizes the RSSM latent state to capture task-relevant spatial structure beyond pixel reconstruction. The resulting structured latent dynamics improve long-horizon imagination fidelity and stabilize policy optimization. Experiments in a driving simulation benchmark demonstrate consistent gains over both model-free and pixel-based world-model baselines in terms of sample efficiency and driving performance. Ablation studies further verify that the proposed design enhances spatial representation quality within the latent space. These results suggest that integrating kinematic grounding into RSSM-based world models provides a scalable and physically grounded paradigm for autonomous driving policy learning.
comment: 6 pages, 5 figures. Under review at IEEE ITSC
☆ Vision-Guided MPPI for Agile Drone Racing: Navigating Arbitrary Gate Poses via Neural Signed Distance Fields
Autonomous drone racing requires the tight coupling of perception, planning, and control under extreme agility. However, recent approaches typically rely on precomputed spatial reference trajectories or explicit 6-DoF gate pose estimation, rendering them brittle to spatial perturbations, unmodeled track changes, and sensor noise. Conversely, end-to-end learning policies frequently overfit to specific track layouts and struggle with zero-shot generalization. To address these fundamental limitations, we propose a fully onboard, vision guided optimal control framework that enables reference-free agile flight through arbitrarily placed and oriented gates. Central to our approach is Gate-SDF, a novel, implicitly learned neural signed distance field. Gate-SDF directly processes raw, noisy depth images to predict a continuous spatial field that provides both collision repulsion and active geometric guidance toward the valid traversal area. We seamlessly integrate this representation into a sampling-based Model Predictive Path Integral (MPPI) controller. By fully exploiting GPU parallelism, the framework evaluates these continuous spatial constraints across thousands of simulated trajectory rollouts simultaneously in real time. Furthermore, our formulation inherently maintains spatial consistency, ensuring robust navigation even under severe visual occlusion during aggressive maneuvers. Extensive simulations and real-world experiments demonstrate that the proposed system achieves high-speed agile flight and successfully navigates unseen tracks subject to severe unmodeled gate displacements and orientation perturbations. Videos are available at https://zhaofangguo.github.io/vision_guided_mppi/
☆ RoTri-Diff: A Spatial Robot-Object Triadic Interaction-Guided Diffusion Model for Bimanual Manipulation ICRA 2026
Bimanual manipulation is a fundamental robotic skill that requires continuous and precise coordination between two arms. While imitation learning (IL) is the dominant paradigm for acquiring this capability, existing approaches, whether robot-centric or object-centric, often overlook the dynamic geometric relationship among the two arms and the manipulated object. This limitation frequently leads to inter-arm collisions, unstable grasps, and degraded performance in complex tasks. To address this, in this paper we explicitly models the Robot-Object Triadic Interaction (RoTri) representation in bimanual systems, by encoding the relative 6D poses between the two arms and the object to capture their spatial triadic relationship and establish continuous triangular geometric constraints. Building on this, we further introduce RoTri-Diff, a diffusion-based imitation learning framework that combines RoTri constraints with robot keyposes and object motion in a hierarchical diffusion process. This enables the generation of stable, coordinated trajectories and robust execution across different modes of bimanual manipulation. Extensive experiments show that our approach outperforms state-of-the-art baselines by 10.2% on 11 representative RLBench2 tasks and achieves stable performance on 4 challenging real-world bimanual tasks. Project website: https://rotri-diff.github.io/.
comment: ICRA 2026
☆ Tutorial on Aided Inertial Navigation Systems: A Modern Treatment Using Lie-Group Theoretical Methods
This tutorial presents a control-oriented introduction to aided inertial navigation systems using a Lie-group formulation centered on the extended Special Euclidean group SE_2(3). The focus is on developing a clear and implementation-oriented geometric framework for fusing inertial measurements with aiding information, while making the role of invariance and symmetry explicit. Recent extensions, including higher-order state representations, synchronous observer designs, and equivariant filtering methods, are discussed as natural continuations of the same underlying principles. The goal is to provide readers with a coherent system-theoretic perspective that supports both understanding and practical use of modern aided inertial navigation methods.
☆ Model-based thermal drift compensation for high-precision hexapod robot actuators
Thermal expansion is a significant source of positioning error in high-precision hexapod robots (Gough-Stewart platforms). Any variation in the temperature of the hexapod's parts induces expansion, which alters their kinematic model and reduces the robot's accuracy and repeatability. These variations may arise from internal heat sources (such as motors, encoders, and electronics) or from environmental changes. In this study, a method is proposed to anticipate and therefore correct the thermal drift of one of the hexapod precision electro-mechanical actuators. This method is based on determining a model that links the expansion state of the actuator at any given moment to the temperature of some well-chosen points on its surface. This model was initially developed theoretically. Its coefficients were then adjusted experimentally on a specific test-bench, based on a rigorous measurement campaign of actuator expansion using a high-precision interferometric measurement system. Experimental validation demonstrates a reduction of thermally induced expansion by more than 80%. This paves the way for thermal drift correction across the entire robot or similar robotics parts.
☆ DexKnot: Generalizable Visuomotor Policy Learning for Dexterous Bag-Knotting Manipulation
Knotting plastic bags is a common task in daily life, yet it is challenging for robots due to the bags' infinite degrees of freedom and complex physical dynamics. Existing methods often struggle in generalization to unseen bag instances or deformations. To address this, we present DexKnot, a framework that combines keypoint affordance with diffusion policy to learn a generalizable bag-knotting policy. Our approach learns a shape-agnostic representation of bags from keypoint correspondence data collected through real-world manual deformation. For an unseen bag configuration, the keypoints can be identified by matching the representation to a reference. These keypoints are then provided to a diffusion transformer, which generates robot action based on a small number of human demonstrations. DexKnot enables effective policy generalization by reducing the dimensionality of observation space into a sparse set of keypoints. Experiments show that DexKnot achieves reliable and consistent knotting performance across a variety of previously unseen instances and deformations.
☆ Efficient Trajectory Optimization for Autonomous Racing via Formula-1 Data-Driven Initialization
Trajectory optimization is a central component of fast and efficient autonomous racing. However practical optimization pipelines remain highly sensitive to initialization and may converge slowly or to suboptimal local solutions when seeded with heuristic trajectories such as the centerline or minimum-curvature paths. To address this limitation, we leverage expert driving behavior as a initialization prior and propose a learning-informed initialization strategy based on real-world Formula 1 telemetry. To this end, we first construct a multi-track Formula~1 trajectory dataset by reconstructing and aligning noisy GPS telemetry to a standardized reference-line representation across 17 tracks. Building on this, we present a neural network that predicts an expert-like raceline offset directly from local track geometry, without explicitly modeling vehicle dynamics or forces. The predicted raceline is then used as an informed seed for a minimum-time optimal control solver. Experiments on all 17 tracks demonstrate that the learned initialization accelerates solver convergence and significantly reduces runtime compared to traditional geometric baselines, while preserving the final optimized lap time.
☆ Learning From Failures: Efficient Reinforcement Learning Control with Episodic Memory
Reinforcement learning has achieved remarkable success in robot learning. However, under challenging exploration and contact-rich dynamics, early-stage training is frequently dominated by premature terminations such as collisions and falls. As a result, learning is overwhelmed by short-horizon, low-return trajectories, which hinder convergence and limit long-horizon exploration. To alleviate this issue, we propose a technique called Failure Episodic Memory Alert (FEMA). FEMA explicitly stores short-horizon failure experiences through an episodic memory module. During interactions, it retrieves similar failure experiences and prevents the robot from recurrently relapsing into unstable states, guiding the policy toward long-horizon trajectories with greater long-term value. FEMA can be combined easily with model-free reinforcement learning algorithms, and yields a substantial sample-efficiency improvement of 33.11% on MuJoCo tasks across several classical RL algorithms. Furthermore, integrating FEMA into a parallelized PPO training pipeline demonstrates its effectiveness on a real-world bipedal robot task.
☆ Towards Scalable Probabilistic Human Motion Prediction with Gaussian Processes for Safe Human-Robot Collaboration IROS 2026
Accurate human motion prediction with well-calibrated uncertainty is critical for safe human-robot collaboration (HRC), where robots must anticipate and react to human movements in real time. We propose a structured multitask variational Gaussian Process (GP) framework for full-body human motion prediction that captures temporal correlations and leverages joint-dimension-level factorization for scalability, while using a continuous 6D rotation representation to preserve kinematic consistency. Evaluated on Human3.6M (H3.6M), our model achieves up to 50 lower kernel density estimate negative log-likelihood (KDE NLL) than strong baselines, a mean continuous ranked probability score (CRPS) of 0.021 m, and deterministic mean angle error (MAE) that is 3-18% higher than competitive deep learning methods. Empirical coverage analysis shows that the fraction of ground-truth outcomes contained within predicted confidence intervals gradually decreases with horizon, remaining conservative for lower-confidence intervals and near-nominal for higher-confidence intervals, with only modest calibration drift at longer horizons. Despite its probabilistic formulation, our model requires only 0.24-0.35 M parameters, roughly eight times fewer than comparable approaches, and exhibits modest inference times, indicating suitability for real-time deployment. Extensive ablation studies further validated the choice of 6D rotation representation and Matern 3/2 + Linear kernel, and guided the selection of the number of inducing points and latent dimensionality. These results demonstrate that scalable GP-based models can deliver competitive accuracy together with reliable and interpretable uncertainty estimates for downstream robotics tasks such as motion planning and collision avoidance.
comment: Submitted to IROS 2026
☆ ACLM: ADMM-Based Distributed Model Predictive Control for Collaborative Loco-Manipulation
Collaborative transportation of heavy payloads via loco-manipulation is a challenging yet essential capability for legged robots operating in complex, unstructured environments. Centralized planning methods, e.g., holistic trajectory optimization, capture dynamic coupling among robots and payloads but scale poorly with system size, limiting real-time applicability. In contrast, hierarchical and fully decentralized approaches often neglect force and dynamic interactions, leading to conservative behavior. This study proposes an Alternating Direction Method of Multipliers (ADMM)-based distributed model predictive control framework for collaborative loco-manipulation with a team of quadruped robots with manipulators. By exploiting the payload-induced coupling structure, the global optimal control problem is decomposed into parallel individual-robot-level subproblems with consensus constraints. The distributed planner operates in a receding-horizon fashion and achieves fast convergence, requiring only a few ADMM iterations per planning cycle. A wrench-aware whole-body controller executes the planned trajectories, tracking both motion and interaction wrenches. Extensive simulations with up to four robots demonstrate scalability, real-time performance, and robustness to model uncertainty.
☆ VLN-Cache: Enabling Token Caching for VLN Models with Visual/Semantic Dynamics Awareness
Vision-and-Language Navigation (VLN) increasingly relies on large vision-language models, but their inference cost conflicts with real-time deployment. Token caching is a promising training-free strategy that avoids redundant computation by reusing stable visual tokens across frames. However, existing methods assume a static camera and fixed semantic focus, assumptions that VLN fundamentally violates. We identify two failure modes: (1) visual dynamics, where viewpoint shift displaces token positions across frames, causing position-wise matching to pair misaligned content; (2) semantic dynamics, where token relevance shifts across task stages as navigation progresses, making cached states stale. We propose VLN-Cache, a visual-dynamic-aware and semantic-dynamic-aware caching framework that introduces view-aligned remapping to recover geometric correspondences and a task-relevance saliency filter to veto reuse at semantic transitions. A layer-adaptive entropy policy further balances the per-layer reuse budget. Experiments on the R2R-CE simulation benchmark show up to 1.52x speedup while maintaining competitive navigation success rates.
☆ The Talking Robot: Distortion-Robust Acoustic Models for Robot-Robot Communication
We present Artoo, a learned acoustic communication system for robots that replaces hand-designed signal processing with end-to-end co-trained neural networks. Our system pairs a lightweight text-to-speech (TTS) transmitter (1.18M parameters) with a conformer-based automatic speech recognition (ASR) receiver (938K parameters), jointly optimized through a differentiable channel. Unlike human speech, robot-to-robot communication is paralinguistics-free: the system need not preserve timbre, prosody, or naturalness, only maximize decoding accuracy under channel distortion. Through a three-phase co-training curriculum, the TTS transmitter learns to produce distortion-robust acoustic encodings that surpass the baseline under noise, achieving 8.3% CER at 0 dB SNR. The entire system requires only 2.1M parameters (8.4 MB) and runs in under 13 ms end-to-end on a CPU, making it suitable for deployment on resource-constrained robotic platforms.
☆ Morphology-Independent Facial Expression Imitation for Human-Face Robots
Accurate facial expression imitation on human-face robots is crucial for achieving natural human-robot interaction. Most existing methods have achieved photorealistic expression imitation through mapping 2D facial landmarks to a robot's actuator commands. Their imitation of landmark trajectories is susceptible to interference from facial morphology, which would lead to a performance drop. In this paper, we propose a morphology-independent expression imitation method that decouples expressions from facial morphology to eliminate morphological influence and produce more realistic expressions for human-face robots. Specifically, we construct an expression decoupling module to learn expression semantics by disentangling the expression representation from the morphology representation in a self-supervised manner. We devise an expression transfer module to map the representations to the robot's actuator commands through a learning objective of perceiving expression errors, producing accurate facial expressions based on the learned expression semantics. To support experimental validation, a custom-designed and highly expressive human-face robot, namely Pengrui, is developed to serve as an experimental platform for realistic expression imitation. Extensive experiments demonstrate that our method enables the human-face robot to reproduce a wide range of human-like expressions effectively. All code and implementation details of the robot will be released.
☆ GuideTWSI: A Diverse Tactile Walking Surface Indicator Dataset from Synthetic and Real-World Images for Blind and Low-Vision Navigation
Tactile Walking Surface Indicators (TWSIs) are safety-critical landmarks that blind and low-vision (BLV) pedestrians use to locate crossings and hazard zones. From our observation sessions with BLV guide dog handlers, trainers, and an O&M specialist, we confirmed the critical importance of reliable and accurate TWSI segmentation for navigation assistance of BLV individuals. Achieving such reliability requires large-scale annotated data. However, TWSIs are severely underrepresented in existing urban perception datasets, and even existing dedicated paving datasets are limited: they lack robot-relevant viewpoints (e.g., egocentric or top-down) and are geographically biased toward East Asian directional bars - raised parallel strips used for continuous guidance along sidewalks. This narrow focus overlooks truncated domes - rows of round bumps used primarily in North America and Europe as detectable warnings at curbs, crossings, and platform edges. As a result, models trained only on bar-centric data struggle to generalize to dome-based warnings, leading to missed detections and false stops in safety-critical environments.
☆ TacDexGrasp: Compliant and Robust Dexterous Grasping with Tactile Feedback
Multi-fingered hands offer great potential for compliant and robust grasping of unknown objects, yet their high-dimensional force control presents a significant challenge. This work addresses two key problems: (1) distributing forces across multiple contacts to counteract an object's weight, and (2) preventing rotational slip caused by gravitational torque when a grasp is distant from the object's center of mass. We address these challenges via tactile feedback and a Second-Order Cone Programming (SOCP)-based controller, without explicit torque modeling or slip detection. Our key insights are (1) rotational slip inevitably induces translational slip at some contact points for a multi-fingered grasp, and (2) the ratio of tangential to normal force at each contact is an effective early stability indicator. By actively constraining this ratio for each finger below the estimated friction coefficient, our controller maintains grasp stability against both translational and rotational slip. Real-world experiments on 12 diverse objects demonstrate the robustness and compliance of our approach.
comment: 8pages, 7 figures
☆ SSP: Safety-guaranteed Surgical Policy via Joint Optimization of Behavioral and Spatial Constraints
The paradigm of robot-assisted surgery is shifting toward data-driven autonomy, where policies learned via Reinforcement Learning (RL) or Imitation Learning (IL) enable the execution of complex tasks. However, these ``black-box" policies often lack formal safety guarantees, a critical requirement for clinical deployment. In this paper, we propose the Safety-guaranteed Surgical Policy (SSP) framework to bridge the gap between data-driven generality and formal safety. We utilize Neural Ordinary Differential Equations (Neural ODEs) to learn an uncertainty-aware dynamics model from demonstration data. This learned model underpins a robust Control Barrier Function (CBF) safety controller, which minimally alters the actions of a surgical policy to ensure strict safety under uncertainty. Our controller enforces two constraint categories: behavioral constraints (restricting the task space of the agent) and spatial constraints (defining surgical no-go zones). We instantiate the SSP framework with surgical policies derived from RL, IL and Control Lyapunov Functions (CLF). Validation on in both the SurRoL simulation and da Vinci Research Kit (dVRK) demonstrates that our method achieves a near-zero constraint violation rate while maintaining high task success rates compared to unconstrained baselines.
☆ Two-Stage Path Following for Mobile Manipulators via Dimensionality-Reduced Graph Search and Numerical Optimization
Efficient path following for mobile manipulators is often hindered by high-dimensional configuration spaces and kinematic constraints. This paper presents a robust two-stage configuration planning framework that decouples the 8-DoF planning problem into a tractable 2-DoF base optimization under a yaw-fixed base planning assumption. In the first stage, the proposed approach utilizes IRM to discretize the task-space path into a multi-layer graph, where an initial feasible path is extracted via a Dijkstra-based dynamic programming approach to ensure computational efficiency and global optimality within the discretized graph. In the second stage, to overcome discrete search quantization, feasible base regions are transformed into convex hulls, enabling subsequent continuous refinement via the L-BFGS algorithm to maximize trajectory smoothness while strictly enforcing reachability constraints. Simulation results demonstrate the theoretical precision of the proposed method by achieving sub-millimeter kinematic accuracy in simulation, and physical experiments on an omnidirectional mobile manipulator further validate the framework's robustness and practical applicability.
☆ Foundational World Models Accurately Detect Bimanual Manipulator Failures
Deploying visuomotor robots at scale is challenging due to the potential for anomalous failures to degrade performance, cause damage, or endanger human life. Bimanual manipulators are no exception; these robots have vast state spaces comprised of high-dimensional images and proprioceptive signals. Explicitly defining failure modes within such state spaces is infeasible. In this work, we overcome these challenges by training a probabilistic, history informed, world model within the compressed latent space of a pretrained vision foundation model (NVIDIA's Cosmos Tokenizer). The model outputs uncertainty estimates alongside its predictions that serve as non-conformity scores within a conformal prediction framework. We use these scores to develop a runtime monitor, correlating periods of high uncertainty with anomalous failures. To test these methods, we use the simulated Push-T environment and the Bimanual Cable Manipulation dataset, the latter of which we introduce in this work. This new dataset features trajectories with multiple synchronized camera views, proprioceptive signals, and annotated failures from a challenging data center maintenance task. We benchmark our methods against baselines from the anomaly detection and out-of-distribution detection literature, and show that our approach considerably outperforms statistical techniques. Furthermore, we show that our approach requires approximately one twentieth of the trainable parameters as the next-best learning-based approach, yet outperforms it by 3.8% in terms of failure detection rate, paving the way toward safely deploying manipulator robots in real-world environments where reliability is non-negotiable.
comment: 8 pages, 5 figures, accepted at the 2026 IEEE International Conference on Robotics and Automation
☆ VSL-Skin: Individually Addressable Phase-Change Voxel Skin for Variable-Stiffness and Virtual Joints Bridging Soft and Rigid Robots ICRA 2026
Soft robots are compliant but often cannot support loads or hold their shape, while rigid robots provide structural strength but are less adaptable. Existing variable-stiffness systems usually operate at the scale of whole segments or patches, which limits precise control over stiffness distribution and virtual joint placement. This paper presents the Variable Stiffness Lattice Skin (VSL-Skin), the first system to enable individually addressable voxel-level morphological control with centimeter-scale precision. The system provides three main capabilities: nearly two orders of magnitude stiffness modulation across axial (15-1200 N/mm), shear (45-850 N/mm), bending (8*10^2 - 3*10^4 N/deg), and torsional modes with centimeter-scale spatial control; the first demonstrated 30% axial compression in phase-change systems while maintaining structural integrity; and autonomous component-level self-repair through thermal cycling, which eliminates fatigue accumulation and enables programmable sacrificial joints for predictable failure management. Selective voxel activation creates six canonical virtual joint types with programmable compliance while preserving structural integrity in non-activated regions. The platform incorporates closed-form design models and finite element analysis for predictive synthesis of stiffness patterns and joint placement. Experimental validation demonstrates 30% axial contraction, thermal switching in 75-second cycles, and cut-to-fit integration that preserves addressability after trimming. The row-column architecture enables platform-agnostic deployment across diverse robotic systems without specialized infrastructure. This framework establishes morphological intelligence as an engineerable system property and advances autonomous reconfigurable robotics.
comment: ICRA 2026
☆ Energy-Efficient Collaborative Transport of Tether-Suspended Payloads via Rotating Equilibrium
Collaborative aerial transportation of tethered payloads is fundamentally limited by space, power, and weight constraints. Conventional approaches rely on static equilibrium conditions, where each vehicle tilts to generate the forces that ensure they maintain a formation geometry that avoids aerodynamic interactions and collision. This horizontal thrust component represents a significant energy penalty compared to the ideal case in which each vehicle produces purely vertical thrust to lift the payload. Operating in tighter tether configurations can minimize this effect, but at the cost of either having to fly the vehicles in closer proximity, which risks collision, or significantly increasing the length of the tether, which increases complexity and reduces potential use-cases. We propose operating the tether-suspended flying system at a rotating equilibrium. By maintaining steady circular motion, centrifugal forces provide the necessary horizontal tether tension, allowing each quadrotor to generate purely vertical thrust and thus reducing the total force (and power) required compared to an equilibrium where the thrusts are not vertical. It also allows for a wider range of tether configurations to be used without sacrificing efficiency. Results demonstrate that rotating equilibria can reduce power consumption relative to static lifting by up to 20%, making collaborative aerial solutions more practically relevant.
comment: 7 pages, 8 figures
☆ Is Your Safe Controller Actually Safe? A Critical Review of CBF Tautologies and Hidden Assumptions
This tutorial provides a critical review of the practical application of Control Barrier Functions (CBFs) in robotic safety. While the theoretical foundations of CBFs are well-established, I identify a recurring gap between the mathematical assumption of a safe controller's existence and its constructive realization in systems with input constraints. I highlight the distinction between candidate and valid CBFs by analyzing the interplay of system dynamics, actuation limits, and class-K functions. I further show that some purported demonstrations of safe robot policies or controllers are limited to passively safe systems, such as single integrators or kinematic manipulators, where safety is already inherited from the underlying physics and even naive geometric hard constraints suffice to prevent collisions. By revisiting simple low-dimensional examples, I show when CBF formulations provide valid safety guarantees and when they fail due to common misuses. I then provide practical guidelines for constructing realizable safety arguments for systems without such passive safety. The goal of this tutorial is to bridge the gap between theoretical guarantees and actual implementation, supported by an open-source interactive web demonstration that visualizes these concepts intuitively.
comment: Interactive web demo: https://cbf.taekyung.me
♻ ☆ SAC-Loco: Safe and Adjustable Compliant Quadrupedal Locomotion
Quadruped robots are designed to achieve agile and robust locomotion by drawing inspiration from legged animals. However, most existing control methods for quadruped robots lack a key capacity observed in animals: the ability to exhibit diverse compliance behaviors while ensuring stability when experiencing external forces. In particular, achieving adjustable compliance while maintaining robust safety under force disturbances remains a significant challenge. In this work, we propose a safety aware compliant locomotion framework that integrates adjustable disturbance compliance with robust failure prevention. We first train a force compliant policy with adjustable compliance levels using a teacher student reinforcement learning framework, allowing deployment without explicit force sensing. To handle disturbances beyond the limits of compliant control, we develop a safety oriented policy for rapid recovery and stabilization. Finally, we introduce a learned safety critic that monitors the robot's safety in real time and coordinates between compliant locomotion and recovery behaviors. Together, this framework enables quadruped robots to achieve smooth force compliance and robust safety under a wide range of external force disturbances.
♻ ☆ Vision-Guided Targeted Grasping and Vibration for Robotic Pollination in Controlled Environments
Robotic pollination offers a promising alternative to manual labor and bumblebee-assisted methods in controlled agriculture, where wind-driven pollination is absent and regulatory restrictions limit the use of commercial pollinators. In this work, we present and validate a vision-guided robotic framework that uses data from an end-effector mounted RGB-D sensor and combines 3D plant reconstruction, targeted grasp planning, and physics-based vibration modeling to enable precise pollination. First, the plant is reconstructed in 3D and registered to the robot coordinate frame to identify obstacle-free grasp poses along the main stem. Second, a discrete elastic rod model predicts the relationship between actuation parameters and flower dynamics, guiding the selection of optimal pollination strategies. Finally, a manipulator with soft grippers grasps the stem and applies controlled vibrations to induce pollen release. End-to-end experiments demonstrate a 92.5\% main-stem grasping success rate, and simulation-guided optimization of vibration parameters further validates the feasibility of our approach, ensuring that the robot can safely and effectively perform pollination without damaging the flower. To our knowledge, this is the first robotic system to jointly integrate vision-based grasping and vibration modeling for automated precision pollination.
comment: YouTube: https://youtu.be/XHLA7pEXhZU; GitHub: https://github.com/StructuresComp/robotic-pollination
♻ ☆ xTED: Cross-Domain Adaptation via Diffusion-Based Trajectory Editing
Reusing pre-collected data from different domains is an appealing solution for decision-making tasks, especially when data in the target domain are limited. Existing cross-domain policy transfer methods mostly aim at learning domain correspondences or corrections to facilitate policy learning, such as learning task/domain-specific discriminators, representations, or policies. This design philosophy often results in heavy model architectures or task/domain-specific modeling, lacking flexibility. This reality makes us wonder: can we directly bridge the domain gaps universally at the data level, instead of relying on complex downstream cross-domain policy transfer procedures? In this study, we propose the Cross-Domain Trajectory EDiting (xTED) framework that employs a specially designed diffusion model for cross-domain trajectory adaptation. Our proposed model architecture effectively captures the intricate dependencies among states, actions, and rewards, as well as the dynamics patterns within target data. Edited by adding noises and denoising with the pre-trained diffusion model, source domain trajectories can be transformed to align with target domain properties while preserving original semantic information. This process effectively corrects underlying domain gaps, enhancing state realism and dynamics reliability in source data, and allowing flexible integration with various single-domain and cross-domain downstream policy learning methods. Despite its simplicity, xTED demonstrates superior performance in extensive simulation and real-robot experiments.
comment: xTED offers a novel, generic, flexible, simple and effective paradigm that casts cross-domain policy adaptation as a data pre-processing problem
♻ ☆ Safe Navigation of Bipedal Robots via Koopman Operator-Based Model Predictive Control
Nonlinearity in dynamics has long been a major challenge in robotics, often causing significant performance degradation in existing control algorithms. For example, the navigation of bipedal robots can exhibit nonlinear behaviors even under simple velocity commands, as their actual dynamics are governed by complex whole-body movements and discrete contacts. In this work, we propose a safe navigation framework inspired by Koopman operator theory. We first train a low-level locomotion policy using deep reinforcement learning, and then capture its low-frequency, base-level dynamics by learning linearized dynamics in a high-dimensional lifted space. Then, our model-predictive controller (MPC) efficiently optimizes control signals via a standard quadratic objective and the linear dynamics constraint in the lifted space. We demonstrate that the Koopman model more accurately predicts bipedal robot trajectories than baseline approaches. We also show that the proposed navigation framework achieves improved safety with better success rates in dense environments with narrow passages.
comment: 9 pages
♻ ☆ ActivePusher: Active Learning and Planning with Residual Physics for Nonprehensile Manipulation ICRA 2026
Planning with learned dynamics models offers a promising approach toward versatile real-world manipulation, particularly in nonprehensile settings such as pushing or rolling, where accurate analytical models are difficult to obtain. However, collecting training data for learning-based methods can be costly and inefficient, as it often relies on randomly sampled interactions that are not necessarily the most informative. Furthermore, learned models tend to exhibit high uncertainty in underexplored regions of the skill space, undermining the reliability of long-horizon planning. To address these challenges, we propose ActivePusher, a novel framework that combines residual-physics modeling with uncertainty-based active learning, to focus data acquisition on the most informative skill parameters. Additionally, ActivePusher seamlessly integrates with model-based kinodynamic planners, leveraging uncertainty estimates to bias control sampling toward more reliable actions. We evaluate our approach in both simulation and real-world environments, and demonstrate that it consistently improves data efficiency and achieves higher planning success rates in comparison to baseline methods. The source code is available at https://github.com/elpis-lab/ActivePusher.
comment: Accepted by the 2026 IEEE International Conference on Robotics & Automation (ICRA 2026)
♻ ☆ Accelerating Robotic Reinforcement Learning with Agent Guidance
Reinforcement Learning (RL) offers a powerful paradigm for autonomous robots to master generalist manipulation skills through trial-and-error. However, its real-world application is stifled by low sample efficiency. Recent Human-in-the-Loop (HIL) methods accelerate training by using human corrections, yet this approach faces a scalability barrier. Reliance on human supervisors imposes a 1:1 supervision ratio that limits scalability, suffers from operator fatigue over extended sessions, and introduces high variance due to inconsistent human proficiency. We present Agent-guided Policy Search (AGPS), a framework that automates the training pipeline by replacing human supervisors with a multimodal agent. Our key insight is that the agent can be viewed as a semantic world model, injecting intrinsic value priors to structure physical exploration. By using tools, the agent provides precise guidance via corrective waypoints and spatial constraints for exploration pruning. We validate our approach on three tasks, ranging from precision insertion to deformable object manipulation. Results demonstrate that AGPS outperforms HIL methods in sample efficiency. This automates the supervision pipeline, unlocking the path to labor-free and scalable robot learning. Project website: https://agps-rl.github.io/agps/.
♻ ☆ DrivingGen: A Comprehensive Benchmark for Generative Video World Models in Autonomous Driving ICLR 2026
Video generation models, as one form of world models, have emerged as one of the most exciting frontiers in AI, promising agents the ability to imagine the future by modeling the temporal evolution of complex scenes. In autonomous driving, this vision gives rise to driving world models: generative simulators that imagine ego and agent futures, enabling scalable simulation, safe testing of corner cases, and rich synthetic data generation. Yet, despite fast-growing research activity, the field lacks a rigorous benchmark to measure progress and guide priorities. Existing evaluations remain limited: generic video metrics overlook safety-critical imaging factors; trajectory plausibility is rarely quantified; temporal and agent-level consistency is neglected; and controllability with respect to ego conditioning is ignored. Moreover, current datasets fail to cover the diversity of conditions required for real-world deployment. To address these gaps, we present DrivingGen, the first comprehensive benchmark for generative driving world models. DrivingGen combines a diverse evaluation dataset curated from both driving datasets and internet-scale video sources, spanning varied weather, time of day, geographic regions, and complex maneuvers, with a suite of new metrics that jointly assess visual realism, trajectory plausibility, temporal coherence, and controllability. Benchmarking 14 state-of-the-art models reveals clear trade-offs: general models look better but break physics, while driving-specific ones capture motion realistically but lag in visual quality. DrivingGen offers a unified evaluation framework to foster reliable, controllable, and deployable driving world models, enabling scalable simulation, planning, and data-driven decision-making.
comment: ICLR 2026 Poster; Project Website: https://drivinggen-bench.github.io/
♻ ☆ Hybrid Diffusion Policies with Projective Geometric Algebra for Efficient Robot Manipulation Learning ICRA 2026
Diffusion policies are a powerful paradigm for robot learning, but their training is often inefficient. A key reason is that networks must relearn fundamental spatial concepts, such as translations and rotations, from scratch for every new task. To alleviate this redundancy, we propose embedding geometric inductive biases directly into the network architecture using Projective Geometric Algebra (PGA). PGA provides a unified algebraic framework for representing geometric primitives and transformations, allowing neural networks to reason about spatial structure more effectively. In this paper, we introduce hPGA-DP, a novel hybrid diffusion policy that capitalizes on these benefits. Our architecture leverages the Projective Geometric Algebra Transformer (P-GATr) as a state encoder and action decoder, while employing established U-Net or Transformer-based modules for the core denoising process. Through extensive experiments and ablation studies in both simulated and real-world environments, we demonstrate that hPGA-DP significantly improves task performance and training efficiency. Notably, our hybrid approach achieves substantially faster convergence compared to both standard diffusion policies and architectures that rely solely on P-GATr. The project website is available at: https://apollo-lab-yale.github.io/26-ICRA-hPGA-website/.
comment: Accepted to ICRA 2026
♻ ☆ Automated Pest Counting in Water Traps through Active Robotic Stirring for Occlusion Handling
Existing image-based pest counting methods rely on single static images and often produce inaccurate results under occlusion. To address this issue, this paper proposes an automated pest counting method in water traps through active robotic stirring. First, an automated robotic arm-based stirring system is developed to redistribute pests and reveal occluded individuals for counting. Then, the effects of different stirring patterns on pest counting performance are investigated. Six stirring patterns are designed and evaluated across different pest density scenarios to identify the optimal one. Finally, a heuristic counting confidence-driven closed-loop control system is proposed for adaptive-speed robotic stirring, adjusting the stirring speed based on the average change rate of counting confidence between consecutive frames. Experimental results show that the four circles is the optimal stirring pattern, achieving the lowest overall mean absolute counting error of 4.384 and the highest overall mean counting confidence of 0.721. Compared with constant-speed stirring, adaptive-speed stirring reduces task execution time by up to 44.7% and achieves more stable performance across different pest density scenarios. Moreover, the proposed pest counting method reduces the mean absolute counting error by up to 3.428 compared to the single static image counting method under high-density scenarios where occlusion is severe.
♻ ☆ ActivePose: Active 6D Object Pose Estimation and Tracking for Robotic Manipulation
Accurate 6-DoF object pose estimation and tracking are critical for reliable robotic manipulation. However, zero-shot methods often fail under viewpoint-induced ambiguities and fixed-camera setups struggle when objects move or become self-occluded. To address these challenges, we propose an active pose estimation pipeline that combines a Vision-Language Model (VLM) with "robotic imagination" to dynamically detect and resolve ambiguities in real time. In an offline stage, we render a dense set of views of the CAD model, compute the FoundationPose entropy for each view, and construct a geometric-aware prompt that includes low-entropy (unambiguous) and high-entropy (ambiguous) examples. At runtime, the system: (1) queries the VLM on the live image for an ambiguity score; (2) if ambiguity is detected, imagines a discrete set of candidate camera poses by rendering virtual views, scores each based on a weighted combination of VLM ambiguity probability and FoundationPose entropy, and then moves the camera to the Next-Best-View (NBV) to obtain a disambiguated pose estimation. Furthermore, since moving objects may leave the camera's field of view, we introduce an active pose tracking module: a diffusion-policy trained via imitation learning, which generates camera trajectories that preserve object visibility and minimize pose ambiguity. Experiments in simulation and real-world show that our approach significantly outperforms classical baselines.
comment: 6D Pose, Diffusion Policy, Robot Learning
♻ ☆ Radio-based Multi-Robot Odometry and Relative Localization
Radio-based methods such as Ultra-Wideband (UWB) and RAdio Detection And Ranging (radar), which have traditionally seen limited adoption in robotics, are experiencing a boost in popularity thanks to their robustness to harsh environmental conditions and cluttered environments. This work proposes a multi-robot UGV-UAV localization system that leverages the two technologies with inexpensive and readily-available sensors, such as Inertial Measurement Units (IMUs) and wheel encoders, to estimate the relative position of an aerial robot with respect to a ground robot. The first stage of the system pipeline includes a nonlinear optimization framework to trilaterate the location of the aerial platform based on UWB range data, and a radar pre-processing module with loosely coupled ego-motion estimation which has been adapted for a multi-robot scenario. Then, the pre-processed radar data as well as the relative transformation are fed to a pose-graph optimization framework with odometry and inter-robot constraints. The system, implemented for the Robotic Operating System (ROS 2) with the Ceres optimizer, has been validated in Software-in-the-Loop (SITL) simulations and in a real-world dataset. The proposed relative localization module outperforms state-of-the-art closed-form methods which are less robust to noise. Our SITL environment includes a custom Gazebo plugin for generating realistic UWB measurements modeled after real data. Conveniently, the proposed factor graph formulation makes the system readily extensible to full Simultaneous Localization And Mapping (SLAM). Finally, all the code and experimental data is publicly available to support reproducibility and to serve as a common open dataset for benchmarking.
♻ ☆ ELHPlan: Efficient Long-Horizon Task Planning for Multi-Agent Collaboration
Large Language Models (LLMs) enable intelligent multi-robot collaboration but face fundamental trade-offs: open-loop methods that compile tasks into formal representations for external executors produce sound plans but lack adaptability in partially observable environments, while iterative methods incur prohibitive computational costs that scale poorly with team size and task complexity. In this paper, we propose Efficient Long-Horizon Planning (ELHPlan), a novel framework that introduces Action Chains, sequences of actions explicitly bound to sub-goal intentions, as the fundamental planning primitive. ELHPlan operates via a cyclical process: 1) constructing intention-bound action sequences, 2) proactively validating for conflicts and feasibility, 3) refining issues through targeted mechanisms, and 4) executing validated actions. This design balances adaptability and efficiency by providing intention-bound action sequences with longer lookahead while avoiding expensive full re-planning. We further advocate comprehensive efficiency metrics, including token consumption and planning time, to more holistically evaluate multi-agent collaboration. Our experiments on benchmarks TDW-MAT and C-WAH demonstrate that ELHPlan achieves comparable task success rates while consuming only 30-40% of the tokens required by state-of-the-art methods. Our research establishes a new efficiency-effectiveness frontier for LLM-based multi-agent planning systems.
♻ ☆ EB-MBD: Emerging-Barrier Model-Based Diffusion for Safe Trajectory Optimization in Highly Constrained Environments ICRA 2026
We propose enforcing constraints on Model-Based Diffusion by introducing emerging barrier functions inspired by interior point methods. We demonstrate that the standard Model-Based Diffusion algorithm can lead to catastrophic performance degradation in highly constrained environments, even on simple 2D systems due to sample inefficiency in the Monte Carlo approximation of the score function. We introduce Emerging-Barrier Model-Based Diffusion (EB-MBD) which uses progressively introduced barrier constraints to avoid these problems, significantly improving solution quality, without expensive projection operations such as projections. We analyze the sampling liveliness of samples at each iteration to inform barrier parameter scheduling choice. We demonstrate results for 2D collision avoidance and a 3D underwater manipulator system and show that our method achieves lower cost solutions than Model-Based Diffusion, and requires orders of magnitude less computation time than projection based methods.
comment: Accepted to ICRA 2026. Code available at https://github.com/acfr/emerging_barrier_mbd
♻ ☆ Green-VLA: Staged Vision-Language-Action Model for Generalist Robots
We introduce Green-VLA, a staged Vision-Language-Action (VLA) framework for real-world deployment on the Green humanoid robot while maintaining generalization across diverse embodiments. Green-VLA follows a five stage curriculum: (L0) foundational VLMs, (L1) multimodal grounding, (R0) multi-embodiment pretraining, (R1) embodiment-specific adaptation, and (R2) reinforcement-learning (RL) policy alignment. We couple a scalable data-processing pipeline (3,000 hours of demonstrations) with temporal alignment and quality filtering, and use a unified, embodiment-aware action interface enabling a single policy to control humanoids, mobile manipulators, and fixed-base arms. At inference, the VLA controller is enhanced with episode-progress prediction, out-of-distribution detection, and joint-prediction-based guidance to improve safety and precise target selection. Experiments on Simpler BRIDGE WidowX and CALVIN ABC-D, as well as real-robot evaluations, show strong generalization and performance gains from RL alignment in success rate, robustness, and long-horizon efficiency.
comment: 22 pages, 14 figures
♻ ☆ Agile in the Face of Delay: Asynchronous End-to-End Learning for Real-World Aerial Navigation
Robust autonomous navigation for Autonomous Aerial Vehicles (AAVs) in complex environments is a critical capability. However, modern end-to-end navigation faces a key challenge: the high-frequency control loop needed for agile flight conflicts with low-frequency perception streams, which are limited by sensor update rates and significant computational cost. This mismatch forces conventional synchronous models into undesirably low control rates. To resolve this, we propose an asynchronous reinforcement learning framework that decouples perception and control, enabling a high-frequency policy to act on the latest IMU state for immediate reactivity, while incorporating perception features asynchronously. To manage the resulting data staleness, we introduce a theoretically-grounded Temporal Encoding Module (TEM) that explicitly conditions the policy on perception delays, a strategy complemented by a two-stage curriculum to ensure stable and efficient training. Validated in extensive simulations, our method was successfully deployed in zero-shot sim-to-real transfer on an onboard NUC, where it sustains a 100~Hz control rate and demonstrates robust, agile navigation in cluttered real-world environments. Our source code will be released for community reference.
♻ ☆ SToRM: Supervised Token Reduction for Multi-modal LLMs toward efficient end-to-end autonomous driving
In autonomous driving, end-to-end (E2E) driving systems that predict control commands directly from sensor data have achieved significant advancements. For safe driving in unexpected scenarios, these systems may additionally rely on human interventions such as natural language instructions. Using a multi-modal large language model (MLLM) facilitates human-vehicle interaction and can improve performance in such scenarios. However, this approach requires substantial computational resources due to its reliance on an LLM and numerous visual tokens from sensor inputs, which are limited in autonomous vehicles. Many MLLM studies have explored reducing visual tokens, but often suffer end-task performance degradation compared to using all tokens. To enable efficient E2E driving while maintaining performance comparable to using all tokens, this paper proposes the first Supervised Token Reduction framework for multi-modal LLMs (SToRM). The proposed framework consists of three key elements. First, a lightweight importance predictor with short-term sliding windows estimates token importance scores. Second, a supervised training approach uses an auxiliary path to obtain pseudo-supervision signals from an all-token LLM pass. Third, an anchor-context merging module partitions tokens into anchors and context tokens, and merges context tokens into relevant anchors to reduce redundancy while minimizing information loss. Experiments on the LangAuto benchmark show that SToRM outperforms state-of-the-art E2E driving MLLMs under the same reduced-token budget, maintaining all-token performance while reducing computational cost by up to 30x.
♻ ☆ Bio-inspired tail oscillation enables robot fast crawling on deformable granular terrains
Deformable substrates such as sand and mud present significant challenges for terrestrial robots due to complex robot-terrain interactions. Inspired by mudskippers, amphibious animals that naturally adjust their tail morphology and movement jointly to navigate such environments, we investigate how tail design and control can jointly enhance flipper-driven locomotion on granular media. Using a bio-inspired robot modeled after the mudskipper, we experimentally compared locomotion performance between idle and actively oscillating tail configurations. Tail oscillation increased robot speed by 67% and reduced body drag by 46%. Shear force measurements revealed that this improvement was enabled by tail oscillation fluidizing the substrate, thereby reducing resistance. Additionally, tail morphology strongly influenced the oscillation strategy: designs with larger horizontal surface areas leveraged the oscillation-reduced shear resistance more effectively by limiting insertion depth. Based on these findings, we present a design principle to inform tail action selection based on substrate strength and tail morphology. Our results offer new insights into tail design and control for improving robot locomotion on deformable substrates, with implications for agricultural robotics, search and rescue, and environmental exploration.
♻ ☆ CDE: Concept-Driven Exploration for Reinforcement Learning
Intelligent exploration remains a critical challenge in reinforcement learning (RL), especially in visual control tasks. Unlike low-dimensional state-based RL, visual RL must extract task-relevant structure from raw pixels, making exploration inefficient. We propose Concept-Driven Exploration (CDE), which leverages a pre-trained vision-language model (VLM) to generate object-centric visual concepts from textual task descriptions as weak, potentially noisy supervisory signals. Rather than directly conditioning on these noisy signals, CDE trains a policy to reconstruct the concepts via an auxiliary objective, learning general representations of the concepts and using reconstruction accuracy as an intrinsic reward to guide exploration toward task-relevant objects. Across five challenging simulated visual manipulation tasks, CDE achieves efficient, targeted exploration and remains robust to both synthetic errors and noisy VLM predictions. Finally, we demonstrate real-world transfer by deploying CDE on a Franka arm, attaining an 80\% success rate in a real-world manipulation task.
comment: Preprint
♻ ☆ VL-Nav: A Neuro-Symbolic Approach for Reasoning-based Vision-Language Navigation
Navigating unseen, large-scale environments based on complex and abstract human instructions remains a formidable challenge for autonomous mobile robots. Addressing this requires robots to infer implicit semantics and efficiently explore large-scale task spaces. However, existing methods, ranging from end-to-end learning to foundation model-based modular architectures, often lack the capability to decompose complex tasks or employ efficient exploration strategies, leading to robot aimless wandering or target recognition failures. To address these limitations, we propose VL-Nav, a neuro-symbolic (NeSy) vision-language navigation system. The proposed system intertwines neural reasoning with symbolic guidance through two core components: (1) a NeSy task planner that leverages a symbolic 3D scene graph and image memory system to enhance the vision language models' (VLMs) neural reasoning capabilities for task decomposition and replanning; and (2) a NeSy exploration system that couples neural semantic cues with the symbolic heuristic function to efficiently gather the task-related information while minimizing unnecessary repeat travel during exploration. Validated on the DARPA TIAMAT Challenge navigation tasks, our system achieved an 83.4% success rate (SR) in indoor environments and 75% in outdoor scenarios. VL-Nav achieved an 86.3% SR in real-world experiments, including a challenging 483-meter run. Finally, we validate the system with complex instructions in a 3D multi-floor scenario.
♻ ☆ Diffusion Stabilizer Policy for Automated Surgical Robot Manipulations ICRA 2026
Intelligent surgical robots have the potential to revolutionize clinical practice by enabling more precise and automated surgical procedures. However, the automation of such robot for surgical tasks remains under-explored compared to recent advancements in solving household manipulation tasks. These successes have been largely driven by (1) advanced models, such as transformers and diffusion models, and (2) large-scale data utilization. Aiming to extend these successes to the domain of surgical robotics, we propose a diffusion-based policy learning framework, called Diffusion Stabilizer Policy (DSP), which enables training with imperfect or even failed trajectories. Our approach consists of two stages: first, we train the diffusion stabilizer policy using only clean data. Then, the policy is continuously updated using a mixture of clean and perturbed data, with filtering based on the prediction error on actions. Comprehensive experiments conducted in various surgical environments demonstrate the superior performance of our method in perturbation-free settings and its robustness when handling perturbed demonstrations.
comment: ICRA 2026
♻ ☆ Vectorized Online POMDP Planning ICRA 2026
Planning under partial observability is an essential capability of autonomous robots. The Partially Observable Markov Decision Process (POMDP) provides a powerful framework for planning under partial observability problems, capturing the stochastic effects of actions and the limited information available through noisy observations. POMDP solving could benefit tremendously from massive parallelization on today's hardware, but parallelizing POMDP solvers has been challenging. Most solvers rely on interleaving numerical optimization over actions with the estimation of their values, which creates dependencies and synchronization bottlenecks between parallel processes that can offset the benefits of parallelization. In this paper, we propose Vectorized Online POMDP Planner (VOPP), a novel parallel online solver that leverages a recent POMDP formulation which analytically solves part of the optimization component, leaving numerical computations to consist of only estimation of expectations. VOPP represents all data structures related to planning as a collection of tensors, and implements all planning steps as fully vectorized computations over this representation. The result is a massively parallel online solver with no dependencies or synchronization bottlenecks between concurrent processes. Experimental results indicate that VOPP is at least $20\times$ more efficient in computing near-optimal solutions compared to an existing state-of-the-art parallel online solver. Moreover, VOPP outperforms state-of-the-art sequential online solvers, while using a planning budget that is $1000\times$ smaller.
comment: 8 pages, 3 figures. Accepted at ICRA 2026
♻ ☆ RetoVLA: Reusing Register Tokens for Spatial Reasoning in Vision-Language-Action Models
Vision-Language-Action (VLA) models have demonstrated robust performance across diverse robotic tasks. However, their high memory and computational demands often limit real-time deployment. While existing model compression techniques reduce the parameter footprint, they often drop in 3D spatial reasoning and scene layout understanding. This work introduces RetoVLA, an architecture designed to maintain spatial awareness in lightweight models by repurposing Register Tokens-learnable parameters originally introduced to mitigate attention artifacts in Vision Transformers. While these tokens are generally discarded once used, we repurpose them for their dense representation of global spatial context. RetoVLA integrates these recycled tokens directly into the action-planning module through a dedicated spatial context injection path. Our proposed design enables the recovery of global context without increasing the total parameter count. Real-world experiments using a 7-DOF manipulator show a 17.1%p improvement in average success rates over the baseline. Our results demonstrate that leveraging internal register tokens provides a highly effective mechanism for developing efficient, spatially-aware robotic agents. A video demonstration is available at: https://youtu.be/2CseBR-snZg
Robotics 126
☆ Underactuated multimodal jumping robot for extraterrestrial exploration ICRA 2026
We present a rolling and jumping underactuated monopedal robot designed to explore multimodal locomotion on low-gravity bodies. It uses only two reaction wheels to control its spatial orientation with two controllers: a balancing controller which can aim the robot's jump direction on the ground, and an aerial reorientation controller which can aim the robot's leg for landing after flight. We demonstrate rolling, targeted jumping and landing, and self-righting using only three actuators total, keeping system size to 0.33m and 1.25kg. Simple switching between locomotion modes enables the system to deal with differing landscapes and environmental conditions.
comment: 8 pages, 14 figures, Accepted for ICRA 2026
☆ SG-DOR: Learning Scene Graphs with Direction-Conditioned Occlusion Reasoning for Pepper Plants
Robotic harvesting in dense crop canopies requires effective interventions that depend not only on geometry, but also on explicit, direction-conditioned relations identifying which organs obstruct a target fruit. We present SG-DOR (Scene Graphs with Direction-Conditioned Occlusion Reasoning), a relational framework that, given instance-segmented organ point clouds, infers a scene graph encoding physical attachments and direction-conditioned occlusion. We introduce an occlusion ranking task for retrieving and ranking candidate leaves for a target fruit and approach direction, and propose a direction-aware graph neural architecture with per-fruit leaf-set attention and union-level aggregation. Experiments on a multi-plant synthetic pepper dataset show improved occlusion prediction (F1=0.73, NDCG@3=0.85) and attachment inference (edge F1=0.83) over strong ablations, yielding a structured relational signal for downstream intervention planning.
☆ CFEAR-Teach-and-Repeat: Fast and Accurate Radar-only Localization ICRA
Reliable localization in prior maps is essential for autonomous navigation, particularly under adverse weather, where optical sensors may fail. We present CFEAR-TR, a teach-and-repeat localization pipeline using a single spinning radar, which is designed for easily deployable, lightweight, and robust navigation in adverse conditions. Our method localizes by jointly aligning live scans to both stored scans from the teach mapping pass, and to a sliding window of recent live keyframes. This ensures accurate and robust pose estimation across different seasons and weather phenomena. Radar scans are represented using a sparse set of oriented surface points, computed from Doppler-compensated measurements. The map is stored in a pose graph that is traversed during localization. Experiments on the held-out test sequences from the Boreas dataset show that CFEAR-TR can localize with an accuracy as low as 0.117 m and 0.096°, corresponding to improvements of up to 63% over the previous state of the art, while running efficiently at 29 Hz. These results substantially narrow the gap to lidar-level localization, particularly in heading estimation. We make the C++ implementation of our work available to the community.
comment: This paper has been accepted for publication in the IEEE International Conference on Robotics and Automation (ICRA), 2026
☆ A Unified Low-Dimensional Design Embedding for Joint Optimization of Shape, Material, and Actuation in Soft Robots
Soft robots achieve functionality through tight coupling among geometry, material composition, and actuation. As a result, effective design optimization requires these three aspects to be considered jointly rather than in isolation. This coupling is computationally challenging: nonlinear large-deformation mechanics increase simulation cost, while contact, collision handling, and non-smooth state transitions limit the applicability of standard gradient-based approaches. We introduce a smooth, low-dimensional design embedding for soft robots that unifies shape morphing, multi-material distribution, and actuation within a single structured parameter space. Shape variation is modeled through continuous deformation maps of a reference geometry, while material properties are encoded as spatial fields. Both are constructed from shared basis functions. This representation enables expressive co-design while drastically reducing the dimensionality of the search space. In our experiments, we show that design expressiveness increases with the number of basis functions, unlike comparable neural network encodings whose representational capacity does not scale predictably with parameter count. We further show that joint co-optimization of shape, material, and actuation using our unified embedding consistently outperforms sequential strategies. All experiments are performed independently of the underlying simulator, confirming compatibility with black-box simulation pipelines. Across multiple dynamic tasks, the proposed embedding surpasses neural network and voxel-based baseline parameterizations while using significantly fewer design parameters. Together, these findings demonstrate that structuring the design space itself enables efficient co-design of soft robots.
comment: This work has been submitted to the IEEE for possible publication
☆ Control Barrier Corridors: From Safety Functions to Safe Sets
Safe autonomy is a critical requirement and a key enabler for robots to operate safely in unstructured complex environments. Control barrier functions and safe motion corridors are two widely used but technically distinct safety methods, functional and geometric, respectively, for safe motion planning and control. Control barrier functions are applied to the safety filtering of control inputs to limit the decay rate of system safety, whereas safe motion corridors are geometrically constructed to define a local safe zone around the system state for use in motion optimization and reference-governor design. This paper introduces a new notion of control barrier corridors, which unifies these two approaches by converting control barrier functions into local safe goal regions for reference goal selection in feedback control systems. We show, with examples on fully actuated systems, kinematic unicycles, and linear output regulation systems, that individual state safety can be extended locally over control barrier corridors for convex barrier functions, provided the control convergence rate matches the barrier decay rate, highlighting a trade-off between safety and reactiveness. Such safe control barrier corridors enable safely reachable persistent goal selection over continuously changing barrier corridors during system motion, which we demonstrate for verifiably safe and persistent path following in autonomous exploration of unknown environments.
comment: 12 pages, 6 figures, an extended preprint version of a conference paper
☆ History-Conditioned Spatio-Temporal Visual Token Pruning for Efficient Vision-Language Navigation
Vision-Language Navigation (VLN) enables robots to follow natural-language instructions in visually grounded environments, serving as a key capability for embodied robotic systems. Recent Vision-Language-Action (VLA) models have demonstrated strong navigation performance, but their high computational cost introduces latency that limits real-time deployment. We propose a training-free spatio-temporal vision token pruning framework tailored to VLA-based VLN. We apply spatial token selection to the current view, alongside spatio-temporal compression for historical memories, enabling efficient long-horizon inference while reducing redundant computation. Leveraging attention-based token importance and query-guided spatio-temporal filtering, the proposed approach preserves navigation-relevant information without retraining or modifying pretrained models, allowing plug-and-play integration into existing VLA systems. Through experiments on standard VLN benchmarks, we confirm that our method significantly outperforms existing pruning strategies. It successfully preserves superior navigation accuracy under extreme pruning scenarios, all while maintaining the highly competitive inference efficiency. Real-world deployment on a Unitree Go2 quadruped robot further validates reliable and low-latency instruction-following navigation under practical robotic constraints. We hope this work helps bridge the gap between large-scale multimodal modeling and efficient, real-time embodied deployment in robotic navigation systems.
☆ Data Analogies Enable Efficient Cross-Embodiment Transfer
Generalist robot policies are trained on demonstrations collected across a wide variety of robots, scenes, and viewpoints. Yet it remains unclear how to best organize and scale such heterogeneous data so that it genuinely improves performance in a given target setting. In this work, we ask: what form of demonstration data is most useful for enabling transfer across robot set-ups? We conduct controlled experiments that vary end-effector morphology, robot platform appearance, and camera perspective, and compare the effects of simply scaling the number of demonstrations against systematically broadening the diversity in different ways. Our simulated experiments show that while perceptual shifts such as viewpoint benefit most from broad diversity, morphology shifts benefit far less from unstructured diversity and instead see the largest gains from data analogies, i.e. paired demonstrations that align scenes, tasks, and/or trajectories across different embodiments. Informed by the simulation results, we improve real-world cross-embodiment transfer success by an average of $22.5\%$ over large-scale, unpaired datasets by changing only the composition of the data.
comment: 14 pages, 11 Figures, 6 Tables
☆ Safe Consensus of Cooperative Manipulation with Hierarchical Event-Triggered Control Barrier Functions
Cooperative transport and manipulation of heavy or bulky payloads by multiple manipulators requires coordinated formation tracking, while simultaneously enforcing strict safety constraints in varying environments with limited communication and real-time computation budgets. This paper presents a distributed control framework that achieves consensus coordination with safety guarantees via hierarchical event-triggered control barrier functions (CBFs). We first develop a consensus-based protocol that relies solely on local neighbor information to enforce both translational and rotational consistency in task space. Building on this coordination layer, we propose a three-level hierarchical event-triggered safety architecture with CBFs, which is integrated with a risk-aware leader selection and smooth switching strategy to reduce online computation. The proposed approach is validated through real-world hardware experiments using two Franka manipulators operating with static obstacles, as well as comprehensive simulations demonstrating scalable multi-arm cooperation with dynamic obstacles. Results demonstrate higher precision cooperation under strict safety constraints, achieving substantially reduced computational cost and communication frequency compared to baseline methods.
comment: 8 pages
☆ Open-Source Based and ETSI Compliant Cooperative, Connected, and Automated Mini-Cars
The automotive sector is following a revolutionary path from vehicles controlled by humans to vehicles that will be fully automated, fully connected, and ultimately fully cooperative. Along this road, new cooperative algorithms and protocols will be designed and field tested, which represents a great challenge in terms of costs. In this context, in particular, moving from simulations to practical experiments requires huge investments that are not always affordable and may become a barrier in some cases. To solve this issue and provide the community with an intermediate step, we here propose the use of 1:10 scaled cooperative, autonomous, and connected mini-cars. The mini-car is equipped with a Jetson Orin board running the open Robot Operating System 2 (ROS2), sensors for autonomous operations, and a Raspberry Pi board for connectivity mounting the open source Open Stack for Car (OScar). A key aspect of the proposal is the use of OScar, which implements a full ETSI cooperative-intelligent transport systems (C-ITS) compliant stack. The feasibility and potential of the proposed platform is here demonstrated through the implementation of a case study where the Day-1 intersection collision warning (ICW) application is implemented and validated.
comment: 5 pages, 6 figures
☆ SuperSuit: An Isomorphic Bimodal Interface for Scalable Mobile Manipulation
High-quality, long-horizon demonstrations are essential for embodied AI, yet acquiring such data for tightly coupled wheeled mobile manipulators remains a fundamental bottleneck. Unlike fixed-base systems, mobile manipulators require continuous coordination between $SE(2)$ locomotion and precise manipulation, exposing limitations in existing teleoperation and wearable interfaces. We present \textbf{SuperSuit}, a bimodal data acquisition framework that supports both robot-in-the-loop teleoperation and active demonstration under a shared kinematic interface. Both modalities produce structurally identical joint-space trajectories, enabling direct data mixing without modifying downstream policies. For locomotion, SuperSuit maps natural human stepping to continuous planar base velocities, eliminating discrete command switches. For manipulation, it employs a strictly isomorphic wearable arm in both modes, while policy training is formulated in a shift-invariant delta-joint representation to mitigate calibration offsets and structural compliance without inverse kinematics. Real-world experiments on long-horizon mobile manipulation tasks show 2.6$\times$ higher demonstration throughput in active mode compared to a teleoperation baseline, comparable policy performance when substituting teleoperation data with active demonstrations at fixed dataset size, and monotonic performance improvement as active data volume increases. These results indicate that consistent kinematic representations across collection modalities enable scalable data acquisition for long-horizon mobile manipulation.
☆ Can we Trust Unreliable Voxels? Exploring 3D Semantic Occupancy Prediction under Label Noise
3D semantic occupancy prediction is a cornerstone of robotic perception, yet real-world voxel annotations are inherently corrupted by structural artifacts and dynamic trailing effects. This raises a critical but underexplored question: can autonomous systems safely rely on such unreliable occupancy supervision? To systematically investigate this issue, we establish OccNL, the first benchmark dedicated to 3D occupancy under occupancy-asymmetric and dynamic trailing noise. Our analysis reveals a fundamental domain gap: state-of-the-art 2D label noise learning strategies collapse catastrophically in sparse 3D voxel spaces, exposing a critical vulnerability in existing paradigms. To address this challenge, we propose DPR-Occ, a principled label noise-robust framework that constructs reliable supervision through dual-source partial label reasoning. By synergizing temporal model memory with representation-level structural affinity, DPR-Occ dynamically expands and prunes candidate label sets to preserve true semantics while suppressing noise propagation. Extensive experiments on SemanticKITTI demonstrate that DPR-Occ prevents geometric and semantic collapse under extreme corruption. Notably, even at 90% label noise, our method achieves significant performance gains (up to 2.57% mIoU and 13.91% IoU) over existing label noise learning baselines adapted to the 3D occupancy prediction task. By bridging label noise learning and 3D perception, OccNL and DPR-Occ provide a reliable foundation for safety-critical robotic perception in dynamic environments. The benchmark and source code will be made publicly available at https://github.com/mylwx/OccNL.
comment: The benchmark and source code will be made publicly available at https://github.com/mylwx/OccNL
☆ Towards Robotic Lake Maintenance: Integrating SONAR and Satellite Data to Assist Human Operators ICMR
Artificial Water Bodies (AWBs) are human-made systems that require continuous monitoring due to their artificial biological processes. These systems demand regular maintenance to manage their ecosystems effectively. As a result of these artificial conditions, underwater vegetation can grow rapidly and must be harvested to preserve the ecological balance. This paper proposes a two-step approach to support targeted weed harvesting for the maintenance of artificial lakes. The first step is the initial detection of Submerged Aquatic Vegetation (SAV), also referred to in this paper as areas of interest, is performed using satellite-derived indices, specifically the Aquatic Plants and Algae (APA) index, which highlights submerged vegetation in water bodies. Subsequently, an Unmanned Surface Vehicle (USV) equipped with multibeam SOund NAvigation and Ranging (SONAR) performs high-resolution bathymetric mapping to locate and quantify aquatic vegetation precisely. This two-stage approach offers an effective human-robot collaboration, where satellite data guides the USV missions and boat skippers leverage detailed SONAR maps for targeted harvesting. This setup narrows the search space and reduces manual workload from human operators, making the harvesting process less labour-intensive for operators. Preliminary results demonstrate the feasibility of integrating satellite imagery and underwater acoustic sensing to improve vegetation management in artificial lakes.
comment: Accepted to and presented at the 2026 IEEE International Conference on Mechatronics and Robotics Engineering (ICMRE)
☆ NOVA: Next-step Open-Vocabulary Autoregression for 3D Multi-Object Tracking in Autonomous Driving
Generalizing across unknown targets is critical for open-world perception, yet existing 3D Multi-Object Tracking (3D MOT) pipelines remain limited by closed-set assumptions and ``semantic-blind'' heuristics. To address this, we propose Next-step Open-Vocabulary Autoregression (NOVA), an innovative paradigm that shifts 3D tracking from traditional fragmented distance-based matching toward generative spatio-temporal semantic modeling. NOVA reformulates 3D trajectories as structured spatio-temporal semantic sequences, enabling the simultaneous encoding of physical motion continuity and deep linguistic priors. By leveraging the autoregressive capabilities of Large Language Models (LLMs), we transform the tracking task into a principled process of next-step sequence completion. This mechanism allows the model to explicitly utilize the hierarchical structure of language space to resolve fine-grained semantic ambiguities and maintain identity consistency across complex long-range sequences through high-level commonsense reasoning. Extensive experiments on nuScenes, V2X-Seq-SPD, and KITTI demonstrate the superior performance of NOVA. Notably, on the nuScenes dataset, NOVA achieves an AMOTA of 22.41% for Novel categories, yielding a significant 20.21% absolute improvement over the baseline. These gains are realized through a compact 0.5B autoregressive model. Code will be available at https://github.com/xifen523/NOVA.
comment: Code will be available at https://github.com/xifen523/NOVA
☆ TaPD: Temporal-adaptive Progressive Distillation for Observation-Adaptive Trajectory Forecasting in Autonomous Driving
Trajectory prediction is essential for autonomous driving, enabling vehicles to anticipate the motion of surrounding agents to support safe planning. However, most existing predictors assume fixed-length histories and suffer substantial performance degradation when observations are variable or extremely short in real-world settings (e.g., due to occlusion or a limited sensing range). We propose TaPD (Temporal-adaptive Progressive Distillation), a unified plug-and-play framework for observation-adaptive trajectory forecasting under variable history lengths. TaPD comprises two cooperative modules: an Observation-Adaptive Forecaster (OAF) for future prediction and a Temporal Backfilling Module (TBM) for explicit reconstruction of the past. OAF is built on progressive knowledge distillation (PKD), which transfers motion pattern knowledge from long-horizon "teachers" to short-horizon "students" via hierarchical feature regression, enabling short observations to recover richer motion context. We further introduce a cosine-annealed distillation weighting scheme to balance forecasting supervision and feature alignment, improving optimization stability and cross-length consistency. For extremely short histories where implicit alignment is insufficient, TBM backfills missing historical segments conditioned on scene evolution, producing context-rich trajectories that strengthen PKD and thereby improve OAF. We employ a decoupled pretrain-reconstruct-finetune protocol to preserve real-motion priors while adapting to backfilled inputs. Extensive experiments on Argoverse 1 and Argoverse 2 show that TaPD consistently outperforms strong baselines across all observation lengths, delivers especially large gains under very short inputs, and improves other predictors (e.g., HiVT) in a plug-and-play manner. Code will be available at https://github.com/zhouhao94/TaPD.
☆ Few-Shot Neural Differentiable Simulator: Real-to-Sim Rigid-Contact Modeling
Accurate physics simulation is essential for robotic learning and control, yet analytical simulators often fail to capture complex contact dynamics, while learning-based simulators typically require large amounts of costly real-world data. To bridge this gap, we propose a few-shot real-to-sim approach that combines the physical consistency of analytical formulations with the representational capacity of graph neural network (GNN)-based models. Using only a small amount of real-world data, our method calibrates analytical simulators to generate large-scale synthetic datasets that capture diverse contact interactions. On this foundation, we introduce a mesh-based GNN that implicitly models rigid-body forward dynamics and derive surrogate gradients for collision detection, achieving full differentiability. Experimental results demonstrate that our approach enables learning-based simulators to outperform differentiable baselines in replicating real-world trajectories. In addition, the differentiable design supports gradient-based optimization, which we validate through simulation-based policy learning in multi-object interaction scenarios. Extensive experiments show that our framework not only improves simulation fidelity with minimal supervision but also increases the efficiency of policy learning. Taken together, these findings suggest that differentiable simulation with few-shot real-world grounding provides a powerful direction for advancing future robotic manipulation and control.
☆ VG3S: Visual Geometry Grounded Gaussian Splatting for Semantic Occupancy Prediction
3D semantic occupancy prediction has become a crucial perception task for comprehensive scene understanding in autonomous driving. While recent advances have explored 3D Gaussian splatting for occupancy modeling to substantially reduce computational overhead, the generation of high-quality 3D Gaussians relies heavily on accurate geometric cues, which are often insufficient in purely vision-centric paradigms. To bridge this gap, we advocate for injecting the strong geometric grounding capability from Vision Foundation Models (VFMs) into occupancy prediction. In this regard, we introduce Visual Geometry Grounded Gaussian Splatting (VG3S), a novel framework that empowers Gaussian-based occupancy prediction with cross-view 3D geometric grounding. Specifically, to fully exploit the rich 3D geometric priors from a frozen VFM, we propose a plug-and-play hierarchical geometric feature adapter, which can effectively transform generic VFM tokens via feature aggregation, task-specific alignment, and multi-scale restructuring. Extensive experiments on the nuScenes occupancy benchmark demonstrate that VG3S achieves remarkable improvements of 12.6% in IoU and 7.5% in mIoU over the baseline. Furthermore, we show that VG3S generalizes seamlessly across diverse VFMs, consistently enhancing occupancy prediction accuracy and firmly underscoring the immense value of integrating priors derived from powerful, pre-trained geometry-grounded VFMs.
☆ KISS-IMU: Self-supervised Inertial Odometry with Motion-balanced Learning and Uncertainty-aware Inference
Inertial measurement units (IMUs), which provide high-frequency linear acceleration and angular velocity measurements, serve as fundamental sensing modalities in robotic systems. Recent advances in deep neural networks have led to remarkable progress in inertial odometry. However, the heavy reliance on ground truth data during training fundamentally limits scalability and generalization to unseen and diverse environments. We propose KISS-IMU, a novel self-supervised inertial odometry framework that eliminates ground truth dependency by leveraging simple LiDAR-based ICP registration and pose graph optimization as a supervisory signal. Our approach embodies two key principles: keeping the IMU stable through motion-aware balanced training and keeping the IMU strong through uncertainty-driven adaptive weighting during inference. To evaluate performance across diverse motion patterns and scenarios, we conducted comprehensive experiments on various real-world platforms, including quadruped robots. Importantly, we train only the IMU network in a self-supervised manner, with LiDAR serving solely as a lightweight supervisory signal rather than requiring additional learnable processes. This design enables the framework to ensure robustness without relying on joint multi-modal learning or ground truth supervision. The supplementary materials are available at https://sparolab.github.io/research/kiss_imu.
comment: 8 pages, 9 figures
☆ DreamToNav: Generalizable Navigation for Robots via Generative Video Planning
We present DreamToNav, a novel autonomous robot framework that uses generative video models to enable intuitive, human-in-the-loop control. Instead of relying on rigid waypoint navigation, users provide natural language prompts (e.g. ``Follow the person carefully''), which the system translates into executable motion. Our pipeline first employs Qwen 2.5-VL-7B-Instruct to refine vague user instructions into precise visual descriptions. These descriptions condition NVIDIA Cosmos 2.5, a state-of-the-art video foundation model, to synthesize a physically consistent video sequence of the robot performing the task. From this synthetic video, we extract a valid kinematic path using visual pose estimation, robot detection and trajectory recovery. By treating video generation as a planning engine, DreamToNav allows robots to visually "dream" complex behaviors before executing them, providing a unified framework for obstacle avoidance and goal-directed navigation without task-specific engineering. We evaluate the approach on both a wheeled mobile robot and a quadruped robot in indoor navigation tasks. DreamToNav achieves a success rate of 76.7%, with final goal errors typically within 0.05-0.10 m and trajectory tracking errors below 0.15 m. These results demonstrate that trajectories extracted from generative video predictions can be reliably executed on physical robots across different locomotion platforms.
comment: Submitted to conference
☆ Dual-Agent Multiple-Model Reinforcement Learning for Event-Triggered Human-Robot Co-Adaptation in Decoupled Task Spaces
This paper presents a shared-control rehabilitation policy for a custom 6-degree-of-freedom (6-DoF) upper-limb robot that decomposes complex reaching tasks into decoupled spatial axes. The patient governs the primary reaching direction using binary commands, while the robot autonomously manages orthogonal corrective motions. Because traditional fixed-frequency control often induces trajectory oscillations due to variable inverse-kinematics execution times, an event-driven progression strategy is proposed. This architecture triggers subsequent control actions only when the end-effector enters an admission sphere centred on the immediate target waypoint, and was validated in a semi-virtual setup linking a physical pressure sensor to a MuJoCo simulation. To optimise human--robot co-adaptation safely and efficiently, this study introduces Dual Agent Multiple Model Reinforcement Learning (DAMMRL). This framework discretises decision characteristics: the human agent selects the admission sphere radius to reflect their inherent speed--accuracy trade-off, while the robot agent dynamically adjusts its 3D Cartesian step magnitudes to complement the user's cognitive state. Trained in simulation and deployed across mixed environments, this event-triggered DAMMRL approach effectively suppresses waypoint chatter, balances spatial precision with temporal efficiency, and significantly improves success rates in object acquisition tasks.
☆ A Hazard-Informed Data Pipeline for Robotics Physical Safety
This report presents a structured Robotics Physical Safety Framework based on explicit asset declaration, systematic vulnerability enumeration, and hazard-driven synthetic data generation. The approach bridges classical risk engineering with modern machine learning pipelines, enabling safety envelope learning grounded in a formalized hazard ontology. The key contribution of this framework is the alignment between classical safety engineering, digital twin simulation, synthetic data generation, and machine learning model training.
comment: 4th International Conference on Automation and Mechatronics Engineering (ICAME 2026)
☆ Sticky-Glance: Robust Intent Recognition for Human Robot Collaboration via Single-Glance
Gaze is a valuable means of communication for impaired people with extremely limited motor capabilities. However, robust gaze-based intent recognition in multi-object environments is challenging due to gaze noise, micro-saccades, viewpoint changes, and dynamic objects. To address this, we propose an object-centric gaze grounding framework that stabilizes intent through a sticky-glance algorithm, jointly modeling geometric distance and direction trends. The inferred intent remains anchored to the object even under short glances with minimal 3 gaze samples, achieving a tracking rate of 0.94 for dynamic targets and selection accuracy of 0.98 for static targets. We further introduce a continuous shared control and multi-modal interaction paradigm, enabling high-readiness control and human-in-loop feedback, thereby reducing task duration for nearly 10 \%. Experiments across dynamic tracking, multi-perspective alignment, a baseline comparison, user studies, and ablation studies demonstrate improved robustness, efficiency, and reduced workload compared to representative baselines.
☆ Multimodal Behavior Tree Generation: A Small Vision-Language Model for Robot Task Planning
Large and small language models have been widely used for robotic task planning. At the same time, vision-language models (VLMs) have successfully tackled problems such as image captioning, scene understanding, and visual question answering. In this work, we combine these two approaches by deploying a compact, open-source multimodal model to generate behavior trees for robotic task planning. The main obstacle to achieving this goal is the lack of an existing dataset that links visual observations and instructions to executable behavior trees. We propose a method to construct such a dataset starting from existing robotic episodes (i.e., Open X-Embodiment), in which a large model serves as a teacher in a multi-stage generation pipeline. We use this dataset to fine-tune VLMs ranging from 500M to 4B parameters via parameter-efficient fine-tuning (PEFT). The generated behavior trees, compatible with the BehaviorTree.CPP library, are evaluated both offline, using structural and lexical metrics, and online through the execution of household tasks in a state-of-the-art embodied simulator. Our results demonstrate that our fine-tuned 4B-parameter VLM approaches the performance of state-of-the-art closed-source models, achieving an 87\% success rate while requiring only a fraction of the computational resources.
☆ Lifelong Embodied Navigation Learning
Embodied navigation agents powered by large language models have shown strong performance on individual tasks but struggle to continually acquire new navigation skills, which suffer from catastrophic forgetting. We formalize this challenge as lifelong embodied navigation learning (LENL), where an agent is required to adapt to a sequence of navigation tasks spanning multiple scenes and diverse user instruction styles, while retaining previously learned knowledge. To tackle this problem, we propose Uni-Walker, a lifelong embodied navigation framework that decouples navigation knowledge into task-shared and task-specific components with Decoder Extension LoRA (DE-LoRA). To learn the shared knowledge, we design a knowledge inheritance strategy and an experts co-activation strategy to facilitate shared knowledge transfer and refinement across multiple navigation tasks. To learn the specific knowledge, we propose an expert subspace orthogonality constraint together and a navigation-specific chain-of-thought reasoning mechanism to capture specific knowledge and enhance instruction-style understanding. Extensive experiments demonstrate the superiority of Uni-Walker for building universal navigation agents with lifelong learning.
comment: 24 pages, 7 figures
☆ Transforming Omnidirectional RGB-LiDAR data into 3D Gaussian Splatting IROS
The demand for large-scale digital twins is rapidly growing in robotics and autonomous driving. However, constructing these environments with 3D Gaussian Splatting (3DGS) usually requires expensive, purpose-built data collection. Meanwhile, deployed platforms routinely collect extensive omnidirectional RGB and LiDAR logs, but a significant portion of these sensor data is directly discarded or strictly underutilized due to transmission constraints and the lack of scalable reuse pipeline. In this paper, we present an omnidirectional RGB-LiDAR reuse pipeline that transforms these archived logs into robust initialization assets for 3DGS. Direct conversion of such raw logs introduces practical bottlenecks: inherent non-linear distortion leads to unreliable Structure-from-Motion (SfM) tracking, and dense, unorganized LiDAR clouds cause computational overhead during 3DGS optimization. To overcome these challenges, our pipeline strategically integrates an ERP-to-cubemap conversion module for deterministic spatial anchoring, alongside PRISM-a color stratified downsampling strategy. By bridging these multi-modal inputs via Fast Point Feature Histograms (FPFH) based global registration and Iterative Closest Point (ICP), our pipeline successfully repurposes a considerable fraction of discarded data into usable SfM geometry. Furthermore, our LiDAR-reinforced initialization consistently enhances the final 3DGS rendering fidelity in structurally complex scenes compared to vision-only baselines. Ultimately, this work provides a deterministic workflow for creating simulation-grade digital twins from standard archived sensor logs.
comment: This work has been submitted to the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) for possible publication
☆ RODEO: RObotic DEcentralized Organization
Robots are improving their autonomy with minimal human supervision. However, auditable actions, transparent decision processes, and new human-robot interaction models are still missing requirements to achieve extended robot autonomy. To tackle these challenges, we propose RODEO (RObotic DEcentralized Organization), a blockchain-based framework that integrates trust and accountability mechanisms for robots. This paper formalizes Decentralized Autonomous Organizations (DAOs) for service robots. First, it provides a ROS-ETH bridge between the DAO and the robots. Second, it offers templates that enable organizations (e.g., companies, universities) to integrate service robots into their operations. Third, it provides proof-verification mechanisms that allow robot actions to be auditable. In our experimental setup, a mobile robot was deployed as a trash collector in a lab scenario. The robot collects trash and uses a smart bin to sort and dispose of it correctly. Then, the robot submits a proof of the successful operation and is compensated in DAO tokens. Finally, the robot re-invests the acquired funds to purchase battery charging services. Data collected in a three day experiment show that the robot doubled its income and reinvested funds to extend its operating time. The proof validation times of approximately one minute ensured verifiable task execution, while the accumulated robot income successfully funded up to 88 hours of future autonomous operation. The results of this research give insights about how robots and organizations can coordinate tasks and payments with auditable execution proofs and on-chain settlement.
comment: 8 pages, 6 figures, Accepted at IEEE International Conference on Robotics & Automation (2026)
☆ Devil is in Narrow Policy: Unleashing Exploration in Driving VLA Models CVPR2026
We identify a fundamental Narrow Policy limitation undermining the performance of autonomous VLA models, where driving Imitation Learning (IL) tends to collapse exploration and limit the potential of subsequent Reinforcement Learning (RL) stages, which often saturate prematurely due to insufficient feedback diversity. Thereby, we propose Curious-VLA, a framework that alleviates the exploit-explore dilemma through a two-stage design. During IL, we introduce a Feasible Trajectory Expansion (FTE) strategy to generate multiple physically valid trajectories and a step-wise normalized trajectory representation to adapt this diverse data. In the RL stage, we present Adaptive Diversity-Aware Sampling (ADAS) that prioritizes high-diversity samples and introduce Spanning Driving Reward (SDR) with a focal style weighting to amplify reward's value span for improving sensitivity to driving quality. On the Navsim benchmark, Curious-VLA achieves SoTA results (PDMS 90.3, EPDMS 85.4) and a Best-of-N PDMS of 94.8, demonstrating its effectiveness in unlocking the exploratory potential of VLA models. Code: https://github.com/Mashiroln/curious_vla.git.
comment: Accepted by CVPR2026 findings
☆ Restoring Linguistic Grounding in VLA Models via Train-Free Attention Recalibration
Vision-Language-Action (VLA) models enable robots to perform manipulation tasks directly from natural language instructions and are increasingly viewed as a foundation for generalist robotic policies. However, their reliability under Out-of-Distribution (OOD) instructions remains underexplored. In this paper, we reveal a critical failure mode in which VLA policies continue executing visually plausible actions even when the language instruction contradicts the scene. We refer to this phenomenon as linguistic blindness, where VLA policies prioritize visual priors over instruction semantics during action generation. To systematically analyze this issue, we introduce ICBench, a diagnostic benchmark constructed from the LIBERO dataset that probes language-action coupling by injecting controlled OOD instruction contradictions while keeping the visual environment unchanged. Evaluations on three representative VLA architectures, including Pi0, Pi0.5 and OpenVLA OFT, show that these models frequently succeed at tasks despite logically impossible instructions, revealing a strong visual bias in action generation. To mitigate this issue, we propose Instruction-Guided Attention Recalibration (IGAR), a train-free inference-time mechanism that rebalances attention distributions to restore the influence of language instructions. IGAR operates without retraining or architectural modification and can be directly applied to existing VLA models. Experiments across 30 LIBERO tasks demonstrate that IGAR substantially reduces erroneous execution under OOD contradictory instructions while preserving baseline task performance. We additionally validate the approach on a real Franka robotic arm, where IGAR effectively prevents manipulation triggered by inconsistent instructions.
☆ TADPO: Reinforcement Learning Goes Off-road ICRA 2026
Off-road autonomous driving poses significant challenges such as navigating unmapped, variable terrain with uncertain and diverse dynamics. Addressing these challenges requires effective long-horizon planning and adaptable control. Reinforcement Learning (RL) offers a promising solution by learning control policies directly from interaction. However, because off-road driving is a long-horizon task with low-signal rewards, standard RL methods are challenging to apply in this setting. We introduce TADPO, a novel policy gradient formulation that extends Proximal Policy Optimization (PPO), leveraging off-policy trajectories for teacher guidance and on-policy trajectories for student exploration. Building on this, we develop a vision-based, end-to-end RL system for high-speed off-road driving, capable of navigating extreme slopes and obstacle-rich terrain. We demonstrate our performance in simulation and, importantly, zero-shot sim-to-real transfer on a full-scale off-road vehicle. To our knowledge, this work represents the first deployment of RL-based policies on a full-scale off-road platform.
comment: 8 pages, 5 figures, 2 tables. Accepted at ICRA 2026
☆ Moving Through Clutter: Scaling Data Collection and Benchmarking for 3D Scene-Aware Humanoid Locomotion via Virtual Reality
Recent advances in humanoid locomotion have enabled dynamic behaviors such as dancing, martial arts, and parkour, yet these capabilities are predominantly demonstrated in open, flat, and obstacle-free settings. In contrast, real-world environments such as homes, offices, and public spaces, are densely cluttered, three-dimensional, and geometrically constrained, requiring scene-aware whole-body coordination, precise balance control, and reasoning over spatial constraints imposed by furniture and household objects. However, humanoid locomotion in cluttered 3D environments remains underexplored, and no public dataset systematically couples full-body human locomotion with the scene geometry that shapes it. To address this gap, we present Moving Through Clutter (MTC), an opensource Virtual Reality (VR) based data collection and evaluation framework for scene-aware humanoid locomotion in cluttered environments. Our system procedurally generates scenes with controllable clutter levels and captures embodiment-consistent, whole-body human motion through immersive VR navigation, which is then automatically retargeted to a humanoid robot model. We further introduce benchmarks that quantify environment clutter level and locomotion performance, including stability and collision safety. Using this framework, we compile a dataset of 348 trajectories across 145 diverse 3D cluttered scenes. The dataset provides a foundation for studying geometry-induced adaptation in humanoid locomotion and developing scene-aware planning and control methods.
☆ MagRobot:An Open Simulator for Magnetically Navigated Robots
Magnetic navigation systems, including magnetic tracking systems and magnetic actuation systems, have shown great potential for occlusion-free localization and remote control of intracorporeal medical devices and robots in minimally invasive medicine, such as capsule endoscopy and cardiovascular intervention. However, the design of magnetically navigated robots remains heavily reliant on experimental prototyping, which is time-consuming and costly. Furthermore, there is a lack of a consistent experimental environment to compare and benchmark the hardware and algorithms across different magnetic navigation systems. To address these challenges, we propose the first universal open-source simulation platform to facilitate research, design and benchmarking of magnetically navigated robots. Our simulator features an intuitive graphical user interface that enables the user to efficiently design, visualize, and analyze magnetic navigation systems for both rigid and soft robots. The proposed simulator is versatile, which can simulate both magnetic actuation and magnetic tracking tasks in diverse medical applications that involve deformable anatomies. The proposed simulator provides an open development environment, where the user can load third-party anatomical models and customize both hardware and algorithms of magnetic navigation systems. The fidelity of the simulator is validated using both phantom and ex vivo experiments of magnetic navigation of a continuum robot and a capsule robot with diverse magnetic actuation setups. Three use cases of the simulator, i.e., bronchoscopy, endovascular intervention, and gastrointestinal endoscopy, are implemented to demonstrate the functionality of the simulator. It is shown that the configuration and algorithms of magnetic navigation systems can be flexibly designed and optimized for better performance using the simulator.
comment: 20 pages, 10 figures
☆ HarvestFlex: Strawberry Harvesting via Vision-Language-Action Policy Adaptation in the Wild
This work presents the first study on transferring vision-language-action (VLA) policies to real greenhouse tabletop strawberry harvesting, a long-horizon, unstructured task challenged by occlusion and specular reflections. We built an end-to-end closed-loop system on the HarvestFlex platform using three-view RGB sensing (two fixed scene views plus a wrist-mounted view) and intentionally avoided depth clouds and explicit geometric calibration. We collected 3.71 h of VR teleoperated demonstrations (227 episodes) and fine-tuned pi_0, pi_0.5, and WALL-OSS with full fine-tuning and LoRA. Under a unified 50 trials real-greenhouse protocol and metrics spanning completion, pi_0.5 with full fine-tuning achieved success rate of 74.0% with 32.6 s/pick and damage rate of 4.1%. Asynchronous inference-control decoupling further improved performance over synchronous deployment. Results showed non-trivial closed-loop picking with fewer than four hours of real data, while remaining limited by close-range observability loss and contact-dynamics mismatch. A demonstration video is available at: https://youtu.be/bN8ZowZKPMI.
☆ Proprioceptive Shape Estimation of Tensegrity Manipulators Using Energy Minimisation ICRA 2026
Shape estimation is fundamental for controlling continuously bending tensegrity manipulators, yet achieving it remains a challenge. Although using exteroceptive sensors makes the implementation straightforward, it is costly and limited to specific environments. Proprioceptive approaches, by contrast, do not suffer from these limitations. So far, several methods have been proposed; however, to our knowledge, there are no proven examples of large-scale tensegrity structures used as manipulators. This paper demonstrates that shape estimation of the entire tensegrity manipulator can be achieved using only the inclination angle information relative to gravity for each strut. Inclination angle information is intrinsic sensory data that can be obtained simply by attaching an inertial measurement unit (IMU) to each strut. Experiments conducted on a five-layer tensegrity manipulator with 20 struts and a total length of 1160 mm demonstrate that the proposed method can estimate the shape with an accuracy of 2.1 \% of the total manipulator length, from arbitrary initial conditions under both static conditions and maintains stable shape estimation under external disturbances.
comment: 8 pages, 10 figures, IEEE ICRA 2026
☆ PROBE: Probabilistic Occupancy BEV Encoding with Analytical Translation Robustness for 3D Place Recognition
We present PROBE (PRobabilistic Occupancy BEV Encoding), a learning-free LiDAR place recognition descriptor that models each BEV cell's occupancy as a Bernoulli random variable. Rather than relying on discrete point-cloud perturbations, PROBE analytically marginalizes over continuous Cartesian translations via the polar Jacobian, yielding a distance-adaptive angular uncertainty $σ_θ= σ_t / r$ in $\mathcal{O}(R \times S)$ time. The primary parameter $σ_t$ represents the expected translational uncertainty in meters, a sensor-independent physical quantity allowing cross-sensor generalization without per-dataset tuning. Pairwise similarity combines a Bernoulli-KL Jaccard with exponential uncertainty gating and FFT-based height cosine similarity for rotation alignment. Evaluated on four datasets spanning four diverse LiDAR types, PROBE achieves the highest accuracy among handcrafted descriptors in multi-session evaluation and competitive single-session performance against both handcrafted and supervised baselines. The source code and supplementary materials are available at https://sites.google.com/view/probe-pr.
comment: 8 pages, 8 figures
☆ How to Model Your Crazyflie Brushless
The Crazyflie quadcopter is widely recognized as a leading platform for nano-quadcopter research. In early 2025, the Crazyflie Brushless was introduced, featuring brushless motors that provide around 50% more thrust compared to the brushed motors of its predecessor, the Crazyflie 2.1. This advancement has opened new opportunities for research in agile nano-quadcopter control. To support researchers utilizing this new platform, this work presents a dynamics model of the Crazyflie Brushless and identifies its key parameters. Through simulations and hardware analyses, we assess the accuracy of our model. We furthermore demonstrate its suitability for reinforcement learning applications by training an end-to-end neural network position controller and learning a backflip controller capable of executing two complete rotations with a vertical movement of just 1.8 meters. This showcases the model's ability to facilitate the learning of controllers and acrobatic maneuvers that successfully transfer from simulation to hardware. Utilizing this application, we investigate the impact of domain randomization on control performance, offering valuable insights into bridging the sim-to-real gap with the presented model. We have open-sourced the entire project, enabling users of the Crazyflie Brushless to swiftly implement and test their own controllers on an accurate simulation platform.
☆ Swooper: Learning High-Speed Aerial Grasping With a Simple Gripper
High-speed aerial grasping presents significant challenges due to the high demands on precise, responsive flight control and coordinated gripper manipulation. In this work, we propose Swooper, a deep reinforcement learning (DRL) based approach that achieves both precise flight control and active gripper control using a single lightweight neural network policy. Training such a policy directly via DRL is nontrivial due to the complexity of coordinating flight and grasping. To address this, we adopt a two-stage learning strategy: we first pre-train a flight control policy, and then fine-tune it to acquire grasping skills. With the carefully designed reward functions and training framework, the entire training process completes in under 60 minutes on a standard desktop with an Nvidia RTX 3060 GPU. To validate the trained policy in the real world, we develop a lightweight quadrotor grasping platform equipped with a simple off-the-shelf gripper, and deploy the policy in a zero-shot manner on the onboard Raspberry Pi 4B computer, where each inference takes only about 1.0 ms. In 25 real-world trials, our policy achieves an 84% grasp success rate and grasping speeds of up to 1.5 m/s without any fine-tuning. This matches the robustness and agility of state-of-the-art classical systems with sophisticated grippers, highlighting the capability of DRL for learning a robust control policy that seamlessly integrates high-speed flight and grasping. The supplementary video is available for more results. Video: https://zikenhuang.github.io/Swooper/.
☆ FTSplat: Feed-forward Triangle Splatting Network
High-fidelity three-dimensional (3D) reconstruction is essential for robotics and simulation. While Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) achieve impressive rendering quality, their reliance on time-consuming per-scene optimization limits real-time deployment. Emerging feed-forward Gaussian splatting methods improve efficiency but often lack explicit, manifold geometry required for direct simulation. To address these limitations, we propose a feed-forward framework for triangle primitive generation that directly predicts continuous triangle surfaces from calibrated multi-view images. Our method produces simulation-ready models in a single forward pass, obviating the need for per-scene optimization or post-processing. We introduce a pixel-aligned triangle generation module and incorporate relative 3D point cloud supervision to enhance geometric learning stability and consistency. Experiments demonstrate that our method achieves efficient reconstruction while maintaining seamless compatibility with standard graphics and robotic simulators.
☆ Iterative Convex Optimization with Control Barrier Functions for Obstacle Avoidance among Polytopes
Obstacle avoidance of polytopic obstacles by polytopic robots is a challenging problem in optimization-based control and trajectory planning. Many existing methods rely on smooth geometric approximations, such as hyperspheres or ellipsoids, which allow differentiable distance expressions but distort the true geometry and restrict the feasible set. Other approaches integrate exact polytope distances into nonlinear model predictive control (MPC), resulting in nonconvex programs that limit real-time performance. In this paper, we construct linear discrete-time control barrier function (DCBF) constraints by deriving supporting hyperplanes from exact closest-point computations between convex polytopes. We then propose a novel iterative convex MPC-DCBF framework, where local linearization of system dynamics and robot geometry ensures convexity of the finite-horizon optimization at each iteration. The resulting formulation reduces computational complexity and enables fast online implementation for safety-critical control and trajectory planning of general nonlinear dynamics. The framework extends to multi-robot and three-dimensional environments. Numerical experiments demonstrate collision-free navigation in cluttered maze scenarios with millisecond-level solve times.
comment: 9 pages, 4 figures
☆ Improved hopping control on slopes for small robots using spring mass modeling
Hopping robots often lose balance on slopes because the tilted ground creates unwanted rotation at landing. This work analyzes that effect using a simple spring mass model and identifies how slope induced impulses destabilize the robot. To address this, we introduce two straightforward fixes, adjusting the bodys touchdown angle based on the slope and applying a small corrective torque before takeoff. Together, these steps effectively cancel the unwanted rotation caused by inclined terrain, allowing the robot to land smoothly and maintain stable hopping even on steep slopes. Moreover, the proposed method remains simple enough to implement on low cost robotic platforms without requiring complex sensing or computation. By combining this analytical model with minimal control actions, this approach provides a practical path toward reliable hopping on uneven terrain. The results from simulation confirm that even small slope aware adjustments can dramatically improve landing stability, making the technique suitable for future autonomous field robots that must navigate natural environments such as hills, rubble, and irregular outdoor landscapes.
☆ Systematic Evaluation of Novel View Synthesis for Video Place Recognition IROS 2026
The generation of synthetic novel views has the potential to positively impact robot navigation in several ways. In image-based navigation, a novel overhead view generated from a scene taken by a ground robot could be used to guide an aerial robot to that location. In Video Place Recognition (VPR), novel views of ground locations from the air can be added that enable a UAV to identify places seen by the ground robot, and similarly, overhead views can be used to generate novel ground views. This paper presents a systematic evaluation of synthetic novel views in VPR using five public VPR image databases and seven typical image similarity methods. We show that for small synthetic additions, novel views improve VPR recognition statistics. We find that for larger additions, the magnitude of viewpoint change is less important than the number of views added and the type of imagery in the dataset.
comment: Submitted to IEEE IROS 2026
☆ AnyCamVLA: Zero-Shot Camera Adaptation for Viewpoint Robust Vision-Language-Action Models
Despite remarkable progress in Vision-Language-Action models (VLAs) for robot manipulation, these large pre-trained models require fine-tuning to be deployed in specific environments. These fine-tuned models are highly sensitive to camera viewpoint changes that frequently occur in unstructured environments. In this paper, we propose a zero-shot camera adaptation framework without additional demonstration data, policy fine-tuning, or architectural modification. Our key idea is to virtually adjust test-time camera observations to match the training camera configuration in real-time. For that, we use a recent feed-forward novel view synthesis model which outputs high-quality target view images, handling both extrinsic and intrinsic parameters. This plug-and-play approach preserves the pre-trained capabilities of VLAs and applies to any RGB-based policy. Through extensive experiments on the LIBERO benchmark, our method consistently outperforms baselines that use data augmentation for policy fine-tuning or additional 3D-aware features for visual input. We further validate that our approach constantly enhances viewpoint robustness in real-world robotic manipulation scenarios, including settings with varying camera extrinsics, intrinsics, and freely moving handheld cameras.
comment: Under review, Project Page: https://heo0224.github.io/AnyCamVLA/
☆ DexEMG: Towards Dexterous Teleoperation System via EMG2Pose Generalization
High-fidelity teleoperation of dexterous robotic hands is essential for bringing robots into unstructured domestic environments. However, existing teleoperation systems often face a trade-off between performance and portability: vision-based capture systems are constrained by costs and line-of-sight requirements, while mechanical exoskeletons are bulky and physically restrictive. In this paper, we present DexEMG, a lightweight and cost-effective teleoperation system leveraging surface electromyography (sEMG) to bridge the gap between human intent and robotic execution. We first collect a synchronized dataset of sEMG signals and hand poses via a MoCap glove to train EMG2Pose, a neural network capable of continuously predicting hand kinematics directly from muscle activity. To ensure seamless control, we develop a robust hand retargeting algorithm that maps the predicted poses onto a multi-fingered dexterous hand in real-time. Experimental results demonstrate that DexEMG achieves high precision in diverse teleoperation tasks. Notably, our system exhibits strong generalization capabilities across novel objects and complex environments without the need for intensive individual-specific recalibration. This work offers a scalable and intuitive interface for both general-purpose robotic manipulation and assistive technologies.
☆ Expert Knowledge-driven Reinforcement Learning for Autonomous Racing via Trajectory Guidance and Dynamics Constraints
Reinforcement learning has demonstrated significant potential in the field of autonomous driving. However, it suffers from defects such as training instability and unsafe action outputs when faced with autonomous racing environments characterized by high dynamics and strong nonlinearities. To this end, this paper proposes a trajectory guidance and dynamics constraints Reinforcement Learning (TraD-RL) method for autonomous racing. The key features of this method are as follows: 1) leveraging the prior expert racing line to construct an augmented state representation and facilitate reward shaping, thereby integrating domain knowledge to stabilize early-stage policy learning; 2) embedding explicit vehicle dynamic priors into a safe operating envelope formulated via control barrier functions to enable safety-constrained learning; and 3) adopting a multi-stage curriculum learning strategy that shifts from expert-guided learning to autonomous exploration, allowing the learned policy to surpass expert-level performance. The proposed method is evaluated in a high-fidelity simulation environment modeled after the Tempelhof Airport Street Circuit. Experimental results demonstrate that TraD-RL effectively improves both lap speed and driving stability of the autonomous racing vehicle, achieving a synergistic optimization of racing performance and safety.
☆ Terrain characterization and locomotion adaptation in a small-scale lizard-inspired robot IROS 2026
Unlike their large-scale counterparts, small-scale robots are largely confined to laboratory environments and are rarely deployed in real-world settings. As robot size decreases, robot-terrain interactions fundamentally change; however, there remains a lack of systematic understanding of what sensory information small-scale robots should acquire and how they should respond when traversing complex natural terrains. To address these challenges, we develop a Small-scale, Intelligent, Lizard-inspired, Adaptive Robot (SILA Bot) capable of adapting to diverse substrates. We use granular media of varying depths as a controlled yet representative terrain paradigm. We show that the optimal body movement pattern (ranging from standing-wave bending that assists limb retraction on flat ground to traveling-wave undulation that generates thrust in deep granular media) can be parameterized and approximated as a linear function of granular depth. Furthermore, proprioceptive signals, such as joint torque, provide sufficient information to estimate granular depth via a K-Nearest Neighbors classifier, achieving 95% accuracy. Leveraging these relationships, we design a simple linear feedback controller that modulates body phase and substantially improves locomotion performance on terrains with unknown depth. Together, these results establish a principled framework for perception and control in small-scale locomotion and enable effective terrain-adaptive locomotion while maintaining low computational complexity.
comment: 7 pages. 9 figures. IROS 2026 Conference
☆ OpenHEART: Opening Heterogeneous Articulated Objects with a Legged Manipulator
Legged manipulators offer high mobility and versatile manipulation. However, robust interaction with heterogeneous articulated objects, such as doors, drawers, and cabinets, remains challenging because of the diverse articulation types of the objects and the complex dynamics of the legged robot. Existing reinforcement learning (RL)-based approaches often rely on high-dimensional sensory inputs, leading to sample inefficiency. In this paper, we propose a robust and sample-efficient framework for opening heterogeneous articulated objects with a legged manipulator. In particular, we propose Sampling-based Abstracted Feature Extraction (SAFE), which encodes handle and panel geometry into a compact low-dimensional representation, improving cross-domain generalization. Additionally, Articulation Information Estimator (ArtIEst) is introduced to adaptively mix proprioception with exteroception to estimate opening direction and range of motion for each object. The proposed framework was deployed to manipulate various heterogeneous articulated objects in simulation and real-world robot systems. Videos can be found on the project website: https://openheart-icra.github.io/OpenHEART/
comment: 8 pages
☆ Hierarchical Latent Action Model ICLR 2026
Latent Action Models (LAMs) enable learning from actionless data for applications ranging from robotic control to interactive world models. However, existing LAMs typically focus on short-horizon frame transitions and capture low-level motion while overlooking longer-term temporal structure. In contrast, actionless videos often contain temporally extended and high-level skills. We present HiLAM, a hierarchical latent action model that discovers latent skills by modeling long-term temporal information. To capture these dependencies across long horizons, we utilize a pretrained LAM as a low-level extractor. This architecture aggregates latent action sequences, which contain the underlying dynamic patterns of the video, into high-level latent skills. Our experiments demonstrate that HiLAM improves over the baseline and exhibits robust dynamic skill discovery.
comment: ICLR 2026 Workshop - 2nd Workshop on World Models: Understanding, Modelling and Scaling
☆ CDF-Glove: A Cable-Driven Force Feedback Glove for Dexterous Teleoperation
High-quality teleoperated demonstrations are a primary bottleneck for imitation learning (IL) in dexterous manipulation. However, haptic feedback provides operators with real-time contact information, enabling real-time finger posture adjustments, and thereby improving demonstration quality. Existing dexterous teleoperation platforms typically omit haptic feedback and remain bulky and expensive. We introduce CDF-Glove, a lightweight and low cost cable-driven force-feedback glove. The real-time state is available for 20 finger degrees of freedom (DoF), of which 16 are directly sensed and 4 are passively coupled (inferred from kinematic constraints). We develop a kinematic model and control stack for the glove, and validate them across multiple robotic hands with diverse kinematics and DoF. The CDF-Glove achieves distal joint repeatability of 0.4 degrees, and delivers about 200 ms force feedback latency, yielding a 4x improvement in task success rate relative to no-feedback teleoperation. We collect two bimanual teleoperation datasets, on which we train and evaluate Diffusion Policy baselines. Compared to kinesthetic teaching, the policies trained in our teleoperated demonstrations increase the average success rate by 55% and reduce the mean completion time by approximately 15.2 seconds (a 47.2% relative reduction). In particular, the CDF-Glove costs approximately US$230. The code and designs are released as open source at https://cdfglove.github.io/.
☆ Task-Level Decisions to Gait Level Control: A Hierarchical Policy Approach for Quadruped Navigation IROS 2026
Real-world quadruped navigation is constrained by a scale mismatch between high-level navigation decisions and low-level gait execution, as well as by instabilities under out-of-distribution environmental changes. Such variations challenge sim-to-real transfer and can trigger falls when policies lack explicit interfaces for adaptation. In this paper, we present a hierarchical policy architecture for quadrupedal navigation, termed Task-level Decision to Gait Control (TDGC). A low-level policy, trained with reinforcement learning in simulation, delivers gait-conditioned locomotion and maps task requirements to a compact set of controllable behavior parameters, enabling robust mode generation and smooth switching. A high-level policy makes task-centric decisions from sparse semantic or geometric terrain cues and translates them into low-level targets, forming a traceable decision pipeline without dense maps or high-resolution terrain reconstruction. Different from end-to-end approaches, our architecture provides explicit interfaces for deployment-time tuning, fault diagnosis, and policy refinement. We introduce a structured curriculum with performance-driven progression that expands environmental difficulty and disturbance ranges. Experiments show higher task success rates on mixed terrains and out-of-distribution tests.
comment: Submitted to IROS 2026
☆ Multi-Robot Trajectory Planning via Constrained Bayesian Optimization and Local Cost Map Learning with STL-Based Conflict Resolution ICRA 2026
We address multi-robot motion planning under Signal Temporal Logic (STL) specifications with kinodynamic constraints. Exact approaches face scalability bottlenecks and limited adaptability, while conventional sampling-based methods require excessive samples to construct optimal trajectories. We propose a two-stage framework integrating sampling-based online learning with formal STL reasoning. At the single-robot level, our constrained Bayesian Optimization-based Tree search (cBOT) planner uses a Gaussian process as a surrogate model to learn local cost maps and feasibility constraints, generating shorter collision-free trajectories with fewer samples. At the multi-robot level, our STL-enhanced Kinodynamic Conflict-Based Search (STL-KCBS) algorithm incorporates STL monitoring into conflict detection and resolution, ensuring specification satisfaction while maintaining scalability and probabilistic completeness. Benchmarking demonstrates improved trajectory efficiency and safety over existing methods. Real-world experiments with autonomous surface vehicles validate robustness and practical applicability in uncertain environments. The STLcBOT Planner will be released as an open-source package, and videos of real-world and simulated experiments are available at https://stlbot.github.io/.
comment: Accepted to ICRA 2026
☆ Feasibility Restoration under Conflicting STL Specifications with Pareto-Optimal Refinement
Signal Temporal Logic (STL) is expressive formal language that specifies spatio-temporal requirements in robotics. Its quantitative robustness semantics can be easily integrated with optimization-based control frameworks. However, STL specifications may become conflicting in real-world applications, where safety rules, traffic regulations, and task objectives can be cannot be satisfied together. In these situations, traditional STL-constrained Model Predictive Control (MPC) becomes infeasible and default to conservative behaviors such as freezing, which can largely increase risks in safety-critical scenarios. In this paper, we proposes a unified two-stage framework that first restores feasibility via minimal relaxation, then refine the feasible solution by formulating it as a value-aware multi-objective optimization problem. Using $\varepsilon$-constraint method, we approximate the Pareto front of the multi-objective optimization, which allows analysis of tradeoffs among competing objectives and counterfactual analysis of alternative actions. We demonstrate that the proposed approach avoids deadlock under conflicting STL specifications and enables interpretable decision-making in safety-critical applications by conducting a case study in autonomous driving.
☆ Failure Mechanisms and Risk Estimation for Legged Robot Locomotion on Granular Slopes
Locomotion on granular slopes such as sand dunes remains a fundamental challenge for legged robots due to reduced shear strength and gravity-induced anisotropic yielding of granular media. Using a hexapedal robot on a tiltable granular bed, we systematically measure locomotion speed together with slope-dependent normal and shear granular resistive forces. While normal penetration resistance remains nearly unchanged with inclination, shear resistance decreases substantially as slope angle increases. Guided by these measurements, we develop a simple robot-terrain interaction model that predicts anchoring timing, step length, and resulting robot speed, as functions of terrain strength and slope angle. The model reveals that slope-induced performance loss is primarily governed by delayed anchoring and increased backward slip rather than excessive sinkage. By extending the model to generalized terrain conditions, we construct failure phase diagrams that identify sinkage- and slippage-induced failure regimes, enabling quantitative risk estimation for locomotion on granular slopes. This physics-informed framework provides predictive insight into terrain-dependent failure mechanisms and offers guidance for safer and more robust robot operation on deformable inclines.
☆ A Contrastive Fewshot RGBD Traversability Segmentation Framework for Indoor Robotic Navigation
Indoor traversability segmentation aims to identify safe, navigable free space for autonomous agents, which is critical for robotic navigation. Pure vision-based models often fail to detect thin obstacles, such as chair legs, which can pose serious safety risks. We propose a multi-modal segmentation framework that leverages RGB images and sparse 1D laser depth information to capture geometric interactions and improve the detection of challenging obstacles. To reduce the reliance on large labeled datasets, we adopt the few-shot segmentation (FSS) paradigm, enabling the model to generalize from limited annotated examples. Traditional FSS methods focus solely on positive prototypes, often leading to overfitting to the support set and poor generalization. To address this, we introduce a negative contrastive learning (NCL) branch that leverages negative prototypes (obstacles) to refine free-space predictions. Additionally, we design a two-stage attention depth module to align 1D depth vectors with RGB images both horizontally and vertically. Extensive experiments on our custom-collected indoor RGB-D traversability dataset demonstrate that our method outperforms state-of-the-art FSS and RGB-D segmentation baselines, achieving up to 9\% higher mIoU under both 1-shot and 5-shot settings. These results highlight the effectiveness of leveraging negative prototypes and sparse depth for robust and efficient traversability segmentation.
☆ LIPP: Load-Aware Informative Path Planning with Physical Sampling
In classical Informative Path Planning (C-IPP), robots are typically modeled as mobile sensors that acquire digital measurements such as images or radiation levels. In this model - since making a measurement leaves the robot's physical state unchanged - traversal costs are determined solely by the path taken. This is a natural assumption for many missions, but does not extend to settings involving physical sample collection, where each collected sample adds mass and increases the energy cost of all subsequent motion. As a result, IPP formulations that ignore this coupling between information gain and load-dependent traversal cost can produce plans that are distance-efficient but energy-suboptimal, collecting fewer samples and less data than the energy budget would permit. In this paper, we introduce Load-aware Informative Path Planning (LIPP ), a generalization of C-IPP that explicitly models this coupling and the resulting order-dependent traversal costs. We formulate LIPP as a Mixed-Integer Quadratic Program (MIQP) that jointly optimizes routing, visitation order, and per-location sampling count under an energy budget. We show that LIPP strictly generalizes C-IPP: as sample unit mass $λ\to 0$, the load-dependent energy model reduces exactly to the classical distance budget constraint, recovering C-IPP as a special case. We further derive theoretical bounds on the path-length increase of LIPP relative to C-IPP, characterizing the trade-off for improved energy efficiency. Finally, through extensive simulations across 2000 diverse mission scenarios, we demonstrate that LIPP matches the behavior of C-IPP at zero sample mass and progressively achieves higher uncertainty reduction per unit energy as sample mass increases.
☆ CN-CBF: Composite Neural Control Barrier Function for Safe Robot Navigation in Dynamic Environments
Safe navigation of autonomous robots remains one of the core challenges in the field, especially in dynamic and uncertain environments. One of the prevalent approaches is safety filtering based on control barrier functions (CBFs), which are easy to deploy but difficult to design. Motivated by the shortcomings of existing learning- and model-based methods, we propose a simple yet effective neural CBF design method for safe robot navigation in dynamic environments. We employ the idea of a composite CBF, where multiple neural CBFs are combined into a single CBF. The individual CBFs are trained via the Hamilton-Jacobi reachability framework to approximate the optimal safe set for single moving obstacles. Additionally, we use the residual neural architecture, which guarantees that the estimated safe set does not intersect with the corresponding failure set. The method is extensively evaluated in simulation experiments for a ground robot and a quadrotor, comparing it against several baseline methods. The results show improved success rates of up to 18\% compared to the best baseline, without increasing the conservativeness of the motion. Also, the method is demonstrated in hardware experiments for both types of robots.
☆ SurgSync: Time-Synchronized Multi-Modal Data Collection Framework and Dataset for Surgical Robotics ICRA
Most existing robotic surgery systems adopt a human-in-the-loop paradigm, often with the surgeon directly teleoperating the robotic system. Adding intelligence to these robots would enable higher-level control, such as supervised autonomy or even full autonomy. However, artificial intelligence (AI) requires large amounts of training data, which is currently lacking. This work proposes SurgSync, a multi-modal data collection framework with offline and online synchronization to support training and real-time inference, respectively. The framework is implemented on a da Vinci Research Kit (dVRK) and introduces (1) dual-mode (online/offline-matching) synchronized recorders, (2) a modern stereo endoscope to achieve image quality on par with clinical systems, and (3) additional sensors such as a side-view camera and a novel capacitive contact sensor to provide ground truth contact data. The framework also incorporates a post-processing toolbox for tasks such as depth estimation, optical flow, and a practical kinematic reprojection method using Gaussian heatmap. User studies with participants of varying skill levels are performed with ex-vivo tissue to provide clinically realistic data, and a network for surgical skill assessment is employed to demonstrate utilization of the collected data. Through the user study experiments, we obtained a dataset of 214 validated instances across multiple canonical training tasks. All software and data are available at surgsync.github.io.
comment: Accepted By International Conference on Robotics and Automation (ICRA), IEEE, 2026. More details can be found at https://surgsync.github.io/
☆ T2Nav Algebraic Topology Aware Temporal Graph Memory and Loop Detection for ZeroShot Visual Navigation
Deploying autonomous agents in real world environments is challenging, particularly for navigation, where systems must adapt to situations they have not encountered before. Traditional learning approaches require substantial amounts of data, constant tuning, and, sometimes, starting over for each new task. That makes them hard to scale and not very flexible. Recent breakthroughs in foundation models, such as large language models and vision language models, enable systems to attempt new navigation tasks without requiring additional training. However, many of these methods only work with specific input types, employ relatively basic reasoning, and fail to fully exploit the details they observe or the structure of the spaces. Here, we introduce T2Nav, a zeroshot navigation system that integrates heterogeneous data and employs graph-based reasoning. By directly incorporating visual information into the graph and matching it to the environment, our approach enables the system to strike a good balance between exploration and goal attainment. This strategy allows robust obstacle avoidance, reliable loop closure detection, and efficient path planning while eliminating redundant exploration patterns. The system demonstrates flexibility by handling goals specified using reference images of target object instances, making it particularly suitable for scenarios in which agents must navigate to visually similar yet spatially distinct instances. Experiments demonstrate that our approach is efficient and adapts well to unknown environments, moving toward practical zero-shot instance-image navigation capabilities.
☆ SysNav: Multi-Level Systematic Cooperation Enables Real-World, Cross-Embodiment Object Navigation
Object navigation (ObjectNav) in real-world environments is a complex problem that requires simultaneously addressing multiple challenges, including complex spatial structure, long-horizon planning and semantic understanding. Recent advances in Vision-Language Models (VLMs) offer promising capabilities for semantic understanding, yet effectively integrating them into real-world navigation systems remains a non-trivial challenge. In this work, we formulate real-world ObjectNav as a system-level problem and introduce SysNav, a three-level ObjectNav system designed for real-world crossembodiment deployment. SysNav decouples semantic reasoning, navigation planning and motion control to ensure robustness and generalizability. At the high-level, we summarize the environment into a structured scene representation and leverage VLMs to provide semantic-grounded navigation guidance. At the mid-level, we introduce a hierarchical room-based navigation strategy that reserves VLM guidance for room-level decisions, which effectively utilizes its reasoning ability while ensuring system efficiency. At the low-level, planned waypoints are executed through different embodiment-specific motion control modules. We deploy our system on three embodiments, a custom-built wheeled robot, the Unitree Go2 quadruped and the Unitree G1 humanoid, and conduct 190 real-world experiments. Our system achieves substantial improvements in both success rate and navigation efficiency. To the best of our knowledge, SysNav is the first system capable of reliably and efficiently completing building-scale long-range object navigation in complex real-world environments. Furthermore, extensive experiments on four simulation benchmarks demonstrate state-of-the-art performance. Project page is available at: https://cmu-vln.github.io/.
☆ Collaborative Planning with Concurrent Synchronization for Operationally Constrained UAV-UGV Teams
Collaborative planning under operational constraints is an essential capability for heterogeneous robot teams tackling complex large-scale real-world tasks. Unmanned Aerial Vehicles (UAVs) offer rapid environmental coverage, but flight time is often limited by energy constraints, whereas Unmanned Ground Vehicles (UGVs) have greater energy capacity to support long-duration missions, but movement is constrained by traversable terrain. Individually, neither can complete tasks such as environmental monitoring. Effective UAV-UGV collaboration therefore requires energy-constrained multi-UAV task planning, traversability-constrained multi-UGV path planning, and crucially, synchronized concurrent co-planning to ensure timely in-mission recharging. To enable these capabilities, we propose Collaborative Planning with Concurrent Synchronization (CoPCS), a learning-based approach that integrates a heterogeneous graph transformer for operationally constrained task encoding with a transformer decoder for joint, synchronized co-planning that enables UAVs and UGVs to act concurrently in a coordinated manner. CoPCS is trained end-to-end under a unified imitation learning paradigm. We conducted extensive experiments to evaluate CoPCS in both robotic simulations and physical robot teams. Experimental results demonstrate that our method provides the novel multi-robot capability of synchronized concurrent co-planning and substantially improves team performance. More details of this work are available on the project website: https://hcrlab.gitlab.io/project/CoPCS.
☆ VertiAdaptor: Online Kinodynamics Adaptation for Vertically Challenging Terrain
Autonomous driving in off-road environments presents significant challenges due to the dynamic and unpredictable nature of unstructured terrain. Traditional kinodynamic models often struggle to generalize across diverse geometric and semantic terrain types, underscoring the need for real-time adaptation to ensure safe and reliable navigation. We propose VertiAdaptor (VA), a novel online adaptation framework that efficiently integrates elevation with semantic embeddings to enable terrain-aware kinodynamic modeling and planning via function encoders. VA learns a kinodynamic space spanned by a set of neural ordinary differential equation basis functions, capturing complex vehicle-terrain interactions across varied environments. After offline training, the proposed approach can rapidly adapt to new, unseen environments by identifying kinodynamics in the learned space through a computationally efficient least-squares calculation. We evaluate VA within the Verti-Bench simulator, built on the Chrono multi-physics engine, and validate its performance both in simulation and on a physical Verti-4-Wheeler platform. Our results demonstrate that VA improves prediction accuracy by up to 23.9% and achieves a 5X faster adaptation time, advancing the robustness and reliability of autonomous robots in complex and evolving off-road environments.
☆ Material Driven HRI Design: Aesthetics as Explainability
Aesthetics - often treated as secondary to function-guides how people interpret robots' roles. A great deal of robot designs - both real and fictitious - use sleek industrial aesthetics. These feature hard glossy plastics, hiding as much of the underlying mechanical and electrical components as possible, resembling something akin to a nude humanoid figure. This leaves robots as something of a blank slate to which end-users apply coverings to, often based on media of fiction and non-fiction alike. We argue that designers can take cues from fashion to design interaction and set appropriate expectations. Rather than viewing appearance as decoration, we propose that color, texture, and material choices function as interaction signals. These signals can invite or discourage touch, clarify a robot's role, and help align user expectations with a robot's actual capabilities. When done thoughtfully, such cues can create familiarity and legibility; when done poorly, they can lead to wrong expectations. This preliminary paper proposes a framework describing how materials can create explainability by signaling expectations for interaction, task, and environment. We use this framework to do a content analysis of 6 robots.
comment: 4 pages, 1 table, 2026 ACM/IEEE Human-Robot Interaction Conference Workshop on Articulating the Value of Design Research for HRI
☆ CAR: Cross-Vehicle Kinodynamics Adaptation via Mobility Representation
Developing autonomous off-road mobility typically requires either extensive, platform-specific data collection or relies on simplified abstractions, such as unicycle or bicycle models, that fail to capture the complex kinodynamics of diverse platforms, ranging from wheeled to tracked vehicles. This limitation hinders scalability across evolving heterogeneous autonomous robot fleets. To address this challenge, we propose Cross-vehicle kinodynamics Adaptation via mobility Representation (CAR), a novel framework that enables rapid mobility transfer to new vehicles. CAR employs a Transformer encoder with Adaptive Layer Normalization to embed vehicle trajectory transitions and physical configurations into a shared mobility latent space. By identifying and extracting commonality from nearest neighbors within this latent space, our approach enables rapid kinodynamics adaptation to novel platforms with minimal data collection and computational overhead. We evaluate CAR using the Verti-Bench simulator, built on the Chrono multi-physics engine, and validate its performance on four distinct physical configurations of the Verti-4-Wheeler platform. With only one minute of new trajectory data, CAR achieves up to 67.2% reduction in prediction error compared to direct neighbor transfer across diverse unseen vehicle configurations, demonstrating the effectiveness of cross-vehicle mobility knowledge transfer in both simulated and real-world environments.
☆ Robodimm: A Physics-Grounded Framework for Automated Actuator Sizing in Scalable Modular Robots
Selecting an appropriate motor-gearbox combination is a critical design task in robotics because it directly affects cost, mass, and dynamic performance. This process is especially challenging in modular robots with closed kinematic chains, where joint torques are coupled and actuator inertia propagates through the mechanism. We present Robodimm, a software framework for automated actuator sizing in scalable robot architectures. By leveraging Pinocchio for dynamics and Pink for inverse kinematics, Robodimm uses a Karush-Kuhn-Tucker (KKT) formulation for constrained inverse dynamics. The platform supports parametric scaling, interactive trajectory programming through jog modes, and a two-round validation workflow that addresses actuator self-weight effects.
comment: 8 pages, 3 figures. Preprint version submitted to arXiv
☆ Nonlinear Performance Degradation of Vision-Based Teleoperation under Network Latency
Teleoperation is increasingly being adopted as a critical fallback for autonomous vehicles. However, the impact of network latency on vision-based, perception-driven control remains insufficiently studied. The present work investigates the nonlinear degradation of closed-loop stability in camera-based lane keeping under varying network delays. To conduct this study, we developed the Latency-Aware Vision Teleoperation testbed (LAVT), a research-oriented ROS 2 framework that enables precise, distributed one-way latency measurement and reproducible delay injection. Using LAVT, we performed 180 closed-loop experiments in simulation across diverse road geometries. Our findings reveal a sharp collapse in stability between 150 ms and 225 ms of one-way perception latency, where route completion rates drop from 100% to below 50% as oscillatory instability and phase-lag effects emerge. We further demonstrate that additional control-channel delay compounds these effects, significantly accelerating system failure even under constant visual latency. By combining this systematic empirical characterization with the LAVT testbed, this work provides quantitative insights into perception-driven instability and establishes a reproducible baseline for future latency-compensation and predictive control strategies. Project page, supplementary video, and code are available at https://bimilab.github.io/paper-LAVT
☆ MotionBits: Video Segmentation through Motion-Level Analysis of Rigid Bodies
Rigid bodies constitute the smallest manipulable elements in the real world, and understanding how they physically interact is fundamental to embodied reasoning and robotic manipulation. Thus, accurate detection, segmentation, and tracking of moving rigid bodies is essential for enabling reasoning modules to interpret and act in diverse environments. However, current segmentation models trained on semantic grouping are limited in their ability to provide meaningful interaction-level cues for completing embodied tasks. To address this gap, we introduce MotionBit, a novel concept that, unlike prior formulations, defines the smallest unit in motion-based segmentation through kinematic spatial twist equivalence, independent of semantics. In this paper, we contribute (1) the MotionBit concept and definition, (2) a hand-labeled benchmark, called MoRiBo, for evaluating moving rigid-body segmentation across robotic manipulation and human-in-the-wild videos, and (3) a learning-free graph-based MotionBits segmentation method that outperforms state-of-the-art embodied perception methods by 37.3\% in macro-averaged mIoU on the MoRiBo benchmark. Finally, we demonstrate the effectiveness of MotionBits segmentation for downstream embodied reasoning and manipulation tasks, highlighting its importance as a fundamental primitive for understanding physical interactions.
comment: 23 pages, 18 figures
☆ RoboCritics: Enabling Reliable End-to-End LLM Robot Programming through Expert-Informed Critics
End-user robot programming grants users the flexibility to re-task robots in situ, yet it remains challenging for novices due to the need for specialized robotics knowledge. Large Language Models (LLMs) hold the potential to lower the barrier to robot programming by enabling task specification through natural language. However, current LLM-based approaches generate opaque, "black-box" code that is difficult to verify or debug, creating tangible safety and reliability risks in physical systems. We present RoboCritics, an approach that augments LLM-based robot programming with expert-informed motion-level critics. These critics encode robotics expertise to analyze motion-level execution traces for issues such as joint speed violations, collisions, and unsafe end-effector poses. When violations are detected, critics surface transparent feedback and offer one-click fixes that forward structured messages back to the LLM, enabling iterative refinement while keeping users in the loop. We instantiated RoboCritics in a web-based interface connected to a UR3e robot and evaluated it in a between-subjects user study (n=18). Compared to a baseline LLM interface, RoboCritics reduced safety violations, improved execution quality, and shaped how participants verified and refined their programs. Our findings demonstrate that RoboCritics enables more reliable and user-centered end-to-end robot programming with LLMs.
comment: 10 pages, 5 figures, Proceedings of the 21st ACM/IEEE International Conference on Human Robot Interaction (HRI 2026)
☆ Receding-Horizon Nullspace Optimization for Actuation-Aware Control Allocation in Omnidirectional UAVs
Fully actuated omnidirectional UAVs enable independent control of forces and torques along all six degrees of freedom, broadening the operational envelope for agile flight and aerial interaction tasks. However, conventional control allocation methods neglect the asymmetric dynamics of the onboard actuators, which can induce oscillatory motor commands and degrade trajectory tracking during dynamic maneuvers. This work proposes a receding-horizon, actuation-aware allocation strategy that explicitly incorporates asymmetric motor dynamics and exploits the redundancy of over-actuated platforms through nullspace optimization. By forward-simulating the closed-loop system over a prediction horizon, the method anticipates actuator-induced oscillations and suppresses them through smooth redistribution of motor commands, while preserving the desired body wrench exactly. The approach is formulated as a constrained optimal control problem solved online via Constrained iterative LQR. Simulation results on the OmniOcta platform demonstrate that the proposed method significantly reduces motor command oscillations compared to a conventional single-step quadratic programming allocator, yielding improved trajectory tracking in both position and orientation.
comment: 8 pages, 8 figures
☆ Learning-Based Robust Control: Unifying Exploration and Distributional Robustness for Reliable Robotics via Free Energy
A key challenge towards reliable robotic control is devising computational models that can both learn policies and guarantee robustness when deployed in the field. Inspired by the free energy principle in computational neuroscience, to address these challenges, we propose a model for policy computation that jointly learns environment dynamics and rewards, while ensuring robustness to epistemic uncertainties. Expounding a distributionally robust free energy principle, we propose a modification to the maximum diffusion learning framework. After explicitly characterizing robustness of our policies to epistemic uncertainties in both environment and reward, we validate their effectiveness on continuous-control benchmarks, via both simulations and real-world experiments involving manipulation with a Franka Research~3 arm. Across simulation and zero-shot deployment, our approach narrows the sim-to-real gap, and enables repeatable tabletop manipulation without task-specific fine-tuning.
☆ A Comprehensive Analysis of the Effects of Network Quality of Service on Robotic Telesurgery
The viability of long-distance telesurgery hinges on reliable network Quality of Service (QoS), yet the impact of realistic network degradations on task performance is not sufficiently understood. This paper presents a comprehensive analysis of how packet loss, delay, and communication loss affect telesurgical task execution. We introduce NetFI, a novel fault injection tool that emulates different network conditions using stochastic QoS models informed by real-world network data. By integrating NetFI with a surgical simulation platform, we conduct a user study involving 15 participants at three proficiency levels, performing a standardized Peg Transfer task under varying levels of packet loss, delay, and communication loss. We analyze the effect of network QoS on overall task performance and the fine-grained motion primitives (MPs) using objective performance and safety metrics and subjective operator's perception of workload. We identify specific MPs vulnerable to network degradation and find strong correlations between proficiency, objective performance, and subjective workload. These findings offer quantitative insights into the operational boundaries of telesurgery. Our open-source tools and annotated dataset provide a foundation for developing robust and network-aware control and mitigation strategies.
comment: Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
☆ BEVLM: Distilling Semantic Knowledge from LLMs into Bird's-Eye View Representations
The integration of Large Language Models (LLMs) into autonomous driving has attracted growing interest for their strong reasoning and semantic understanding abilities, which are essential for handling complex decision-making and long-tail scenarios. However, existing methods typically feed LLMs with tokens from multi-view and multi-frame images independently, leading to redundant computation and limited spatial consistency. This separation in visual processing hinders accurate 3D spatial reasoning and fails to maintain geometric coherence across views. On the other hand, Bird's-Eye View (BEV) representations learned from geometrically annotated tasks (e.g., object detection) provide spatial structure but lack the semantic richness of foundation vision encoders. To bridge this gap, we propose BEVLM, a framework that connects a spatially consistent and semantically distilled BEV representation with LLMs. Through extensive experiments, we show that BEVLM enables LLMs to reason more effectively in cross-view driving scenes, improving accuracy by 46%, by leveraging BEV features as unified inputs. Furthermore, by distilling semantic knowledge from LLMs into BEV representations, BEVLM significantly improves closed-loop end-to-end driving performance by 29% in safety-critical scenarios.
comment: 4 figures, 6 tables in the main paper, 32 pages in total
☆ Fly360: Omnidirectional Obstacle Avoidance within Drone View
Obstacle avoidance in unmanned aerial vehicles (UAVs), as a fundamental capability, has gained increasing attention with the growing focus on spatial intelligence. However, current obstacle-avoidance methods mainly depend on limited field-of-view sensors and are ill-suited for UAV scenarios which require full-spatial awareness when the movement direction differs from the UAV's heading. This limitation motivates us to explore omnidirectional obstacle avoidance for panoramic drones with full-view perception. We first study an under explored problem setting in which a UAV must generate collision-free motion in environments with obstacles from arbitrary directions, and then construct a benchmark that consists of three representative flight tasks. Based on such settings, we propose Fly360, a two-stage perception-decision pipeline with a fixed random-yaw training strategy. At the perception stage, panoramic RGB observations are input and converted into depth maps as a robust intermediate representation. For the policy network, it is lightweight and used to output body-frame velocity commands from depth inputs. Extensive simulation and real-world experiments demonstrate that Fly360 achieves stable omnidirectional obstacle avoidance and outperforms forward-view baselines across all tasks. Our model is available at https://zxkai.github.io/fly360/
comment: 16 pages, 10 figures
☆ Uncertainty-Aware Adaptive Dynamics For Underwater Vehicle-Manipulator Robots
Accurate and adaptive dynamic models are critical for underwater vehicle-manipulator systems where hydrodynamic effects induce time-varying parameters. This paper introduces a novel uncertainty-aware adaptive dynamics model framework that remains linear in lumped vehicle and manipulator parameters, and embeds convex physical consistency constraints during online estimation. Moving horizon estimation is used to stack horizon regressors, enforce realizable inertia, damping, friction, and hydrostatics, and quantify uncertainty from parameter evolution. Experiments on a BlueROV2 Heavy with a 4-DOF manipulator demonstrate rapid convergence and calibrated predictions. Manipulator fits achieve R2 = 0.88 to 0.98 with slopes near unity, while vehicle surge, heave, and roll are reproduced with good fidelity under stronger coupling and noise. Median solver time is approximately 0.023 s per update, confirming online feasibility. A comparison against a fixed parameter model shows consistent reductions in MAE and RMSE across degrees of freedom. Results indicate physically plausible parameters and confidence intervals with near 100% coverage, enabling reliable feedforward control and simulation in underwater environments.
☆ Unified Learning of Temporal Task Structure and Action Timing for Bimanual Robot Manipulation
Temporal task structure is fundamental for bimanual manipulation: a robot must not only know that one action precedes or overlaps another, but also when each action should occur and how long it should take. While symbolic temporal relations enable high-level reasoning about task structure and alternative execution sequences, concrete timing parameters are equally essential for coordinating two hands at the execution level. Existing approaches address these two levels in isolation, leaving a gap between high-level task planning and low-level movement synchronization. This work presents an approach for learning both symbolic and subsymbolic temporal task constraints from human demonstrations and deriving executable, temporally parametrized plans for bimanual manipulation. Our contributions are (i) a 3-dimensional representation of timings between two actions with methods based on multivariate Gaussian Mixture Models to represent temporal relationships between actions on a subsymbolic level, (ii) a method based on the Davis-Putnam-Logemann-Loveland (DPLL) algorithm that finds and ranks all contradiction-free assignments of Allen relations to action pairs, representing different modes of a task, and (iii) an optimization-based planning system that combines the identified symbolic and subsymbolic temporal task constraints to derive temporally parametrized plans for robot execution. We evaluate our approach on several datasets, demonstrating that our method generates temporally parametrized plans closer to human demonstrations than the most characteristic demonstration baseline.
comment: This work has been submitted to the IEEE for possible publication
☆ A Multi-Layer Sim-to-Real Framework for Gaze-Driven Assistive Neck Exoskeletons ICRA
Dropped head syndrome, caused by neck muscle weakness from neurological diseases, severely impairs an individual's ability to support and move their head, causing pain and making everyday tasks challenging. Our long-term goal is to develop an assistive powered neck exoskeleton that restores natural movement. However, predicting a user's intended head movement remains a key challenge. We leverage virtual reality (VR) to collect coupled eye and head movement data from healthy individuals to train models capable of predicting head movement based solely on eye gaze. We also propose a novel multi-layer controller selection framework, where head control strategies are evaluated across decreasing levels of abstraction -- from simulation and VR to a physical neck exoskeleton. This pipeline effectively rejects poor-performing controllers early, identifying two novel gaze-driven models that achieve strong performance when deployed on the physical exoskeleton. Our results reveal that no single controller is universally preferred, highlighting the necessity for personalization in gaze-driven assistive control. Our work demonstrates the utility of VR-based evaluation for accelerating the development of intuitive, safe, and personalized assistive robots.
comment: IEEE International Conference on Robotics & Automation (ICRA), 2026. Equal Contribution from the first two authors
☆ Spatial Calibration of Diffuse LiDARs
Diffuse direct time-of-flight LiDARs report per-pixel depth histograms formed by aggregating photon returns over a wide instantaneous field of view, violating the single-ray assumption behind standard LiDAR-RGB calibration. We present a simple spatial calibration procedure that estimates, for each diffuse LiDAR pixel, its footprint (effective support region) and relative spatial sensitivity in a co-located RGB image plane. Using a scanned retroreflective patch with background subtraction, we recover per-pixel response maps that provide an explicit LiDAR-to-RGB correspondence for cross-modal alignment and fusion. We demonstrate the method on the ams OSRAM TMF8828.
☆ HybridMimic: Hybrid RL-Centroidal Control for Humanoid Motion Mimicking
Motion mimicking, i.e., encouraging the control policy to mimic human motion, facilitates the learning of complex tasks via reinforcement learning (RL) for humanoid robots. Although standard RL frameworks demonstrate impressive locomotion agility, they often bypass explicit reasoning about robot dynamics during deployment, which is a design choice that can lead to physically infeasible commands when the robot encounters out-of-distribution environments. By integrating model-based principles, hybrid approaches can improve performance; however, existing methods typically rely on predefined contact timing, limiting their versatility. This paper introduces HybridMimic, a framework in which a learned policy dynamically modulates a centroidal-model-based controller by predicting continuous contact states and desired centroidal velocities. This architecture exploits the physical grounding of centroidal dynamics to generate feedforward torques that remain feasible even under domain shift. Using physics-informed rewards, the policy is trained to efficiently utilize the centroidal controller's optimization by outputting precise control targets and reference torques. Through hardware experiments on the Booster T1 humanoid, HybridMimic reduces the average base position tracking error by 13\% compared to a state-of-the-art RL baseline, demonstrating the robustness of dynamics-aware deployment.
☆ Stability-Guided Exploration for Diverse Motion Generation
Scaling up datasets is highly effective in improving the performance of deep learning models, including in the field of robot learning. However, data collection still proves to be a bottleneck. Approaches relying on collecting human demonstrations are labor-intensive and inherently limited: they tend to be narrow, task-specific, and fail to adequately explore the full space of feasible states. Synthetic data generation could remedy this, but current techniques mostly rely on local trajectory optimization and fail to find diverse solutions. In this work, we propose a novel method capable of finding diverse long-horizon manipulations through black-box simulation. We achieve this by combining an RRT-style search with sampling-based MPC, together with a novel sampling scheme that guides the exploration toward stable configurations. Specifically, we sample from a manifold of stable states while growing a search tree directly through simulation, without restricting the planner to purely stable motions. We demonstrate the method's ability to discover diverse manipulation strategies, including pushing, grasping, pivoting, throwing, and tool use, across different robot morphologies, without task-specific guidance.
☆ Gradient-based Nested Co-Design of Aerodynamic Shape and Control for Winged Robots
Designing aerial robots for specialized tasks, from perching to payload delivery, requires tailoring their aerodynamic shape to specific mission requirements. For tasks involving wide flight envelopes, the usual sequential process of first determining the shape and then the motion planner is likely to be suboptimal due to the inherent nonlinear interactions between them. This limitation has been motivating co-design research, which involves jointly optimizing the aerodynamic shape and the motion planner. In this paper, we present a general-purpose, gradient-based, nested co-design framework where the motion planner solves an optimal control problem and the aerodynamic forces used in the dynamics model are determined by a neural surrogate model. This enables us to model complex subsonic flow conditions encountered in aerial robotics and to overcome the limited applicability of existing co-design methods. These limitations stem from the simplifying assumptions they require for computational tractability to either the planner or the aerodynamics. We validate our method on two complex dynamic tasks for fixed-wing gliders: perching and a short landing. Our optimized designs improve task performance compared to an evolutionary baseline in a fraction of the computation time.
☆ Robotic Foundation Models for Industrial Control: A Comprehensive Survey and Readiness Assessment Framework
Robotic foundation models (RFMs) are emerging as a promising route towards flexible, instruction- and demonstration-driven robot control, however, a critical investigation of their industrial applicability is still lacking. This survey gives an extensive overview over the RFM-landscape and analyses, driven by concrete implications, how industrial domains and use cases shape the requirements of RFMs, with particular focus on collaborative robot platforms, heterogeneous sensing and actuation, edge-computing constraints, and safety-critical operation. We synthesise industrial deployment perspectives into eleven interdependent implications and operationalise them into an assessment framework comprising a catalogue of 149 concrete criteria, spanning both model capabilities and ecosystem requirements. Using this framework, we evaluate 324 manipulation-capable RFMs via 48,276 criterion-level decisions obtained via a conservative LLM-assisted evaluation pipeline, validated against expert judgements. The results indicate that industrial maturity is limited and uneven: even the highest-rated models satisfy only a fraction of criteria and typically exhibit narrow implication-specific peaks rather than integrated coverage. We conclude that progress towards industry-grade RFMs depends less on isolated benchmark successes than on systematic incorporation of safety, real-time feasibility, robust perception, interaction, and cost-effective system integration into auditable deployment stacks.
☆ Improved Constrained Generation by Bridging Pretrained Generative Models
Constrained generative modeling is fundamental to applications such as robotic control and autonomous driving, where models must respect physical laws and safety-critical constraints. In real-world settings, these constraints rarely take the form of simple linear inequalities, but instead complex feasible regions that resemble road maps or other structured spatial domains. We propose a constrained generation framework that generates samples directly within such feasible regions while preserving realism. Our method fine-tunes a pretrained generative model to enforce constraints while maintaining generative fidelity. Experimentally, our method exhibits characteristics distinct from existing fine-tuning and training-free constrained baselines, revealing a new compromise between constraint satisfaction and sampling quality.
♻ ☆ Whole-Body Model-Predictive Control of Legged Robots with MuJoCo ICRA 2026
We demonstrate the surprising real-world effectiveness of a very simple approach to whole-body model-predictive control (MPC) of quadruped and humanoid robots: the iterative LQR (iLQR) algorithm with MuJoCo dynamics and finite-difference approximated derivatives. Building upon the previous success of model-based behavior synthesis and control of locomotion and manipulation tasks with MuJoCo in simulation, we show that these policies can easily generalize to the real world with few sim-to-real considerations. Our baseline method achieves real-time whole-body MPC on a variety of hardware experiments, including dynamic quadruped locomotion, quadruped walking on two legs, and full-sized humanoid bipedal locomotion. We hope this easy-to-reproduce hardware baseline lowers the barrier to entry for real-world whole-body MPC research and contributes to accelerating research velocity in the community. Our code and experiment videos will be available online at:https://johnzhang3.github.io/mujoco_ilqr
comment: to appear at ICRA 2026
♻ ☆ CAPS: Context-Aware Priority Sampling for Enhanced Imitation Learning in Autonomous Driving ICRA 2026
In this paper, we introduce Context-Aware Priority Sampling (CAPS), a novel method designed to enhance data efficiency in learning-based autonomous driving systems. CAPS addresses the challenge of imbalanced datasets in imitation learning by leveraging Vector Quantized Variational Autoencoders (VQ-VAEs). In this way, we can get structured and interpretable data representations, which help to reveal meaningful patterns in the data. These patterns are used to group the data into clusters, with each sample being assigned a cluster ID. The cluster IDs are then used to re-balance the dataset, ensuring that rare yet valuable samples receive higher priority during training. We evaluate our method through closed-loop experiments in the CARLA simulator. The results on Bench2Drive scenarios demonstrate the effectiveness of CAPS in enhancing model generalization, with substantial improvements in both driving score and success rate.
comment: Accepted at IEEE International Conference on Robotics & Automation (ICRA 2026)
♻ ☆ ROScopter: A Multirotor Autopilot based on ROSflight 2.0
ROScopter is a lean multirotor autopilot built for researchers. ROScopter seeks to accelerate simulation and hardware testing of research code with an architecture that is both easy to understand and simple to modify. ROScopter is designed to interface with ROSflight 2.0 and runs entirely on an onboard flight computer, leveraging the features of ROS 2 to improve modularity. This work describes the architecture of ROScopter and how it can be used to test application code in both simulated and hardware environments. Hardware results of the default ROScopter behavior are presented, showing that ROScopter achieves similar performance to another state-of-the-art autopilot for basic waypoint-following maneuvers, but with a significantly reduced and more modular code-base.
comment: Submitted to the 2026 International Conference on Unmanned Aerial Systems
♻ ☆ ROSplane 2.0: A Fixed-Wing Autopilot for Research
Unmanned aerial vehicle (UAV) research requires the integration of cutting-edge technology into existing autopilot frameworks. This process can be arduous, requiring extensive resources, time, and detailed knowledge of the existing system. ROSplane is a lean, open-source fixed-wing autonomy stack built by researchers for researchers. It is designed to accelerate research by providing clearly defined interfaces with an easily modifiable framework. Built around ROS 2, ROSplane allows for rapid integration of low or high-level control, path planning, or estimation algorithms. A focus on lean, easily-understood code and extensive documentation lowers the barrier to entry for researchers. Recent developments to ROSplane improve its capacity to accelerate UAV research, including the transition from ROS 1 to ROS 2, enhanced estimation and control algorithms, increased modularity, and an improved aerodynamic modeling pipeline. This aerodynamic modeling pipeline significantly reduces the effort of transitioning from simulation to real-world testing without requiring costly system identification or computational fluid dynamics tools. ROSplane's architecture reduces the effort required to integrate new research tools and methods, expediting hardware experimentation.
comment: Submitted to the 2026 International Conference on Unmanned Aerial Systems
♻ ☆ ROSflight 2.0: Lean ROS 2-Based Autopilot for Unmanned Aerial Vehicles
ROSflight is a lean, open-source autopilot ecosystem for unmanned aerial vehicles (UAVs). Designed by researchers for researchers, it is built to lower the barrier to entry to UAV research and accelerate the transition from simulation to hardware experiments by maintaining a lean (not full-featured), well-documented, and modular codebase. This publication builds on previous treatments and describes significant additions to the architecture that improve the modularity and usability of ROSflight, including the transition from ROS 1 to ROS 2, supported hardware, low-level actuator mixing, and the simulation environment. We believe that these changes improve the usability of ROSflight and enable ROSflight to accelerate research in areas like advanced-air mobility. Hardware results are provided, showing that ROSflight is able to control a multirotor over a serial connection at 400 Hz while closing all control loops on the companion computer.
comment: Submitted to the 2026 International Conference on Unmanned Aerial Systems
♻ ☆ ROS-related Robotic Systems Development with V-model-based Application of MeROS Metamodel
Systems built on the Robot Operating System (ROS) are increasingly easy to assemble, yet hard to govern and reliably coordinate. Beyond the sheer number of subsystems involved, the difficulty stems from their diversity and interaction depth. In this paper, we use a compact heterogeneous robotic system (HeROS), combining mobile and manipulation capabilities, as a demonstration vehicle under dynamically changing tasks. Notably, all its subsystems are powered by ROS. The use of compatible interfaces and other ROS integration capabilities simplifies the construction of such systems. However, this only addresses part of the complexity: the semantic coherence and structural traceability are even more important for precise coordination and call for deliberate engineering methods. The Model-Based Systems Engineering (MBSE) discipline, which emerged from the experience of complexity management in large-scale engineering domains, offers the methodological foundations needed. Despite their strengths in complementary aspects of robotics systems engineering, the lack of a unified approach to integrate ROS and MBSE hinders the full potential of these tools. Motivated by the anticipated impact of such a synergy in robotics practice, we propose a structured methodology based on MeROS - a SysML metamodel created specifically to put the ROS-based systems into the focus of the MBSE workflow. As its methodological backbone, we adapt the well-known V-model to this context, illustrating how complex robotic systems can be designed with traceability and validation capabilities embedded into their lifecycle using practices familiar to engineering teams.
comment: 22 pages
♻ ☆ InsSo3D: Inertial Navigation System and 3D Sonar SLAM for turbid environment inspection
This paper presents InsSo3D, an accurate and efficient method for large-scale 3D Simultaneous Localisation and Mapping (SLAM) using a 3D Sonar and an Inertial Navigation System (INS). Unlike traditional sonar, which produces 2D images containing range and azimuth information but lacks elevation information, 3D Sonar produces a 3D point cloud, which therefore does not suffer from elevation ambiguity. We introduce a robust and modern SLAM framework adapted to the 3D Sonar data using INS as prior, detecting loop closure and performing pose graph optimisation. We evaluated InsSo3D performance inside a test tank with access to ground truth data and in an outdoor flooded quarry. Comparisons to reference trajectories and maps obtained from an underwater motion tracking system and visual Structure From Motion (SFM) demonstrate that InsSo3D efficiently corrects odometry drift. The average trajectory error is below 21cm during a 50-minute-long mission, producing a map of 10m by 20m with a 9cm average reconstruction error, enabling safe inspection of natural or artificial underwater structures even in murky water conditions.
♻ ☆ VISO: Robust Underwater Visual-Inertial-Sonar SLAM with Photometric Rendering for Dense 3D Reconstruction
Visual challenges in underwater environments significantly hinder the accuracy of vision-based localisation and the high-fidelity dense reconstruction. In this paper, we propose VISO, a robust underwater SLAM system that fuses a stereo camera, an inertial measurement unit (IMU), and a 3D sonar to achieve accurate 6-DoF localisation and enable efficient dense 3D reconstruction with high photometric fidelity. We introduce a coarse-to-fine online calibration approach for extrinsic parameters estimation between the 3D sonar and the camera. Additionally, a photometric rendering strategy is proposed for the 3D sonar point cloud to enrich the sonar map with visual information. Extensive experiments in a laboratory tank and an open lake demonstrate that VISO surpasses current state-of-the-art underwater and visual-based SLAM algorithms in terms of localisation robustness and accuracy, while also exhibiting real-time dense 3D reconstruction performance comparable to the offline dense mapping method.
♻ ☆ Taxonomy-aware Dynamic Motion Generation on Hyperbolic Manifolds ICRA
Human-like motion generation for robots often draws inspiration from biomechanical studies, which often categorize complex human motions into hierarchical taxonomies. While these taxonomies provide rich structural information about how movements relate to one another, this information is frequently overlooked in motion generation models, leading to a disconnect between the generated motions and their underlying hierarchical structure. This paper introduces the \ac{gphdm}, a novel approach that learns latent representations preserving both the hierarchical structure of motions and their temporal dynamics to ensure physical consistency. Our model achieves this by extending the dynamics prior of the Gaussian Process Dynamical Model (GPDM) to the hyperbolic manifold and integrating it with taxonomy-aware inductive biases. Building on this geometry- and taxonomy-aware frameworks, we propose three novel mechanisms for generating motions that are both taxonomically-structured and physically-consistent: two probabilistic recursive approaches and a method based on pullback-metric geodesics. Experiments on generating realistic motion sequences on the hand grasping taxonomy show that the proposed GPHDM faithfully encodes the underlying taxonomy and temporal dynamics, and it generates novel physically-consistent trajectories.
comment: Accepted for publication in IEEE Conference on Robotics and Automation (ICRA), 8 pages, 6 figures, 1 table
♻ ☆ Contact-Safe Reinforcement Learning with ProMP Reparameterization and Energy Awareness
Reinforcement learning (RL) approaches based on Markov Decision Processes (MDPs) are predominantly applied in the robot joint space, often relying on limited task-specific information and partial awareness of the 3D environment. In contrast, episodic RL has demonstrated advantages over traditional MDP-based methods in terms of trajectory consistency, task awareness, and overall performance in complex robotic tasks. Moreover, traditional step-wise and episodic RL methods often neglect the contact-rich information inherent in task-space manipulation, especially considering the contact-safety and robustness. In this work, contact-rich manipulation tasks are tackled using a task-space, energy-safe framework, where reliable and safe task-space trajectories are generated through the combination of Proximal Policy Optimization (PPO) and movement primitives. Furthermore, an energy-aware Cartesian Impedance Controller objective is incorporated within the proposed framework to ensure safe interactions between the robot and the environment. Our experimental results demonstrate that the proposed framework outperforms existing methods in handling tasks on various types of surfaces in 3D environments, achieving high success rates as well as smooth trajectories and energy-safe interactions.
comment: 8 pages
♻ ☆ FALCON: Future-Aware Learning with Contextual Object-Centric Pretraining for UAV Action Recognition
We introduce FALCON, a unified self-supervised video pretraining approach for UAV action recognition from raw RGB aerial footage, requiring no additional preprocessing at inference. UAV videos exhibit severe spatial imbalance: large, cluttered backgrounds dominate the field of view, causing reconstruction-based pretraining to waste capacity on uninformative regions and under-learn action-relevant human/object cues. FALCON addresses this by integrating object-aware masked autoencoding with object-centric dual-horizon future reconstruction. Using detections only during pretraining, we construct objectness priors that (i) enforce balanced token visibility during masking and (ii) concentrate reconstruction supervision on action-relevant regions, preventing learning from being dominated by background appearance. To promote temporal dynamics learning, we further reconstruct short- and long-horizon future content within an object-centric supervision region, injecting anticipatory temporal supervision that is robust to noisy aerial context. Across UAV benchmarks, FALCON improves top-1 accuracy by 2.9\% on NEC-Drone and 5.8\% on UAV-Human with a ViT-B backbone, while achieving 2$\times$--5$\times$ faster inference than supervised approaches that rely on heavy test-time augmentation.
♻ ☆ Decision-Driven Semantic Object Exploration for Legged Robots via Confidence-Calibrated Perception and Topological Subgoal Selection
Conventional navigation pipelines for legged robots remain largely geometry-centric, relying on dense SLAM representations that are fragile under rapid motion and offer limited support for semantic decision making in open-world exploration. In this work, we focus on decision-driven semantic object exploration, where the primary challenge is not map consistency but how noisy and heterogeneous semantic observations can be transformed into stable and executable exploration decisions. We propose a vision-based approach that explicitly addresses this problem through confidence-calibrated semantic evidence arbitration, a controlled-growth semantic topological memory, and a semantic utility-driven subgoal selection mechanism. These components enable the robot to accumulate task-relevant semantic knowledge over time and select exploration targets that balance semantic relevance, reliability, and reachability, without requiring dense geometric reconstruction. Extensive experiments in both simulation and real-world environments demonstrate that the proposed mechanisms consistently improve the quality of semantic decision inputs, subgoal selection accuracy, and overall exploration performance on legged robots.
♻ ☆ OmniDP: Beyond-FOV Large-Workspace Humanoid Manipulation with Omnidirectional 3D Perception
The deployment of humanoid robots for dexterous manipulation in unstructured environments remains challenging due to perceptual limitations that constrain the effective workspace. In scenarios where physical constraints prevent the robot from repositioning itself, maintaining omnidirectional awareness becomes far more critical than color or semantic information.While recent advances in visuomotor policy learning have improved manipulation capabilities, conventional RGB-D solutions suffer from narrow fields of view (FOV) and self-occlusion, requiring frequent base movements that introduce motion uncertainty and safety risks. Existing approaches to expanding perception, including active vision systems and third-view cameras, introduce mechanical complexity, calibration dependencies, and latency that hinder reliable real-time performance. In this work, We propose OmniDP, an end-to-end LiDAR-driven 3D visuomotor policy that enables robust manipulation in large workspaces. Our method processes panoramic point clouds through a Time-Aware Attention Pooling mechanism, efficiently encoding sparse 3D data while capturing temporal dependencies. This 360° perception allows the robot to interact with objects across wide areas without frequent repositioning. To support policy learning, we develop a whole-body teleoperation system for efficient data collection on full-body coordination. Extensive experiments in simulation and real-world environments show that OmniDP achieves robust performance in large-workspace and cluttered scenarios, outperforming baselines that rely on egocentric depth cameras.
comment: 8 pages, 6 figures
♻ ☆ AIM-SLAM: Dense Monocular SLAM via Adaptive and Informative Multi-View Keyframe Prioritization with Foundation Model
Recent advances in geometric foundation models have emerged as a promising alternative for addressing the challenge of dense reconstruction in monocular visual simultaneous localization and mapping (SLAM). Although geometric foundation models enable SLAM to leverage variable input views, the previous methods remain confined to two-view pairs or fixed-length inputs without sufficient deliberation of geometric context for view selection. To tackle this problem, we propose AIM-SLAM, a dense monocular SLAM framework that exploits an adaptive and informative multi-view keyframe prioritization with dense pointmap predictions from visual geometry grounded transformer (VGGT). Specifically, we introduce the selective information- and geometric-aware multi-view adaptation (SIGMA) module, which employs voxel overlap and information gain to retrieve a candidate set of keyframes and adaptively determine its size. Furthermore, we formulate a joint multi-view Sim(3) optimization that enforces consistent alignment across selected views, substantially improving pose estimation accuracy. The effectiveness of AIM-SLAM is demonstrated on real-world datasets, where it achieves state-of-the-art performance in both pose estimation and dense reconstruction. Our system supports ROS integration, with code is available at https://aimslam.github.io/.
comment: 8 pages
♻ ☆ C*: A Coverage Path Planning Algorithm for Unknown Environments using Rapidly Covering Graphs
The paper presents a novel sample-based algorithm, called C*, for real-time coverage path planning (CPP) of unknown environments. C* is built upon the concept of a Rapidly Covering Graph (RCG), which is incrementally constructed during robot navigation via progressive sampling of the search space. By using efficient sampling and pruning techniques, the RCG is constructed to be a minimum-sufficient graph, where its nodes and edges form the potential waypoints and segments of the coverage trajectory, respectively. The RCG tracks the coverage progress, generates the coverage trajectory and helps the robot to escape from the dead-end situations. To minimize coverage time, C* produces the desired back-and-forth coverage pattern, while adapting to the TSP-based optimal coverage of local isolated regions, called coverage holes, which are surrounded by obstacles and covered regions. It is analytically proven that C* provides complete coverage of unknown environments. The algorithmic simplicity and low computational complexity of C* make it easy to implement and suitable for real-time on-board applications. The performance of C* is validated by 1) extensive high-fidelity simulations and 2) laboratory experiments using an autonomous robot. C* yields near optimal trajectories, and a comparative evaluation with seven existing CPP methods demonstrates significant improvements in performance in terms of coverage time, number of turns, trajectory length, and overlap ratio, while preventing the formation of coverage holes. Finally, C* is comparatively evaluated on two different CPP applications using 1) energy-constrained robots and 2) multi-robot teams.
♻ ☆ Robustness-Aware Tool Selection and Manipulation Planning with Learned Energy-Informed Guidance ICRA
Humans subconsciously choose robust ways of selecting and using tools, for example, choosing a ladle over a flat spatula to serve meatballs. However, robustness under external disturbances remains underexplored in robotic tool-use planning. This paper presents a robustness-aware method that jointly selects tools and plans contact-rich manipulation trajectories, explicitly optimizing for robustness against disturbances. At the core of our method is an energy-based robustness metric that guides the planner toward robust manipulation behaviors. We formulate a hierarchical optimization pipeline that first identifies a tool and configuration that optimizes robustness, and then plans a corresponding manipulation trajectory that maintains robustness throughout execution. We evaluate our method across three representative tool-use tasks. Simulation and real-world results demonstrate that our method consistently selects robust tools and generates disturbance-resilient manipulation plans.
comment: IEEE International Conference on Robotics and Automation (ICRA), 2026
♻ ☆ Safe Autonomous Lane Changing: Planning with Dynamic Risk Fields and Time-Varying Convex Space Generation
This paper presents a novel trajectory planning pipeline for complex driving scenarios like autonomous lane changing, by integrating risk-aware planning with guaranteed collision avoidance into a unified optimization framework. We first construct a dynamic risk fields (DRF) that captures both the static and dynamic collision risks from surrounding vehicles. Then, we develop a rigorous strategy for generating time-varying convex feasible spaces that ensure kinematic feasibility and safety requirements. The trajectory planning problem is formulated as a finite-horizon optimal control problem and solved using a constrained iterative Linear Quadratic Regulator (iLQR) algorithm that jointly optimizes trajectory smoothness, control effort, and risk exposure while maintaining strict feasibility. Extensive simulations demonstrate that our method outperforms traditional approaches in terms of safety and efficiency, achieving collision-free trajectories with shorter lane-changing distances (28.59 m) and times (2.84 s) while maintaining smooth and comfortable acceleration patterns. In dense roundabout environments the planner further demonstrates robust adaptability, producing larger safety margins, lower jerk, and superior curvature smoothness compared with APF, MPC, and RRT based baselines. These results confirm that the integrated DRF with convex feasible space and constrained iLQR solver provides a balanced solution for safe, efficient, and comfortable trajectory generation in dynamic and interactive traffic scenarios.
♻ ☆ Safe Model Predictive Diffusion with Shielding ICRA
Generating safe, kinodynamically feasible, and optimal trajectories for complex robotic systems is a central challenge in robotics. This paper presents Safe Model Predictive Diffusion (Safe MPD), a training-free diffusion planner that unifies a model-based diffusion framework with a safety shield to generate trajectories that are both kinodynamically feasible and safe by construction. By enforcing feasibility and safety on all samples during the denoising process, our method avoids the common pitfalls of post-processing corrections, such as computational intractability and loss of feasibility. We validate our approach on challenging non-convex planning problems, including kinematic and acceleration-controlled tractor-trailer systems. The results show that it substantially outperforms existing safety strategies in success rate and safety, while achieving sub-second computation times.
comment: 2026 IEEE International Conference on Robotics and Automation (ICRA). Project page: https://www.taekyung.me/safe-mpd
♻ ☆ (MGS)$^2$-Net: Unifying Micro-Geometric Scale and Macro-Geometric Structure for Cross-View Geo-Localization
Cross-view geo-localization (CVGL) is pivotal for GNSS-denied UAV navigation but remains brittle under the drastic geometric misalignment between oblique aerial views and orthographic satellite references. Existing methods predominantly operate within a 2D manifold, neglecting the underlying 3D geometry where view-dependent vertical facades (macro-structure) and scale variations (micro-scale) severely corrupt feature alignment. To bridge this gap, we propose (MGS)$^2$, a geometry-grounded framework. The core of our innovation is the Macro-Geometric Structure Filtering (MGSF) module. Unlike pixel-wise matching sensitive to noise, MGSF leverages dilated geometric gradients to physically filter out high-frequency facade artifacts while enhancing the view-invariant horizontal plane, directly addressing the domain shift. To guarantee robust input for this structural filtering, we explicitly incorporate a Micro-Geometric Scale Adaptation (MGSA) module. MGSA utilizes depth priors to dynamically rectify scale discrepancies via multi-branch feature fusion. Furthermore, a Geometric-Appearance Contrastive Distillation (GACD) loss is designed to strictly discriminate against oblique occlusions. Extensive experiments demonstrate that (MGS)$^2$ achieves state-of-the-art performance, recording a Recall@1 of 97.5\% on University-1652 and 97.02\% on SUES-200. Furthermore, the framework exhibits superior cross-dataset generalization against geometric ambiguity. The code is available at: \href{https://github.com/GabrielLi1473/MGS-Net}{https://github.com/GabrielLi1473/MGS-Net}.
♻ ☆ FindAnything: Open-Vocabulary and Object-Centric Mapping for Robot Exploration in Any Environment
Geometrically accurate and semantically expressive map representations have proven invaluable for robot deployment and task planning in unknown environments. Nevertheless, real-time, open-vocabulary semantic understanding of large-scale unknown environments still presents open challenges, mainly due to computational requirements. In this paper we present FindAnything, an open-world mapping framework that incorporates vision-language information into dense volumetric submaps. Thanks to the use of vision-language features, FindAnything combines pure geometric and open-vocabulary semantic information for a higher level of understanding. It proposes an efficient storage of open-vocabulary information through the aggregation of features at the object level. Pixelwise vision-language features are aggregated based on eSAM segments, which are in turn integrated into object-centric volumetric submaps, providing a mapping from open-vocabulary queries to 3D geometry that is scalable also in terms of memory usage. We demonstrate that FindAnything performs on par with the state-of-the-art in terms of semantic accuracy while being substantially faster and more memory-efficient, allowing its deployment in large-scale environments and on resourceconstrained devices, such as MAVs. We show that the real-time capabilities of FindAnything make it useful for downstream tasks, such as autonomous MAV exploration in a simulated Search and Rescue scenario. Project Page: https://ethz-mrl.github.io/findanything/.
comment: 11 pages, 5 figures
♻ ☆ Bridging Simulation and Usability: A User-Friendly Framework for Scenario Generation in CARLA
Autonomous driving promises safer roads, reduced congestion, and improved mobility, yet validating these systems across diverse conditions remains a major challenge. Real-world testing is expensive, time-consuming, and sometimes unsafe, making large-scale validation impractical. In contrast, simulation environments offer a scalable and cost-effective alternative for rigorous verification and validation. A critical component of the validation process is scenario generation, which involves designing and configuring traffic scenarios to evaluate autonomous systems' responses to various events and uncertainties. However, existing scenario generation tools often require programming knowledge, limiting accessibility for non-technical users. To address this limitation, we present an interactive, no-code framework for scenario generation. Our framework features a graphical interface that enables users to create, modify, save, load, and execute scenarios without needing coding expertise or detailed simulation knowledge. Unlike script-based tools such as Scenic or ScenarioRunner, our approach lowers the barrier to entry and supports a broader user base. Central to our framework is a graph-based scenario representation that facilitates structured management, supports both manual and automated generation, and enables integration with deep learning-based scenario and behavior generation methods. In automated mode, the framework can randomly sample parameters such as actor types, behaviors, and environmental conditions, allowing the generation of diverse and realistic test datasets. By simplifying the scenario generation process, this framework supports more efficient testing workflows and increases the accessibility of simulation-based validation for researchers, engineers, and policymakers.
comment: Paper is accepted in IEEE International Automated Vehicle Validation Conference (IAVVC 2025)
♻ ☆ Diverse and Adaptive Behavior Curriculum for Autonomous Driving: A Student-Teacher Framework with Multi-Agent RL IROS 2025
Autonomous driving faces challenges in navigating complex real-world traffic, requiring safe handling of both common and critical scenarios. Reinforcement learning (RL), a prominent method in end-to-end driving, enables agents to learn through trial and error in simulation. However, RL training often relies on rule-based traffic scenarios, limiting generalization. Additionally, current scenario generation methods focus heavily on critical scenarios, neglecting a balance with routine driving behaviors. Curriculum learning, which progressively trains agents on increasingly complex tasks, is a promising approach to improving the robustness and coverage of RL driving policies. However, existing research mainly emphasizes manually designed curricula, focusing on scenery and actor placement rather than traffic behavior dynamics. This work introduces a novel student-teacher framework for automatic curriculum learning. The teacher, a graph-based multi-agent RL component, adaptively generates traffic behaviors across diverse difficulty levels. An adaptive mechanism adjusts task difficulty based on student performance, ensuring exposure to behaviors ranging from common to critical. The student, though exchangeable, is realized as a deep RL agent with partial observability, reflecting real-world perception constraints. Results demonstrate the teacher's ability to generate diverse traffic behaviors. The student, trained with automatic curricula, outperformed agents trained on rule-based traffic, achieving higher rewards and exhibiting balanced, assertive driving.
comment: First and Second authors contributed equally; Paper accepted in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025)
♻ ☆ VEGA: Electric Vehicle Navigation Agent via Physics-Informed Neural Operator and Proximal Policy Optimization IROS
We present VEGA, a vehicle-adaptive energy-aware routing system for electric vehicles (EVs) that integrates physics-informed parameter estimation with RL-based charge-aware path planning. VEGA consists of two copupled modules: (1) a physics-informed neural operator (PINO) that estimates vehicle-specific physical parameters-drag, rolling resistance, mass, motor and regenerative-braking efficiencies, and auxiliary load-from short windows of onboard speed and acceleration data; (2) a Proximal Policy Optimization (PPO) agent that navigates a charger-annotated road graph, jointly selecting routes and charging stops under state-of-charge constraints. The agent is initialized via behavior cloning from an A* teacher and fine-tuned with cirriculum-guided PPO on the full U.S. highway network with Tesla Supercharger locations. On a cross-country San Francisco-to-New York route (~4,860km), VEGA produces a feasible 20-stop plan with 56.12h total trip time and minimum SoC 11.41%. Against the controlled Energy-aware A* baseline, the distance and driving-time gaps are small (-8.49km and +0.37h), while inference is >20x faster. The learned policy generalizes without retraining to road networks in France and Japan.
comment: This work has been submitted to the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) for possible publication
♻ ☆ Graph-based Online Lidar Odometry with Retrospective Map Refinement
Lidar-only odometry aims to estimate the trajectory of a mobile platform from a stream of lidar scans. Traditional scan-to map approaches register each scan against a single, evolving map, which propagates registration errors over time. To mitigate this, we propose a multitude-of-maps approach where the current scan is registered against multiple overlapping submaps instead of a single static map. By optimizing the resulting constraints in a pose graph, our method enables not only precise estimation of the current pose but also retrospective refinement of the submaps' anchor points, which improves short-term consistency and long-term accuracy. We demonstrate that our approach achieves competitive and often superior accuracy on a variety of automotive datasets while maintaining real-time performance. Ablation studies confirm the critical role of multiple registrations and retrospective refinement of the map as core factors for our accuracy gains. Code and raw results are available on our public GitHub at https://github.com/Fusion-Goettingen/IROS_2026_Kurda_Graph.
♻ ☆ Beyond Imitation: Reinforcement Learning-Based Sim-Real Co-Training for VLA Models
Simulation offers a scalable and low-cost way to enrich vision-language-action (VLA) training, reducing reliance on expensive real-robot demonstrations. However, most sim-real co-training methods rely on supervised fine-tuning (SFT), which treats simulation as a static source of demonstrations and does not exploit large-scale closed-loop interaction. Consequently, real-world gains and generalization are often limited. In this paper, we propose an \underline{\textit{RL}}-based sim-real \underline{\textit{Co}}-training \modify{(RL-Co)} framework that leverages interactive simulation while preserving real-world capabilities. Our method follows a generic two-stage design: we first warm-start the policy with SFT on a mixture of real and simulated demonstrations, then fine-tune it with reinforcement learning in simulation while adding an auxiliary supervised loss on real-world data to anchor the policy and mitigate catastrophic forgetting. We evaluate our framework on four real-world tabletop manipulation tasks using two representative VLA architectures, OpenVLA and $π_{0.5}$, and observe consistent improvements over real-only fine-tuning and SFT-based co-training, including +24% real-world success on OpenVLA and +20% on $π_{0.5}$. Beyond higher success rates, RL co-training yields stronger generalization to unseen task variations and substantially improved real-world data efficiency, providing a practical and scalable pathway for leveraging simulation to enhance real-robot deployment.
♻ ☆ Language Conditioning Improves Accuracy of Aircraft Goal Prediction in Non-Towered Airspace
Autonomous aircraft must safely operate in non-towered airspace, where coordination relies on voice-based communication among human pilots. Safe operation requires an aircraft to predict the intent, and corresponding goal location, of other aircraft. This paper introduces a multimodal framework for aircraft goal prediction that integrates natural language understanding with spatial reasoning to improve autonomous decision-making in such environments. We leverage automatic speech recognition and large language models to transcribe and interpret pilot radio calls, identify aircraft, and extract discrete intent labels. These intent labels are fused with observed trajectories to condition a temporal convolutional network and Gaussian mixture model for probabilistic goal prediction. Our method significantly reduces goal prediction error compared to baselines that rely solely on motion history, demonstrating that language-conditioned prediction increases prediction accuracy. Experiments on a real-world dataset from a non-towered airport validate the approach and highlight its potential to enable socially aware, language-conditioned robotic motion planning.
comment: The last two authors advised equally. Accepted to the 2026 IEEE International Conference on Robotics and Automation. 8 pages, 6 figures
♻ ☆ RAG-Driver: Generalisable Driving Explanations with Retrieval-Augmented In-Context Learning in Multi-Modal Large Language Model
We need to trust robots that use often opaque AI methods. They need to explain themselves to us, and we need to trust their explanation. In this regard, explainability plays a critical role in trustworthy autonomous decision-making to foster transparency and acceptance among end users, especially in complex autonomous driving. Recent advancements in Multi-Modal Large Language models (MLLMs) have shown promising potential in enhancing the explainability as a driving agent by producing control predictions along with natural language explanations. However, severe data scarcity due to expensive annotation costs and significant domain gaps between different datasets makes the development of a robust and generalisable system an extremely challenging task. Moreover, the prohibitively expensive training requirements of MLLM and the unsolved problem of catastrophic forgetting further limit their generalisability post-deployment. To address these challenges, we present RAG-Driver, a novel retrieval-augmented multi-modal large language model that leverages in-context learning for high-performance, explainable, and generalisable autonomous driving. By grounding in retrieved expert demonstration, we empirically validate that RAG-Driver achieves state-of-the-art performance in producing driving action explanations, justifications, and control signal prediction. More importantly, it exhibits exceptional zero-shot generalisation capabilities to unseen environments without further training endeavours.
comment: 14 pages, 6 figures
♻ ☆ RoboPocket: Improve Robot Policies Instantly with Your Phone
Scaling imitation learning is fundamentally constrained by the efficiency of data collection. While handheld interfaces have emerged as a scalable solution for in-the-wild data acquisition, they predominantly operate in an open-loop manner: operators blindly collect demonstrations without knowing the underlying policy's weaknesses, leading to inefficient coverage of critical state distributions. Conversely, interactive methods like DAgger effectively address covariate shift but rely on physical robot execution, which is costly and difficult to scale. To reconcile this trade-off, we introduce RoboPocket, a portable system that enables Robot-Free Instant Policy Iteration using single consumer smartphones. Its core innovation is a Remote Inference framework that visualizes the policy's predicted trajectory via Augmented Reality (AR) Visual Foresight. This immersive feedback allows collectors to proactively identify potential failures and focus data collection on the policy's weak regions without requiring a physical robot. Furthermore, we implement an asynchronous Online Finetuning pipeline that continuously updates the policy with incoming data, effectively closing the learning loop in minutes. Extensive experiments demonstrate that RoboPocket adheres to data scaling laws and doubles the data efficiency compared to offline scaling strategies, overcoming their long-standing efficiency bottleneck. Moreover, our instant iteration loop also boosts sample efficiency by up to 2$\times$ in distributed environments a small number of interactive corrections per person. Project page and videos: https://robo-pocket.github.io.
comment: Project page: https://robo-pocket.github.io
♻ ☆ Symmetry-Breaking in Multi-Agent Navigation: Winding Number-Aware MPC with a Learned Topological Strategy
In distributed multi-agent navigation without explicit communication, agents can fall into symmetry-induced deadlocks because each agent must autonomously decide how to pass others. To address this problem, we propose WNumMPC, a hierarchical navigation method that quantifies cooperative symmetry-breaking strategies via a topological invariant, the winding number, and learns such strategies through reinforcement learning. The learning-based Planner outputs continuous-valued signed target winding numbers and dynamic importance weights to prioritize critical interactions in dense crossings. Then, the model-based Controller generates collision-free and efficient motions based on the strategy and weights provided by the Planner. Simulation and real-world robot experiments indicate that WNumMPC effectively avoids deadlocks and collisions and achieves better performance than the baselines, particularly in dense and symmetry-prone scenarios. These experiments also suggest that explicitly leveraging winding numbers yields robust sim-to-real transfer with minimal performance degradation. The code for the experiments is available at https://github.com/omron-sinicx/WNumMPC.
comment: 12 pages, 7 figures
♻ ☆ APEX: Learning Adaptive High-Platform Traversal for Humanoid Robots
Humanoid locomotion has advanced rapidly with deep reinforcement learning (DRL), enabling robust feet-based traversal over uneven terrain. Yet platforms beyond leg length remain largely out of reach because current RL training paradigms often converge to jumping-like solutions that are high-impact, torque-limited, and unsafe for real-world deployment. To address this gap, we propose APEX, a system for perceptive, climbing-based high-platform traversal that composes terrain-conditioned behaviors: climb-up and climb-down at vertical edges, walking or crawling on the platform, and stand-up and lie-down for posture reconfiguration. Central to our approach is a generalized ratchet progress reward for learning contact-rich, goal-reaching maneuvers. It tracks the best-so-far task progress and penalizes non-improving steps, providing dense yet velocity-free supervision that enables efficient exploration under strong safety regularization. Based on this formulation, we train LiDAR-based full-body maneuver policies and reduce the sim-to-real perception gap through a dual strategy: modeling mapping artifacts during training and applying filtering and inpainting to elevation maps during deployment. Finally, we distill all six skills into a single policy that autonomously selects behaviors and transitions based on local geometry and commands. Experiments on a 29-DoF Unitree G1 humanoid demonstrate zero-shot sim-to-real traversal of 0.8 meter platforms (approximately 114% of leg length), with robust adaptation to platform height and initial pose, as well as smooth and stable multi-skill transitions.
comment: Project Website: https://apex-humanoid.github.io/
♻ ☆ Phys4D: Fine-Grained Physics-Consistent 4D Modeling from Video Diffusion
Recent video diffusion models have achieved impressive capabilities as large-scale generative world models. However, these models often struggle with fine-grained physical consistency, exhibiting physically implausible dynamics over time. In this work, we present \textbf{Phys4D}, a pipeline for learning physics-consistent 4D world representations from video diffusion models. Phys4D adopts \textbf{a three-stage training paradigm} that progressively lifts appearance-driven video diffusion models into physics-consistent 4D world representations. We first bootstrap robust geometry and motion representations through large-scale pseudo-supervised pretraining, establishing a foundation for 4D scene modeling. We then perform physics-grounded supervised fine-tuning using simulation-generated data, enforcing temporally consistent 4D dynamics. Finally, we apply simulation-grounded reinforcement learning to correct residual physical violations that are difficult to capture through explicit supervision. To evaluate fine-grained physical consistency beyond appearance-based metrics, we introduce a set of \textbf{4D world consistency evaluation} that probe geometric coherence, motion stability, and long-horizon physical plausibility. Experimental results demonstrate that Phys4D substantially improves fine-grained spatiotemporal and physical consistency compared to appearance-driven baselines, while maintaining strong generative performance. Our project page is available at https://sensational-brioche-7657e7.netlify.app/
♻ ☆ Bi-AQUA: Bilateral Control-Based Imitation Learning for Underwater Robot Arms via Lighting-Aware Action Chunking with Transformers
Underwater robotic manipulation remains challenging because lighting variation, color attenuation, scattering, and reduced visibility can severely degrade visuomotor policies. We present Bi-AQUA, the first underwater bilateral control-based imitation learning framework for robot arms that explicitly models lighting within the policy. Bi-AQUA integrates transformer-based bilateral action chunking with a hierarchical lighting-aware design composed of a label-free Lighting Encoder, FiLM-based visual feature modulation, and a lighting token for action conditioning. This design enables adaptation to static and dynamically changing underwater illumination while preserving the force-sensitive advantages of bilateral control, which are particularly important in long-horizon and contact-rich manipulation. Real-world experiments on underwater pick-and-place, drawer closing, and peg extraction tasks show that Bi-AQUA outperforms a bilateral baseline without lighting modeling and achieves robust performance under seen, unseen, and changing lighting conditions. These results highlight the importance of combining explicit lighting modeling with force-aware bilateral imitation learning for reliable underwater manipulation. For additional material, please check: https://mertcookimg.github.io/bi-aqua
♻ ☆ Safe-SAGE: Social-Semantic Adaptive Guidance for Safe Engagement through Laplace-Modulated Poisson Safety Functions
Traditional safety-critical control methods, such as control barrier functions, suffer from semantic blindness, exhibiting the same behavior around obstacles regardless of contextual significance. This limitation leads to the uniform treatment of all obstacles, despite their differing semantic meanings. We present Safe-SAGE (Social-Semantic Adaptive Guidance for Safe Engagement), a unified framework that bridges the gap between high-level semantic understanding and low-level safety-critical control through a Poisson safety function (PSF) modulated using a Laplace guidance field. Our approach perceives the environment by fusing multi-sensor point clouds with vision-based instance segmentation and persistent object tracking to maintain up-to-date semantics beyond the camera's field of view. A multi-layer safety filter is then used to modulate system inputs to achieve safe navigation using this semantic understanding of the environment. This safety filter consists of both a model predictive control layer and a control barrier function layer. Both layers utilize the PSF and flux modulation of the guidance field to introduce varying levels of conservatism and multi-agent passing norms for different obstacles in the environment. Our framework enables legged robots to safely navigate semantically rich, dynamic environments with context-dependent safety margins.
comment: 8 pages
♻ ☆ Sample-Based Hybrid Mode Control: Asymptotically Optimal Switching of Algorithmic and Non-Differentiable Control Modes
This paper investigates a sample-based solution to the hybrid mode control problem across non-differentiable and algorithmic hybrid modes. Our approach reasons about a set of hybrid control modes as an integer-based optimization problem where we select what mode to apply, when to switch to another mode, and the duration for which we are in a given control mode. A sample-based variation is derived to efficiently search the integer domain for optimal solutions. We find our formulation yields strong performance guarantees that can be applied to a number of robotics-related tasks. In addition, our approach is able to synthesize complex algorithms and policies to compound behaviors and achieve challenging tasks. Last, we demonstrate the effectiveness of our approach in real-world robotic examples that require reactive switching between long-term planning and high-frequency control.
♻ ☆ Learning Robust Control Policies for Inverted Pose on Miniature Blimp Robots ICRA 2026
The ability to achieve and maintain inverted poses is essential for unlocking the full agility of miniature blimp robots (MBRs). However, developing reliable inverted control strategies for MBRs remains challenging due to their complex and underactuated dynamics. To address this challenge, we propose a novel framework that enables robust control policy learning for inverted pose on MBRs. The proposed framework consists of three core stages. First, a high-fidelity three-dimensional (3D) simulation environment is constructed and calibrated using real-world MBR motion data. Second, a robust inverted control policy is trained in simulation using a modified Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm combined with a domain randomization strategy. Third, a mapping layer is designed to bridge the sim-to-real gap and facilitate real-world deployment of the learned policy. Comprehensive evaluations in the simulation environment demonstrate that the learned policy achieves a higher success rate compared to the energy-shaping controller. Furthermore, experimental results confirm that the learned policy with a mapping layer enables an MBR to achieve and maintain a fully inverted pose in real-world settings.
comment: Accepted in ICRA 2026
♻ ☆ EchoVLA: Synergistic Declarative Memory for VLA-Driven Mobile Manipulation
Recent progress in Vision-Language-Action (VLA) models has enabled embodied agents to interpret multimodal instructions and perform complex tasks. However, existing VLAs are mostly confined to short-horizon, table-top manipulation, lacking the memory and reasoning capability required for mobile manipulation, where agents must coordinate navigation and manipulation under changing spatial contexts. In this work, we present EchoVLA, a memory-aware VLA model for mobile manipulation. EchoVLA incorporates a synergistic declarative memory inspired by the human brain, consisting of a scene memory that maintains a collection of spatial-semantic maps and an episodic memory that stores task-level experiences with multimodal contextual features. The two memories are individually stored, updated, and retrieved based on current observations, task history, and instructions, and their retrieved representations are fused via coarse- and fine-grained attention to guide base-arm diffusion policies. To support large-scale training, we further introduce MoMani, an automated benchmark that generates expert-level trajectories through multimodal large language model (MLLM)-guided planning and feedback-driven refinement, supplemented with real-robot demonstrations. Comprehensive simulated and real-world results demonstrate that EchoVLA substantially improves overall performance, e.g., it achieves the highest success rates of 0.52 on manipulation/navigation tasks and 0.31 on mobile manipulation tasks in simulation, exceeding the strong baseline $π_{0.5}$ by +0.20 and +0.11, respectively.
♻ ☆ AURASeg: Attention-guided Upsampling with Residual-Assistive Boundary Refinement for Onboard Robot Drivable-Area Segmentation
Free space ground segmentation is essential to navigate autonomous robots, recognize drivable zones, and traverse efficiently. Fine-grained features remain challenging for existing segmentation models, particularly for robots in indoor, outdoor and road-scene environments. These difficulties arise from ineffective multi-scale processing, sub-optimal boundary refinement, and limited feature representation. To address this, we propose Attention-guided Upsampling with Residual-Assistive Boundary Refinement (AURASeg), a ground-plane drivable area segmentation framework designed to improve boundary precision while preserving strong region accuracy under edge-deployment constraints. Built on ResNet backbone, we propose (i) a Residual Boundary Refinement Module (RBRM) that enhances edge delineation through boundary-assistive feature refinement, and (ii) Attention Progressive Upsampling Decoder (APUD) blocks that fuse multi-level features using residual fusion of attention modules; additionally, we integrate (iii) a lightweight ASPPLite module to capture multi-scale context with minimal overhead. Extensive experiments on CARL-D, the Ground Mobile Robot Perception (GMRPD) dataset, and a custom Gazebo indoor dataset show that AURASeg consistently outperforms strong baselines, with notable gains in boundary metrics. Finally, we demonstrate on-device deployment on a Jetson Nano powered Kobuki TurtleBot, validating practical edge-inference feasibility. Code is omitted for anonymity and will be released upon acceptance.
comment: 6 pages, 4 figures, 4 tables
♻ ☆ Integrated Hierarchical Decision-Making in Inverse Kinematic Planning and Control
This work presents a novel and efficient non-linear programming framework that tightly integrates hierarchical decision-making with inverse kinematic planning and control. Decision-making plays a central role in many aspects of robotics, from sparse inverse kinematic control with a minimal number of joints, to inverse kinematic planning while simultaneously selecting a discrete end-effector location from multiple candidates. Current approaches often rely on heavy computations using mixed-integer non-linear programming, separate decision-making from inverse kinematics (some times approximated by reachability methods), or employ efficient but less accurate $\ell_1$-norm formulations of linear sparse programming, without addressing the underlying non-linear problem formulations. In contrast, the proposed sparse hierarchical non-linear programming solver is efficient, versatile, and accurate by exploiting sparse hierarchical structure and leveraging the rarely used $\ell_0$-norm in robotics. The solver efficiently addresses complex non-linear hierarchical decision-making problems, such as inverse kinematic planning with simultaneous prioritized selection of end-effector locations from a large set of candidates, or inverse kinematic control with simultaneous selection of bi-manual grasp locations on a randomly rotated box.
♻ ☆ Phys2Real: Fusing VLM Priors with Interactive Online Adaptation for Uncertainty-Aware Sim-to-Real Manipulation ICRA
Learning robotic manipulation policies directly in the real world can be expensive and time-consuming. While reinforcement learning (RL) policies trained in simulation present a scalable alternative, effective sim-to-real transfer remains challenging, particularly for tasks that require precise dynamics. To address this, we propose Phys2Real, a real-to-sim-to-real RL pipeline that combines vision-language model (VLM)-inferred physical parameter estimates with interactive adaptation through uncertainty-aware fusion. Our approach consists of three core components: (1) high-fidelity geometric reconstruction with 3D Gaussian splatting, (2) VLM-inferred prior distributions over physical parameters, and (3) online physical parameter estimation from interaction data. Phys2Real conditions policies on interpretable physical parameters, refining VLM predictions with online estimates via ensemble-based uncertainty quantification. On planar pushing tasks of a T-block with varying center of mass (CoM) and a hammer with an off-center mass distribution, Phys2Real achieves substantial improvements over a domain randomization baseline: 100% vs 79% success rate for the bottom-weighted T-block, 57% vs 23% in the challenging top-weighted T-block, and 15% faster average task completion for hammer pushing. Ablation studies indicate that the combination of VLM and interaction information is essential for success. Project website: https://phys2real.github.io/.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
♻ ☆ XR-DT: Extended Reality-Enhanced Digital Twin for Safe Motion Planning via Human-Aware Model Predictive Path Integral Control
As mobile robots increasingly operate alongside humans in shared workspaces, ensuring safe, efficient, and interpretable Human-Robot Interaction (HRI) has become a pressing challenge. While substantial progress has been devoted to human behavior prediction, limited attention has been paid to how humans perceive, interpret, and trust robots' inferences and how robots plan safe and efficient trajectories based on predicted human behaviors. To address these challenges, this paper presents XR-DT, an eXtended Reality-enhanced Digital Twin framework for mobile robots, which bridges physical and virtual spaces to enable bi-directional understanding between humans and robots. Our hierarchical XR-DT architecture integrates augmented-, virtual-, and mixed-reality layers, fusing real-time sensor data, simulated environments in the Unity game engine, and human feedback captured through wearable XR devices. Within this framework, we design a novel Human-Aware Model Predictive Path Integral (HA-MPPI) control model, an MPPI-based motion planner that incorporates ATLAS (Attention-based Trajectory Learning with Anticipatory Sensing), a multi-modal Transformer model designed for egocentric human trajectory prediction via XR headsets. Extensive real-world experimental results demonstrate accurate human trajectory prediction, and safe and efficient robot navigation, validating the HA-MPPI's effectiveness within the XR-DT framework. By embedding human behavior, environmental dynamics, and robot navigation into the XR-DT framework, our system enables interpretable, trustworthy, and adaptive HRI.
comment: 8 pages, 6 figures, 3 tables
♻ ☆ Real-Time Learning of Predictive Dynamic Obstacle Models for Robotic Motion Planning ICRA
Autonomous systems often must predict the motions of nearby agents from partial and noisy data. This paper asks and answers the question: "can we learn, in real-time, a nonlinear predictive model of another agent's motions?" Our online framework denoises and forecasts such dynamics using a modified sliding-window Hankel Dynamic Mode Decomposition (Hankel-DMD). Partial noisy measurements are embedded into a Hankel matrix, while an associated Page matrix enables singular-value hard thresholding (SVHT) to estimate the effective rank. A Cadzow projection enforces structured low-rank consistency, yielding a denoised trajectory and local noise variance estimates. From this representation, a time-varying Hankel-DMD lifted linear predictor is constructed for multi-step forecasts. The residual analysis provides variance-tracking signals that can support downstream estimators and risk-aware planning. We validate the approach in simulation under Gaussian and heavy-tailed noise, and experimentally on a dynamic crane testbed. Results show that the method achieves stable variance-aware denoising and short-horizon prediction suitable for integration into real-time control frameworks.
comment: 10 pages, 6 figures, submitted to IEEE International Conference on Robotics and Automation (ICRA) 2025
♻ ☆ Large-Language-Model-Guided State Estimation for Partially Observable Task and Motion Planning
Robot planning in partially observable environments, where not all objects are known or visible, is a challenging problem, as it requires reasoning under uncertainty through partially observable Markov decision processes. During the execution of a computed plan, a robot may unexpectedly observe task-irrelevant objects, which are typically ignored by naive planners. In this work, we propose incorporating two types of common-sense knowledge: (1) certain objects are more likely to be found in specific locations; and (2) similar objects are likely to be co-located, while dissimilar objects are less likely to be found together. Manually engineering such knowledge is complex, so we explore leveraging the powerful common-sense reasoning capabilities of large language models (LLMs). Our planning and execution framework, CoCo-TAMP, introduces a hierarchical state estimation that uses LLM-guided information to shape the belief over task-relevant objects, enabling efficient solutions to long-horizon task and motion planning problems. In experiments, CoCo-TAMP achieves an average reduction of 62.7% in planning and execution time in simulation, and 72.6% in real-world demonstrations, compared to a baseline that does not incorporate either type of common-sense knowledge.
♻ ☆ ExpReS-VLA: Specializing Vision-Language-Action Models Through Experience Replay and Retrieval ICRA
Vision-Language-Action (VLA) models like OpenVLA demonstrate impressive zero-shot generalization across robotic manipulation tasks but struggle to adapt to specific deployment environments where consistent high performance on a limited set of tasks is more valuable than broad generalization. We present EXPierence replayed, REtrieval augmented, Specialized VLA (ExpReS-VLA), a method that enables rapid on-device adaptation of pre-trained VLAs to target domains while preventing catastrophic forgetting through compressed experience replay and retrieval-augmented generation. Our approach maintains a memory-efficient buffer by storing extracted embeddings from OpenVLA's frozen vision backbone, reducing storage requirements by 97% compared to raw image-action pairs. During deployment, ExpReS-VLA retrieves the $k$ most similar past experiences using cosine similarity to augment training batches, while a prioritized experience replay buffer preserves recently successful trajectories. To leverage failed attempts, we introduce Thresholded Hybrid Contrastive Loss (THCL), enabling the model to learn from both successful and unsuccessful demonstrations. Experiments on the LIBERO benchmark show improvements from 82.6% to 93.1% on spatial reasoning and 61% to 72.3% on long-horizon tasks over base OpenVLA, with gains across architectures including $π_0$ (+3.2 points) and OpenVLA-OFT (+1.7 points). Physical robot experiments across five tasks demonstrate 98% success on both in-distribution and out-of-distribution conditions, improving from 84.7% and 32% respectively for naive fine-tuning. Adaptation completes in 31 seconds using 12 demonstrations on a single RTX 5090.
comment: 8 pages, 4 figures, 3 tables, accepted to International Conference on Robotics and Automation (ICRA) 2026
♻ ☆ Assigning Multi-Robot Tasks to Multitasking Robots
One simplifying assumption in existing and well-performing task allocation methods is that the robots are single-tasking: each robot operates on a single task at any given time. While this assumption is harmless to make in some situations, it can be inefficient or even infeasible in others. In this paper, we consider assigning multi-robot tasks to multitasking robots. The key contribution is a novel task allocation framework that incorporates the consideration of physical constraints introduced by multitasking. This is in contrast to the existing work where such constraints are largely ignored. After formulating the problem, we propose a compilation to weighted MAX-SAT, which allows us to leverage existing solvers for a solution. A more efficient greedy heuristic is then introduced. For evaluation, we first compare our methods with a modern baseline that is efficient for single-tasking robots to validate the benefits of multitasking in synthetic domains. Then, using a site-clearing scenario in simulation, we further illustrate the complex task interaction considered by the multitasking robots in our approach to demonstrate its performance. Finally, we demonstrate a physical experiment to show how multitasking enabled by our approach can benefit task efficiency in a realistic setting.
♻ ☆ Preference-Conditioned Multi-Objective RL for Integrated Command Tracking and Force Compliance in Humanoid Locomotion
Humanoid locomotion requires not only accurate command tracking for navigation but also compliant responses to external forces during human interaction. Despite significant progress, existing RL approaches mainly emphasize robustness, yielding policies that resist external forces but lack compliance particularly challenging for inherently unstable humanoids. In this work, we address this by formulating humanoid locomotion as a multi-objective optimization problem that balances command tracking and external force compliance. We introduce a preference-conditioned multi-objective RL (MORL) framework that enables a single omnidirectional locomotion policy to trade off between command following and force compliance via a user-specified preference input. External forces are modeled via velocity-resistance factor for consistent reward design, and training leverages an encoder-decoder structure that infers task-relevant privileged features from deployable observations. We validate our approach in both simulation and real-world experiments on a humanoid robot. Experimental results in simulation and on hardware show that the framework trains stably and enables deployable preference-conditioned humanoid locomotion.
♻ ☆ Diffusion-SAFE: Diffusion-Native Human-to-Robot Driving Handover for Shared Autonomy
Shared autonomy in driving requires anticipating human behavior, flagging risk before it becomes unavoidable, and transferring control safely and smoothly. We propose Diffusion-SAFE, a closed-loop framework built on two diffusion models: an evaluator that predicts multimodal human-intent action sequences for probabilistic risk detection, and a safety-guided copilot that steers its denoising process toward safe regions using the gradient of a map-based safety certificate. When risk is detected, control is transferred through partial diffusion: the human plan is forward-noised to an intermediate level and denoised by the safety-guided copilot. The forward-diffusion ratio $ρ$ acts as a continuous takeover knob-small $ρ$ keeps the output close to human intent, while increasing $ρ$ shifts authority toward the copilot, avoiding the mixed-unsafe pitfall of action-level blending. Unlike methods relying on hand-crafted score functions, our diffusion formulation supports both safety evaluation and plan generation directly from demonstrations. We evaluate Diffusion-SAFE in simulation and on a real ROS-based race car, achieving 93.0%/87.0% (sim/real) handover success rates with smooth transitions.
♻ ☆ PAD-TRO: Projection-Augmented Diffusion for Direct Trajectory Optimization
Recently, diffusion models have gained popularity and attention in trajectory optimization due to their capability of modeling multi-modal probability distributions. However, addressing nonlinear equality constraints, i.e, dynamic feasibility, remains a great challenge in diffusion-based trajectory optimization. Recent diffusion-based trajectory optimization frameworks rely on a single-shooting style approach where the denoised control sequence is applied to forward propagate the dynamical system, which cannot explicitly enforce constraints on the states and frequently leads to sub-optimal solutions. In this work, we propose a novel direct trajectory optimization approach via model-based diffusion, which directly generates a sequence of states. To ensure dynamic feasibility, we propose a gradient-free projection mechanism that is incorporated into the reverse diffusion process. Our results show that, compared to a recent state-of-the-art baseline, our approach leads to zero dynamic feasibility error and approximately 4x higher success rate in a quadrotor waypoint navigation scenario involving dense static obstacles.
comment: Final manuscript. Accepted for publication at the 2026 American Control Conference
♻ ☆ ViLAM: Distilling Vision-Language Reasoning into Attention Maps for Social Robot Navigation
We introduce ViLAM, a novel method for distilling vision-language reasoning from large Vision-Language Models (VLMs) into spatial attention maps for socially compliant robot navigation. Unlike traditional methods that rely on expert demonstrations or human-annotated datasets, ViLAM performs knowledge distillation and fine-tuning at the intermediate layer representation (attention) level by aligning attention maps from a pretrained vision-action model with socially guided attention maps derived from a large VLM. These distilled attention maps highlight key navigational regions in a scene and serve as socially informed spatial cost maps for motion planning. To achieve this, we introduce a novel attention-level distillation loss that fuses knowledge from both sources, generating augmented attention maps with enhanced social awareness. These refined attention maps are then used as a traversability costmap within a socially aware local planner for navigation. We validate our approach through real-world experiments on a Husky wheeled robot, and demonstrate 14.2% - 50% improvements in success rate over existing methods.
Robotics 118
☆ RoboPocket: Improve Robot Policies Instantly with Your Phone
Scaling imitation learning is fundamentally constrained by the efficiency of data collection. While handheld interfaces have emerged as a scalable solution for in-the-wild data acquisition, they predominantly operate in an open-loop manner: operators blindly collect demonstrations without knowing the underlying policy's weaknesses, leading to inefficient coverage of critical state distributions. Conversely, interactive methods like DAgger effectively address covariate shift but rely on physical robot execution, which is costly and difficult to scale. To reconcile this trade-off, we introduce RoboPocket, a portable system that enables Robot-Free Instant Policy Iteration using single consumer smartphones. Its core innovation is a Remote Inference framework that visualizes the policy's predicted trajectory via Augmented Reality (AR) Visual Foresight. This immersive feedback allows collectors to proactively identify potential failures and focus data collection on the policy's weak regions without requiring a physical robot. Furthermore, we implement an asynchronous Online Finetuning pipeline that continuously updates the policy with incoming data, effectively closing the learning loop in minutes. Extensive experiments demonstrate that RoboPocket adheres to data scaling laws and doubles the data efficiency compared to offline scaling strategies, overcoming their long-standing efficiency bottleneck. Moreover, our instant iteration loop also boosts sample efficiency by up to 2$\times$ in distributed environments a small number of interactive corrections per person. Project page and videos: https://robo-pocket.github.io.
comment: Project page: https://robo-pocket.github.io
☆ Safe-SAGE: Social-Semantic Adaptive Guidance for Safe Engagement through Laplace-Modulated Poisson Safety Functions
Traditional safety-critical control methods, such as control barrier functions, suffer from semantic blindness, exhibiting the same behavior around obstacles regardless of contextual significance. This limitation leads to the uniform treatment of all obstacles, despite their differing semantic meanings. We present Safe-SAGE (Social-Semantic Adaptive Guidance for Safe Engagement), a unified framework that bridges the gap between high-level semantic understanding and low-level safety-critical control through a Poisson safety function (PSF) modulated using a Laplace guidance field. Our approach perceives the environment by fusing multi-sensor point clouds with vision-based instance segmentation and persistent object tracking to maintain up-to-date semantics beyond the camera's field of view. A multi-layer safety filter is then used to modulate system inputs to achieve safe navigation using this semantic understanding of the environment. This safety filter consists of both a model predictive control layer and a control barrier function layer. Both layers utilize the PSF and flux modulation of the guidance field to introduce varying levels of conservatism and multi-agent passing norms for different obstacles in the environment. Our framework enables legged robots to navigate semantically rich, dynamic environments with context-dependent safety margins while maintaining rigorous safety guarantees.
☆ cuRoboV2: Dynamics-Aware Motion Generation with Depth-Fused Distance Fields for High-DoF Robots
Effective robot autonomy requires motion generation that is safe, feasible, and reactive. Current methods are fragmented: fast planners output physically unexecutable trajectories, reactive controllers struggle with high-fidelity perception, and existing solvers fail on high-DoF systems. We present cuRoboV2, a unified framework with three key innovations: (1) B-spline trajectory optimization that enforces smoothness and torque limits; (2) a GPU-native TSDF/ESDF perception pipeline that generates dense signed distance fields covering the full workspace, unlike existing methods that only provide distances within sparsely allocated blocks, up to 10x faster and in 8x less memory than the state-of-the-art at manipulation scale, with up to 99% collision recall; and (3) scalable GPU-native whole-body computation, namely topology-aware kinematics, differentiable inverse dynamics, and map-reduce self-collision, that achieves up to 61x speedup while also extending to high-DoF humanoids (where previous GPU implementations fail). On benchmarks, cuRoboV2 achieves 99.7% success under 3kg payload (where baselines achieve only 72--77%), 99.6% collision-free IK on a 48-DoF humanoid (where prior methods fail entirely), and 89.5% retargeting constraint satisfaction (vs. 61% for PyRoki); these collision-free motions yield locomotion policies with 21% lower tracking error than PyRoki and 12x lower cross-seed variance than mink. A ground-up codebase redesign for discoverability enabled LLM coding assistants to author up to 73% of new modules, including hand-optimized CUDA kernels, demonstrating that well-structured robotics code can unlock productive human--LLM collaboration. Together, these advances provide a unified, dynamics-aware motion generation stack that scales from single-arm manipulators to full humanoids.
comment: cuRoboV2 Technical Report
☆ Observing and Controlling Features in Vision-Language-Action Models
Vision-Language-Action Models (VLAs) have shown remarkable progress towards embodied intelligence. While their architecture partially resembles that of Large Language Models (LLMs), VLAs exhibit higher complexity due to their multi-modal inputs/outputs and often hybrid nature of transformer and diffusion heads. This is part of the reason why insights from mechanistic interpretability in LLMs, which explain how the internal model representations relate to their output behavior, do not trivially transfer to VLA counterparts. In this work, we propose to close this gap by introducing and analyzing two main concepts: feature-observability and feature-controllability. In particular, we first study features that are linearly encoded in representation space, and show how they can be observed by means of a linear classifier. Then, we use a minimal linear intervention grounded in optimal control to accurately place internal representations and steer the VLA's output towards a desired region. Our results show that targeted, lightweight interventions can reliably steer a robot's behavior while preserving closed-loop capabilities. We demonstrate on different VLA architectures ($π_{0.5}$ and OpenVLA) through simulation experiments that VLAs possess interpretable internal structure amenable to online adaptation without fine-tuning, enabling real-time alignment with user preferences and task requirements.
☆ Residual RL--MPC for Robust Microrobotic Cell Pushing Under Time-Varying Flow
Contact-rich micromanipulation in microfluidic flow is challenging because small disturbances can break pushing contact and induce large lateral drift. We study planar cell pushing with a magnetic rolling microrobot that tracks a waypoint-sampled reference curve under time-varying Poiseuille flow. We propose a hybrid controller that augments a nominal MPC with a learned residual policy trained by SAC. The policy outputs a bounded 2D velocity correction that is contact-gated, so residual actions are applied only during robot--cell contact, preserving reliable approach behavior and stabilizing learning. All methods share the same actuation interface and speed envelope for fair comparisons. Experiments show improved robustness and tracking accuracy over pure MPC and PID under nonstationary flow, with generalization from a clover training curve to unseen circle and square trajectories. A residual-bound sweep identifies an intermediate correction limit as the best trade-off, which we use in all benchmarks.
comment: 8 pages, 8 figures
☆ Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model CVPR 2026
World models provide a powerful framework for simulating environment dynamics conditioned on actions or instructions, enabling downstream tasks such as action planning or policy learning. Recent approaches leverage world models as learned simulators, but its application to decision-time planning remains computationally prohibitive for real-time control. A key bottleneck lies in latent representations: conventional tokenizers encode each observation into hundreds of tokens, making planning both slow and resource-intensive. To address this, we propose CompACT, a discrete tokenizer that compresses each observation into as few as 8 tokens, drastically reducing computational cost while preserving essential information for planning. An action-conditioned world model that occupies CompACT tokenizer achieves competitive planning performance with orders-of-magnitude faster planning, offering a practical step toward real-world deployment of world models.
comment: CVPR 2026
☆ PhysiFlow: Physics-Aware Humanoid Whole-Body VLA via Multi-Brain Latent Flow Matching and Robust Tracking
In the domain of humanoid robot control, the fusion of Vision-Language-Action (VLA) with whole-body control is essential for semantically guided execution of real-world tasks. However, existing methods encounter challenges in terms of low VLA inference efficiency or an absence of effective semantic guidance for whole-body control, resulting in instability in dynamic limb-coordinated tasks. To bridge this gap, we present a semantic-motion intent guided, physics-aware multi-brain VLA framework for humanoid whole-body control. A series of experiments was conducted to evaluate the performance of the proposed framework. The experimental results demonstrated that the framework enabled reliable vision-language-guided full-body coordination for humanoid robots.
☆ ROScopter: A Multirotor Autopilot based on ROSflight 2.0
ROScopter is a lean multirotor autopilot built for researchers. ROScopter seeks to accelerate simulation and hardware testing of research code with an architecture that is both easy to understand and simple to modify. ROScopter is designed to interface with ROSflight 2.0 and runs entirely on an onboard flight computer, leveraging the features of ROS 2 to improve modularity. This work describes the architecture of ROScopter and how it can be used to test application code in both simulated and hardware environments. Hardware results of the default ROScopter behavior are presented, showing that ROScopter achieves similar performance to another state-of-the-art autopilot for basic waypoint-following maneuvers, but with a significantly reduced and more modular code-base.
☆ Loop Closure via Maximal Cliques in 3D LiDAR-Based SLAM
Reliable loop closure detection remains a critical challenge in 3D LiDAR-based SLAM, especially under sensor noise, environmental ambiguity, and viewpoint variation conditions. RANSAC is often used in the context of loop closures for geometric model fitting in the presence of outliers. However, this approach may fail, leading to map inconsistency. We introduce a novel deterministic algorithm, CliReg, for loop closure validation that replaces RANSAC verification with a maximal clique search over a compatibility graph of feature correspondences. This formulation avoids random sampling and increases robustness in the presence of noise and outliers. We integrated our approach into a real- time pipeline employing binary 3D descriptors and a Hamming distance embedding binary search tree-based matching. We evaluated it on multiple real-world datasets featuring diverse LiDAR sensors. The results demonstrate that our proposed technique consistently achieves a lower pose error and more reliable loop closures than RANSAC, especially in sparse or ambiguous conditions. Additional experiments on 2D projection-based maps confirm its generality across spatial domains, making our approach a robust and efficient alternative for loop closure detection.
comment: Accepted in the 2025 European Conference on Mobile Robots (ECMR). This is the author's version of the work
☆ Accelerating Sampling-Based Control via Learned Linear Koopman Dynamics
This paper presents an efficient model predictive path integral (MPPI) control framework for systems with complex nonlinear dynamics. To improve the computational efficiency of classic MPPI while preserving control performance, we replace the nonlinear dynamics used for trajectory propagation with a learned linear deep Koopman operator (DKO) model, enabling faster rollout and more efficient trajectory sampling. The DKO dynamics are learned directly from interaction data, eliminating the need for analytical system models. The resulting controller, termed MPPI-DK, is evaluated in simulation on pendulum balancing and surface vehicle navigation tasks, and validated on hardware through reference-tracking experiments on a quadruped robot. Experimental results demonstrate that MPPI-DK achieves control performance close to MPPI with true dynamics while substantially reducing computational cost, enabling efficient real-time control on robotic platforms.
☆ OpenFrontier: General Navigation with Visual-Language Grounded Frontiers
Open-world navigation requires robots to make decisions in complex everyday environments while adapting to flexible task requirements. Conventional navigation approaches often rely on dense 3D reconstruction and hand-crafted goal metrics, which limits their generalization across tasks and environments. Recent advances in vision--language navigation (VLN) and vision--language--action (VLA) models enable end-to-end policies conditioned on natural language, but typically require interactive training, large-scale data collection, or task-specific fine-tuning with a mobile agent. We formulate navigation as a sparse subgoal identification and reaching problem and observe that providing visual anchoring targets for high-level semantic priors enables highly efficient goal-conditioned navigation. Based on this insight, we select navigation frontiers as semantic anchors and propose OpenFrontier, a training-free navigation framework that seamlessly integrates diverse vision--language prior models. OpenFrontier enables efficient navigation with a lightweight system design, without dense 3D mapping, policy training, or model fine-tuning. We evaluate OpenFrontier across multiple navigation benchmarks and demonstrate strong zero-shot performance, as well as effective real-world deployment on a mobile robot.
☆ Omni-Manip: Beyond-FOV Large-Workspace Humanoid Manipulation with Omnidirectional 3D Perception
The deployment of humanoid robots for dexterous manipulation in unstructured environments remains challenging due to perceptual limitations that constrain the effective workspace. In scenarios where physical constraints prevent the robot from repositioning itself, maintaining omnidirectional awareness becomes far more critical than color or semantic information. While recent advances in visuomotor policy learning have improved manipulation capabilities, conventional RGB-D solutions suffer from narrow fields of view (FOV) and self-occlusion, requiring frequent base movements that introduce motion uncertainty and safety risks. Existing approaches to expanding perception, including active vision systems and third-view cameras, introduce mechanical complexity, calibration dependencies, and latency that hinder reliable real-time performance. In this work, We propose Omni-Manip, an end-to-end LiDAR-driven 3D visuomotor policy that enables robust manipulation in large workspaces. Our method processes panoramic point clouds through a Time-Aware Attention Pooling mechanism, efficiently encoding sparse 3D data while capturing temporal dependencies. This 360° perception allows the robot to interact with objects across wide areas without frequent repositioning. To support policy learning, we develop a whole-body teleoperation system for efficient data collection on full-body coordination. Extensive experiments in simulation and real-world environments show that Omni-Manip achieves robust performance in large-workspace and cluttered scenarios, outperforming baselines that rely on egocentric depth cameras.
comment: 8 pages, 6 figures
☆ CT-Enabled Patient-Specific Simulation and Contact-Aware Robotic Planning for Cochlear Implantation
Robotic cochlear-implant (CI) insertion requires precise prediction and regulation of contact forces to minimize intracochlear trauma and prevent failure modes such as locking and buckling. Aligned with the integration of advanced medical imaging and robotics for autonomous, precision interventions, this paper presents a unified CT-to-simulation pipeline for contact-aware insertion planning and validation. We develop a low-dimensional, differentiable Cosserat-rod model of the electrode array coupled with frictional contact and pseudo-dynamics regularization to ensure continuous stick-slip transitions. Patient-specific cochlear anatomy is reconstructed from CT imaging and encoded via an analytic parametrization of the scala-tympani lumen, enabling efficient and differentiable contact queries through closest-point projection. Based on a differentiated equilibrium-constraint formulation, we derive an online direction-update law under an RCM-like constraint that suppresses lateral insertion forces while maintaining axial advancement. Simulations and benchtop experiments validate deformation and force trends, demonstrating reduced locking/buckling risk and improved insertion depth. The study highlights how CT-based imaging enhances modeling, planning, and safety capabilities in robot-assisted inner-ear procedures.
☆ UltraDexGrasp: Learning Universal Dexterous Grasping for Bimanual Robots with Synthetic Data ICRA
Grasping is a fundamental capability for robots to interact with the physical world. Humans, equipped with two hands, autonomously select appropriate grasp strategies based on the shape, size, and weight of objects, enabling robust grasping and subsequent manipulation. In contrast, current robotic grasping remains limited, particularly in multi-strategy settings. Although substantial efforts have targeted parallel-gripper and single-hand grasping, dexterous grasping for bimanual robots remains underexplored, with data being a primary bottleneck. Achieving physically plausible and geometrically conforming grasps that can withstand external wrenches poses significant challenges. To address these issues, we introduce UltraDexGrasp, a framework for universal dexterous grasping with bimanual robots. The proposed data-generation pipeline integrates optimization-based grasp synthesis with planning-based demonstration generation, yielding high-quality and diverse trajectories across multiple grasp strategies. With this framework, we curate UltraDexGrasp-20M, a large-scale, multi-strategy grasp dataset comprising 20 million frames across 1,000 objects. Based on UltraDexGrasp-20M, we further develop a simple yet effective grasp policy that takes point clouds as input, aggregates scene features via unidirectional attention, and predicts control commands. Trained exclusively on synthetic data, the policy achieves robust zero-shot sim-to-real transfer and consistently succeeds on novel objects with varied shapes, sizes, and weights, attaining an average success rate of 81.2% in real-world universal dexterous grasping. To facilitate future research on grasping with bimanual robots, we open-source the data generation pipeline at https://github.com/InternRobotics/UltraDexGrasp.
comment: Published at International Conference on Robotics and Automation (ICRA) 2026
☆ Constraint-Free Static Modeling of Continuum Parallel Robot
Continuum parallel robots (CPR) combine rigid actuation mechanisms with multiple elastic rods in a closed-loop topology, making forward statics challenging when rigid--continuum junctions are enforced by explicit kinematic constraints. Such constraint-based formulations typically introduce additional algebraic variables and complicate both numerical solution and downstream control. This paper presents a geometric exact, configuration-based and constraint-free static model of CPR that remains valid under geometrically nonlinear, large-deformation and large-rotation conditions. Connectivity constraints are eliminated by kinematic embedding, yielding a reduced unconstrained problem. Each rod of CPR is discretized by nodal poses on SE(3), while the element-wise strain field is reconstructed through a linear strain parameterization. A fourth-order Magnus approximation yields an explicit and geometrically consistent mapping between element end poses and the strain. Rigid attachments at the motor-driven base and the end-effector platforms are handled through kinematic embeddings. Based on total potential energy and virtual work, we derive assembly-ready residuals and explicit Newton tangents, and solve the resulting nonlinear equilibrium equations using a Riemannian Newton iteration on the product manifold. Experiments on a three-servomotor, six-rod prototype validate the model by showing good agreement between simulation and measurements for both unloaded motions and externally loaded cases.
☆ Latent Policy Steering through One-Step Flow Policies
Offline reinforcement learning (RL) allows robots to learn from offline datasets without risky exploration. Yet, offline RL's performance often hinges on a brittle trade-off between (1) return maximization, which can push policies outside the dataset support, and (2) behavioral constraints, which typically require sensitive hyperparameter tuning. Latent steering offers a structural way to stay within the dataset support during RL, but existing offline adaptations commonly approximate action values using latent-space critics learned via indirect distillation, which can lose information and hinder convergence. We propose Latent Policy Steering (LPS), which enables high-fidelity latent policy improvement by backpropagating original-action-space Q-gradients through a differentiable one-step MeanFlow policy to update a latent-action-space actor. By eliminating proxy latent critics, LPS allows an original-action-space critic to guide end-to-end latent-space optimization, while the one-step MeanFlow policy serves as a behavior-constrained generative prior. This decoupling yields a robust method that works out-of-the-box with minimal tuning. Across OGBench and real-world robotic tasks, LPS achieves state-of-the-art performance and consistently outperforms behavioral cloning and strong latent steering baselines.
comment: Project Webpage : https://jellyho.github.io/LPS/
☆ Iterative On-Policy Refinement of Hierarchical Diffusion Policies for Language-Conditioned Manipulation
Hierarchical policies for language-conditioned manipulation decompose tasks into subgoals, where a high-level planner guides a low-level controller. However, these hierarchical agents often fail because the planner generates subgoals without considering the actual limitations of the controller. Existing solutions attempt to bridge this gap via intermediate modules or shared representations, but they remain limited by their reliance on fixed offline datasets. We propose HD-ExpIt, a framework for iterative fine-tuning of hierarchical diffusion policies via environment feedback. HD-ExpIt organizes training into a self-reinforcing cycle: it utilizes diffusion-based planning to autonomously discover successful behaviors, which are then distilled back into the hierarchical policy. This loop enables both components to improve while implicitly grounding the planner in the controller's actual capabilities without requiring explicit proxy models. Empirically, HD-ExpIt significantly improves hierarchical policies trained solely on offline data, achieving state-of-the-art performance on the long-horizon CALVIN benchmark among methods trained from scratch.
☆ From Code to Road: A Vehicle-in-the-Loop and Digital Twin-Based Framework for Central Car Server Testing in Autonomous Driving
Simulation is one of the most essential parts in the development stage of automotive software. However, purely virtual simulations often struggle to accurately capture all real-world factors due to limitations in modeling. To address this challenge, this work presents a test framework for automotive software on the centralized E/E architecture, which is a central car server in our case, based on Vehicle-in-the-Loop (ViL) and digital twin technology. The framework couples a physical test vehicle on a dynamometer test bench with its synchronized virtual counterpart in a simulation environment. Our approach provides a safe, reproducible, realistic, and cost-effective platform for validating autonomous driving algorithms with a centralized architecture. This test method eliminates the need to test individual physical ECUs and their communication protocols separately. In contrast to traditional ViL methods, the proposed framework runs the full autonomous driving software directly on the vehicle hardware after the simulation process, eliminating flashing and intermediate layers while enabling seamless virtual-physical integration and accurately reflecting centralized E/E behavior. In addition, incorporating mixed testing in both simulated and physical environments reduces the need for full hardware integration during the early stages of automotive development. Experimental case studies demonstrate the effectiveness of the framework in different test scenarios. These findings highlight the potential to reduce development and integration efforts for testing autonomous driving pipelines in the future.
comment: 8 pages; Accepted for publication at the 37th IEEE Intelligent Vehicles Symposium (IV), Detroit, MI, United States, June 22-25, 2026
☆ Curve-Induced Dynamical Systems on Riemannian Manifolds and Lie Groups
Deploying robots in household environments requires safe, adaptable, and interpretable behaviors that respect the geometric structure of tasks. Often represented on Lie groups and Riemannian manifolds, this includes poses on SE(3) or symmetric positive definite matrices encoding stiffness or damping matrices. In this context, dynamical system-based approaches offer a natural framework for generating such behavior, providing stability and convergence while remaining responsive to changes in the environment. We introduce Curve-induced Dynamical systems on Smooth Manifolds (CDSM), a real-time framework for constructing dynamical systems directly on Riemannian manifolds and Lie groups. The proposed approach constructs a nominal curve on the manifold, and generates a dynamical system which combines a tangential component that drives motion along the curve and a normal component that attracts the state toward the curve. We provide a stability analysis of the resulting dynamical system and validate the method quantitatively. On an S2 benchmark, CDSM demonstrates improved trajectory accuracy, reduced path deviation, and faster generation and query times compared to state-of-the-art methods. Finally, we demonstrate the practical applicability of the framework on both a robotic manipulator, where poses on SE(3) and damping matrices on SPD(n) are adapted online, and a mobile manipulator.
comment: Preprint, 14 pages, video linked in the paper, Saray Bakker and Martin Schonger contributed equally as first authors and are listed alphabetically
☆ Rethinking the Role of Collaborative Robots in Rehabilitation
Current research on collaborative robots (cobots) in physical rehabilitation largely focuses on repeated motion training for people undergoing physical therapy (PuPT), even though these sessions include phases that could benefit from robotic collaboration and assistance. Meanwhile, access to physical therapy remains limited for people with disabilities and chronic illnesses. Cobots could support both PuPT and therapists, and improve access to therapy, yet their broader potential remains underexplored. We propose extending the scope of cobots by imagining their role in assisting therapists and PuPT before, during, and after a therapy session. We discuss how cobot assistance may lift access barriers by promoting ability-based therapy design and helping therapists manage their time and effort. Finally, we highlight challenges to realizing these roles, including advancing user-state understanding, ensuring safety, and integrating cobots into therapists' workflow. This view opens new research questions and opportunities to draw from the HRI community's advances in assistive robotics.
comment: 5 pages, 1 figure
☆ Digital Twin Driven Textile Classification and Foreign Object Recognition in Automated Sorting Systems
The increasing demand for sustainable textile recycling requires robust automation solutions capable of handling deformable garments and detecting foreign objects in cluttered environments. This work presents a digital twin driven robotic sorting system that integrates grasp prediction, multi modal perception, and semantic reasoning for real world textile classification. A dual arm robotic cell equipped with RGBD sensing, capacitive tactile feedback, and collision-aware motion planning autonomously separates garments from an unsorted basket, transfers them to an inspection zone, and classifies them using state of the art Visual Language Models (VLMs). We benchmark nine VLM s from five model families on a dataset of 223 inspection scenarios comprising shirts, socks, trousers, underwear, foreign objects (including garments outside of the aforementioned classes), and empty scenes. The evaluation assesses per class accuracy, hallucination behavior, and computational performance under practical hardware constraints. Results show that the Qwen model family achieves the highest overall accuracy (up to 87.9 %), with strong foreign object detection performance, while lighter models such as Gemma3 offer competitive speed accuracy trade offs for edge deployment. A digital twin combined with MoveIt enables collision aware path planning and integrates segmented 3D point clouds of inspected garments into the virtual environment for improved manipulation reliability. The presented system demonstrates the feasibility of combining semantic VLM reasoning with conventional grasp detection and digital twin technology for scalable, autonomous textile sorting in realistic industrial settings.
comment: 10 pages,single column, 5 figures, preprint for Photomet Edumet 2026 (Klagenfurt, Austria)
☆ Critic in the Loop: A Tri-System VLA Framework for Robust Long-Horizon Manipulation
Balancing high-level semantic reasoning with low-level reactive control remains a core challenge in visual robotic manipulation. While Vision-Language Models (VLMs) excel at cognitive planning, their inference latency precludes real-time execution. Conversely, fast Vision-Language-Action (VLA) models often lack the semantic depth required for complex, long-horizon tasks. To bridge this gap, we introduce Critic in the Loop, an adaptive hierarchical framework driven by dynamic VLM-Expert scheduling. At its core is a bionic Tri-System architecture comprising a VLM brain for global reasoning, a VLA cerebellum for reactive execution, and a lightweight visual Critic. By continuously monitoring the workspace, the Critic dynamically routes control authority. It sustains rapid closed-loop execution via the VLA for routine subtasks, and adaptively triggers the VLM for replanning upon detecting execution anomalies such as task stagnation or failures. Furthermore, our architecture seamlessly integrates human-inspired rules to intuitively break infinite retry loops. This visually-grounded scheduling minimizes expensive VLM queries, while substantially enhancing system robustness and autonomy in out-of-distribution (OOD) scenarios. Comprehensive experiments on challenging, long-horizon manipulation benchmarks reveal that our approach achieves state-of-the-art performance.
☆ Lifelong Language-Conditioned Robotic Manipulation Learning
Traditional language-conditioned manipulation agent sequential adaptation to new manipulation skills leads to catastrophic forgetting of old skills, limiting dynamic scene practical deployment. In this paper, we propose SkillsCrafter, a novel robotic manipulation framework designed to continually learn multiple skills while reducing catastrophic forgetting of old skills. Specifically, we propose a Manipulation Skills Adaptation to retain the old skills knowledge while inheriting the shared knowledge between new and old skills to facilitate learning of new skills. Meanwhile, we perform the singular value decomposition on the diverse skill instructions to obtain common skill semantic subspace projection matrices, thereby recording the essential semantic space of skills. To achieve forget-less and generalization manipulation, we propose a Skills Specialization Aggregation to compute inter-skills similarity in skill semantic subspaces, achieving aggregation of the previously learned skill knowledge for any new or unknown skill. Extensive experiments demonstrate the effectiveness and superiority of our proposed SkillsCrafter.
comment: 14 pages, 7 figures
☆ Act, Think or Abstain: Complexity-Aware Adaptive Inference for Vision-Language-Action Models
Current research on Vision-Language-Action (VLA) models predominantly focuses on enhancing generalization through established reasoning techniques. While effective, these improvements invariably increase computational complexity and inference latency. Furthermore, these mechanisms are typically applied indiscriminately, resulting in the inefficient allocation of resources for trivial tasks while simultaneously failing to provide the uncertainty estimation necessary to prevent catastrophic failure on out-of-distribution tasks. Inspired by human cognition, we propose an adaptive framework that dynamically routes VLA execution based on the complexity of the perceived state. Our approach transforms the VLA's vision-language backbone into an active detection tool by projecting latent embeddings into an ensemble of parametric and non-parametric estimators. This allows the system to execute known tasks immediately (Act), reason about ambiguous scenarios (Think), and preemptively halt execution when encountering significant physical or semantic anomalies (Abstain). In our empirical analysis, we observe a phenomenon where visual embeddings alone are superior for inferring task complexity due to the semantic invariance of language. Evaluated on the LIBERO and LIBERO-PRO benchmarks as well as on a real robot, our vision-only configuration achieves 80% F1-Score using as little as 5% of training data, establishing itself as a reliable and efficient task complexity detector.
☆ SeedPolicy: Horizon Scaling via Self-Evolving Diffusion Policy for Robot Manipulation
Imitation Learning (IL) enables robots to acquire manipulation skills from expert demonstrations. Diffusion Policy (DP) models multi-modal expert behaviors but suffers performance degradation as observation horizons increase, limiting long-horizon manipulation. We propose Self-Evolving Gated Attention (SEGA), a temporal module that maintains a time-evolving latent state via gated attention, enabling efficient recurrent updates that compress long-horizon observations into a fixed-size representation while filtering irrelevant temporal information. Integrating SEGA into DP yields Self-Evolving Diffusion Policy (SeedPolicy), which resolves the temporal modeling bottleneck and enables scalable horizon extension with moderate overhead. On the RoboTwin 2.0 benchmark with 50 manipulation tasks, SeedPolicy outperforms DP and other IL baselines. Averaged across both CNN and Transformer backbones, SeedPolicy achieves 36.8% relative improvement in clean settings and 169% relative improvement in randomized challenging settings over the DP. Compared to vision-language-action models such as RDT with 1.2B parameters, SeedPolicy achieves competitive performance with one to two orders of magnitude fewer parameters, demonstrating strong efficiency and scalability. These results establish SeedPolicy as a state-of-the-art imitation learning method for long-horizon robotic manipulation. Code is available at: https://github.com/Youqiang-Gui/SeedPolicy.
comment: 16 pages, 13 figures
☆ Decoupling Task and Behavior: A Two-Stage Reward Curriculum in Reinforcement Learning for Robotics
Deep Reinforcement Learning is a promising tool for robotic control, yet practical application is often hindered by the difficulty of designing effective reward functions. Real-world tasks typically require optimizing multiple objectives simultaneously, necessitating precise tuning of their weights to learn a policy with the desired characteristics. To address this, we propose a two-stage reward curriculum where we decouple task-specific objectives from behavioral terms. In our method, we first train the agent on a simplified task-only reward function to ensure effective exploration before introducing the full reward that includes auxiliary behavior-related terms such as energy efficiency. Further, we analyze various transition strategies and demonstrate that reusing samples between phases is critical for training stability. We validate our approach on the DeepMind Control Suite, ManiSkill3, and a mobile robot environment, modified to include auxiliary behavioral objectives. Our method proves to be simple yet effective, substantially outperforming baselines trained directly on the full reward while exhibiting higher robustness to specific reward weightings.
☆ SPIRIT: Perceptive Shared Autonomy for Robust Robotic Manipulation under Deep Learning Uncertainty
Deep learning (DL) has enabled impressive advances in robotic perception, yet its limited robustness and lack of interpretability hinder reliable deployment in safety critical applications. We propose a concept termed perceptive shared autonomy, in which uncertainty estimates from DL based perception are used to regulate the level of autonomy. Specifically, when the robot's perception is confident, semi-autonomous manipulation is enabled to improve performance; when uncertainty increases, control transitions to haptic teleoperation for maintaining robustness. In this way, high-performing but uninterpretable DL methods can be integrated safely into robotic systems. A key technical enabler is an uncertainty aware DL based point cloud registration approach based on the so called Neural Tangent Kernels (NTK). We evaluate perceptive shared autonomy on challenging aerial manipulation tasks through a user study of 15 participants and realization of mock-up industrial scenarios, demonstrating reliable robotic manipulation despite failures in DL based perception. The resulting system, named SPIRIT, improves both manipulation performance and system reliability. SPIRIT was selected as a finalist of a major industrial innovation award.
comment: 19 pages, 14 figures
☆ GaussTwin: Unified Simulation and Correction with Gaussian Splatting for Robotic Digital Twins ICRA 2026
Digital twins promise to enhance robotic manipulation by maintaining a consistent link between real-world perception and simulation. However, most existing systems struggle with the lack of a unified model, complex dynamic interactions, and the real-to-sim gap, which limits downstream applications such as model predictive control. Thus, we propose GaussTwin, a real-time digital twin that combines position-based dynamics with discrete Cosserat rod formulations for physically grounded simulation, and Gaussian splatting for efficient rendering and visual correction. By anchoring Gaussians to physical primitives and enforcing coherent SE(3) updates driven by photometric error and segmentation masks, GaussTwin achieves stable prediction-correction while preserving physical fidelity. Through experiments in both simulation and on a Franka Research 3 platform, we show that GaussTwin consistently improves tracking accuracy and robustness compared to shape-matching and rigid-only baselines, while also enabling downstream tasks such as push-based planning. These results highlight GaussTwin as a step toward unified, physically meaningful digital twins that can support closed-loop robotic interaction and learning.
comment: 8 pages, 4 figures, 3 tables, ICRA 2026
☆ AIM-SLAM: Dense Monocular SLAM via Adaptive and Informative Multi-View Keyframe Prioritization with Foundation Model
Recent advances in geometric foundation models have emerged as a promising alternative for addressing the challenge of dense reconstruction in monocular visual simultaneous localization and mapping (SLAM). Although geometric foundation models enable SLAM to leverage variable input views, the previous methods remain confined to two-view pairs or fixed-length inputs without sufficient deliberation of geometric context for view selection. To tackle this problem, we propose AIM-SLAM, a dense monocular SLAM framework that exploits an adaptive and informative multi-view keyframe prioritization with dense pointmap predictions from visual geometry grounded transformer (VGGT). Specifically, we introduce the selective information- and geometric-aware multi-view adaptation (SIGMA) module, which employs voxel overlap and information gain to retrieve a candidate set of keyframes and adaptively determine its size. Furthermore, we formulate a joint multi-view Sim(3) optimization that enforces consistent alignment across selected views, substantially improving pose estimation accuracy. The effectiveness of AIM-SLAM is demonstrated on real-world datasets, where it achieves state-of-the-art performance in both pose estimation and dense reconstruction. Our system supports ROS integration, with code is available at https://aimslam.github.io/.
comment: 8 pages
☆ VinePT-Map: Pole-Trunk Semantic Mapping for Resilient Autonomous Robotics in Vineyards
Reliable long-term deployment of autonomous robots in agricultural environments remains challenging due to perceptual aliasing, seasonal variability, and the dynamic nature of crop canopies. Vineyards, characterized by repetitive row structures and significant visual changes across phenological stages, represent a pivotal field challenge, limiting the robustness of conventional feature-based localization and mapping approaches. This paper introduces VinePT-Map, a semantic mapping framework that leverages vine trunks and support poles as persistent structural landmarks to enable season-agnostic and resilient robot localization. The proposed method formulates the mapping problem as a factor graph, integrating GPS, IMU, and RGB-D observations through robust geometrical constraints that exploit vineyard structure. An efficient perception pipeline based on instance segmentation and tracking, combined with a clustering filter for outlier rejection and pose refinement, enables accurate landmark detection using low-cost sensors and onboard computation. To validate the pipeline, we present a multi-season dataset for trunk and pole segmentation and tracking. Extensive field experiments conducted across diverse seasons demonstrate the robustness and accuracy of the proposed approach, highlighting its suitability for long-term autonomous operation in agricultural environments.
☆ CoIn3D: Revisiting Configuration-Invariant Multi-Camera 3D Object Detection CVPR 2026
Multi-camera 3D object detection (MC3D) has attracted increasing attention with the growing deployment of multi-sensor physical agents, such as robots and autonomous vehicles. However, MC3D models still struggle to generalize to unseen platforms with new multi-camera configurations. Current solutions simply employ a meta-camera for unified representation but lack comprehensive consideration. In this paper, we revisit this issue and identify that the devil lies in spatial prior discrepancies across source and target configurations, including different intrinsics, extrinsics, and array layouts. To address this, we propose CoIn3D, a generalizable MC3D framework that enables strong transferability from source configurations to unseen target ones. CoIn3D explicitly incorporates all identified spatial priors into both feature embedding and image observation through spatial-aware feature modulation (SFM) and camera-aware data augmentation (CDA), respectively. SFM enriches feature space by integrating four spatial representations, such as focal length, ground depth, ground gradient, and Plücker coordinate. CDA improves observation diversity under various configurations via a training-free dynamic novel-view image synthesis scheme. Extensive experiments demonstrate that CoIn3D achieves strong cross-configuration performance on landmark datasets such as NuScenes, Waymo, and Lyft, under three dominant MC3D paradigms represented by BEVDepth, BEVFormer, and PETR.
comment: Accepted to CVPR 2026 main track
☆ Direct Contact-Tolerant Motion Planning With Vision Language Models
Navigation in cluttered environments often requires robots to tolerate contact with movable or deformable objects to maintain efficiency. Existing contact-tolerant motion planning (CTMP) methods rely on indirect spatial representations (e.g., prebuilt map, obstacle set), resulting in inaccuracies and a lack of adaptiveness to environmental uncertainties. To address this issue, we propose a direct contact-tolerant (DCT) planner, which integrates vision-language models (VLMs) into direct point perception and navigation, including two key components. The first one is VLM point cloud partitioner (VPP), which performs contact-tolerance reasoning in image space using VLM, caches inference masks, propagates them across frames using odometry, and projects them onto the current scan to generate a contact-aware point cloud. The second innovation is VPP guided navigation (VGN), which formulates CTMP as a perception-to-control optimization problem under direct contact-aware point cloud constraints, which is further solved by a specialized deep neural network (DNN). We implement DCT in Isaac Sim and a real car-like robot, demonstrating that DCT achieves robust and efficient navigation in cluttered environments with movable obstacles, outperforming representative baselines across diverse metrics. The code is available at: https://github.com/ChrisLeeUM/DCT.
☆ Observer Design for Augmented Reality-based Teleoperation of Soft Robots
Although virtual and augmented reality are gaining traction as teleoperation tools for various types of robots, including manipulators and mobile robots, they are not being used for soft robots. The inherent difficulties of modelling soft robots mean that combining accurate and computationally efficient representations is very challenging. This paper presents an augmented reality interface for teleoperating these devices. The developed system consists of Microsoft HoloLens 2 glasses and a central computer responsible for calculations. Validation is performed on PETER, a highly modular pneumatic manipulator. Using data collected from sensors, the computer estimates the robot's position based on the physics of the virtual reality programme. Errors obtained are on the order of 5% of the robot's length, demonstrating that augmented reality facilitates operator interaction with soft manipulators and can be integrated into the control loop.
☆ Person Detection and Tracking from an Overhead Crane LiDAR
This paper investigates person detection and tracking in an industrial indoor workspace using a LiDAR mounted on an overhead crane. The overhead viewpoint introduces a strong domain shift from common vehicle-centric LiDAR benchmarks, and limited availability of suitable public training data. Henceforth, we curate a site-specific overhead LiDAR dataset with 3D human bounding-box annotations and adapt selected candidate 3D detectors under a unified training and evaluation protocol. We further integrate lightweight tracking-by-detection using AB3DMOT and SimpleTrack to maintain person identities over time. Detection performance is reported with distance-sliced evaluation to quantify the practical operating envelope of the sensing setup. The best adapted detector configurations achieve average precision (AP) up to 0.84 within a 5.0 m horizontal radius, increasing to 0.97 at 1.0 m, with VoxelNeXt and SECOND emerging as the most reliable backbones across this range. The acquired results contribute in bridging the domain gap between standard driving datasets and overhead sensing for person detection and tracking. We also report latency measurements, highlighting practical real-time feasibility. Finally, we release our dataset and implementations in GitHub to support further research
comment: 8 pages, 7 figures, 4 tables. Submitted to Ubiquitous Robots (UR) 2026. Code: https://github.com/nilushacj/O-LiPeDeT-Overhead-LiDAR-Person-Detection-and-Tracking
☆ Integrated cooperative localization of heterogeneous measurement swarm: A unified data-driven method
The cooperative localization (CL) problem in heterogeneous robotic systems with different measurement capabilities is investigated in this work. In practice, heterogeneous sensors lead to directed and sparse measurement topologies, whereas most existing CL approaches rely on multilateral localization with restrictive multi-neighbor geometric requirements. To overcome this limitation, we enable pairwise relative localization (RL) between neighboring robots using only mutual measurement and odometry information. A unified data-driven adaptive RL estimator is first developed to handle heterogeneous and unidirectional measurements. Based on the convergent RL estimates, a distributed pose-coupling CL strategy is then designed, which guarantees CL under a weakly connected directed measurement topology, representing the least restrictive condition among existing results. The proposed method is independent of specific control tasks and is validated through a formation control application and real-world experiments.
☆ U-OBCA: Uncertainty-Aware Optimization-Based Collision Avoidance via Wasserstein Distributionally Robust Chance Constraints
Uncertainties arising from localization error, trajectory prediction errors of the moving obstacles and environmental disturbances pose significant challenges to robot's safe navigation. Existing uncertainty-aware planners often approximate polygon-shaped robots and obstacles using simple geometric primitives such as circles or ellipses. Though computationally convenient, these approximations substantially shrink the feasible space, leading to overly conservative trajectories and even planning failure in narrow environments. In addition, many such methods rely on specific assumptions about noise distributions, which may not hold in practice and thus limit their performance guarantees. To address these limitations, we extend the Optimization-Based Collision Avoidance (OBCA) framework to an uncertainty-aware formulation, termed \emph{U-OBCA}. The proposed method explicitly accounts for the collision risk between polygon-shaped robots and obstacles by formulating OBCA-based chance constraints, and hence avoiding geometric simplifications and reducing unnecessary conservatism. These probabilistic constraints are further tightened into deterministic nonlinear constraints under mild distributional assumptions, which can be solved efficiently by standard numerical optimization solvers. The proposed approach is validated through theoretical analysis, numerical simulations and real-world experiments. The results demonstrate that U-OBCA significantly mitigates the conservatism in trajectory planning and achieves higher navigation efficiency compared to existing baseline methods, particularly in narrow and cluttered environments.
☆ Beyond the Patch: Exploring Vulnerabilities of Visuomotor Policies via Viewpoint-Consistent 3D Adversarial Object ICRA 2026
Neural network-based visuomotor policies enable robots to perform manipulation tasks but remain susceptible to perceptual attacks. For example, conventional 2D adversarial patches are effective under fixed-camera setups, where appearance is relatively consistent; however, their efficacy often diminishes under dynamic viewpoints from moving cameras, such as wrist-mounted setups, due to perspective distortions. To proactively investigate potential vulnerabilities beyond 2D patches, this work proposes a viewpoint-consistent adversarial texture optimization method for 3D objects through differentiable rendering. As optimization strategies, we employ Expectation over Transformation (EOT) with a Coarse-to-Fine (C2F) curriculum, exploiting distance-dependent frequency characteristics to induce textures effective across varying camera-object distances. We further integrate saliency-guided perturbations to redirect policy attention and design a targeted loss that persistently drives robots toward adversarial objects. Our comprehensive experiments show that the proposed method is effective under various environmental conditions, while confirming its black-box transferability and real-world applicability.
comment: 8 pages, 10 figures, Accepted to ICRA 2026. Project page: https://chan-mi-lee.github.io/3DAdvObj/
☆ VPWEM: Non-Markovian Visuomotor Policy with Working and Episodic Memory
Imitation learning from human demonstrations has achieved significant success in robotic control, yet most visuomotor policies still condition on single-step observations or short-context histories, making them struggle with non-Markovian tasks that require long-term memory. Simply enlarging the context window incurs substantial computational and memory costs and encourages overfitting to spurious correlations, leading to catastrophic failures under distribution shift and violating real-time constraints in robotic systems. By contrast, humans can compress important past experiences into long-term memories and exploit them to solve tasks throughout their lifetime. In this paper, we propose VPWEM, a non-Markovian visuomotor policy equipped with working and episodic memories. VPWEM retains a sliding window of recent observation tokens as short-term working memory, and introduces a Transformer-based contextual memory compressor that recursively converts out-of-window observations into a fixed number of episodic memory tokens. The compressor uses self-attention over a cache of past summary tokens and cross-attention over a cache of historical observations, and is trained jointly with the policy. We instantiate VPWEM on diffusion policies to exploit both short-term and episode-wide information for action generation with nearly constant memory and computation per step. Experiments demonstrate that VPWEM outperforms state-of-the-art baselines including diffusion policies and vision-language-action (VLA) models by more than 20% on the memory-intensive manipulation tasks in MIKASA and achieves an average 5% improvement on the mobile manipulation benchmark MoMaRT. Code is available at https://github.com/HarryLui98/code_vpwem.
☆ Causally Robust Reward Learning from Reason-Augmented Preference Feedback ICLR
Preference-based reward learning is widely used for shaping agent behavior to match a user's preference, yet its sparse binary feedback makes it especially vulnerable to causal confusion. The learned reward often latches onto spurious features that merely co-occur with preferred trajectories during training, collapsing when those correlations disappear or reverse at test time. We introduce ReCouPLe, a lightweight framework that uses natural language rationales to provide the missing causal signal. Each rationale is treated as a guiding projection axis in an embedding space, training the model to score trajectories based on features aligned with that axis while de-emphasizing context that is unrelated to the stated reason. Because the same rationales (e.g., "avoids collisions", "completes the task faster") can appear across multiple tasks, ReCouPLe naturally reuses the same causal direction whenever tasks share semantics, and transfers preference knowledge to novel tasks without extra data or language-model fine-tuning. Our learned reward model can ground preferences on the articulated reason, aligning better with user intent and generalizing beyond spurious features. ReCouPLe outperforms baselines by up to 1.5x in reward accuracy under distribution shifts, and 2x in downstream policy performance in novel tasks. We have released our code at https://github.com/mj-hwang/ReCouPLe
comment: Published in International Conference on Learning Representations (ICLR) 2026
☆ Hyperbolic Multiview Pretraining for Robotic Manipulation CVPR 2026
3D-aware visual pretraining has proven effective in improving the performance of downstream robotic manipulation tasks. However, existing methods are constrained to Euclidean embedding spaces, whose flat geometry limits their ability to model structural relations among embeddings. As a result, they struggle to learn structured embeddings that are essential for robust spatial perception in robotic applications. To this end, we propose HyperMVP, a self-supervised framework for \underline{Hyper}bolic \underline{M}ulti\underline{V}iew \underline{P}retraining. Hyperbolic space offers geometric properties well suited for capturing structural relations. Methodologically, we extend the masked autoencoder paradigm and design a GeoLink encoder to learn multiview hyperbolic representations. The pretrained encoder is then finetuned with visuomotor policies on manipulation tasks. In addition, we introduce 3D-MOV, a large-scale dataset comprising multiple types of 3D point clouds to support pretraining. We evaluate HyperMVP on COLOSSEUM, RLBench, and real-world scenarios, where it consistently outperforms strong baselines across diverse tasks and perturbation settings. Our results highlight the potential of 3D-aware pretraining in a non-Euclidean space for learning robust and generalizable robotic manipulation policies.
comment: This paper was submitted to CVPR 2026 and was recommended for Findings, but the authors have withdrawn it and are currently adding more content to submit it elsewhere
☆ Task-Relevant and Irrelevant Region-Aware Augmentation for Generalizable Vision-Based Imitation Learning in Agricultural Manipulation
Vision-based imitation learning has shown promise for robotic manipulation; however, its generalization remains limited in practical agricultural tasks. This limitation stems from scarce demonstration data and substantial visual domain gaps caused by i) crop-specific appearance diversity and ii) background variations. To address this limitation, we propose Dual-Region Augmentation for Imitation Learning (DRAIL), a region-aware augmentation framework designed for generalizable vision-based imitation learning in agricultural manipulation. DRAIL explicitly separates visual observations into task-relevant and task-irrelevant regions. The task-relevant region is augmented in a domain-knowledge-driven manner to preserve essential visual characteristics, while the task-irrelevant region is aggressively randomized to suppress spurious background correlations. By jointly handling both sources of visual variation, DRAIL promotes learning policies that rely on task-essential features rather than incidental visual cues. We evaluate DRAIL on diffusion policy-based visuomotor controllers through robot experiments on artificial vegetable harvesting and real lettuce defective leaf picking preparation tasks. The results show consistent improvements in success rates under unseen visual conditions compared to baseline methods. Further attention analysis and representation generalization metrics indicate that the learned policies rely more on task-essential visual features, resulting in enhanced robustness and generalization.
☆ On the Strengths and Weaknesses of Data for Open-set Embodied Assistance
Embodied foundation models are increasingly performant in real-world domains such as robotics or autonomous driving. These models are often deployed in interactive or assistive settings, where it is important that these assistive models generalize to new users and new tasks. Diverse interactive data generation offers a promising avenue for providing data-efficient generalization capabilities for interactive embodied foundation models. In this paper, we investigate the generalization capabilities of a multimodal foundation model fine-tuned on diverse interactive assistance data in a synthetic domain. We explore generalization along two axes: a) assistance with unseen categories of user behavior and b) providing guidance in new configurations not encountered during training. We study a broad capability called \textbf{Open-Set Corrective Assistance}, in which the model needs to inspect lengthy user behavior and provide assistance through either corrective actions or language-based feedback. This task remains unsolved in prior work, which typically assumes closed corrective categories or relies on external planners, making it a challenging testbed for evaluating the limits of assistive data. To support this task, we generate synthetic assistive datasets in Overcooked and fine-tune a LLaMA-based model to evaluate generalization to novel tasks and user behaviors. Our approach provides key insights into the nature of assistive datasets required to enable open-set assistive intelligence. In particular, we show that performant models benefit from datasets that cover different aspects of assistance, including multimodal grounding, defect inference, and exposure to diverse scenarios.
☆ Diffusion Policy through Conditional Proximal Policy Optimization
Reinforcement learning (RL) has been extensively employed in a wide range of decision-making problems, such as games and robotics. Recently, diffusion policies have shown strong potential in modeling multi-modal behaviors, enabling more diverse and flexible action generation compared to the conventional Gaussian policy. Despite various attempts to combine RL with diffusion, a key challenge is the difficulty of computing action log-likelihood under the diffusion model. This greatly hinders the direct application of diffusion policies in on-policy reinforcement learning. Most existing methods calculate or approximate the log-likelihood through the entire denoising process in the diffusion model, which can be memory- and computationally inefficient. To overcome this challenge, we propose a novel and efficient method to train a diffusion policy in an on-policy setting that requires only evaluating a simple Gaussian probability. This is achieved by aligning the policy iteration with the diffusion process, which is a distinct paradigm compared to previous work. Moreover, our formulation can naturally handle entropy regularization, which is often difficult to incorporate into diffusion policies. Experiments demonstrate that the proposed method produces multimodal policy behaviors and achieves superior performance on a variety of benchmark tasks in both IsaacLab and MuJoCo Playground.
☆ Data-Driven Control of a Magnetically Actuated Fish-Like Robot
Magnetically actuated fish-like robots offer promising solutions for underwater exploration due to their miniaturization and agility; however, precise control remains a significant challenge because of nonlinear fluid dynamics, flexible fin hysteresis, and the variable-duration control steps inherent to the actuation mechanism. This paper proposes a comprehensive data-driven control framework to address these complexities without relying on analytical modeling. Our methodology comprises three core components: 1) developing a forward dynamics model (FDM) using a neural network trained on real-world experimental data to capture state transitions under varying time steps; 2) integrating this FDM into a gradient-based model predictive control (G-MPC) architecture to optimize control inputs for path following; and 3) applying imitation learning to approximate the G-MPC policy, thereby reducing the computational cost for real-time implementation. We validate the approach through simulations utilizing the identified dynamics model. The results demonstrate that the G-MPC framework achieves accurate path convergence with minimal root mean square error (RMSE), and the imitation learning controller (ILC) effectively replicates this performance. This study highlights the potential of data-driven control strategies for the precise navigation of miniature, fish-like soft robots.
comment: Author's version of the paper presented at AROB-ISBC 2026
☆ LLM-Guided Decentralized Exploration with Self-Organizing Robot Teams
When individual robots have limited sensing capabilities or insufficient fault tolerance, it becomes necessary for multiple robots to form teams during exploration, thereby increasing the collective observation range and reliability. Traditionally, swarm formation has often been managed by a central controller; however, from the perspectives of robustness and flexibility, it is preferable for the swarm to operate autonomously even in the absence of centralized control. In addition, the determination of exploration targets for each team is crucial for efficient exploration in such multi-team exploration scenarios. This study therefore proposes an exploration method that combines (1) an algorithm for self-organization, enabling the autonomous and dynamic formation of multiple teams, and (2) an algorithm that allows each team to autonomously determine its next exploration target (destination). In particular, for (2), this study explores a novel strategy based on large language models (LLMs), while classical frontier-based methods and deep reinforcement learning approaches have been widely studied. The effectiveness of the proposed method was validated through simulations involving tens to hundreds of robots.
comment: Author's version of the paper presented at AROB-ISBC 2026
☆ Adaptive Policy Switching of Two-Wheeled Differential Robots for Traversing over Diverse Terrains
Exploring lunar lava tubes requires robots to traverse without human intervention. Because pre-trained policies cannot fully cover all possible terrain conditions, our goal is to enable adaptive policy switching, where the robot selects an appropriate terrain-specialized model based on its current terrain features. This study investigates whether terrain types can be estimated effectively using posture-related observations collected during navigation. We fine-tuned a pre-trained policy using Proximal Policy Optimization (PPO), and then collected the robot's 3D orientation data as it moved across flat and rough terrain in a simulated lava-tube environment. Our analysis revealed that the standard deviation of the robot's pitch data shows a clear difference between these two terrain types. Using Gaussian mixture models (GMM), we evaluated terrain classification across various window sizes. An accuracy of more than 98% was achieved when using a 70-step window. The result suggests that short-term orientation data are sufficient for reliable terrain estimation, providing a foundation for adaptive policy switching.
comment: Author's version of the paper presented at AROB-ISBC 2026
☆ Designing and Validating a Self-Aligning Tool Changer for Modular Reconfigurable Manipulation Robots
Modular reconfigurable robots require reliable mechanisms for automated module exchange, but conventional rigid active couplings often fail due to inevitable positioning and orientational errors. To address this, we propose a misalignment-tolerant tool-changing system. The hardware features a motor-driven coupling utilizing passive self-alignment geometries, specifically chamfered receptacles and triangular lead-in guides, to robustly compensate for angular and lateral misalignments without complex force sensors. To make this autonomous exchange practically feasible, the mechanism is complemented by a compact rotating tool exchange station for efficient module storage. Real-world autonomous tool-picking experiments validate that the self-aligning features successfully absorb execution errors, enabling highly reliable robotic tool reconfiguration.
comment: 6 pages, 13 figures
☆ Gait Generation Balancing Joint Load and Mobility for Legged Modular Robots with Easily Detachable Joints
While modular robots offer versatility, excessive joint torque during locomotion poses a significant risk of mechanical failure, especially for detachable joints. To address this, we propose an optimization framework using the NSGA-III algorithm. Unlike conventional approaches that prioritize mobility alone, our method derives Pareto optimal solutions to minimize joint load while maintaining necessary locomotion speed and stability. Simulations and physical experiments demonstrate that our approach successfully generates gait motions for diverse environments, such as slopes and steps, ensuring structural integrity without compromising overall mobility.
comment: 6 pages, 7 figures
☆ Design, Mapping, and Contact Anticipation with 3D-printed Whole-Body Tactile and Proximity Sensors ICRA
Robots operating in dynamic and shared environments benefit from anticipating contact before it occurs. We present GenTact-Prox, a fully 3D-printed artificial skin that integrates tactile and proximity sensing for contact detection and anticipation. The artificial skin platform is modular in design, procedurally generated to fit any robot morphology, and can cover the whole body of a robot. The skin achieved detection ranges of up to 18 cm during evaluation. To characterize how robots perceive nearby space through this skin, we introduce a data-driven framework for mapping the Perisensory Space -- the body-centric volume of space around the robot where sensors provide actionable information for contact anticipation. We demonstrate this approach on a Franka Research 3 robot equipped with five GenTact-Prox units, enabling online object-aware operation and contact prediction.
comment: This work was accepted at the International Conference on Robotics and Automation (ICRA) 2026
☆ LEGS-POMDP: Language and Gesture-Guided Object Search in Partially Observable Environments
To assist humans in open-world environments, robots must interpret ambiguous instructions to locate desired objects. Foundation model-based approaches excel at multimodal grounding, but they lack a principled mechanism for modeling uncertainty in long-horizon tasks. In contrast, Partially Observable Markov Decision Processes (POMDPs) provide a systematic framework for planning under uncertainty but are often limited in supported modalities and rely on restrictive environment assumptions. We introduce LanguagE and Gesture-Guided Object Search in Partially Observable Environments (LEGS-POMDP), a modular POMDP system that integrates language, gesture, and visual observations for open-world object search. Unlike prior work, LEGS-POMDP explicitly models two sources of partial observability: uncertainty over the target object's identity and its spatial location. In simulation, multimodal fusion significantly outperforms unimodal baselines, achieving an average success rate of 89\% across challenging environments and object categories. Finally, we demonstrate the full system on a quadruped mobile manipulator, where real-world experiments qualitatively validate robust multimodal perception and uncertainty reduction under ambiguous instructions.
comment: 10 pages, 8 figures, accepted at ACM/IEEE International Conference on Human-Robot Interaction (HRI 2026)
☆ Selecting Spots by Explicitly Predicting Intention from Motion History Improves Performance in Autonomous Parking
In many applications of social navigation, existing works have shown that predicting and reasoning about human intentions can help robotic agents make safer and more socially acceptable decisions. In this work, we study this problem for autonomous valet parking (AVP), where an autonomous vehicle ego agent must drop off its passengers, explore the parking lot, find a parking spot, negotiate for the spot with other vehicles, and park in the spot without human supervision. Specifically, we propose an AVP pipeline that selects parking spots by explicitly predicting where other agents are going to park from their motion history using learned models and probabilistic belief maps. To test this pipeline, we build a simulation environment with reactive agents and realistic modeling assumptions on the ego agent, such as occlusion-aware observations, and imperfect trajectory prediction. Simulation experiments show that our proposed method outperforms existing works that infer intentions from future predicted motion or embed them implicitly in end-to-end models, yielding better results in prediction accuracy, social acceptance, and task completion. Our key insight is that, in parking, where driving regulations are more lax, explicit intention prediction is crucial for reasoning about diverse and ambiguous long-term goals, which cannot be reliably inferred from short-term motion prediction alone, but can be effectively learned from motion history.
comment: 8 pages, 4 figures
☆ EmboAlign: Aligning Video Generation with Compositional Constraints for Zero-Shot Manipulation
Video generative models (VGMs) pretrained on large-scale internet data can produce temporally coherent rollout videos that capture rich object dynamics, offering a compelling foundation for zero-shot robotic manipulation. However, VGMs often produce physically implausible rollouts, and converting their pixel-space motion into robot actions through geometric retargeting further introduces cumulative errors from imperfect depth estimation and keypoint tracking. To address these challenges, we present \method{}, a data-free framework that aligns VGM outputs with compositional constraints generated by vision-language models (VLMs) at inference time. The key insight is that VLMs offer a capability complementary to VGMs: structured spatial reasoning that can identify the physical constraints critical to the success and safety of manipulation execution. Given a language instruction, \method{} uses a VLM to automatically extract a set of compositional constraints capturing task-specific requirements, which are then applied at two stages: (1) constraint-guided rollout selection, which scores and filters a batch of VGM rollouts to retain the most physically plausible candidate, and (2) constraint-based trajectory optimization, which uses the selected rollout as initialization and refines the robot trajectory under the same constraint set to correct retargeting errors. We evaluate \method{} on six real-robot manipulation tasks requiring precise, constraint-sensitive execution, improving the overall success rate by 43.3\% points over the strongest baseline without any task-specific training data.
☆ Safe-Night VLA: Seeing the Unseen via Thermal-Perceptive Vision-Language-Action Models for Safety-Critical Manipulation
Current Vision-Language-Action (VLA) models rely primarily on RGB perception, preventing them from capturing modalities such as thermal signals that are imperceptible to conventional visual sensors. Moreover, end-to-end generative policies lack explicit safety constraints, making them fragile when encountering obstacles and novel scenarios outside the training distribution. To address these limitations, we propose Safe-Night VLA, a multimodal manipulation framework that enables robots to see the unseen while enforcing rigorous safety constraints for thermal-aware manipulation in unstructured environments. Specifically, Safe-Night VLA integrates long-wave infrared thermal perception into a pre-trained vision-language backbone, enabling semantic reasoning grounded in thermodynamic properties. To ensure safe execution under out-of-distribution conditions, we incorporate a safety filter via control barrier functions, which provide deterministic workspace constraint enforcement during policy execution. We validate our framework through real-world experiments on a Franka manipulator, introducing a novel evaluation paradigm featuring temperature-conditioned manipulation, subsurface target localization, and reflection disambiguation, while maintaining constrained execution at inference time. Results demonstrate that Safe-Night VLA outperforms RGB-only baselines and provide empirical evidence that foundation models can effectively leverage non-visible physical modalities for robust manipulation.
☆ Vision-Language System using Open-Source LLMs for Gestures in Medical Interpreter Robots
Effective communication is vital in healthcare, especially across language barriers, where non-verbal cues and gestures are critical. This paper presents a privacy-preserving vision-language framework for medical interpreter robots that detects specific speech acts (consent and instruction) and generates corresponding robotic gestures. Built on locally deployed open-source models, the system utilizes a Large Language Model (LLM) with few-shot prompting for intent detection. We also introduce a novel dataset of clinical conversations annotated for speech acts and paired with gesture clips. Our identification module achieved 0.90 accuracy, 0.93 weighted precision, and a 0.91 weighted F1-Score. Our approach significantly improves computational efficiency and, in user studies, outperforms the speech-gesture generation baseline in human-likeness while maintaining comparable appropriateness.
☆ Environment-Aware Path Generation for Robotic Additive Manufacturing of Structures
Robotic Additive Manufacturing (AM) has emerged as a scalable and customizable construction method in the last decade. However, current AM design methods rely on pre-conceived (A priori) toolpath of the structure, often developed via offline slicing software. Moreover, considering the dynamic construction environments involving obstacles on terrestrial and extraterrestrial environments, there is a need for online path generation methods. Here, an environment-aware path generation framework (PGF) is proposed for the first time in which structures are designed in an online fashion by utilizing four path planning (PP) algorithms (two search-based and two sampling-based). To evaluate the performance of the proposed PGF in different obstacle arrangements (periodic, random) for two types of structures (closed and open), structural (path roughness, turns, offset, Root Mean Square Error (RMSE), deviation) and computational (run time) performance metrics are developed. Most challenging environments (i.e., dense with high number of obstacles) are considered to saturate the feasibility limits of PP algorithms. The capability of each of the four path planners used in the PGF in finding a feasible path is assessed. Finally, the effectiveness of the proposed structural performance metrics is evaluated individually and comparatively, and most essential metrics necessary for evaluation of toolpath of the resulting structures are prescribed. Consequently, the most promising path planners in challenging environments are identified for robotic additive manufacturing applications.
☆ Introducing the transitional autonomous vehicle lane-changing dataset: Empirical Experiments
Transitional autonomous vehicles (tAVs), which operate beyond SAE Level 1-2 automation but short of full autonomy, are increasingly sharing the road with human-driven vehicles (HDVs). As these systems interact during complex maneuvers such as lane changes, new patterns may emerge with implications for traffic stability and safety. Assessing these dynamics, particularly during mandatory lane changes, requires high-resolution trajectory data, yet datasets capturing tAV lane-changing behavior are scarce. This study introduces the North Carolina Transitional Autonomous Vehicle Lane-Changing (NC-tALC) Dataset, a high-fidelity trajectory dataset designed to characterize tAV interactions during lane-changing maneuvers. The dataset includes two controlled experimental series. In the first, tAV lane-changing experiments, a tAV executes lane changes in the presence of adaptive cruise control (ACC) equipped target vehicles, enabling analysis of lane-changing execution. In the second, tAV responding experiments, two tAVs act as followers and respond to cut-in maneuvers initiated by another tAV, enabling analysis of follower response dynamics. The dataset contains 152 trials (72 lane-changing and 80 responding trials) sampled at 20 Hz with centimeter-level RTK-GPS accuracy. The NC-tALC dataset provides a rigorous empirical foundation for evaluating tAV decision-making and interaction dynamics in controlled mandatory lane-changing scenarios.
☆ Contact-Grounded Policy: Dexterous Visuotactile Policy with Generative Contact Grounding
Contact-Grounded Policy (CGP) enables fine-grained, contact-rich dexterous manipulation by grounding multi-point contacts through predicting the actual robot state and tactile feedback, and by using a learned contact-consistency mapping to convert these predictions into controller-executable targets for a compliance controller. CGP supports both dense tactile arrays and vision-based tactile sensors mounted on the hand. We collect demonstrations via teleoperation in both simulation and on a physical robot, and evaluate CGP across multiple dexterous manipulation tasks.
☆ TransMASK: Masked State Representation through Learned Transformation
Humans train robots to complete tasks in one environment, and expect robots to perform those same tasks in new environments. As humans, we know which aspects of the environment (i.e., the state) are relevant to the task. But there are also things that do not matter; e.g., the color of the table or the presence of clutter in the background. Ideally, the robot's policy learns to ignore these irrelevant state components. Achieving this invariance improves generalization: the robot knows not to factor irrelevant variables into its control decisions, making the policy more robust to environment changes. In this paper we therefore propose a self-supervised method to learn a mask which, when multiplied by the observed state, transforms that state into a latent representation that is biased towards relevant elements. Our method -- which we call TransMASK -- can be combined with a variety of imitation learning frameworks (such as diffusion policies) without any additional labels or alterations to the loss function. To achieve this, we recognize that the learned policy updates to better match the human's true policy. This true policy only depends on the relevant parts of the state; hence, as the gradients pass back through the learned policy and our proposed mask, they increase the value for elements that cause the robot to better imitate the human. We can therefore train TransMASK at the same time as we learn the policy. By normalizing the magnitude of each row in TransMASK, we force the mask to align with the Jacobian of the expert policy: columns that correspond to relevant states have large magnitudes, while columns for irrelevant states approach zero magnitude. We compare our approach to other methods that extract relevant states for downstream imitation learning. See our project website: https://collab.me.vt.edu/TransMASK/
☆ Relational Semantic Reasoning on 3D Scene Graphs for Open World Interactive Object Search
Open-world interactive object search in household environments requires understanding semantic relationships between objects and their surrounding context to guide exploration efficiently. Prior methods either rely on vision-language embeddings similarity, which does not reliably capture task-relevant relational semantics, or large language models (LLMs), which are too slow and costly for real-time deployment. We introduce SCOUT: Scene Graph-Based Exploration with Learned Utility for Open-World Interactive Object Search, a novel method that searches directly over 3D scene graphs by assigning utility scores to rooms, frontiers, and objects using relational exploration heuristics such as room-object containment and object-object co-occurrence. To make this practical without sacrificing open-vocabulary generalization, we propose an offline procedural distillation framework that extracts structured relational knowledge from LLMs into lightweight models for on-robot inference. Furthermore, we present SymSearch, a scalable symbolic benchmark for evaluating semantic reasoning in interactive object search tasks. Extensive evaluations across symbolic and simulation environments show that SCOUT outperforms embedding similarity-based methods and matches LLM-level performance while remaining computationally efficient. Finally, real-world experiments demonstrate effective transfer to physical environments, enabling open-world interactive object search under realistic sensing and navigation constraints.
☆ RFM-HRI : A Multimodal Dataset of Medical Robot Failure, User Reaction and Recovery Preferences for Item Retrieval Tasks
While robots deployed in real-world environments inevitably experience interaction failures, understanding how users respond through verbal and non-verbal behaviors remains under-explored in human-robot interaction (HRI). This gap is particularly significant in healthcare-inspired settings, where interaction failures can directly affect task performance and user trust. We present the Robot Failures in Medical HRI (RFM-HRI) Dataset, a multimodal dataset capturing dyadic interactions between humans and robots embodied in crash carts, where communication failures are systematically induced during item retrieval tasks. Through Wizard-of-Oz studies with 41 participants across laboratory and hospital settings, we recorded responses to four failure types (speech, timing, comprehension, and search) derived from three years of crash-cart robot interaction data. The dataset contains 214 interaction samples including facial action units, head pose, speech transcripts, and post-interaction self-reports. Our analysis shows that failures significantly degrade affective valence and reduce perceived control compared to successful interactions. Failures are strongly associated with confusion, annoyance, and frustration, while successful interactions are characterized by surprise, relief, and confidence in task completion. Emotional responses also evolve across repeated failures, with confusion decreasing and frustration increasing over time. This work contributes (1) a publicly available multimodal dataset (RFM-HRI), (2) analysis of user responses to different failure types and preferred recovery strategies, and (3) a crash-cart retrieval scenario enabling systematic comparison of recovery strategies with implications for safety-critical failure recovery. Our findings provide a foundation for failure detection and recovery methods in embodied HRI.
☆ Control Lyapunov Functions for Underactuated Soft Robots
Soft and soft-rigid hybrid robots are inherently underactuated and operate under tight actuator limits, making task-space control with stability guarantees challenging. Common nonlinear strategies for soft robots (e.g., those based on PD control) often rely on the assumption of full actuation with no actuator limits. This paper presents a general control framework for task-space regulation and tracking of underactuated soft robots under bounded inputs. The method enforces a rapidly exponentially stabilizing control Lyapunov function as a convex inequality constraint while simultaneously satisfying underactuated full-body dynamics and actuator bounds. We validate the approach in simulation on several platforms spanning increasing underactuation: a simple two link tendon-driven "finger", a trimmed helicoid manipulator, and a highly underactuated spiral robot. We compare against a number of baseline methods from the literature. Results show improved task-space accuracy and consistent Lyapunov convergence under input limits, achieving superior set-point and trajectory-tracking performance.
comment: 8 pages, 5 figures, 2 tables. Submitted for publication to a conference
☆ RACAS: Controlling Diverse Robots With a Single Agentic System
Many robotic platforms expose an API through which external software can command their actuators and read their sensors. However, transitioning from these low-level interfaces to high-level autonomous behaviour requires a complicated pipeline, whose components demand distinct areas of expertise. Existing approaches to bridging this gap either require retraining for every new embodiment or have only been validated across structurally similar platforms. We introduce RACAS (Robot-Agnostic Control via Agentic Systems), a cooperative agentic architecture in which three LLM/VLM-based modules (Monitors, a Controller, and a Memory Curator) communicate exclusively through natural language to provide closed-loop robot control. RACAS requires only a natural language description of the robot, a definition of available actions, and a task specification; no source code, model weights, or reward functions need to be modified to move between platforms. We evaluate RACAS on several tasks using a wheeled ground robot, a recently published novel multi-jointed robotic limb, and an underwater vehicle. RACAS consistently solved all assigned tasks across these radically different platforms, demonstrating the potential of agentic AI to substantially reduce the barrier to prototyping robotic solutions.
comment: 7 pages in main text + 1 page of appendices + 1 page of references, 5 figures in main text + 1 figure in appendices, 2 tables in main text
☆ From Decoupled to Coupled: Robustness Verification for Learning-based Keypoint Detection with Joint Specifications
Keypoint detection underpins many vision tasks, including pose estimation, viewpoint recovery, and 3D reconstruction, yet modern neural models remain vulnerable to small input perturbations. Despite its importance, formal robustness verification for keypoint detectors is largely unexplored due to high-dimensional inputs and continuous coordinate outputs. We propose the first coupled robustness verification framework for heatmap-based keypoint detectors that bounds the joint deviation across all keypoints, capturing their interdependencies and downstream task requirements. Unlike prior decoupled, classification-style approaches that verify each keypoint independently and yield conservative guarantees, our method verifies collective behavior. We formulate verification as a falsification problem using a mixed-integer linear program (MILP) that combines reachable heatmap sets with a polytope encoding joint deviation constraints. Infeasibility certifies robustness, while feasibility provides counterexamples, and we prove the method is sound: if it certifies the model as robust, then the keypoint detection model is guaranteed to be robust. Experiments show that our coupled approach achieves high verified rates and remains effective under strict error thresholds where decoupled methods fail.
comment: 21 pages, 4 figures, 9 tables. arXiv admin note: text overlap with arXiv:2408.00117
☆ Task Parameter Extrapolation via Learning Inverse Tasks from Forward Demonstrations
Generalizing skill policies to novel conditions remains a key challenge in robot learning. Imitation learning methods, while data-efficient, are largely confined to the training region and consistently fail on input data outside it, leading to unpredictable policy failures. Alternatively, transfer learning approaches offer methods for trajectory generation robust to both changes in environment or tasks, but they remain data-hungry and lack accuracy in zero-shot generalization. We address these challenges by framing the problem in the context of task inversion learning and proposing a novel joint learning approach to achieve accurate and efficient knowledge transfer. Our method constructs a common representation of the forward and inverse tasks, and leverages auxiliary forward demonstrations from novel configurations to successfully execute the corresponding inverse tasks, without any direct supervision. We show the extrapolation capabilities of our framework via ablation studies and experiments in simulated and real-world environments that require complex manipulation skills with a diverse set of objects and tools, where we outperform diffusion-based alternatives.
☆ PRISM: Personalized Refinement of Imitation Skills for Manipulation via Human Instructions
This paper presents PRISM: an instruction-conditioned refinement method for imitation policies in robotic manipulation. This approach bridges Imitation Learning (IL) and Reinforcement Learning (RL) frameworks into a seamless pipeline, such that an imitation policy on a broad generic task, generated from a set of user-guided demonstrations, can be refined through reinforcement to generate new unseen fine-grain behaviours. The refinement process follows the Eureka paradigm, where reward functions for RL are iteratively generated from an initial natural-language task description. Presented approach, builds on top of this mechanism to adapt a refined IL policy of a generic task to new goal configurations and the introduction of constraints by adding also human feedback correction on intermediate rollouts, enabling policy reusability and therefore data efficiency. Results for a pick-and-place task in a simulated scenario show that proposed method outperforms policies without human feedback, improving robustness on deployment and reducing computational burden.
comment: 10 pages, 3 figures, Accepted for publication at European Robotics Forum 2026
♻ ☆ CBF-RL: Safety Filtering Reinforcement Learning in Training with Control Barrier Functions
Reinforcement learning (RL), while powerful and expressive, can often prioritize performance at the expense of safety. Yet safety violations can lead to catastrophic outcomes in real-world deployments. Control Barrier Functions (CBFs) offer a principled method to enforce dynamic safety -- traditionally deployed online via safety filters. While the result is safe behavior, the fact that the RL policy does not have knowledge of the CBF can lead to conservative behaviors. This paper proposes CBF-RL, a framework for generating safe behaviors with RL by enforcing CBFs in training. CBF-RL has two key attributes: (1) minimally modifying a nominal RL policy to encode safety constraints via a CBF term, (2) and safety filtering of the policy rollouts in training. Theoretically, we prove that continuous-time safety filters can be deployed via closed-form expressions on discrete-time roll-outs. Practically, we demonstrate that CBF-RL internalizes the safety constraints in the learned policy -- both enforcing safer actions and biasing towards safer rewards -- enabling safe deployment without the need for an online safety filter. We validate our framework through ablation studies on navigation tasks and on the Unitree G1 humanoid robot, where CBF-RL enables safer exploration, faster convergence, and robust performance under uncertainty, enabling the humanoid robot to avoid obstacles and climb stairs safely in real-world settings without a runtime safety filter.
comment: 8 pages
♻ ☆ SpikeATac: A Multimodal Tactile Finger with Taxelized Dynamic Sensing for Dexterous Manipulation ICRA 2026
In this work, we introduce SpikeATac, a multimodal tactile finger combining a taxelized and highly sensitive dynamic response (PVDF) with a static transduction method (capacitive) for multimodal touch sensing. Named for its `spiky' response, SpikeATac's 16-taxel PVDF film sampled at 4 kHz provides fast, sensitive dynamic signals to the very onset and breaking of contact. We characterize the sensitivity of the different modalities, and show that SpikeATac provides the ability to stop quickly and delicately when grasping fragile, deformable objects. Beyond parallel grasping, we show that SpikeATac can be used in a learning-based framework to achieve new capabilities on a dexterous multifingered robot hand. We use a learning recipe that combines reinforcement learning from human feedback with tactile-based rewards to fine-tune the behavior of a policy to modulate force. Our hardware platform and learning pipeline together enable a difficult dexterous and contact-rich task that has not previously been achieved: in-hand manipulation of fragile objects. Videos are available at https://roamlab.github.io/spikeatac/ .
comment: 8 pages, 8 figures, ICRA 2026
♻ ☆ Quadrotor Navigation using Reinforcement Learning with Privileged Information
This paper presents a reinforcement learning-based quadrotor navigation method that leverages efficient differentiable simulation, novel loss functions, and privileged information to navigate around large obstacles. Prior learning-based methods perform well in scenes that exhibit narrow obstacles, but struggle when the goal location is blocked by large walls or terrain. In contrast, the proposed method utilizes time-of-arrival (ToA) maps as privileged information and a yaw alignment loss to guide the robot around large obstacles. The policy is evaluated in photo-realistic simulation environments containing large obstacles, sharp corners, and dead-ends. Our approach achieves an 86% success rate and outperforms baseline strategies by 34%. We deploy the policy onboard a custom quadrotor in outdoor cluttered environments both during the day and night. The policy is validated across 20 flights, covering 589 meters without collisions at speeds up to 4 m/s.
♻ ☆ Ask, Reason, Assist: Robot Collaboration via Natural Language and Temporal Logic
Increased robot deployment, such as in warehousing, has revealed a need for collaboration among heterogeneous robot teams to resolve unforeseen conflicts. To this end, we propose a peer-to-peer coordination protocol that enables robots to request and provide help without a central task allocator. The process begins when a robot detects a conflict and uses a Large Language Model (LLM) to decide whether external assistance is required. If so, it crafts and broadcasts a natural language (NL) help request. Potential helper robots reason over the request and respond with offers of assistance, including information about the effect on their ongoing tasks. Helper reasoning is implemented via an LLM grounded in Signal Temporal Logic (STL) using a Backus-Naur Form (BNF) grammar, ensuring syntactically valid NL-to-STL translations, which are then solved as a Mixed Integer Linear Program (MILP). Finally, the requester robot selects a helper by reasoning over the expected increase in system-level total task completion time. We evaluated our framework through experiments comparing different helper-selection strategies and found that considering multiple offers allows the requester to minimize added makespan. Our approach significantly outperforms heuristics such as selecting the nearest available candidate helper robot, and achieves performance comparable to a centralized "Oracle" baseline but without heavy information demands.
comment: arXiv admin note: substantial text overlap with arXiv:2505.13376
♻ ☆ ROVER: Regulator-Driven Robust Temporal Verification of Black-Box Robot Policies
We present a novel, regulator-driven approach for the temporal verification of black-box autonomous robot policies, inspired by real-world certification processes where regulators often evaluate observable behavior without access to model internals. Central to our method is a regulator-in-the-loop approach that evaluates execution traces from black-box policies against temporal safety requirements. These requirements, expressed as prioritized Signal Temporal Logic (STL) specifications, characterize behavior changes over time and encode domain knowledge into the verification process. We use Total Robustness Value (TRV) and Largest Robustness Value (LRV) to quantify average performance and worst-case adherence, and introduce Average Violation Robustness Value (AVRV) to measure average specification violation. Together, these metrics guide targeted retraining and iterative model improvement. Our approach accommodates diverse temporal safety requirements (e.g., lane-keeping, delayed acceleration, and turn smoothness), capturing persistence, sequencing, and response across two distinct domains (virtual racing game and mobile robot navigation). Across six STL specifications in both scenarios, regulator-guided retraining increased satisfaction rates by an average of 43.8%, with consistent improvement in average performance (TRV) and reduced violation severity (LRV) in half of the specifications. Finally, real-world validation on a TurtleBot3 robot demonstrates a 27% improvement in smooth-navigation satisfaction, yielding smoother paths and stronger compliance with STL-defined temporal safety requirements.
♻ ☆ LHM-Humanoid: Learning a Unified Policy for Long-Horizon Humanoid Whole-Body Loco-Manipulation in Diverse Messy Environments
We introduce LHM-Humanoid, a benchmark and learning framework for long-horizon whole-body humanoid loco-manipulation in diverse, cluttered scenes. In our setting, multiple objects are displaced from their intended locations and may obstruct navigation; a humanoid agent must repeatedly (i) walk to a target, (ii) pick it up with diverse whole-body postures under balance constraints, (iii) carry it while navigating around obstacles, and (iv) place it at a designated goal -- all within a single continuous episode and without any environment reset. This task simultaneously demands cross-scene generalization and unified one-policy control: layouts, obstacle arrangements, object category/mass/shape/color and object start/goal poses vary substantially even within a room category, requiring a single general policy that directly outputs actions rather than invoking pre-trained skill libraries. Our dataset spans four room types (bedroom, living room, kitchen, and warehouse), comprising 350 diverse scenes/tasks with 79 objects (25 movable targets). Since no scene-specific ground-truth motion sequences are provided, we learn goal-conditioned teacher policies via reinforcement learning and distill them into a single end-to-end student policy using DAgger. We further distill this unified policy into a vision-language-action (VLA) model driven by egocentric RGB observations and natural language. Experiments in Isaac Gym demonstrate that LHM-Humanoid substantially outperforms end-to-end RL baselines and prior humanoid loco-manipulation methods on both seen and unseen scenes, exhibiting strong long-horizon robustness and cross-scene generalization.
♻ ☆ Conflict-Based Search as a Protocol: A Multi-Agent Motion Planning Protocol for Heterogeneous Agents, Solvers, and Independent Tasks ICRA 2026
Imagine the future construction site, hospital, or office with dozens of robots bought from different manufacturers. How can we enable these different robots to effectively move in a shared environment, given that each robot may have its own independent motion planning system? This work shows how we can get efficient collision-free movements between algorithmically heterogeneous agents by using Conflict-Based Search (Sharon et al. 2015) as a protocol. At its core, the CBS Protocol requires one specific single-agent motion planning API; finding a collision-free path that satisfies certain space-time constraints. Given such an API, CBS uses a central planner to find collision-free paths - independent of how the API is implemented. We demonstrate how this protocol enables multi-agent motion planning for a heterogeneous team of agents completing independent tasks with a variety of single-agent planners including: Heuristic Search (e.g., A*), Sampling Based Search (e.g., RRT), Optimization (e.g., Direct Collocation), Diffusion, and Reinforcement Learning.
comment: Published at ICRA 2026, Project webpage: https://rishi-v.github.io/CBS-Protocol/
♻ ☆ Kinodynamic Task and Motion Planning using VLM-guided and Interleaved Sampling
Task and Motion Planning (TAMP) integrates high-level task planning with low-level motion feasibility, but existing methods are costly in long-horizon problems due to excessive motion sampling. While LLMs provide commonsense priors, they lack 3D spatial reasoning and cannot ensure geometric or dynamic feasibility. We propose a kinodynamic TAMP planner based on a hybrid state tree that uniformly represents symbolic and numeric states during planning, enabling task and motion decisions to be jointly decided. Kinodynamic constraints embedded in the TAMP problem are verified by an off-the-shelf motion planner and physics simulator, and a VLM guides exploring a TAMP solution and backtracks the search based on visual rendering of the states. Experiments on the simulated domains and in the real world show 32.14% - 1166.67% increased average success rates compared to traditional and LLM-based TAMP planners and reduced planning time on complex problems, with ablations further highlighting the benefits of VLM backtracking. More details are available at https://graphics.ewha.ac.kr/kinodynamicTAMP/.
♻ ☆ Runge-Kutta Approximations for Direct Coning Compensation Applying Lie Theory
The integration of gyroscope measurements is an essential task for most navigation systems. Modern vehicles typically use strapdown systems, such that gyro integration requires coning compensation to account for the sensor's rotation during the integration. Many coning compensation algorithms have been developed and a few are reviewed. This work introduces a new class of coning correction algorithm built directly from the classical Runge-Kutta integration routines. A simple case is shown to collapse to one of the most popular coning algorithms and a clear procedure for generating higher-order algorithms is presented.
comment: Accepted manuscript. AIAA JGCD
♻ ☆ Diffusion-Based Impedance Learning for Contact-Rich Manipulation Tasks
Learning-based methods excel at robot motion generation but remain limited in contact-rich physical interaction. Impedance control provides stable and safe contact behavior but requires task-specific tuning of stiffness and damping parameters. We present Diffusion-Based Impedance Learning, a framework that bridges these paradigms by combining generative modeling with energy-consistent impedance control. A Transformer-based Diffusion Model, conditioned via cross-attention on measured external wrenches, reconstructs simulated Zero-Force Trajectories (sZFTs) that represent contact-consistent equilibrium behavior. A SLERP-based quaternion noise scheduler preserves geometric consistency for rotations on the unit sphere. The reconstructed sZFT is used by an energy-based estimator to adapt impedance online through directional stiffness and damping modulation. Trained on parkour and robot-assisted therapy demonstrations collected via Apple Vision Pro teleoperation, the model achieves sub-millimeter positional and sub-degree rotational accuracy using only tens of thousands of samples. Deployed in realtime torque control on a KUKA LBR iiwa, the approach enables smooth obstacle traversal and generalizes to unseen tasks, achieving 100% success in multi-geometry peg-in-hole insertion.
comment: 15 pages, 12 figures
♻ ☆ Viewpoint Matters: Dynamically Optimizing Viewpoints with Masked Autoencoder for Visual Manipulation
Robotic manipulation continues to be a challenge, and imitation learning (IL) enables robots to learn tasks from expert demonstrations. Current IL methods typically rely on fixed camera setups, where cameras are manually positioned in static locations, imposing significant limitations on adaptability and coverage. Inspired by human active perception, where humans dynamically adjust their viewpoint to capture the most relevant and least noisy information, we propose MAE-Select, a novel framework for active viewpoint selection in single-camera robotic systems. MAE-Select fully leverages pre-trained multi-view masked autoencoder representations and dynamically selects the next most informative viewpoint at each time chunk without requiring labeled viewpoints. Extensive experiments demonstrate that MAE-Select improves the capabilities of single-camera systems and, in some cases, even surpasses multi-camera setups. The project will be available at https://mae-select.github.io.
♻ ☆ PeRoI: A Pedestrian-Robot Interaction Dataset for Learning Avoidance, Neutrality, and Attraction Behaviors in Social Navigation
Robots are increasingly being deployed in public spaces such as shopping malls, sidewalks, and hospitals, where safe and socially aware navigation depends on anticipating how pedestrians respond to their presence. However, existing datasets rarely capture the full spectrum of robot-induced reactions, e.g., avoidance, neutrality, attraction, which limits progress in modeling these interactions. In this paper, we present the Pedestrian-Robot Interaction~(PeRoI) dataset that captures pedestrian motions categorized into attraction, neutrality, and repulsion across two outdoor sites under three controlled conditions: no robot present, with stationary robot, and with moving robot. This design explicitly reveals how pedestrian behavior varies across robot contexts, and we provide qualitative and quantitative comparisons to established state-of-the-art datasets. Building on these data, we propose the Neural Robot Social Force Model~(NeuRoSFM), an extension of the Social Force Model that integrates neural networks to augment inter-human dynamics with learned components and explicit robot-induced forces to better predict pedestrian motion in vicinity of robots. We evaluate NeuRoSFM by generating trajectories on multiple real-world datasets. The results demonstrate improved modeling of pedestrian-robot interactions, leading to better prediction accuracy, and highlight the value of our dataset and method for advancing socially aware navigation strategies in human-centered environments.
♻ ☆ Vision Language Model-based Testing of Industrial Autonomous Mobile Robots
PAL Robotics, in Spain, builds a variety of Autonomous Mobile Robots (AMRs), which are deployed in diverse environments (e.g., warehouses, retail spaces, and offices), where they work alongside humans. Given that human behavior can be unpredictable and that AMRs may not have been trained to handle all possible unknown and uncertain behaviors, it is important to test AMRs under a wide range of human interactions to ensure their safe behavior. Moreover, testing in real environments with actual AMRs and humans is often costly, impractical, and potentially hazardous (e.g., it could result in human injury). To this end, we propose a Vision Language Model (VLM)-based testing approach (RVSG) for industrial AMRs developed together with PAL Robotics. Based on the functional and safety requirements, RVSG uses the VLM to generate diverse human behaviors that violate these requirements. We evaluated RVSG with several requirements and navigation routes in a simulator using the latest AMR from PAL Robotics. Our results show that, compared with the baseline, RVSG can effectively generate requirement-violating scenarios. Moreover, RVSG-generated scenarios increase variability in robot behavior, thereby helping reveal their uncertain behaviors.
♻ ☆ Least Restrictive Hyperplane Control Barrier Functions
Control Barrier Functions (CBFs) can provide provable safety guarantees for dynamic systems. However, finding a valid CBF for a system of interest is often non-trivial, especially for systems having low computational resources, higher-order dynamics, and moving close to obstacles of complex shape. A common solution to this problem is to use a purely distance-based CBF. In this paper, we study Hyperplane CBFs (H-CBFs), where a hyperplane separates the agent from the obstacle. First, we note that the common distance-based CBF is a special case of an H-CBF where the hyperplane is a supporting hyperplane of the obstacle that is orthogonal to a line between the agent and the obstacle. Then we show that a less conservative CBF can be found by optimising over the orientation of the supporting hyperplane, in order to find the Least Restrictive Hyperplane CBF. This enables us to maintain the safety guarantees while allowing controls that are closer to the desired ones, especially when moving fast and passing close to obstacles. We illustrate the approach on a double integrator dynamical system with acceleration constraints, moving through a group of arbitrarily shaped static and moving obstacles.
♻ ☆ Towards Exploratory and Focused Manipulation with Bimanual Active Perception: A New Problem, Benchmark and Strategy ICRA 2026
Recently, active vision has reemerged as an important concept for manipulation, since visual occlusion occurs more frequently when main cameras are mounted on the robot heads. We reflect on the visual occlusion issue and identify its essence as the absence of information useful for task completion. Inspired by this, we come up with the more fundamental problem of Exploratory and Focused Manipulation (EFM). The proposed problem is about actively collecting information to complete challenging manipulation tasks that require exploration or focus. As an initial attempt to address this problem, we establish the EFM-10 benchmark that consists of 4 categories of tasks that align with our definition (10 tasks in total). We further come up with a Bimanual Active Perception (BAP) strategy, which leverages one arm to provide active vision and another arm to provide force sensing while manipulating. Based on this idea, we collect a dataset named BAPData for the tasks in EFM-10. With the dataset, we successfully verify the effectiveness of the BAP strategy in an imitation learning manner. We hope that the EFM-10 benchmark along with the BAP strategy can become a cornerstone that facilitates future research towards this direction. Project website: EFManipulation.github.io.
comment: ICRA 2026
♻ ☆ EmboTeam: Grounding LLM Reasoning into Reactive Behavior Trees via PDDL for Embodied Multi-Robot Collaboration
In embodied artificial intelligence, enabling heterogeneous robot teams to execute long-horizon tasks from high-level instructions remains a critical challenge. While large language models (LLMs) show promise in instruction parsing and preliminary planning, they exhibit limitations in long-term reasoning and dynamic multi-robot coordination. We propose EmboTeam, a novel embodied multi-robot task planning framework that addresses these issues through a three-stage cascaded architecture: 1) It leverages an LLM to parse instructions and generate Planning Domain Definition Language (PDDL) problem descriptions, thereby transforming commands into formal planning problems; 2) It combines the semantic reasoning of LLMs with the search capabilities of a classical planner to produce optimized action sequences; 3) It compiles the resulting plan into behavior trees for reactive control. The framework supports dynamically sized heterogeneous robot teams via a shared blackboard mechanism for communication and state synchronization. To validate our approach, we introduce the MACE-THOR benchmark dataset, comprising 42 complex tasks across 8 distinct household layouts. Experiments show EmboTeam improves the task success rate from 12% to 55% and goal condition recall from 32% to 72% over the LaMMA-P baseline.
♻ ☆ 3D Dynamics-Aware Manipulation: Endowing Manipulation Policies with 3D Foresight ICRA 2026
The incorporation of world modeling into manipulation policy learning has pushed the boundary of manipulation performance. However, existing efforts simply model the 2D visual dynamics, which is insufficient for robust manipulation when target tasks involve prominent depth-wise movement. To address this, we present a 3D dynamics-aware manipulation framework that seamlessly integrates 3D world modeling and policy learning. Three self-supervised learning tasks (current depth estimation, future RGB-D prediction, 3D flow prediction) are introduced within our framework, which complement each other and endow the policy model with 3D foresight. Extensive experiments on simulation and the real world show that 3D foresight can greatly boost the performance of manipulation policies without sacrificing inference speed. Code is available at https://github.com/Stardust-hyx/3D-Foresight.
comment: ICRA 2026
♻ ☆ Infinite-Dimensional Closed-Loop Inverse Kinematics for Soft Robots via Neural Operators
For fully actuated rigid robots, kinematic inversion is a purely geometric problem, efficiently solved by closed-loop inverse kinematics (CLIK) schemes that compute joint configurations to position the robot body in space. For underactuated soft robots, however, not all configurations are attainable through control action, making kinematic inversion extremely challenging. Extensions of CLIK address this by introducing end-to-end mappings from actuation to task space for the controller to operate on, but typically assume finite dimensions of the underlying virtual configuration space. In this work, we formulate CLIK in the infinite-dimensional domain to reason about the entire soft robot shape while solving tasks. We do this by composing an actuation-to-shape map with a shape-to-task map, deriving the differential end-to-end kinematics via an infinite-dimensional chain rule, and thereby obtaining a Jacobian-based CLIK algorithm. Since this actuation-to-shape mapping is rarely available in closed form, we propose to learn it using differentiable neural operator networks. We first present an analytical study on a constant-curvature segment, and then apply the neural version of the algorithm to a three-fiber soft robotic arm whose underlying model relies on morphoelasticity and active filament theory.
♻ ☆ TEMPO-VINE: A Multi-Temporal Sensor Fusion Dataset for Localization and Mapping in Vineyards
In recent years, precision agriculture has been introducing groundbreaking innovations in the field, with a strong focus on automation. However, research studies in robotics and autonomous navigation often rely on controlled simulations or isolated field trials. The absence of a realistic common benchmark represents a significant limitation for the diffusion of robust autonomous systems under real complex agricultural conditions. Vineyards pose significant challenges due to their dynamic nature, and they are increasingly drawing attention from both academic and industrial stakeholders interested in automation. In this context, we introduce the TEMPO-VINE dataset, a large-scale multi-temporal dataset specifically designed for evaluating sensor fusion, simultaneous localization and mapping (SLAM), and place recognition techniques within operational vineyard environments. TEMPO-VINE is the first multi-modal public dataset that brings together data from heterogeneous LiDARs of different price levels, AHRS, RTK-GPS, and cameras in real trellis and pergola vineyards, with multiple rows exceeding 100 m in length. In this work, we address a critical gap in the landscape of agricultural datasets by providing researchers with a comprehensive data collection and ground truth trajectories in different seasons, vegetation growth stages, terrain and weather conditions. The sequence paths with multiple runs and revisits will foster the development of sensor fusion, localization, mapping and place recognition solutions for agricultural fields. The dataset, the processing tools and the benchmarking results are available on the webpage.
♻ ☆ FreeTacMan: Robot-free Visuo-Tactile Data Collection System for Contact-rich Manipulation
Enabling robots with contact-rich manipulation remains a pivotal challenge in robot learning, which is substantially hindered by the data collection gap, including its inefficiency and limited sensor setup. While prior work has explored handheld paradigms, their rod-based mechanical structures remain rigid and unintuitive, providing limited tactile feedback and posing challenges for operators. Motivated by the dexterity and force feedback of human motion, we propose FreeTacMan, a human-centric and robot-free data collection system for accurate and efficient robot manipulation. Concretely, we design a wearable gripper with visuo-tactile sensors for data collection, which can be worn by human fingers for intuitive control. A high-precision optical tracking system is introduced to capture end-effector poses while synchronizing visual and tactile feedback simultaneously. We leverage FreeTacMan to collect a large-scale multimodal dataset, comprising over 3000k paired visuo-tactile images with end-effector poses, 10k demonstration trajectories across 50 diverse contact-rich manipulation tasks. FreeTacMan achieves multiple improvements in data collection performance over prior works and enables effective policy learning from self-collected datasets. By open-sourcing the hardware and the dataset, we aim to facilitate reproducibility and support research in visuo-tactile manipulation.
♻ ☆ Responsibility and Engagement -- Evaluating Interactions in Social Robot Navigation ICRA
In Social Robot Navigation (SRN), the availability of meaningful metrics is crucial for evaluating trajectories from human-robot interactions. In the SRN context, such interactions often relate to resolving conflicts between two or more agents. Correspondingly, the shares to which agents contribute to the resolution of such conflicts are important. This paper builds on recent work, which proposed a Responsibility metric capturing such shares. We extend this framework in two directions: First, we model the conflict buildup phase by introducing a time normalization. Second, we propose the related Engagement metric, which captures how the agents' actions intensify a conflict. In a comprehensive series of simulated scenarios with dyadic, group and crowd interactions, we show that the metrics carry meaningful information about the cooperative resolution of conflicts in interactions. They can be used to assess behavior quality and foresightedness. We extensively discuss applicability, design choices and limitations of the proposed metrics.
comment: Accepted at the 2026 IEEE International Conference on Robotics & Automation (ICRA)
♻ ☆ EgoTraj-Bench: Towards Robust Trajectory Prediction Under Ego-view Noisy Observations
Reliable trajectory prediction from an ego-centric perspective is crucial for robotic navigation in human-centric environments. However, existing methods typically assume noiseless observation histories, failing to account for the perceptual artifacts inherent in first-person vision, such as occlusions, ID switches, and tracking drift. This discrepancy between training assumptions and deployment reality severely limits model robustness. To bridge this gap, we introduce EgoTraj-Bench, built upon TBD dataset, which is the first real-world benchmark that aligns noisy, first-person visual histories with clean, bird's-eye-view future trajectories, enabling robust learning under realistic perceptual constraints. Building on this benchmark, we propose BiFlow, a dual-stream flow matching model that concurrently denoises historical observations and forecasts future motion. To better model agent intent, BiFlow incorporates our EgoAnchor mechanism, which conditions the prediction decoder on distilled historical features via feature modulation. Extensive experiments show that BiFlow achieves state-of-the-art performance, reducing minADE and minFDE by 10-15% on average and demonstrating superior robustness. We anticipate that our benchmark and model will provide a critical foundation for robust real-world ego-centric trajectory prediction. The benchmark library is available at: https://github.com/zoeyliu1999/EgoTraj-Bench.
♻ ☆ Efficient Path Generation with Curvature Guarantees by Mollification
Path generation, the process of converting high-level mission specifications, such as sequences of waypoints from a path planner, into smooth, executable paths, is a fundamental challenge in mobile robotics. Most path following and trajectory tracking algorithms require the desired path to be defined by at least twice continuously differentiable functions to guarantee key properties such as global convergence, especially for nonholonomic robots like unicycles with speed constraints. Consequently, path generation methods must bridge the gap between convenient but non-differentiable planning outputs, such as piecewise linear segments, and the differentiability requirements imposed by downstream control algorithms. While techniques such as spline interpolation or optimization-based methods are commonly used to smooth non-differentiable paths or create feasible ones from sequences of waypoints, they either produce unnecessarily complex trajectories or are computationally expensive. In this work, we present a method to regularize non-differentiable functions and generate feasible paths through mollification. Specifically, we approximate an arbitrary path with a differentiable function that can converge to it with arbitrary precision. Additionally, we provide a systematic method for bounding the curvature of generated paths, which we demonstrate by applying it to paths resulting from linking a sequence of waypoints with segments. The proposed approach is analytically shown to be computationally more efficient than standard interpolation methods, enabling real-time implementation on microcontrollers, while remaining compatible with standard trajectory tracking and path following algorithms.
♻ ☆ RoboPARA: Dual-Arm Robot Planning with Parallel Allocation and Recomposition Across Tasks ICLR 2026
Dual-arm robots play a crucial role in improving efficiency and flexibility in complex multitasking scenarios.While existing methods have achieved promising results in task planning, they often fail to fully optimize task parallelism, limiting the potential of dual-arm collaboration.To address this issue, we propose RoboPARA, a novel large language model (LLM)-driven framework for dual-arm task parallelism planning.RoboPARA employs a two-stage process: (1) Dependency Graph-based Planning Candidates Generation, which constructs directed acyclic graphs (DAGs) to model task dependencies and eliminate redundancy, and (2) Graph Re-Traversal-based Dual-Arm Parallel Planning, which optimizes DAG traversal to maximize parallelism while maintaining task coherence.In addition, we introduce the Cross-Scenario Dual-Arm Parallel Task dataset (X-DAPT dataset), the first dataset specifically designed to evaluate dual-arm task parallelism across diverse scenarios and difficulty levels.Extensive experiments demonstrate that RoboPARA significantly outperforms existing planning methods, achieving higher efficiency and reliability, particularly in complex task combinations.Our code is publicly available at https://github.com/AiDuanshiying/RoboPARA.
comment: Accepted to ICLR 2026
♻ ☆ Risk-Aware Autonomous Driving with Linear Temporal Logic Specifications
Human drivers naturally balance the risks of different concerns while driving, including traffic rule violations, minor accidents, and fatalities. However, achieving the same behavior in autonomous driving systems remains an open problem. This paper extends a risk metric that has been verified in human-like driving studies to encompass more complex driving scenarios specified by linear temporal logic (LTL) that go beyond just collision risks. This extension incorporates the timing and severity of events into LTL specifications, thereby reflecting a human-like risk awareness. Without sacrificing expressivity for traffic rules, we adopt LTL specifications composed of safety and co-safety formulas, allowing the control synthesis problem to be reformulated as a reachability problem. By leveraging occupation measures, we further formulate a linear programming (LP) problem for this LTL-based risk metric. Consequently, the synthesized policy balances different types of driving risks, including both collision risks and traffic rule violations. The effectiveness of the proposed approach is validated by three typical traffic scenarios in Carla simulator.
♻ ☆ MOSAIC: Modular Scalable Autonomy for Intelligent Coordination of Heterogeneous Robotic Teams
Mobile robots have become indispensable for exploring hostile environments, such as in space or disaster relief scenarios, but often remain limited to teleoperation by a human operator. This restricts the deployment scale and requires near-continuous low-latency communication between the operator and the robot. We present MOSAIC: a scalable autonomy framework for multi-robot scientific exploration using a unified mission abstraction based on Points of Interest (POIs) and multiple layers of autonomy, enabling supervision by a single operator. The framework dynamically allocates exploration and measurement tasks based on each robot's capabilities, leveraging team-level redundancy and specialization to enable continuous operation. We validated the framework in a space-analog field experiment emulating a lunar prospecting scenario, involving a heterogeneous team of five robots and a single operator. Despite the complete failure of one robot during the mission, the team completed 82.3% of assigned tasks at an Autonomy Ratio of 86%, while the operator workload remained at only 78.2%. These results demonstrate that the proposed framework enables robust, scalable multi-robot scientific exploration with limited operator intervention. We further derive practical lessons learned in robot interoperability, networking architecture, team composition, and operator workload management to inform future multi-robot exploration missions.
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ LAP: Fast LAtent Diffusion Planner for Autonomous Driving
Diffusion models have demonstrated strong capabilities for modeling human-like driving behaviors in autonomous driving, but their iterative sampling process induces substantial latency, and operating directly on raw trajectory points forces the model to spend capacity on low-level kinematics, rather than high-level multi-modal semantics. To address these limitations, we propose LAtent Planner (LAP), a framework that plans in a VAE-learned latent space that disentangles high-level intents from low-level kinematics, enabling our planner to capture rich, multi-modal driving strategies. To bridge the representational gap between the high-level semantic planning space and the vectorized scene context, we introduce an intermediate feature alignment mechanism that facilitates robust information fusion. Notably, LAP can produce high-quality plans in one single denoising step, substantially reducing computational overhead. Through extensive evaluations on the large-scale nuPlan benchmark, LAP achieves state-of-the-art closed-loop performance among learning-based planning methods, while demonstrating an inference speed-up of at most 10x over previous SOTA approaches.
♻ ☆ Interpretable Multimodal Gesture Recognition for Drone and Mobile Robot Teleoperation via Log-Likelihood Ratio Fusion
Human operators are still frequently exposed to hazardous environments such as disaster zones and industrial facilities, where intuitive and reliable teleoperation of mobile robots and Unmanned Aerial Vehicles (UAVs) is essential. In this context, hands-free teleoperation enhances operator mobility and situational awareness, thereby improving safety in hazardous environments. While vision-based gesture recognition has been explored as one method for hands-free teleoperation, its performance often deteriorates under occlusions, lighting variations, and cluttered backgrounds, limiting its applicability in real-world operations. To overcome these limitations, we propose a multimodal gesture recognition framework that integrates inertial data (accelerometer, gyroscope, and orientation) from Apple Watches on both wrists with capacitive sensing signals from custom gloves. We design a late fusion strategy based on the log-likelihood ratio (LLR), which not only enhances recognition performance but also provides interpretability by quantifying modality-specific contributions. To support this research, we introduce a new dataset of 20 distinct gestures inspired by aircraft marshalling signals, comprising synchronized RGB video, IMU, and capacitive sensor data. Experimental results demonstrate that our framework achieves performance comparable to a state-of-the-art vision-based baseline while significantly reducing computational cost, model size, and training time, making it well suited for real-time robot control. We therefore underscore the potential of sensor-based multimodal fusion as a robust and interpretable solution for gesture-driven mobile robot and drone teleoperation.
♻ ☆ Balancing Progress and Safety: A Novel Risk-Aware Objective for RL in Autonomous Driving
Reinforcement Learning (RL) is a promising approach for achieving autonomous driving due to robust decision-making capabilities. RL learns a driving policy through trial and error in traffic scenarios, guided by a reward function that combines the driving objectives. The design of such reward function has received insufficient attention, yielding ill-defined rewards with various pitfalls. Safety, in particular, has long been regarded only as a penalty for collisions. This leaves the risks associated with actions leading up to a collision unaddressed, limiting the applicability of RL in real-world scenarios. To address these shortcomings, our work focuses on enhancing the reward formulation by defining a set of driving objectives and structuring them hierarchically. Furthermore, we discuss the formulation of these objectives in a normalized manner to transparently determine their contribution to the overall reward. Additionally, we introduce a novel risk-aware objective for various driving interactions based on a two-dimensional ellipsoid function and an extension of Responsibility-Sensitive Safety (RSS) concepts. We evaluate the efficacy of our proposed reward in unsignalized intersection scenarios with varying traffic densities. The approach decreases collision rates by 21\% on average compared to baseline rewards and consistently surpasses them in route progress and cumulative reward, demonstrating its capability to promote safer driving behaviors while maintaining high-performance levels.
comment: Accepted in the 36th IEEE Intelligent vehicles Symposium (IV 2025)
♻ ☆ Automatic Curriculum Learning for Driving Scenarios: Towards Robust and Efficient Reinforcement Learning
This paper addresses the challenges of training end-to-end autonomous driving agents using Reinforcement Learning (RL). RL agents are typically trained in a fixed set of scenarios and nominal behavior of surrounding road users in simulations, limiting their generalization and real-life deployment. While domain randomization offers a potential solution by randomly sampling driving scenarios, it frequently results in inefficient training and sub-optimal policies due to the high variance among training scenarios. To address these limitations, we propose an automatic curriculum learning framework that dynamically generates driving scenarios with adaptive complexity based on the agent's evolving capabilities. Unlike manually designed curricula that introduce expert bias and lack scalability, our framework incorporates a ``teacher'' that automatically generates and mutates driving scenarios based on their learning potential -- an agent-centric metric derived from the agent's current policy -- eliminating the need for expert design. The framework enhances training efficiency by excluding scenarios the agent has mastered or finds too challenging. We evaluate our framework in a reinforcement learning setting where the agent learns a driving policy from camera images. Comparative results against baseline methods, including fixed scenario training and domain randomization, demonstrate that our approach leads to enhanced generalization, achieving higher success rates: +9% in low traffic density, +21% in high traffic density, and faster convergence with fewer training steps. Our findings highlight the potential of ACL in improving the robustness and efficiency of RL-based autonomous driving agents.
comment: Accepted in the 36th IEEE Intelligent Vehicles Symposium (IV 2025)
♻ ☆ MachaGrasp: Morphology-Aware Cross-Embodiment Dexterous Hand Articulation Generation for Grasping
Dexterous grasping with multi-fingered hands remains challenging due to high-dimensional articulations and the cost of optimization-based pipelines. Existing end-to-end methods require training on large-scale datasets for specific hands, limiting their ability to generalize across different embodiments. We propose MachaGrasp, an eigengrasp-based, end-to-end framework for cross-embodiment grasp generation. From a hand's morphology description, we derive a morphology embedding and an eigengrasp set. Conditioned on these, together with the object point cloud and wrist pose, an amplitude predictor regresses articulation coefficients in a low-dimensional space, which are decoded into full joint articulations. Articulation learning is supervised with a Kinematic-Aware Articulation Loss (KAL) that emphasizes fingertip-relevant motions and injects morphology-specific structure. In simulation on unseen objects across three dexterous hands, MachaGrasp attains a 91.9% average grasp success rate with less than 0.4 seconds inference per grasp. With few-shot adaptation to an unseen hand, it achieves 85.6% success on unseen objects in simulation, and real-world experiments on this few-shot-generalized hand achieve an 87% success rate. The code and additional materials are available on our project website https://connor-zh.github.io/MachaGrasp.
♻ ☆ Distant Object Localisation from Noisy Image Segmentation Sequences
3D object localisation based on a sequence of camera measurements is essential for safety-critical surveillance tasks, such as drone-based wildfire monitoring. Localisation of objects detected with a camera can typically be solved with specialised sensor configurations or 3D scene reconstruction. However, in the context of distant objects or tasks limited by the amount of available computational resources, neither solution is feasible. In this paper, we show that the task can be solved with either multi-view triangulation or particle filters, with the latter also providing shape and uncertainty estimates. We studied the solutions using 3D simulation and drone-based image segmentation sequences with global navigation satellite system (GNSS) based camera pose estimates. The results suggest that combining the proposed methods with pre-existing image segmentation models and drone-carried computational resources yields a reliable system for drone-based wildfire monitoring. The proposed solutions are independent of the detection method, also enabling quick adaptation to similar tasks.
♻ ☆ Collaborative Learning of Local 3D Occupancy Prediction and Versatile Global Occupancy Mapping ICRA 2026
Vision-based 3D semantic occupancy prediction is vital for autonomous driving, enabling unified modeling of static infrastructure and dynamic agents. Global occupancy maps serve as long-term memory priors, providing valuable historical context that enhances local perception. This is particularly important in challenging scenarios such as occlusion or poor illumination, where current and nearby observations may be unreliable or incomplete. Priors aggregated from previous traversals under better conditions help fill gaps and enhance the robustness of local 3D occupancy prediction. In this paper, we propose Long-term Memory Prior Occupancy (LMPOcc), a plug-and-play framework that incorporates global occupancy priors to boost local prediction and simultaneously updates global maps with new observations. To realize the information gain from global priors, we design an efficient and lightweight Current-Prior Fusion module that adaptively integrates prior and current features. Meanwhile, we introduce a model-agnostic prior format to enable continual updating of global occupancy and ensure compatibility across diverse prediction baselines. LMPOcc achieves state-of-the-art local occupancy prediction performance validated on the Occ3D-nuScenes benchmark, especially on static semantic categories. Furthermore, we verify LMPOcc's capability to build large-scale global occupancy maps through multi-vehicle crowdsourcing, and utilize occupancy-derived dense depth to support the construction of 3D open-vocabulary maps. Our method opens up a new paradigm for continuous global information updating and storage, paving the way towards more comprehensive and scalable scene understanding in large outdoor environments.
comment: Accepted by ICRA 2026
♻ ☆ GUIDE: A Diffusion-Based Autonomous Robot Exploration Framework Using Global Graph Inference
Autonomous exploration in structured and complex indoor environments remains a challenging task, as existing methods often struggle to appropriately model unobserved space and plan globally efficient paths. To address these limitations, we propose GUIDE, a novel exploration framework that synergistically combines global graph inference with diffusion-based decision-making. We introduce a region-evaluation global graph representation that integrates both observed environmental data and predictions of unexplored areas, enhanced by a region-level evaluation mechanism to prioritize reliable structural inferences while discounting uncertain predictions. Building upon this enriched representation, a diffusion policy network generates stable, foresighted action sequences with significantly reduced denoising steps. Extensive simulations and real-world deployments demonstrate that GUIDE consistently outperforms state-of-the-art methods, achieving up to 18.3% faster coverage completion and a 34.9% reduction in redundant movements.
♻ ☆ Environment-Aware Learning of Smooth GNSS Covariance Dynamics for Autonomous Racing ICRA
Ensuring accurate and stable state estimation is a challenging task crucial to safety-critical domains such as high-speed autonomous racing, where measurement uncertainty must be both adaptive to the environment and temporally smooth for control. In this work, we develop a learning-based framework, LACE, capable of directly modeling the temporal dynamics of GNSS measurement covariance. We model the covariance evolution as an exponentially stable dynamical system where a deep neural network (DNN) learns to predict the system's process noise from environmental features through an attention mechanism. By using contraction-based stability and systematically imposing spectral constraints, we formally provide guarantees of exponential stability and smoothness for the resulting covariance dynamics. We validate our approach on an AV-24 autonomous racecar, demonstrating improved localization performance and smoother covariance estimates in challenging, GNSS-degraded environments. Our results highlight the promise of dynamically modeling the perceived uncertainty in state estimation problems that are tightly coupled with control sensitivity.
comment: 8 pages, Accepted to IEEE International Conference on Robotics and Automation (ICRA) 2026
♻ ☆ Seeing the Bigger Picture: 3D Latent Mapping for Mobile Manipulation Policy Learning ICRA 2026
In this paper, we demonstrate that mobile manipulation policies utilizing a 3D latent map achieve stronger spatial and temporal reasoning than policies relying solely on images. We introduce Seeing the Bigger Picture (SBP), an end-to-end policy learning approach that operates directly on a 3D map of latent features. In SBP, the map extends perception beyond the robot's current field of view and aggregates observations over long horizons. Our mapping approach incrementally fuses multiview observations into a grid of scene-specific latent features. A pre-trained, scene-agnostic decoder reconstructs target embeddings from these features and enables online optimization of the map features during task execution. A policy, trainable with behavior cloning or reinforcement learning, treats the latent map as a state variable and uses global context from the map obtained via a 3D feature aggregator. We evaluate SBP on scene-level mobile manipulation and sequential tabletop manipulation tasks. Our experiments demonstrate that SBP (i) reasons globally over the scene, (ii) leverages the map as long-horizon memory, and (iii) outperforms image-based policies in both in-distribution and novel scenes, e.g., improving the success rate by 15% for the sequential manipulation task.
comment: ICRA 2026, project page: https://existentialrobotics.org/sbp_page/
♻ ☆ NeuralRemaster: Phase-Preserving Diffusion for Structure-Aligned Generation
Standard diffusion corrupts data using Gaussian noise whose Fourier coefficients have random magnitudes and random phases. While effective for unconditional or text-to-image generation, corrupting phase components destroys spatial structure, making it ill-suited for tasks requiring geometric consistency, such as re-rendering, simulation enhancement, and image-to-image translation. We introduce Phase-Preserving Diffusion (φ-PD), a model-agnostic reformulation of the diffusion process that preserves input phase while randomizing magnitude, enabling structure-aligned generation without architectural changes or additional parameters. We further propose Frequency-Selective Structured (FSS) noise, which provides continuous control over structural rigidity via a single frequency-cutoff parameter. φ-PD adds no inference-time cost and is compatible with any diffusion model for images or videos. Across photorealistic and stylized re-rendering, as well as sim-to-real enhancement for driving planners, φ-PD produces controllable, spatially aligned results. When applied to the CARLA simulator, φ-PD significantly improves sim-to-real planner transfer performance. The method is complementary to existing conditioning approaches and broadly applicable to image-to-image and video-to-video generation. Videos, additional examples, and code are available on our \href{https://yuzeng-at-tri.github.io/ppd-page/}{project page}.
♻ ☆ Seeing Through Uncertainty: A Free-Energy Approach for Real-Time Perceptual Adaptation in Robust Visual Navigation
Navigation in the natural world is a feat of adaptive inference, where biological organisms maintain goal-directed behaviour despite noisy and incomplete sensory streams. Central to this ability is the Free Energy Principle (FEP), which posits that perception is a generative process where the brain minimises Variational Free Energy (VFE) to maintain accurate internal models of the world. While Deep Neural Networks (DNNs) have served as powerful analogues for biological brains, they typically lack the real-time plasticity required to handle abrupt sensory shifts. We introduce FEP-Nav, a biologically-inspired framework that implements real-time perceptual adaptation for robust visual navigation. By decomposing VFE into its constituent components--prediction error and Bayesian surprise--we propose a dual-mechanism architecture: a Top-down Decoder that provides an internal expectation of uncorrupted sensory input, and Adaptive Normalisation that dynamically aligns shifted feature distributions with prior beliefs. Theoretically, we demonstrate that this integration of reconstruction and normalisation provides a formal mechanism for minimising VFE during inference without the need for gradient-based updates. Evaluations across a diverse suite of simulated and real-world visual corruptions demonstrate that FEP-Nav facilitates a substantial recovery of navigation performance, consistently exceeding the capabilities of both non-adaptive baselines and strong adaptive methods. We show that bridging machine learning with the brain's variational principles offers a robust strategy for autonomous behaviour, enabling robots to remain functional under sensory conditions that typically degrade the performance of standard adaptive models.
♻ ☆ Whole-Body Safe Control of Robotic Systems with Koopman Neural Dynamics
Controlling robots with strongly nonlinear, high-dimensional dynamics remains challenging, as direct nonlinear optimization with safety constraints is often intractable in real time. The Koopman operator offers a way to represent nonlinear systems linearly in a lifted space, enabling the use of efficient linear control. We propose a data-driven framework that learns a Koopman embedding and operator from data, and integrates the resulting linear model with the Safe Set Algorithm (SSA). This allows the tracking and safety constraints to be solved in a single quadratic program (QP), ensuring feasibility and optimality without a separate safety filter. We validate the method on a Kinova Gen3 manipulator and a Go2 quadruped, showing accurate tracking and obstacle avoidance.
♻ ☆ Learning Agile Gate Traversal via Analytical Optimal Policy Gradient
Traversing narrow gates presents a significant challenge and has become a standard benchmark for evaluating agile and precise quadrotor flight. Traditional modularized autonomous flight stacks require extensive design and parameter tuning, while end-to-end reinforcement learning (RL) methods often suffer from low sample efficiency, limited interpretability, and degraded disturbance rejection under unseen perturbations. In this work, we present a novel hybrid framework that adaptively fine-tunes model predictive control (MPC) parameters online using outputs from a neural network (NN) trained offline. The NN jointly predicts a reference pose and cost function weights, conditioned on the coordinates of the gate corners and the current drone state. To achieve efficient training, we derive analytical policy gradients not only for the MPC module but also for an optimization-based gate traversal detection module. Hardware experiments demonstrate agile and accurate gate traversal with peak accelerations of $30\ \mathrm{m/s^2}$, as well as recovery within $0.85\ \mathrm{s}$ following body-rate disturbances exceeding $1146\ \mathrm{deg/s}$.
comment: 8 pages, 8 figures
♻ ☆ MarketGen: A Scalable Simulation Platform with Auto-Generated Embodied Supermarket Environments
The development of embodied agents for complex commercial environments is hindered by a critical gap in existing robotics datasets and benchmarks, which primarily focus on household or tabletop settings with short-horizon tasks. To address this limitation, we introduce MarketGen, a scalable simulation platform with automatic scene generation for complex supermarket environments. MarketGen features a novel agent-based Procedural Content Generation (PCG) framework. It uniquely supports multi-modal inputs (text and reference images) and integrates real-world design principles to automatically generate complete, structured, and realistic supermarkets. We also provide an extensive and diverse 3D asset library with a total of 1100+ supermarket goods and parameterized facilities assets. Building on this generative foundation, we propose a novel benchmark for assessing supermarket agents, featuring two daily tasks in a supermarket: (1) Checkout Unloading: long-horizon tabletop tasks for cashier agents, and (2) In-Aisle Item Collection: complex mobile manipulation tasks for salesperson agents. We validate our platform and benchmark through extensive experiments, including the deployment of a modular agent system and successful sim-to-real transfer. MarketGen provides a comprehensive framework to accelerate research in embodied AI for complex commercial applications.
comment: Project Page: https://xuhu0529.github.io/MarketGen
♻ ☆ Distributed UAV Formation Control Robust to Relative Pose Measurement Noise
A technique that allows a Formation-Enforcing Control (FEC) derived from graph rigidity theory to interface with a realistic relative localization system onboard lightweight Unmanned Aerial Vehicles (UAVs) is proposed in this paper. The proposed methodology enables reliable real-world deployment of UAVs in tight formations using relative localization systems burdened by non-negligible sensory noise. Such noise otherwise causes undesirable oscillations and drifts in sensor-based formations, and this effect is not sufficiently addressed in existing FEC algorithms. The proposed solution is based on decomposition of the gradient descent-based FEC command into interpretable elements, and then modifying these individually based on the estimated distribution of sensory noise, such that the resulting action limits the probability of overshooting the desired formation. The behavior of the system was analyzed and the practicality of the proposed solution was compared to pure gradient-descent in real-world experiments where it presented significantly better performance in terms of oscillations, deviation from the desired state
comment: Submitted to Robotics and Autonomous Systems journal on May 10. 2025 (Revision on February 27. 2026)
♻ ☆ DDP-WM: Disentangled Dynamics Prediction for Efficient World Models
World models are essential for autonomous robotic planning. However, the substantial computational overhead of existing dense Transformerbased models significantly hinders real-time deployment. To address this efficiency-performance bottleneck, we introduce DDP-WM, a novel world model centered on the principle of Disentangled Dynamics Prediction (DDP). We hypothesize that latent state evolution in observed scenes is heterogeneous and can be decomposed into sparse primary dynamics driven by physical interactions and secondary context-driven background updates. DDP-WM realizes this decomposition through an architecture that integrates efficient historical processing with dynamic localization to isolate primary dynamics. By employing a crossattention mechanism for background updates, the framework optimizes resource allocation and provides a smooth optimization landscape for planners. Extensive experiments demonstrate that DDP-WM achieves significant efficiency and performance across diverse tasks, including navigation, precise tabletop manipulation, and complex deformable or multi-body interactions. Specifically, on the challenging Push-T task, DDP-WM achieves an approximately 9 times inference speedup and improves the MPC success rate from 90% to98% compared to state-of-the-art dense models. The results establish a promising path for developing efficient, high-fidelity world models. Codes is available at https://hcplab-sysu.github.io/DDP-WM/.
comment: Efficient and high-fidelity world model. Code is available at https://hcplab-sysu.github.io/DDP-WM
♻ ☆ Learning Physical Systems: Symplectification via Gauge Fixing in Dirac Structures
Physics-informed deep learning has achieved remarkable progress by embedding geometric priors, such as Hamiltonian symmetries and variational principles, into neural networks, enabling structure-preserving models that extrapolate with high accuracy. However, in systems with dissipation and holonomic constraints, ubiquitous in legged locomotion and multibody robotics, the canonical symplectic form becomes degenerate, undermining the very invariants that guarantee stability and long-term prediction. In this work, we tackle this foundational limitation by introducing Presymplectification Networks (PSNs), the first framework to learn the symplectification lift via Dirac structures, restoring a non-degenerate symplectic geometry by embedding constrained systems into a higher-dimensional manifold. Our architecture combines a recurrent encoder with a flow-matching objective to learn the augmented phase-space dynamics end-to-end. We then attach a lightweight Symplectic Network (SympNet) to forecast constrained trajectories while preserving energy, momentum, and constraint satisfaction. We demonstrate our method on the dynamics of the ANYmal quadruped robot, a challenging contact-rich, multibody system. To the best of our knowledge, this is the first framework that effectively bridges the gap between constrained, dissipative mechanical systems and symplectic learning, unlocking a whole new class of geometric machine learning models, grounded in first principles yet adaptable from data.
comment: Presented at Equivariant Systems: Theory and Applications in State Estimation, Artificial Intelligence and Control, Robotics: Science and Systems (RSS) 2025 Workshop, 6 Pages, 3 Figures
♻ ☆ In-Hand Manipulation of Articulated Tools with Dexterous Robot Hands with Sim-to-Real Transfer
Reinforcement learning (RL) and sim-to-real transfer have advanced rigid-object manipulation. However, policies remain brittle for articulated mechanisms due to contact-rich dynamics that require both stable grasping and simultaneous free in-hand articulation. Furthermore, articulated objects and robot hands exhibit under-modeled joint phenomena such as friction, stiction, and backlash in real life that can increase the sim-to-real gap, and robot hands still fall short of idealized tactile sensing, both in terms of coverage, sensitivity, and specificity. In this paper, we present an original approach to learning dexterous in-hand manipulation of articulated tools that has reduced articulation and kinematic redundancy relative to the human hand. Our approach augments a simulation-trained base policy with a sensor-driven refinement learned from hardware demonstrations. This refinement conditions on proprioception and target articulation states while fusing whole-hand tactile and force-torque feedback with the policy's action intent through cross-attention. The resulting controller adapts online to instance-specific articulation properties, stabilizes contact interactions, and regulates internal forces under perturbations. We validate our method across diverse real-world tools, including scissors, pliers, minimally invasive surgical instruments, and staplers, demonstrating robust sim-to-real transfer, improved disturbance resilience, and generalization across structurally related articulated tools without precise physical modeling.
♻ ☆ Dependent Reachable Sets for the Constant Bearing Pursuit Strategy
This paper introduces a novel reachability problem for the scenario involving two agents, where one agent follows another agent using a feedback strategy. The geometry of the reachable set for an agent, termed \emph{dependent reachable set}, is characterized using the constant bearing pursuit strategy as a case study. Key theoretical results are presented that provide geometric bounds for the associated dependent reachable set. Simulation results are presented to empirically establish the shape of the dependent reachable set. In the process, an original optimization problem is formulated and analyzed for the constant bearing pursuit strategy.
comment: This work has been submitted to a journal for possible publication
♻ ☆ Push Anything: Single- and Multi-Object Pushing From First Sight with Contact-Implicit MPC ICRA 2026
Non-prehensile manipulation of diverse objects remains a core challenge in robotics, driven by unknown physical properties and the complexity of contact-rich interactions. Recent advances in contact-implicit model predictive control (CI-MPC), with contact reasoning embedded directly in the trajectory optimization, have shown promise in tackling the task efficiently and robustly. However, demonstrations have been limited to narrowly curated examples. In this work, we showcase the broader capabilities of CI-MPC through precise planar pushing tasks over a wide range of object geometries, including multi-object domains. These scenarios demand reasoning over numerous inter-object and object-environment contacts to strategically manipulate and de-clutter the environment, challenges that were intractable for prior CI-MPC methods. To achieve this, we introduce Consensus Complementarity Control Plus (C3+), an enhanced CI-MPC algorithm integrated into a complete pipeline spanning object scanning, mesh reconstruction, and hardware execution. Compared to its predecessor C3, C3+ achieves substantially faster solve times, enabling real-time performance even in multi-object pushing tasks. On hardware, our system achieves overall 98% success rate across 33 objects, reaching pose goals within tight tolerances. The average time-to-goal is approximately 0.5, 1.6, 3.2, and 5.3 minutes for 1-, 2-, 3-, and 4-object tasks, respectively. Project page: https://dairlab.github.io/push-anything.
comment: Presented at ICRA 2026; 8 pages, 8 figures. Hien Bui, Yufeiyang Gao, and Haoran Yang contributed equally to this work
♻ ☆ CAVER: Curious Audiovisual Exploring Robot
Multimodal audiovisual perception can enable new avenues for robotic manipulation, from better material classification to the imitation of demonstrations for which only audio signals are available (e.g., playing a tune by ear). However, to unlock such multimodal potential, robots need to learn the correlations between an object's visual appearance and the sound it generates when they interact with it. Such an active sensorimotor experience requires new interaction capabilities, representations, and exploration methods to guide the robot in efficiently building increasingly rich audiovisual knowledge. In this work, we present CAVER, a novel robot that builds and utilizes rich audiovisual representations of objects. CAVER includes three novel contributions: 1) a novel 3D printed end-effector, attachable to parallel grippers, that excites objects' audio responses, 2) an audiovisual representation that combines local and global appearance information with sound features, and 3) an exploration algorithm that uses and builds the audiovisual representation in a curiosity-driven manner that prioritizes interacting with high uncertainty objects to obtain good coverage of surprising audio with fewer interactions. We demonstrate that CAVER builds rich representations in different scenarios more efficiently than several exploration baselines, and that the learned audiovisual representation leads to significant improvements in material classification and the imitation of audio-only human demonstrations. https://caver-bot.github.io/
comment: 9 pages, 6 figures
♻ ☆ GLIDE: A Coordinated Aerial-Ground Framework for Search and Rescue in Unknown Environments
We present a cooperative aerial-ground search-and-rescue (SAR) framework that pairs two unmanned aerial vehicles (UAVs) with an unmanned ground vehicle (UGV) to achieve rapid victim localization and obstacle-aware navigation in unknown environments. We dub this framework Guided Long-horizon Integrated Drone Escort (GLIDE), highlighting the UGV's reliance on UAV guidance for long-horizon planning. In our framework, a goal-searching UAV executes real-time onboard victim detection and georeferencing to nominate goals for the ground platform, while a terrain-scouting UAV flies ahead of the UGV's planned route to provide mid-level traversability updates. The UGV fuses aerial cues with local sensing to perform time-efficient A* planning and continuous replanning as information arrives. Additionally, we present a hardware demonstration (using a GEM e6 golf cart as the UGV and two X500 UAVs) to evaluate end-to-end SAR mission performance and include simulation ablations to assess the planning stack in isolation from detection. Empirical results demonstrate that explicit role separation across UAVs, coupled with terrain scouting and guided planning, improves reach time and navigation safety in time-critical SAR missions.
♻ ☆ Generative Predictive Control: Flow Matching Policies for Dynamic and Difficult-to-Demonstrate Tasks ICRA 2026
Generative control policies have recently unlocked major progress in robotics. These methods produce action sequences via diffusion or flow matching, with training data provided by demonstrations. But existing methods come with two key limitations: they require expert demonstrations, which can be difficult to obtain, and they are limited to relatively slow, quasi-static tasks. In this paper, we leverage a tight connection between sampling-based predictive control and generative modeling to address each of these issues. In particular, we introduce generative predictive control, a supervised learning framework for tasks with fast dynamics that are easy to simulate but difficult to demonstrate. We then show how trained flow-matching policies can be warm-started at inference time, maintaining temporal consistency and enabling high-frequency feedback. We believe that generative predictive control offers a complementary approach to existing behavior cloning methods, and hope that it paves the way toward generalist policies that extend beyond quasi-static demonstration-oriented tasks.
comment: ICRA 2026
♻ ☆ Indicating Robot Vision Capabilities with Augmented Reality
Research indicates that humans can mistakenly assume that robots and humans have the same field of view, possessing an inaccurate mental model of robots. This misperception may lead to failures during human-robot collaboration tasks where robots might be asked to complete impossible tasks about out-of-view objects. The issue is more severe when robots do not have a chance to scan the scene to update their world model while focusing on assigned tasks. To help align humans' mental models of robots' vision capabilities, we propose four field-of-view indicators in augmented reality and conducted a human-subjects experiment (N=41) to evaluate them in a collaborative assembly task regarding accuracy, confidence, task efficiency, and workload. These indicators span a spectrum of positions: two at robot's eye and head space -- deepening eye socket and adding blocks to two sides of the eyes (i.e., egocentric), and two anchoring in the robot's task space -- adding extended blocks from the sides of eyes to the table and placing blocks directly on the tables (i.e., allocentric). Results showed that, when placed directly in the task space, the allocentric indicator yields the highest accuracy, although with a delay in interpreting the robot's field of view. When placed at the robot's eyes, the egocentric indicator of deeper eye sockets, possible for physical alteration, also increased accuracy. In all indicators, participants' confidence was high while cognitive load remained low. Finally, we contribute six guidelines for practitioners to apply our augmented reality indicators or physical alterations to align humans' mental models with robots' vision capabilities.
♻ ☆ MiDAS: A Multimodal Data Acquisition System and Dataset for Robot-Assisted Minimally Invasive Surgery
Background: Robot-assisted minimally invasive surgery (RMIS) research increasingly relies on multimodal data, yet access to proprietary robot telemetry remains a major barrier. We introduce MiDAS, an open-source, platform-agnostic system enabling time-synchronized, non-invasive multimodal data acquisition across surgical robotic platforms. Methods: MiDAS integrates electromagnetic and RGB-D hand tracking, foot pedal sensing, and surgical video capturing without requiring proprietary robot interfaces. We validated MiDAS on the open-source Raven-II and the clinical da Vinci Xi by collecting multimodal datasets of peg transfer and hernia repair suturing tasks performed by surgical residents. Correlation analysis and downstream gesture recognition experiments were conducted. Results: External hand and foot sensing closely approximated internal robot kinematics and non-invasive motion signals achieved gesture recognition performance comparable to proprietary telemetry. Conclusion: MiDAS enables reproducible multimodal RMIS data collection and is released with annotated datasets, including the first multimodal dataset capturing hernia repair suturing on high-fidelity simulation models.
comment: 29 pages, 17 figures
♻ ☆ OA-Bug: An Olfactory-Auditory Augmented Bug Algorithm for Swarm Robots in a Denied Environment IROS
Searching in a denied environment is challenging for swarm robots as no assistance from GNSS, mapping, data sharing, and central processing is allowed. However, using olfactory and auditory signals to cooperate like animals could be an important way to improve the collaboration of swarm robots. In this paper, an Olfactory-Auditory augmented Bug algorithm (OA-Bug) is proposed for a swarm of autonomous robots to explore a denied environment. A simulation environment is built to measure the performance of OA-Bug. The coverage of the search task can reach 96.93% using OA-Bug, which is significantly improved compared with a similar algorithm, SGBA. Furthermore, experiments are conducted on real swarm robots to prove the validity of OA-Bug. Results show that OA-Bug can improve the performance of swarm robots in a denied environment. Video: https://youtu.be/vj9cRiSm9eM.
comment: 7 pages, 6 figures, accepted by 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)