MyArxiv
Robotics 41
☆ One Hand to Rule Them All: Canonical Representations for Unified Dexterous Manipulation
Dexterous manipulation policies today largely assume fixed hand designs, severely restricting their generalization to new embodiments with varied kinematic and structural layouts. To overcome this limitation, we introduce a parameterized canonical representation that unifies a broad spectrum of dexterous hand architectures. It comprises a unified parameter space and a canonical URDF format, offering three key advantages. 1) The parameter space captures essential morphological and kinematic variations for effective conditioning in learning algorithms. 2) A structured latent manifold can be learned over our space, where interpolations between embodiments yield smooth and physically meaningful morphology transitions. 3) The canonical URDF standardizes the action space while preserving dynamic and functional properties of the original URDFs, enabling efficient and reliable cross-embodiment policy learning. We validate these advantages through extensive analysis and experiments, including grasp policy replay, VAE latent encoding, and cross-embodiment zero-shot transfer. Specifically, we train a VAE on the unified representation to obtain a compact, semantically rich latent embedding, and develop a grasping policy conditioned on the canonical representation that generalizes across dexterous hands. We demonstrate, through simulation and real-world tasks on unseen morphologies (e.g., 81.9% zero-shot success rate on 3-finger LEAP Hand), that our framework unifies both the representational and action spaces of structurally diverse hands, providing a scalable foundation for cross-hand learning toward universal dexterous manipulation.
comment: Project Page: https://zhenyuwei2003.github.io/OHRA/
☆ EgoScale: Scaling Dexterous Manipulation with Diverse Egocentric Human Data
Human behavior is among the most scalable sources of data for learning physical intelligence, yet how to effectively leverage it for dexterous manipulation remains unclear. While prior work demonstrates human to robot transfer in constrained settings, it is unclear whether large scale human data can support fine grained, high degree of freedom dexterous manipulation. We present EgoScale, a human to dexterous manipulation transfer framework built on large scale egocentric human data. We train a Vision Language Action (VLA) model on over 20,854 hours of action labeled egocentric human video, more than 20 times larger than prior efforts, and uncover a log linear scaling law between human data scale and validation loss. This validation loss strongly correlates with downstream real robot performance, establishing large scale human data as a predictable supervision source. Beyond scale, we introduce a simple two stage transfer recipe: large scale human pretraining followed by lightweight aligned human robot mid training. This enables strong long horizon dexterous manipulation and one shot task adaptation with minimal robot supervision. Our final policy improves average success rate by 54% over a no pretraining baseline using a 22 DoF dexterous robotic hand, and transfers effectively to robots with lower DoF hands, indicating that large scale human motion provides a reusable, embodiment agnostic motor prior.
☆ Learning Humanoid End-Effector Control for Open-Vocabulary Visual Loco-Manipulation
Visual loco-manipulation of arbitrary objects in the wild with humanoid robots requires accurate end-effector (EE) control and a generalizable understanding of the scene via visual inputs (e.g., RGB-D images). Existing approaches are based on real-world imitation learning and exhibit limited generalization due to the difficulty in collecting large-scale training datasets. This paper presents a new paradigm, HERO, for object loco-manipulation with humanoid robots that combines the strong generalization and open-vocabulary understanding of large vision models with strong control performance from simulated training. We achieve this by designing an accurate residual-aware EE tracking policy. This EE tracking policy combines classical robotics with machine learning. It uses a) inverse kinematics to convert residual end-effector targets into reference trajectories, b) a learned neural forward model for accurate forward kinematics, c) goal adjustment, and d) replanning. Together, these innovations help us cut down the end-effector tracking error by 3.2x. We use this accurate end-effector tracker to build a modular system for loco-manipulation, where we use open-vocabulary large vision models for strong visual generalization. Our system is able to operate in diverse real-world environments, from offices to coffee shops, where the robot is able to reliably manipulate various everyday objects (e.g., mugs, apples, toys) on surfaces ranging from 43cm to 92cm in height. Systematic modular and end-to-end tests in simulation and the real world demonstrate the effectiveness of our proposed design. We believe the advances in this paper can open up new ways of training humanoid robots to interact with daily objects.
comment: Project page: https://hero-humanoid.github.io/
☆ Learning to unfold cloth: Scaling up world models to deformable object manipulation
Learning to manipulate cloth is both a paradigmatic problem for robotic research and a problem of immediate relevance to a variety of applications ranging from assistive care to the service industry. The complex physics of the deformable object makes this problem of cloth manipulation nontrivial. In order to create a general manipulation strategy that addresses a variety of shapes, sizes, fold and wrinkle patterns, in addition to the usual problems of appearance variations, it becomes important to carefully consider model structure and their implications for generalisation performance. In this paper, we present an approach to in-air cloth manipulation that uses a variation of a recently proposed reinforcement learning architecture, DreamerV2. Our implementation modifies this architecture to utilise surface normals input, in addition to modiying the replay buffer and data augmentation procedures. Taken together these modifications represent an enhancement to the world model used by the robot, addressing the physical complexity of the object being manipulated by the robot. We present evaluations both in simulation and in a zero-shot deployment of the trained policies in a physical robot setup, performing in-air unfolding of a variety of different cloth types, demonstrating the generalisation benefits of our proposed architecture.
comment: 8 pages, 5 figures, 3 tables
☆ Towards Autonomous Robotic Kidney Ultrasound: Spatial-Efficient Volumetric Imaging via Template Guided Optimal Pivoting
Medical ultrasound (US) imaging is a frontline tool for the diagnosis of kidney diseases. However, traditional freehand imaging procedure suffers from inconsistent, operator-dependent outcomes, lack of 3D localization information, and risks of work-related musculoskeletal disorders. While robotic ultrasound (RUS) systems offer the potential for standardized, operator-independent 3D kidney data acquisition, the existing scanning methods lack the ability to determine the optimal imaging window for efficient imaging. As a result, the scan is often blindly performed with excessive probe footprint, which frequently leads to acoustic shadowing and incomplete organ coverage. Consequently, there is a critical need for a spatially efficient imaging technique that can maximize the kidney coverage through minimum probe footprint. Here, we propose an autonomous workflow to achieve efficient kidney imaging via template-guided optimal pivoting. The system first performs an explorative imaging to generate partial observations of the kidney. This data is then registered to a kidney template to estimate the organ pose. With the kidney localized, the robot executes a fixed-point pivoting sweep where the imaging plane is aligned with the kidney long axis to minimize the probe translation. The proposed method was validated in simulation and in-vivo. Simulation results indicate that a 60% exploration ratio provides optimal balance between kidney localization accuracy and scanning efficiency. In-vivo evaluation on two male subjects demonstrates a kidney localization accuracy up to 7.36 mm and 13.84 degrees. Moreover, the optimal pivoting approach shortened the probe footprint by around 75 mm when compared with the baselines. These results valid our approach of leveraging anatomical templates to align the probe optimally for volumetric sweep.
☆ Sensor Query Schedule and Sensor Noise Covariances for Accuracy-constrained Trajectory Estimation
Trajectory estimation involves determining the trajectory of a mobile robot by combining prior knowledge about its dynamic model with noisy observations of its state obtained using sensors. The accuracy of such a procedure is dictated by the system model fidelity and the sensor parameters, such as the accuracy of the sensor (as represented by its noise covariance) and the rate at which it can generate observations, referred to as the sensor query schedule. Intuitively, high-rate measurements from accurate sensors lead to accurate trajectory estimation. However, cost and resource constraints limit the sensor accuracy and its measurement rate. Our work's novel contribution is the estimation of sensor schedules and sensor covariances necessary to achieve a specific estimation accuracy. Concretely, we focus on estimating: (i) the rate or schedule with which a sensor of known covariance must generate measurements to achieve specific estimation accuracy, and alternatively, (ii) the sensor covariance necessary to achieve specific estimation accuracy for a given sensor update rate. We formulate the problem of estimating these sensor parameters as semidefinite programs, which can be solved by off-the-shelf solvers. We validate our approach in simulation and real experiments by showing that the sensor schedules and the sensor covariances calculated using our proposed method achieve the desired trajectory estimation accuracy. Our method also identifies scenarios where certain estimation accuracy is unachievable with the given system and sensor characteristics.
☆ Decentralized and Fully Onboard: Range-Aided Cooperative Localization and Navigation on Micro Aerial Vehicles
Controlling a team of robots in a coordinated manner is challenging because centralized approaches (where all computation is performed on a central machine) scale poorly, and globally referenced external localization systems may not always be available. In this work, we consider the problem of range-aided decentralized localization and formation control. In such a setting, each robot estimates its relative pose by combining data only from onboard odometry sensors and distance measurements to other robots in the team. Additionally, each robot calculates the control inputs necessary to collaboratively navigate an environment to accomplish a specific task, for example, moving in a desired formation while monitoring an area. We present a block coordinate descent approach to localization that does not require strict coordination between the robots. We present a novel formulation for formation control as inference on factor graphs that takes into account the state estimation uncertainty and can be solved efficiently. Our approach to range-aided localization and formation-based navigation is completely decentralized, does not require specialized trajectories to maintain formation, and achieves decimeter-level positioning and formation control accuracy. We demonstrate our approach through multiple real experiments involving formation flights in diverse indoor and outdoor environments.
☆ VIGOR: Visual Goal-In-Context Inference for Unified Humanoid Fall Safety
Reliable fall recovery is critical for humanoids operating in cluttered environments. Unlike quadrupeds or wheeled robots, humanoids experience high-energy impacts, complex whole-body contact, and large viewpoint changes during a fall, making recovery essential for continued operation. Existing methods fragment fall safety into separate problems such as fall avoidance, impact mitigation, and stand-up recovery, or rely on end-to-end policies trained without vision through reinforcement learning or imitation learning, often on flat terrain. At a deeper level, fall safety is treated as monolithic data complexity, coupling pose, dynamics, and terrain and requiring exhaustive coverage, limiting scalability and generalization. We present a unified fall safety approach that spans all phases of fall recovery. It builds on two insights: 1) Natural human fall and recovery poses are highly constrained and transferable from flat to complex terrain through alignment, and 2) Fast whole-body reactions require integrated perceptual-motor representations. We train a privileged teacher using sparse human demonstrations on flat terrain and simulated complex terrains, and distill it into a deployable student that relies only on egocentric depth and proprioception. The student learns how to react by matching the teacher's goal-in-context latent representation, which combines the next target pose with the local terrain, rather than separately encoding what it must perceive and how it must act. Results in simulation and on a real Unitree G1 humanoid demonstrate robust, zero-shot fall safety across diverse non-flat environments without real-world fine-tuning. The project page is available at https://vigor2026.github.io/
☆ Reactive Motion Generation With Particle-Based Perception in Dynamic Environments
Reactive motion generation in dynamic and unstructured scenarios is typically subject to essentially static perception and system dynamics. Reliably modeling dynamic obstacles and optimizing collision-free trajectories under perceptive and control uncertainty are challenging. This article focuses on revealing tight connection between reactive planning and dynamic mapping for manipulators from a model-based perspective. To enable efficient particle-based perception with expressively dynamic property, we present a tensorized particle weight update scheme that explicitly maintains obstacle velocities and covariance meanwhile. Building upon this dynamic representation, we propose an obstacle-aware MPPI-based planning formulation that jointly propagates robot-obstacle dynamics, allowing future system motion to be predicted and evaluated under uncertainty. The model predictive method is shown to significantly improve safety and reactivity with dynamic surroundings. By applying our complete framework in simulated and noisy real-world environments, we demonstrate that explicit modeling of robot-obstacle dynamics consistently enhances performance over state-of-the-art MPPI-based perception-planning baselines avoiding multiple static and dynamic obstacles.
comment: This paper has 20 pages, 15 figures, and 3 tables
☆ RoboGene: Boosting VLA Pre-training via Diversity-Driven Agentic Framework for Real-World Task Generation
The pursuit of general-purpose robotic manipulation is hindered by the scarcity of diverse, real-world interaction data. Unlike data collection from web in vision or language, robotic data collection is an active process incurring prohibitive physical costs. Consequently, automated task curation to maximize data value remains a critical yet under-explored challenge. Existing manual methods are unscalable and biased toward common tasks, while off-the-shelf foundation models often hallucinate physically infeasible instructions. To address this, we introduce RoboGene, an agentic framework designed to automate the generation of diverse, physically plausible manipulation tasks across single-arm, dual-arm, and mobile robots. RoboGene integrates three core components: diversity-driven sampling for broad task coverage, self-reflection mechanisms to enforce physical constraints, and human-in-the-loop refinement for continuous improvement. We conduct extensive quantitative analysis and large-scale real-world experiments, collecting datasets of 18k trajectories and introducing novel metrics to assess task quality, feasibility, and diversity. Results demonstrate that RoboGene significantly outperforms state-of-the-art foundation models (e.g., GPT-4o, Gemini 2.5 Pro). Furthermore, real-world experiments show that VLA models pre-trained with RoboGene achieve higher success rates and superior generalization, underscoring the importance of high-quality task generation. Our project is available at https://robogene-boost-vla.github.io.
☆ Dynamic Modeling and MPC for Locomotion of Tendon-Driven Soft Quadruped
SLOT (Soft Legged Omnidirectional Tetrapod), a tendon-driven soft quadruped robot with 3D-printed TPU legs, is presented to study physics-informed modeling and control of compliant legged locomotion using only four actuators. Each leg is modeled as a deformable continuum using discrete Cosserat rod theory, enabling the capture of large bending deformations, distributed elasticity, tendon actuation, and ground contact interactions. A modular whole-body modeling framework is introduced, in which compliant leg dynamics are represented through physically consistent reaction forces applied to a rigid torso, providing a scalable interface between continuum soft limbs and rigid-body locomotion dynamics. This formulation allows efficient whole-body simulation and real-time control without sacrificing physical fidelity. The proposed model is embedded into a convex model predictive control framework that optimizes ground reaction forces over a 0.495 s prediction horizon and maps them to tendon actuation through a physics-informed force-angle relationship. The resulting controller achieves asymptotic stability under diverse perturbations. The framework is experimentally validated on a physical prototype during crawling and walking gaits, achieving high accuracy with less than 5 mm RMSE in center of mass trajectories. These results demonstrate a generalizable approach for integrating continuum soft legs into model-based locomotion control, advancing scalable and reusable modeling and control methods for soft quadruped robots.
☆ Markerless 6D Pose Estimation and Position-Based Visual Servoing for Endoscopic Continuum Manipulators
Continuum manipulators in flexible endoscopic surgical systems offer high dexterity for minimally invasive procedures; however, accurate pose estimation and closed-loop control remain challenging due to hysteresis, compliance, and limited distal sensing. Vision-based approaches reduce hardware complexity but are often constrained by limited geometric observability and high computational overhead, restricting real-time closed-loop applicability. This paper presents a unified framework for markerless stereo 6D pose estimation and position-based visual servoing of continuum manipulators. A photo-realistic simulation pipeline enables large-scale automatic training with pixel-accurate annotations. A stereo-aware multi-feature fusion network jointly exploits segmentation masks, keypoints, heatmaps, and bounding boxes to enhance geometric observability. To enforce geometric consistency without iterative optimization, a feed-forward rendering-based refinement module predicts residual pose corrections in a single pass. A self-supervised sim-to-real adaptation strategy further improves real-world performance using unlabeled data. Extensive real-world validation achieves a mean translation error of 0.83 mm and a mean rotation error of 2.76° across 1,000 samples. Markerless closed-loop visual servoing driven by the estimated pose attains accurate trajectory tracking with a mean translation error of 2.07 mm and a mean rotation error of 7.41°, corresponding to 85% and 59% reductions compared to open-loop control, together with high repeatability in repeated point-reaching tasks. To the best of our knowledge, this work presents the first fully markerless pose-estimation-driven position-based visual servoing framework for continuum manipulators, enabling precise closed-loop control without physical markers or embedded sensing.
comment: 20 pages, 13 figures, 7 tables
☆ Docking and Persistent Operations for a Resident Underwater Vehicle
Our understanding of the oceans remains limited by sparse and infrequent observations, primarily because current methods are constrained by the high cost and logistical effort of underwater monitoring, relying either on sporadic surveys across broad areas or on long-term measurements at fixed locations. To overcome these limitations, monitoring systems must enable persistent and autonomous operations without the need for continuous surface support. Despite recent advances, resident underwater vehicles remain uncommon due to persistent challenges in autonomy, robotic resilience, and mechanical robustness, particularly under long-term deployment in harsh and remote environments. This work addresses these problems by presenting the development, deployment, and operation of a resident infrastructure using a docking station with a mini-class Remotely Operated Vehicle (ROV) at 90m depth. The ROVis equipped with enhanced onboard processing and perception, allowing it to autonomously navigate using USBL signals, dock via ArUco marker-based visual localisation fused through an Extended Kalman Filter, and carry out local inspection routines. The system demonstrated a 90% autonomous docking success rate and completed full inspection missions within four minutes, validating the integration of acoustic and visual navigation in real-world conditions. These results show that reliable, untethered operations at depth are feasible, highlighting the potential of resident ROV systems for scalable, cost-effective underwater monitoring.
☆ System Identification under Constraints and Disturbance: A Bayesian Estimation Approach
We introduce a Bayesian system identification (SysID) framework for jointly estimating robot's state trajectories and physical parameters with high accuracy. It embeds physically consistent inverse dynamics, contact and loop-closure constraints, and fully featured joint friction models as hard, stage-wise equality constraints. It relies on energy-based regressors to enhance parameter observability, supports both equality and inequality priors on inertial and actuation parameters, enforces dynamically consistent disturbance projections, and augments proprioceptive measurements with energy observations to disambiguate nonlinear friction effects. To ensure scalability, we derive a parameterized equality-constrained Riccati recursion that preserves the banded structure of the problem, achieving linear complexity in the time horizon, and develop computationally efficient derivatives. Simulation studies on representative robotic systems, together with hardware experiments on a Unitree B1 equipped with a Z1 arm, demonstrate faster convergence, lower inertial and friction estimation errors, and improved contact consistency compared to forward-dynamics and decoupled identification baselines. When deployed within model predictive control frameworks, the resulting models yield measurable improvements in tracking performance during locomotion over challenging environments.
☆ Articulated 3D Scene Graphs for Open-World Mobile Manipulation
Semantics has enabled 3D scene understanding and affordance-driven object interaction. However, robots operating in real-world environments face a critical limitation: they cannot anticipate how objects move. Long-horizon mobile manipulation requires closing the gap between semantics, geometry, and kinematics. In this work, we present MoMa-SG, a novel framework for building semantic-kinematic 3D scene graphs of articulated scenes containing a myriad of interactable objects. Given RGB-D sequences containing multiple object articulations, we temporally segment object interactions and infer object motion using occlusion-robust point tracking. We then lift point trajectories into 3D and estimate articulation models using a novel unified twist estimation formulation that robustly estimates revolute and prismatic joint parameters in a single optimization pass. Next, we associate objects with estimated articulations and detect contained objects by reasoning over parent-child relations at identified opening states. We also introduce the novel Arti4D-Semantic dataset, which uniquely combines hierarchical object semantics including parent-child relation labels with object axis annotations across 62 in-the-wild RGB-D sequences containing 600 object interactions and three distinct observation paradigms. We extensively evaluate the performance of MoMa-SG on two datasets and ablate key design choices of our approach. In addition, real-world experiments on both a quadruped and a mobile manipulator demonstrate that our semantic-kinematic scene graphs enable robust manipulation of articulated objects in everyday home environments. We provide code and data at: https://momasg.cs.uni-freiburg.de.
☆ Dual-Quadruped Collaborative Transportation in Narrow Environments via Safe Reinforcement Learning
Collaborative transportation, where multiple robots collaboratively transport a payload, has garnered significant attention in recent years. While ensuring safe and high-performance inter-robot collaboration is critical for effective task execution, it is difficult to pursue in narrow environments where the feasible region is extremely limited. To address this challenge, we propose a novel approach for dual-quadruped collaborative transportation via safe reinforcement learning (RL). Specifically, we model the task as a fully cooperative constrained Markov game, where collision avoidance is formulated as constraints. We introduce a cost-advantage decomposition method that enforces the sum of team constraints to remain below an upper bound, thereby guaranteeing task safety within an RL framework. Furthermore, we propose a constraint allocation method that assigns shared constraints to individual robots to maximize the overall task reward, encouraging autonomous task-assignment among robots, thereby improving collaborative task performance. Simulation and real-time experimental results demonstrate that the proposed approach achieves superior performance and a higher success rate in dual-quadruped collaborative transportation compared to existing methods.
☆ SCAR: Satellite Imagery-Based Calibration for Aerial Recordings
We introduce SCAR, a method for long-term auto-calibration refinement of aerial visual-inertial systems that exploits georeferenced satellite imagery as a persistent global reference. SCAR estimates both intrinsic and extrinsic parameters by aligning aerial images with 2D--3D correspondences derived from publicly available orthophotos and elevation models. In contrast to existing approaches that rely on dedicated calibration maneuvers or manually surveyed ground control points, our method leverages external geospatial data to detect and correct calibration degradation under field deployment conditions. We evaluate our approach on six large-scale aerial campaigns conducted over two years under diverse seasonal and environmental conditions. Across all sequences, SCAR consistently outperforms established baselines (Kalibr, COLMAP, VINS-Mono), reducing median reprojection error by a large margin, and translating these calibration gains into substantially lower visual localization rotation errors and higher pose accuracy. These results demonstrate that SCAR provides accurate, robust, and reproducible calibration over long-term aerial operations without the need for manual intervention.
☆ Machine Learning Driven Prediction of the Behavior of Biohybrid Actuators
Skeletal muscle-based biohybrid actuators have proved to be a promising component in soft robotics, offering efficient movement. However, their intrinsic biological variability and nonlinearity pose significant challenges for controllability and predictability. To address these issues, this study investigates the application of supervised learning, a form of machine learning, to model and predict the behavior of biohybrid machines (BHMs), focusing on a muscle ring anchored on flexible polymer pillars. First, static prediction models (i.e., random forest and neural network regressors) are trained to estimate the maximum exerted force achieved from input variables such as muscle sample, electrical stimulation parameters, and baseline exerted force. Second, a dynamic modeling framework, based on Long Short-Term Memory networks, is developed to serve as a digital twin, replicating the time series of exerted forces observed in response to electrical stimulation. Both modeling approaches demonstrate high predictive accuracy. The best performance of the static models is characterized by R2 of 0.9425, whereas the dynamic model achieves R2 of 0.9956. The static models can enable optimization of muscle actuator performance for targeted applications and required force outcomes, while the dynamic model provides a foundation for developing robustly adaptive control strategies in future biohybrid robotic systems.
☆ Markerless Robot Detection and 6D Pose Estimation for Multi-Agent SLAM ICRA 2026
The capability of multi-robot SLAM approaches to merge localization history and maps from different observers is often challenged by the difficulty in establishing data association. Loop closure detection between perceptual inputs of different robotic agents is easily compromised in the context of perceptual aliasing, or when perspectives differ significantly. For this reason, direct mutual observation among robots is a powerful way to connect partial SLAM graphs, but often relies on the presence of calibrated arrays of fiducial markers (e.g., AprilTag arrays), which severely limits the range of observations and frequently fails under sharp lighting conditions, e.g., reflections or overexposure. In this work, we propose a novel solution to this problem leveraging recent advances in Deep-Learning-based 6D pose estimation. We feature markerless pose estimation as part of a decentralized multi-robot SLAM system and demonstrate the benefit to the relative localization accuracy among the robotic team. The solution is validated experimentally on data recorded in a test field campaign on a planetary analogous environment.
comment: Accepted contribution to ICRA 2026
☆ Nonplanar Model Predictive Control for Autonomous Vehicles with Recursive Sparse Gaussian Process Dynamics
This paper proposes a nonplanar model predictive control (MPC) framework for autonomous vehicles operating on nonplanar terrain. To approximate complex vehicle dynamics in such environments, we develop a geometry-aware modeling approach that learns a residual Gaussian Process (GP). By utilizing a recursive sparse GP, the framework enables real-time adaptation to varying terrain geometry. The effectiveness of the learned model is demonstrated in a reference-tracking task using a Model Predictive Path Integral (MPPI) controller. Validation within a custom Isaac Sim environment confirms the framework's capability to maintain high tracking accuracy on challenging 3D surfaces.
comment: 6 pages, 5 figures. Accepted to IEEE Intelligent Vehicles Symposium (IV), 2026
☆ SIT-LMPC: Safe Information-Theoretic Learning Model Predictive Control for Iterative Tasks ICRA 2026
Robots executing iterative tasks in complex, uncertain environments require control strategies that balance robustness, safety, and high performance. This paper introduces a safe information-theoretic learning model predictive control (SIT-LMPC) algorithm for iterative tasks. Specifically, we design an iterative control framework based on an information-theoretic model predictive control algorithm to address a constrained infinite-horizon optimal control problem for discrete-time nonlinear stochastic systems. An adaptive penalty method is developed to ensure safety while balancing optimality. Trajectories from previous iterations are utilized to learn a value function using normalizing flows, which enables richer uncertainty modeling compared to Gaussian priors. SIT-LMPC is designed for highly parallel execution on graphics processing units, allowing efficient real-time optimization. Benchmark simulations and hardware experiments demonstrate that SIT-LMPC iteratively improves system performance while robustly satisfying system constraints.
comment: 8 pages, 5 figures. Published in IEEE RA-L, vol. 11, no. 1, Jan. 2026. Presented at ICRA 2026
☆ World Model Failure Classification and Anomaly Detection for Autonomous Inspection
Autonomous inspection robots for monitoring industrial sites can reduce costs and risks associated with human-led inspection. However, accurate readings can be challenging due to occlusions, limited viewpoints, or unexpected environmental conditions. We propose a hybrid framework that combines supervised failure classification with anomaly detection, enabling classification of inspection tasks as a success, known failure, or anomaly (i.e., out-of-distribution) case. Our approach uses a world model backbone with compressed video inputs. This policy-agnostic, distribution-free framework determines classifications based on two decision functions set by conformal prediction (CP) thresholds before a human observer does. We evaluate the framework on gauge inspection feeds collected from office and industrial sites and demonstrate real-time deployment on a Boston Dynamics Spot. Experiments show over 90% accuracy in distinguishing between successes, failures, and OOD cases, with classifications occurring earlier than a human observer. These results highlight the potential for robust, anticipatory failure detection in autonomous inspection tasks or as a feedback signal for model training to assess and improve the quality of training data. Project website: https://autoinspection-classification.github.io
☆ Image Measurement Method for Automatic Insertion of Forks into Inclined Pallet
In order to insert a fork into a hole of a pallet by a forklift located in front of a pallet, it is necessary to control the height position, reach position, and tilt angle of the fork to match the position and orientation of the hole of the pallet. In order to make AGF (Autonomous Guided Forklift) do this automatically, we propose an image measurement method to measure the pitch inclination of the pallet in the camera coordinate system from an image obtained by using a wide-angle camera. In addition, we propose an image measurement method to easily acquire the calibration information between the camera coordinate system and the fork coordinate system necessary to apply the measurements in the camera coordinate system to the fork control. In the experiment space, a wide-angle camera was fixed at the backrest of a reach type forklift. The wide-angle images taken by placing a pallet in front of the camera were processed. As a result of evaluating the error by comparing the image measurement value with the hand measurement value when changing the pitch inclination angle of the pallet, the relative height of the pallet and the fork, and whether the pallet is loaded or not, it was confirmed that the error was within the allowable range for safely inserting the fork.
comment: Accepted and published in IEEE ICARCV 2022
☆ Reactive Slip Control in Multifingered Grasping: Hybrid Tactile Sensing and Internal-Force Optimization ICRA
We present a hybrid learning and model-based approach that adapts internal grasp forces to halt in-hand slip on 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, we update internal forces computed in the null space of the grasp via a quadratic program that preserves the object wrench while enforcing actuation limits. The pipeline yields a theoretical sensing-to-command latency of 35-40 ms, with 5 ms for PzR-based contact and geometry updates and about 4 ms for the quadratic program solve. In controlled trials, slip onset is detected at 20ms. We demonstrate closed-loop stabilization on multifingered grasps under external perturbations. Augmenting efficient analytic force control with learned tactile cues yields both robustness and rapid reactions, as confirmed in our end-to-end evaluation. Measured delays are dominated by the experimental data path rather than actual computation. The analysis outlines a clear route to sub-50 ms closed-loop stabilization.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA), 2026
♻ ☆ Elements of Robot Morphology: Supporting Designers in Robot Form Exploration
Robot morphology, the form, shape, and structure of robots, is a key design space in human-robot interaction (HRI), shaping how robots function, express themselves, and interact with people. Yet, despite its importance, little is known about how design frameworks can guide systematic form exploration. To address this gap, we introduce Elements of Robot Morphology, a framework that identifies five fundamental elements: perception, articulation, end effectors, locomotion, and structure. Derived from an analysis of existing robots, the framework supports structured exploration of diverse robot forms. To operationalize the framework, we developed Morphology Exploration Blocks (MEB), a set of tangible blocks that enable hands-on, collaborative experimentation with robot morphologies. We evaluate the framework and toolkit through a case study and design workshops, showing how they support analysis, ideation, reflection, and collaborative robot design.
comment: 10 pages, 5 figures, Proceedings of the 21st ACM/IEEE International Conference on Human-Robot Interaction (HRI '26)
♻ ☆ View Invariant Learning for Vision-Language Navigation in Continuous Environments
Vision-Language Navigation in Continuous Environments (VLNCE), where an agent follows instructions and moves freely to reach a destination, is a key research problem in embodied AI. However, most navigation policies are sensitive to viewpoint changes, i.e., variations in camera height and viewing angle that alter the agent's observation. In this paper, we introduce a generalized scenario, V2-VLNCE (VLNCE with Varied Viewpoints), and propose VIL (View Invariant Learning), a view-invariant post-training strategy that enhances the robustness of existing navigation policies to changes in camera viewpoint. VIL employs a contrastive learning framework to learn sparse and view-invariant features. Additionally, we introduce a teacher-student framework for the Waypoint Predictor Module, a core component of most VLNCE baselines, where a view-dependent teacher model distills knowledge into a view-invariant student model. We employ an end-to-end training paradigm to jointly optimize these components, thus eliminating the cost for individual module training. Empirical results show that our method outperforms state-of-the-art approaches on V2-VLNCE by 8-15% measured on Success Rate for two standard benchmark datasets R2R-CE and RxR-CE. Furthermore, we evaluate VIL under the standard VLNCE setting and find that, despite being trained for varied viewpoints, it often still improves performance. On the more challenging RxR-CE dataset, our method also achieved state-of-the-art performance across all metrics when compared to other map-free methods. This suggests that adding VIL does not diminish the standard viewpoint performance and can serve as a plug-and-play post-training method.
comment: This paper is accepted to RA-L 2026
♻ ☆ 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
♻ ☆ IMPACT: Behavioral Intention-aware Multimodal Trajectory Prediction with Adaptive Context Trimming
While most prior research has focused on improving the precision of multimodal trajectory predictions, the explicit modeling of multimodal behavioral intentions (e.g., yielding, overtaking) remains relatively underexplored. This paper proposes a unified framework that jointly predicts both behavioral intentions and trajectories to enhance prediction accuracy, interpretability, and efficiency. Specifically, we employ a shared context encoder for both intention and trajectory predictions, thereby reducing structural redundancy and information loss. Moreover, we address the lack of ground-truth behavioral intention labels in mainstream datasets (Waymo, Argoverse) by auto-labeling these datasets, thus advancing the community's efforts in this direction. We further introduce a vectorized occupancy prediction module that infers the probability of each map polyline being occupied by the target vehicle's future trajectory. By leveraging these intention and occupancy prediction priors, our method conducts dynamic, modality-dependent pruning of irrelevant agents and map polylines in the decoding stage, effectively reducing computational overhead and mitigating noise from non-critical elements. Our approach ranks first among LiDAR-free methods on the Waymo Motion Dataset and achieves first place on the Waymo Interactive Prediction Dataset. Remarkably, even without model ensembling, our single-model framework improves the soft mean average precision (softmAP) by 10 percent compared to the second-best method in the Waymo Interactive Prediction Leaderboard. Furthermore, the proposed framework has been successfully deployed on real vehicles, demonstrating its practical effectiveness in real-world applications.
comment: accepted by IEEE Robotics and Automation Letters
♻ ☆ Ultra-wideband Time Difference of Arrival Indoor Localization: From Sensor Placement to System Evaluation
Wireless indoor localization has attracted significant research interest due to its high accuracy, low cost, lightweight design, and low power consumption. Specifically, ultra-wideband (UWB) time difference of arrival (TDOA)-based localization has emerged as a scalable positioning solution for mobile robots, consumer electronics, and wearable devices, featuring good accuracy and reliability. While UWB TDOA-based localization systems rely on the deployment of UWB radio sensors as positioning landmarks, existing works often assume these placements are predetermined or study the sensor placement problem alone without evaluating it in practical scenarios. In this article, we bridge this gap by approaching the UWB TDOA localization from a system-level perspective, integrating sensor placement as a key component and conducting practical evaluation in real-world scenarios. Through extensive real-world experiments, we demonstrate the accuracy and robustness of our localization system, comparing its performance to the theoretical lower bounds. Using a challenging multi-room environment as a case study, we illustrate the full system construction process, from sensor placement optimization to real-world deployment. Our evaluation, comprising a cumulative total of 39 minutes of real-world experiments involving up to five agents and covering 2608 meters across four distinct scenarios, provides valuable insights and guidelines for constructing UWB TDOA localization systems.
♻ ☆ SurgRAW: Multi-Agent Workflow with Chain of Thought Reasoning for Robotic Surgical Video Analysis
Robotic-assisted surgery (RAS) is central to modern surgery, driving the need for intelligent systems with accurate scene understanding. Most existing surgical AI methods rely on isolated, task-specific models, leading to fragmented pipelines with limited interpretability and no unified understanding of RAS scene. Vision-Language Models (VLMs) offer strong zero-shot reasoning, but struggle with hallucinations, domain gaps and weak task-interdependency modeling. To address the lack of unified data for RAS scene understanding, we introduce SurgCoTBench, the first reasoning-focused benchmark in RAS, covering 14256 QA pairs with frame-level annotations across five major surgical tasks. Building on SurgCoTBench, we propose SurgRAW, a clinically aligned Chain-of-Thought (CoT) driven agentic workflow for zero-shot multi-task reasoning in surgery. SurgRAW employs a hierarchical reasoning workflow where an orchestrator divides surgical scene understanding into two reasoning streams and directs specialized agents to generate task-level reasoning, while higher-level agents capture workflow interdependencies or ground output clinically. Specifically, we propose a panel discussion mechanism to ensure task-specific agents collaborate synergistically and leverage on task interdependencies. Similarly, we incorporate a retrieval-augmented generation module to enrich agents with surgical knowledge and alleviate domain gaps in general VLMs. We design task-specific CoT prompts grounded in surgical domain to ensure clinically aligned reasoning, reduce hallucinations and enhance interpretability. Extensive experiments show that SurgRAW surpasses mainstream VLMs and agentic systems and outperforms a supervised model by 14.61% accuracy. Dataset and code is available at https://github.com/jinlab-imvr/SurgRAW.git .
♻ ☆ FreqPolicy: Efficient Flow-based Visuomotor Policy via Frequency Consistency NeurIPS 2025
Generative modeling-based visuomotor policies have been widely adopted in robotic manipulation, attributed to their ability to model multimodal action distributions. However, the high inference cost of multi-step sampling limits its applicability in real-time robotic systems. Existing approaches accelerate sampling in generative modeling-based visuomotor policies by adapting techniques originally developed to speed up image generation. However, a major distinction exists: image generation typically produces independent samples without temporal dependencies, while robotic manipulation requires generating action trajectories with continuity and temporal coherence. To this end, we propose FreqPolicy, a novel approach that first imposes frequency consistency constraints on flow-based visuomotor policies. Our work enables the action model to capture temporal structure effectively while supporting efficient, high-quality one-step action generation. Concretely, we introduce a frequency consistency constraint objective that enforces alignment of frequency-domain action features across different timesteps along the flow, thereby promoting convergence of one-step action generation toward the target distribution. In addition, we design an adaptive consistency loss to capture structural temporal variations inherent in robotic manipulation tasks. We assess FreqPolicy on 53 tasks across 3 simulation benchmarks, proving its superiority over existing one-step action generators. We further integrate FreqPolicy into the vision-language-action (VLA) model and achieve acceleration without performance degradation on 40 tasks of LIBERO. Besides, we show efficiency and effectiveness in real-world robotic scenarios with an inference frequency of 93.5 Hz.
comment: NeurIPS 2025
♻ ☆ AMBER: A tether-deployable gripping crawler with compliant microspines for canopy manipulation
This paper presents an aerially deployable crawler designed for adaptive locomotion and manipulation within tree canopies. The system combines compliant microspine-based tracks, a dual-track rotary gripper, and an elastic tail, enabling secure attachment and stable traversal across branches of varying curvature and inclination. Experiments demonstrate reliable gripping up to 90$^\circ$ body roll and inclination, while effective climbing on branches inclined up to 67.5$^\circ$, achieving a maximum speed of 0.55 body lengths per second on horizontal branches. The compliant tracks allow yaw steering of up to 10$^\circ$, enhancing maneuverability on irregular surfaces. Power measurements show efficient operation with a dimensionless cost of transport over an order of magnitude lower than typical hovering power consumption in aerial robots. The crawler provides a robust, low-power platform for environmental sampling and in-canopy sensing. The aerial deployment is demonstrated at a conceptual and feasibility level, while full drone-crawler integration is left as future work.
♻ ☆ Inverting Non-Injective Functions with Twin Neural Network Regression
Non-injective functions are not globally invertible. However, they can often be restricted to locally injective subdomains where the inversion is well-defined. In many settings a preferred solution can be selected even when multiple valid preimages exist or input and output dimensions differ. This manuscript describes a natural reformulation of the inverse learning problem for non-injective functions as a collection of locally invertible problems. More precisely, Twin Neural Network Regression is trained to predict local inverse corrections around known anchor points. By anchoring predictions to points within the same locally invertible region, the method consistently selects a valid branch of the inverse. In contrast to current probabilistic state-of-the art inversion methods, Inverse Twin Neural Network Regression is a deterministic framework for resolving multi-valued inverse mappings. I demonstrate the approach on problems that are defined by mathematical equations or by data, including multi-solution toy problems and robot arm inverse kinematics.
♻ ☆ Vision and Language: Novel Representations and Artificial intelligence for Driving Scene Safety Assessment and Autonomous Vehicle Planning
Vision-language models (VLMs) have recently emerged as powerful representation learning systems that align visual observations with natural language concepts, offering new opportunities for semantic reasoning in safety-critical autonomous driving. This paper investigates how vision-language representations support driving scene safety assessment and decision-making when integrated into perception, prediction, and planning pipelines. We study three complementary system-level use cases. First, we introduce a lightweight, category-agnostic hazard screening approach leveraging CLIP-based image-text similarity to produce a low-latency semantic hazard signal. This enables robust detection of diverse and out-of-distribution road hazards without explicit object detection or visual question answering. Second, we examine the integration of scene-level vision-language embeddings into a transformer-based trajectory planning framework using the Waymo Open Dataset. Our results show that naively conditioning planners on global embeddings does not improve trajectory accuracy, highlighting the importance of representation-task alignment and motivating the development of task-informed extraction methods for safety-critical planning. Third, we investigate natural language as an explicit behavioral constraint on motion planning using the doScenes dataset. In this setting, passenger-style instructions grounded in visual scene elements suppress rare but severe planning failures and improve safety-aligned behavior in ambiguous scenarios. Taken together, these findings demonstrate that vision-language representations hold significant promise for autonomous driving safety when used to express semantic risk, intent, and behavioral constraints. Realizing this potential is fundamentally an engineering problem requiring careful system design and structured grounding rather than direct feature injection.
♻ ☆ Efficient Robot Design with Multi-Objective Black-Box Optimization and Large Language Models
Various methods for robot design optimization have been developed so far. These methods are diverse, ranging from numerical optimization to black-box optimization. While numerical optimization is fast, it is not suitable for cases involving complex structures or discrete values, leading to frequent use of black-box optimization instead. However, black-box optimization suffers from low sampling efficiency and takes considerable sampling iterations to obtain good solutions. In this study, we propose a method to enhance the efficiency of robot body design based on black-box optimization by utilizing large language models (LLMs). In parallel with the sampling process based on black-box optimization, sampling is performed using LLMs, which are provided with problem settings and extensive feedback. We demonstrate that this method enables more efficient exploration of design solutions and discuss its characteristics and limitations.
comment: Accepted to IEEE Access, website: https://haraduka.github.io/urdf-llm-opt/ , video: https://www.youtube.com/watch?v=N9iMjx7of1w
♻ ☆ BPP: Long-Context Robot Imitation Learning by Focusing on Key History Frames
Many robot tasks require attending to the history of past observations. For example, finding an item in a room requires remembering which places have already been searched. However, the best-performing robot policies typically condition only on the current observation, limiting their applicability to such tasks. Naively conditioning on past observations often fails due to spurious correlations: policies latch onto incidental features of training histories that do not generalize to out-of-distribution trajectories upon deployment. We analyze why policies latch onto these spurious correlations and find that this problem stems from limited coverage over the space of possible histories during training, which grows exponentially with horizon. Existing regularization techniques provide inconsistent benefits across tasks, as they do not fundamentally address this coverage problem. Motivated by these findings, we propose Big Picture Policies (BPP), an approach that conditions on a minimal set of meaningful keyframes detected by a vision-language model. By projecting diverse rollouts onto a compact set of task-relevant events, BPP substantially reduces distribution shift between training and deployment, without sacrificing expressivity. We evaluate BPP on four challenging real-world manipulation tasks and three simulation tasks, all requiring history conditioning. BPP achieves 70% higher success rates than the best comparison on real-world evaluations. Videos are available at https://bigpicturepolicies.github.io/
♻ ☆ Query-Based Adaptive Aggregation for Multi-Dataset Joint Training Toward Universal Visual Place Recognition ICRA 2026
Deep learning methods for Visual Place Recognition (VPR) have advanced significantly, largely driven by large-scale datasets. However, most existing approaches are trained on a single dataset, which can introduce dataset-specific inductive biases and limit model generalization. While multi-dataset joint training offers a promising solution for developing universal VPR models, divergences among training datasets can saturate the limited information capacity in feature aggregation layers, leading to suboptimal performance. To address these challenges, we propose Query-based Adaptive Aggregation (QAA), a novel feature aggregation technique that leverages learned queries as reference codebooks to effectively enhance information capacity without significant computational or parameter complexity. We show that computing the Cross-query Similarity (CS) between query-level image features and reference codebooks provides a simple yet effective way to generate robust descriptors. Our results demonstrate that QAA outperforms state-of-the-art models, achieving balanced generalization across diverse datasets while maintaining peak performance comparable to dataset-specific models. Ablation studies further explore QAA's mechanisms and scalability. Visualizations reveal that the learned queries exhibit diverse attention patterns across datasets. Project page: http://xjh19971.github.io/QAA.
comment: 8 pages, 4 figures, accepted at ICRA 2026
♻ ☆ RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics CVPR 2025
Spatial understanding is a crucial capability that enables robots to perceive their surroundings, reason about their environment, and interact with it meaningfully. In modern robotics, these capabilities are increasingly provided by vision-language models. However, these models face significant challenges in spatial reasoning tasks, as their training data are based on general-purpose image datasets that often lack sophisticated spatial understanding. For example, datasets frequently do not capture reference frame comprehension, yet effective spatial reasoning requires understanding whether to reason from ego-, world-, or object-centric perspectives. To address this issue, we introduce RoboSpatial, a large-scale dataset for spatial understanding in robotics. It consists of real indoor and tabletop scenes, captured as 3D scans and egocentric images, and annotated with rich spatial information relevant to robotics. The dataset includes 1M images, 5k 3D scans, and 3M annotated spatial relationships, and the pairing of 2D egocentric images with 3D scans makes it both 2D- and 3D- ready. Our experiments show that models trained with RoboSpatial outperform baselines on downstream tasks such as spatial affordance prediction, spatial relationship prediction, and robot manipulation.
comment: CVPR 2025 (Oral); Project Website: https://chanh.ee/RoboSpatial
♻ ☆ Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment
The long-standing vision of general-purpose robots hinges on their ability to understand and act upon natural language instructions. Vision-Language-Action (VLA) models have made remarkable progress toward this goal, yet their generated actions can still misalign with the given instructions. In this paper, we investigate test-time verification as a means to shrink the "intention-action gap." We first characterize the test-time scaling laws for embodied instruction following and demonstrate that jointly scaling the number of rephrased instructions and generated actions greatly increases test-time sample diversity, often recovering correct actions more efficiently than scaling each dimension independently. To capitalize on these scaling laws, we present CoVer, a contrastive verifier for vision-language-action alignment, and show that our architecture scales gracefully with additional computational resources and data. We then introduce CoVer-VLA, a hierarchical test-time verification pipeline using the trained verifier. At deployment, our framework precomputes a diverse set of rephrased instructions from a Vision-Language-Model (VLM), repeatedly generates action candidates for each instruction, and then uses the verifier to select the optimal high-level prompt and low-level action chunks. Compared to scaling policy pre-training on the same data, our verification approach yields 22% gains in-distribution and 13% out-of-distribution on the SIMPLER benchmark, with a further 45% improvement in real-world experiments. On the PolaRiS benchmark, CoVer-VLA achieves 14% gains in task progress and 9% in success rate.
♻ ☆ Zero-Shot UAV Navigation in Forests via Relightable 3D Gaussian Splatting
UAV navigation in unstructured outdoor environments using passive monocular vision is hindered by the substantial visual domain gap between simulation and reality. While 3D Gaussian Splatting enables photorealistic scene reconstruction from real-world data, existing methods inherently couple static lighting with geometry, severely limiting policy generalization to dynamic real-world illumination. In this paper, we propose a novel end-to-end reinforcement learning framework designed for effective zero-shot transfer to unstructured outdoors. Within a high-fidelity simulation grounded in real-world data, our policy is trained to map raw monocular RGB observations directly to continuous control commands. To overcome photometric limitations, we introduce Relightable 3D Gaussian Splatting, which decomposes scene components to enable explicit, physically grounded editing of environmental lighting within the neural representation. By augmenting training with diverse synthesized lighting conditions ranging from strong directional sunlight to diffuse overcast skies, we compel the policy to learn robust, illumination-invariant visual features. Extensive real-world experiments demonstrate that a lightweight quadrotor achieves robust, collision-free navigation in complex forest environments at speeds up to 10 m/s, exhibiting significant resilience to drastic lighting variations without fine-tuning.
comment: 12 pages, 8 figures
♻ ☆ SkillWrapper: Generative Predicate Invention for Task-level Planning
Generalizing from individual skill executions to solving long-horizon tasks remains a core challenge in building autonomous agents. A promising direction is learning high-level, symbolic abstractions of the low-level skills of the agents, enabling reasoning and planning independent of the low-level state space. Among possible high-level representations, object-centric skill abstraction with symbolic predicates has been proven to be efficient because of its compatibility with domain-independent planners. Recent advances in foundation models have made it possible to generate symbolic predicates that operate on raw sensory inputs, a process we call generative predicate invention, to facilitate downstream abstraction learning. However, it remains unclear which formal properties the learned representations must satisfy, and how they can be learned to guarantee these properties. In this paper, we address both questions by presenting a formal theory of generative predicate invention for skill abstraction, resulting in symbolic operators that can be used for provably sound and complete planning. Within this framework, we propose SkillWrapper, a method that leverages foundation models to actively collect robot data and learn human-interpretable, plannable representations of black-box skills, using only RGB image observations. Our extensive empirical evaluation in simulation and on real robots shows that SkillWrapper learns abstract representations that enable solving unseen, long-horizon tasks in the real world with black-box skills.
Robotics 43
☆ ScenicRules: An Autonomous Driving Benchmark with Multi-Objective Specifications and Abstract Scenarios
Developing autonomous driving systems for complex traffic environments requires balancing multiple objectives, such as avoiding collisions, obeying traffic rules, and making efficient progress. In many situations, these objectives cannot be satisfied simultaneously, and explicit priority relations naturally arise. Also, driving rules require context, so it is important to formally model the environment scenarios within which such rules apply. Existing benchmarks for evaluating autonomous vehicles lack such combinations of multi-objective prioritized rules and formal environment models. In this work, we introduce ScenicRules, a benchmark for evaluating autonomous driving systems in stochastic environments under prioritized multi-objective specifications. We first formalize a diverse set of objectives to serve as quantitative evaluation metrics. Next, we design a Hierarchical Rulebook framework that encodes multiple objectives and their priority relations in an interpretable and adaptable manner. We then construct a compact yet representative collection of scenarios spanning diverse driving contexts and near-accident situations, formally modeled in the Scenic language. Experimental results show that our formalized objectives and Hierarchical Rulebooks align well with human driving judgments and that our benchmark effectively exposes agent failures with respect to the prioritized objectives. Our benchmark can be accessed at https://github.com/BerkeleyLearnVerify/ScenicRules/.
comment: 16 pages, 14 figures, 7 tables. Extended version of paper accepted to 2026 IEEE Intelligent Vehicles Symposium (IV 2026). ScenicRules benchmark available at https://github.com/BerkeleyLearnVerify/ScenicRules
☆ The Impact of Class Uncertainty Propagation in Perception-Based Motion Planning
Autonomous vehicles (AVs) are being increasingly deployed in urban environments. In order to operate safely and reliably, AVs need to account for the inherent uncertainty associated with perceiving the world through sensor data and incorporate that into their decision-making process. Uncertainty-aware planners have recently been developed to account for upstream perception and prediction uncertainty. However, such planners may be sensitive to prediction uncertainty miscalibration, the magnitude of which has not yet been characterized. Towards this end, we perform a detailed analysis on the impact that perceptual uncertainty propagation and calibration has on perception-based motion planning. We do so by comparing two novel prediction-planning pipelines with varying levels of uncertainty propagation on the recently-released nuPlan planning benchmark. We study the impact of upstream uncertainty calibration using closed-loop evaluation on the nuPlan challenge scenarios. We find that the method incorporating upstream uncertainty propagation demonstrates superior generalization to complex closed-loop scenarios.
comment: 8 pages, 10 figures. Code and results available at: https://github.com/aivanovic1/uncertainty-trajectron
☆ ODYN: An All-Shifted Non-Interior-Point Method for Quadratic Programming in Robotics and AI
We introduce ODYN, a novel all-shifted primal-dual non-interior-point quadratic programming (QP) solver designed to efficiently handle challenging dense and sparse QPs. ODYN combines all-shifted nonlinear complementarity problem (NCP) functions with proximal method of multipliers to robustly address ill-conditioned and degenerate problems, without requiring linear independence of the constraints. It exhibits strong warm-start performance and is well suited to both general-purpose optimization, and robotics and AI applications, including model-based control, estimation, and kernel-based learning methods. We provide an open-source implementation and benchmark ODYN on the Maros-Mészáros test set, demonstrating state-of-the-art convergence performance in small-to-high-scale problems. The results highlight ODYN's superior warm-starting capabilities, which are critical in sequential and real-time settings common in robotics and AI. These advantages are further demonstrated by deploying ODYN as the backend of an SQP-based predictive control framework (OdynSQP), as the implicitly differentiable optimization layer for deep learning (ODYNLayer), and the optimizer of a contact-dynamics simulation (ODYNSim).
☆ The human intention. A taxonomy attempt and its applications to robotics
Despite a surge in robotics research dedicated to inferring and understanding human intent, a universally accepted definition remains elusive since existing works often equate human intention with specific task-related goals. This article seeks to address this gap by examining the multifaceted nature of intention. Drawing on insights from psychology, it attempts to consolidate a definition of intention into a comprehensible framework for a broader audience. The article classifies different types of intention based on psychological and communication studies, offering guidance to researchers shifting from pure technical enhancements to a more human-centric perspective in robotics. It then demonstrates how various robotics studies can be aligned with these intention categories. Finally, through in-depth analyses of collaborative search and object transport use cases, the article underscores the significance of considering the diverse facets of human intention.
comment: Original version submitted to the International Journal of Social Robotics. Final version available on the SORO website
☆ Hybrid Model Predictive Control with Physics-Informed Neural Network for Satellite Attitude Control
Reliable spacecraft attitude control depends on accurate prediction of attitude dynamics, particularly when model-based strategies such as Model Predictive Control (MPC) are employed, where performance is limited by the quality of the internal system model. For spacecraft with complex dynamics, obtaining accurate physics-based models can be difficult, time-consuming, or computationally heavy. Learning-based system identification presents a compelling alternative; however, models trained exclusively on data frequently exhibit fragile stability properties and limited extrapolation capability. This work explores Physics-Informed Neural Networks (PINNs) for modeling spacecraft attitude dynamics and contrasts it with a conventional data-driven approach. A comprehensive dataset is generated using high-fidelity numerical simulations, and two learning methodologies are investigated: a purely data-driven pipeline and a physics-regularized approach that incorporates prior knowledge into the optimization process. The results indicate that embedding physical constraints during training leads to substantial improvements in predictive reliability, achieving a 68.17% decrease in mean relative error relative. When deployed within an MPC architecture, the physics-informed models yield superior closed-loop tracking performance and improved robustness to uncertainty. Furthermore, a hybrid control formulation that merges the learned nonlinear dynamics with a nominal linear model enables consistent steady-state convergence and significantly faster response, reducing settling times by 61.52%-76.42% under measurement noise and reaction wheel friction.
comment: Paper in peer-review. Copyright notice may change
☆ Dex4D: Task-Agnostic Point Track Policy for Sim-to-Real Dexterous Manipulation
Learning generalist policies capable of accomplishing a plethora of everyday tasks remains an open challenge in dexterous manipulation. In particular, collecting large-scale manipulation data via real-world teleoperation is expensive and difficult to scale. While learning in simulation provides a feasible alternative, designing multiple task-specific environments and rewards for training is similarly challenging. We propose Dex4D, a framework that instead leverages simulation for learning task-agnostic dexterous skills that can be flexibly recomposed to perform diverse real-world manipulation tasks. Specifically, Dex4D learns a domain-agnostic 3D point track conditioned policy capable of manipulating any object to any desired pose. We train this 'Anypose-to-Anypose' policy in simulation across thousands of objects with diverse pose configurations, covering a broad space of robot-object interactions that can be composed at test time. At deployment, this policy can be zero-shot transferred to real-world tasks without finetuning, simply by prompting it with desired object-centric point tracks extracted from generated videos. During execution, Dex4D uses online point tracking for closed-loop perception and control. Extensive experiments in simulation and on real robots show that our method enables zero-shot deployment for diverse dexterous manipulation tasks and yields consistent improvements over prior baselines. Furthermore, we demonstrate strong generalization to novel objects, scene layouts, backgrounds, and trajectories, highlighting the robustness and scalability of the proposed framework.
comment: Project page: https://dex4d.github.io/
☆ Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching
While recent advances in humanoid locomotion have achieved stable walking on varied terrains, capturing the agility and adaptivity of highly dynamic human motions remains an open challenge. In particular, agile parkour in complex environments demands not only low-level robustness, but also human-like motion expressiveness, long-horizon skill composition, and perception-driven decision-making. In this paper, we present Perceptive Humanoid Parkour (PHP), a modular framework that enables humanoid robots to autonomously perform long-horizon, vision-based parkour across challenging obstacle courses. Our approach first leverages motion matching, formulated as nearest-neighbor search in a feature space, to compose retargeted atomic human skills into long-horizon kinematic trajectories. This framework enables the flexible composition and smooth transition of complex skill chains while preserving the elegance and fluidity of dynamic human motions. Next, we train motion-tracking reinforcement learning (RL) expert policies for these composed motions, and distill them into a single depth-based, multi-skill student policy, using a combination of DAgger and RL. Crucially, the combination of perception and skill composition enables autonomous, context-aware decision-making: using only onboard depth sensing and a discrete 2D velocity command, the robot selects and executes whether to step over, climb onto, vault or roll off obstacles of varying geometries and heights. We validate our framework with extensive real-world experiments on a Unitree G1 humanoid robot, demonstrating highly dynamic parkour skills such as climbing tall obstacles up to 1.25m (96% robot height), as well as long-horizon multi-obstacle traversal with closed-loop adaptation to real-time obstacle perturbations.
☆ Solving Parameter-Robust Avoid Problems with Unknown Feasibility using Reinforcement Learning ICLR 2026
Recent advances in deep reinforcement learning (RL) have achieved strong results on high-dimensional control tasks, but applying RL to reachability problems raises a fundamental mismatch: reachability seeks to maximize the set of states from which a system remains safe indefinitely, while RL optimizes expected returns over a user-specified distribution. This mismatch can result in policies that perform poorly on low-probability states that are still within the safe set. A natural alternative is to frame the problem as a robust optimization over a set of initial conditions that specify the initial state, dynamics and safe set, but whether this problem has a solution depends on the feasibility of the specified set, which is unknown a priori. We propose Feasibility-Guided Exploration (FGE), a method that simultaneously identifies a subset of feasible initial conditions under which a safe policy exists, and learns a policy to solve the reachability problem over this set of initial conditions. Empirical results demonstrate that FGE learns policies with over 50% more coverage than the best existing method for challenging initial conditions across tasks in the MuJoCo simulator and the Kinetix simulator with pixel observations.
comment: ICLR 2026. The project page can be found at https://oswinso.xyz/fge
☆ FAST-EQA: Efficient Embodied Question Answering with Global and Local Region Relevancy WACV 2026
Embodied Question Answering (EQA) combines visual scene understanding, goal-directed exploration, spatial and temporal reasoning under partial observability. A central challenge is to confine physical search to question-relevant subspaces while maintaining a compact, actionable memory of observations. Furthermore, for real-world deployment, fast inference time during exploration is crucial. We introduce FAST-EQA, a question-conditioned framework that (i) identifies likely visual targets, (ii) scores global regions of interest to guide navigation, and (iii) employs Chain-of-Thought (CoT) reasoning over visual memory to answer confidently. FAST-EQA maintains a bounded scene memory that stores a fixed-capacity set of region-target hypotheses and updates them online, enabling robust handling of both single and multi-target questions without unbounded growth. To expand coverage efficiently, a global exploration policy treats narrow openings and doors as high-value frontiers, complementing local target seeking with minimal computation. Together, these components focus the agent's attention, improve scene coverage, and improve answer reliability while running substantially faster than prior approaches. On HMEQA and EXPRESS-Bench, FAST-EQA achieves state-of-the-art performance, while performing competitively on OpenEQA and MT-HM3D.
comment: WACV 2026
☆ Robot-Assisted Social Dining as a White Glove Service
Robot-assisted feeding enables people with disabilities who require assistance eating to enjoy a meal independently and with dignity. However, existing systems have only been tested in-lab or in-home, leaving in-the-wild social dining contexts (e.g., restaurants) largely unexplored. Designing a robot for such contexts presents unique challenges, such as dynamic and unsupervised dining environments that a robot needs to account for and respond to. Through speculative participatory design with people with disabilities, supported by semi-structured interviews and a custom AI-based visual storyboarding tool, we uncovered ideal scenarios for in-the-wild social dining. Our key insight suggests that such systems should: embody the principles of a white glove service where the robot (1) supports multimodal inputs and unobtrusive outputs; (2) has contextually sensitive social behavior and prioritizes the user; (3) has expanded roles beyond feeding; (4) adapts to other relationships at the dining table. Our work has implications for in-the-wild and group contexts of robot-assisted feeding.
comment: 20 pages, 9 figures. Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI '26)
☆ RaCo: Ranking and Covariance for Practical Learned Keypoints
This paper introduces RaCo, a lightweight neural network designed to learn robust and versatile keypoints suitable for a variety of 3D computer vision tasks. The model integrates three key components: the repeatable keypoint detector, a differentiable ranker to maximize matches with a limited number of keypoints, and a covariance estimator to quantify spatial uncertainty in metric scale. Trained on perspective image crops only, RaCo operates without the need for covisible image pairs. It achieves strong rotational robustness through extensive data augmentation, even without the use of computationally expensive equivariant network architectures. The method is evaluated on several challenging datasets, where it demonstrates state-of-the-art performance in keypoint repeatability and two-view matching, particularly under large in-plane rotations. Ultimately, RaCo provides an effective and simple strategy to independently estimate keypoint ranking and metric covariance without additional labels, detecting interpretable and repeatable interest points. The code is available at https://github.com/cvg/RaCo.
☆ MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction
Humanoid motion control has witnessed significant breakthroughs in recent years, with deep reinforcement learning (RL) emerging as a primary catalyst for achieving complex, human-like behaviors. However, the high dimensionality and intricate dynamics of humanoid robots make manual motion design impractical, leading to a heavy reliance on expensive motion capture (MoCap) data. These datasets are not only costly to acquire but also frequently lack the necessary geometric context of the surrounding physical environment. Consequently, existing motion synthesis frameworks often suffer from a decoupling of motion and scene, resulting in physical inconsistencies such as contact slippage or mesh penetration during terrain-aware tasks. In this work, we present MeshMimic, an innovative framework that bridges 3D scene reconstruction and embodied intelligence to enable humanoid robots to learn coupled "motion-terrain" interactions directly from video. By leveraging state-of-the-art 3D vision models, our framework precisely segments and reconstructs both human trajectories and the underlying 3D geometry of terrains and objects. We introduce an optimization algorithm based on kinematic consistency to extract high-quality motion data from noisy visual reconstructions, alongside a contact-invariant retargeting method that transfers human-environment interaction features to the humanoid agent. Experimental results demonstrate that MeshMimic achieves robust, highly dynamic performance across diverse and challenging terrains. Our approach proves that a low-cost pipeline utilizing only consumer-grade monocular sensors can facilitate the training of complex physical interactions, offering a scalable path toward the autonomous evolution of humanoid robots in unstructured environments.
comment: 17 pages, 6 figures
☆ Lifelong Scalable Multi-Agent Realistic Testbed and A Comprehensive Study on Design Choices in Lifelong AGV Fleet Management Systems
We present Lifelong Scalable Multi-Agent Realistic Testbed (LSMART), an open-source simulator to evaluate any Multi-Agent Path Finding (MAPF) algorithm in a Fleet Management System (FMS) with Automated Guided Vehicles (AGVs). MAPF aims to move a group of agents from their corresponding starting locations to their goals. Lifelong MAPF (LMAPF) is a variant of MAPF that continuously assigns new goals for agents to reach. LMAPF applications, such as autonomous warehouses, often require a centralized, lifelong system to coordinate the movement of a fleet of robots, typically AGVs. However, existing works on MAPF and LMAPF often assume simplified kinodynamic models, such as pebble motion, as well as perfect execution and communication for AGVs. Prior work has presented SMART, a software capable of evaluating any MAPF algorithms while considering agent kinodynamics, communication delays, and execution uncertainties. However, SMART is designed for MAPF, not LMAPF. Generalizing SMART to an FMS requires many more design choices. First, an FMS parallelizes planning and execution, raising the question of when to plan. Second, given planners with varying optimality and differing agent-model assumptions, one must decide how to plan. Third, when the planner fails to return valid solutions, the system must determine how to recover. In this paper, we first present LSMART, an open-source simulator that incorporates all these considerations to evaluate any MAPF algorithms in an FMS. We then provide experiment results based on state-of-the-art methods for each design choice, offering guidance on how to effectively design centralized lifelong AGV Fleet Management Systems. LSMART is available at https://smart-mapf.github.io/lifelong-smart.
☆ Estimating Human Muscular Fatigue in Dynamic Collaborative Robotic Tasks with Learning-Based Models ICRA 2026
Assessing human muscle fatigue is critical for optimizing performance and safety in physical human-robot interaction(pHRI). This work presents a data-driven framework to estimate fatigue in dynamic, cyclic pHRI using arm-mounted surface electromyography(sEMG). Subject-specific machine-learning regression models(Random Forest, XGBoost, and Linear Regression predict the fraction of cycles to fatigue(FCF) from three frequency-domain and one time-domain EMG features, and are benchmarked against a convolutional neural network(CNN) that ingests spectrograms of filtered EMG. Framing fatigue estimation as regression (rather than classification) captures continuous progression toward fatigue, supporting earlier detection, timely intervention, and adaptive robot control. In experiments with ten participants, a collaborative robot under admittance control guided repetitive lateral (left-right) end-effector motions until muscular fatigue. Average FCF RMSE across participants was 20.8+/-4.3% for the CNN, 23.3+/-3.8% for Random Forest, 24.8+/-4.5% for XGBoost, and 26.9+/-6.1% for Linear Regression. To probe cross-task generalization, one participant additionally performed unseen vertical (up-down) and circular repetitions; models trained only on lateral data were tested directly and largely retained accuracy, indicating robustness to changes in movement direction, arm kinematics, and muscle recruitment, while Linear Regression deteriorated. Overall, the study shows that both feature-based ML and spectrogram-based DL can estimate remaining work capacity during repetitive pHRI, with the CNN delivering the lowest error and the tree-based models close behind. The reported transfer to new motion patterns suggests potential for practical fatigue monitoring without retraining for every task, improving operator protection and enabling fatigue-aware shared autonomy, for safer fatigue-adaptive pHRI control.
comment: ICRA 2026 Original Contribution, Vienne, Austria
☆ Spatially-Aware Adaptive Trajectory Optimization with Controller-Guided Feedback for Autonomous Racing ICRA 2026
We present a closed-loop framework for autonomous raceline optimization that combines NURBS-based trajectory representation, CMA-ES global trajectory optimization, and controller-guided spatial feedback. Instead of treating tracking errors as transient disturbances, our method exploits them as informative signals of local track characteristics via a Kalman-inspired spatial update. This enables the construction of an adaptive, acceleration-based constraint map that iteratively refines trajectories toward near-optimal performance under spatially varying track and vehicle behavior. In simulation, our approach achieves a 17.38% lap time reduction compared to a controller parametrized with maximum static acceleration. On real hardware, tested with different tire compounds ranging from high to low friction, we obtain a 7.60% lap time improvement without explicitly parametrizing friction. This demonstrates robustness to changing grip conditions in real-world scenarios.
comment: Accepted at ICRA 2026
☆ World Action Models are Zero-shot Policies
State-of-the-art Vision-Language-Action (VLA) models excel at semantic generalization but struggle to generalize to unseen physical motions in novel environments. We introduce DreamZero, a World Action Model (WAM) built upon a pretrained video diffusion backbone. Unlike VLAs, WAMs learn physical dynamics by predicting future world states and actions, using video as a dense representation of how the world evolves. By jointly modeling video and action, DreamZero learns diverse skills effectively from heterogeneous robot data without relying on repetitive demonstrations. This results in over 2x improvement in generalization to new tasks and environments compared to state-of-the-art VLAs in real robot experiments. Crucially, through model and system optimizations, we enable a 14B autoregressive video diffusion model to perform real-time closed-loop control at 7Hz. Finally, we demonstrate two forms of cross-embodiment transfer: video-only demonstrations from other robots or humans yield a relative improvement of over 42% on unseen task performance with just 10-20 minutes of data. More surprisingly, DreamZero enables few-shot embodiment adaptation, transferring to a new embodiment with only 30 minutes of play data while retaining zero-shot generalization.
comment: Project page: https://dreamzero0.github.io/
☆ SpecFuse: A Spectral-Temporal Fusion Predictive Control Framework for UAV Landing on Oscillating Marine Platforms
Autonomous landing of Uncrewed Aerial Vehicles (UAVs) on oscillating marine platforms is severely constrained by wave-induced multi-frequency oscillations, wind disturbances, and prediction phase lags in motion prediction. Existing methods either treat platform motion as a general random process or lack explicit modeling of wave spectral characteristics, leading to suboptimal performance under dynamic sea conditions. To address these limitations, we propose SpecFuse: a novel spectral-temporal fusion predictive control framework that integrates frequency-domain wave decomposition with time-domain recursive state estimation for high-precision 6-DoF motion forecasting of Uncrewed Surface Vehicles (USVs). The framework explicitly models dominant wave harmonics to mitigate phase lags, refining predictions in real time via IMU data without relying on complex calibration. Additionally, we design a hierarchical control architecture featuring a sampling-based HPO-RRT* algorithm for dynamic trajectory planning under non-convex constraints and a learning-augmented predictive controller that fuses data-driven disturbance compensation with optimization-based execution. Extensive validations (2,000 simulations + 8 lake experiments) show our approach achieves a 3.2 cm prediction error, 4.46 cm landing deviation, 98.7% / 87.5% success rates (simulation / real-world), and 82 ms latency on embedded hardware, outperforming state-of-the-art methods by 44%-48% in accuracy. Its robustness to wave-wind coupling disturbances supports critical maritime missions such as search and rescue and environmental monitoring. All code, experimental configurations, and datasets will be released as open-source to facilitate reproducibility.
comment: 8 pages, 5 figures, 4 tables
☆ Grip as Needed, Glide on Demand: Ultrasonic Lubrication for Robotic Locomotion ICRA
Friction is the essential mediator of terrestrial locomotion, yet in robotic systems it is almost always treated as a passive property fixed by surface materials and conditions. Here, we introduce ultrasonic lubrication as a method to actively control friction in robotic locomotion. By exciting resonant structures at ultrasonic frequencies, contact interfaces can dynamically switch between "grip" and "slip" states, enabling locomotion. We developed two friction control modules, a cylindrical design for lumen-like environments and a flat-plate design for external surfaces, and integrated them into bio-inspired systems modeled after inchworm and wasp ovipositor locomotion. Both systems achieved bidirectional locomotion with nearly perfect locomotion efficiencies that exceeded 90%. Friction characterization experiments further demonstrated substantial friction reduction across various surfaces, including rigid, soft, granular, and biological tissue interfaces, under dry and wet conditions, and on surfaces with different levels of roughness, confirming the broad applicability of ultrasonic lubrication to locomotion tasks. These findings establish ultrasonic lubrication as a viable active friction control mechanism for robotic locomotion, with the potential to reduce design complexity and improve efficiency of robotic locomotion systems.
comment: Accepted for publication in the 2026 IEEE International Conference on Robotics and Automation (ICRA) in Vienna
☆ Constraining Streaming Flow Models for Adapting Learned Robot Trajectory Distributions
Robot motion distributions often exhibit multi-modality and require flexible generative models for accurate representation. Streaming Flow Policies (SFPs) have recently emerged as a powerful paradigm for generating robot trajectories by integrating learned velocity fields directly in action space, enabling smooth and reactive control. However, existing formulations lack mechanisms for adapting trajectories post-training to enforce safety and task-specific constraints. We propose Constraint-Aware Streaming Flow (CASF), a framework that augments streaming flow policies with constraint-dependent metrics that reshape the learned velocity field during execution. CASF models each constraint, defined in either the robot's workspace or configuration space, as a differentiable distance function that is converted into a local metric and pulled back into the robot's control space. Far from restricted regions, the resulting metric reduces to the identity; near constraint boundaries, it smoothly attenuates or redirects motion, effectively deforming the underlying flow to maintain safety. This allows trajectories to be adapted in real time, ensuring that robot actions respect joint limits, avoid collisions, and remain within feasible workspaces, while preserving the multi-modal and reactive properties of streaming flow policies. We demonstrate CASF in simulated and real-world manipulation tasks, showing that it produces constraint-satisfying trajectories that remain smooth, feasible, and dynamically consistent, outperforming standard post-hoc projection baselines.
comment: 8 pages, 8 figure
☆ VLM-DEWM: Dynamic External World Model for Verifiable and Resilient Vision-Language Planning in Manufacturing
Vision-language model (VLM) shows promise for high-level planning in smart manufacturing, yet their deployment in dynamic workcells faces two critical challenges: (1) stateless operation, they cannot persistently track out-of-view states, causing world-state drift; and (2) opaque reasoning, failures are difficult to diagnose, leading to costly blind retries. This paper presents VLM-DEWM, a cognitive architecture that decouples VLM reasoning from world-state management through a persistent, queryable Dynamic External World Model (DEWM). Each VLM decision is structured into an Externalizable Reasoning Trace (ERT), comprising action proposal, world belief, and causal assumption, which is validated against DEWM before execution. When failures occur, discrepancy analysis between predicted and observed states enables targeted recovery instead of global replanning. We evaluate VLM-DEWM on multi-station assembly, large-scale facility exploration, and real-robot recovery under induced failures. Compared to baseline memory-augmented VLM systems, VLM DEWM improves state-tracking accuracy from 56% to 93%, increases recovery success rate from below 5% to 95%, and significantly reduces computational overhead through structured memory. These results establish VLM-DEWM as a verifiable and resilient solution for long-horizon robotic operations in dynamic manufacturing environments.
☆ Selective Perception for Robot: Task-Aware Attention in Multimodal VLA
In robotics, Vision-Language-Action (VLA) models that integrate diverse multimodal signals from multi-view inputs have emerged as an effective approach. However, most prior work adopts static fusion that processes all visual inputs uniformly, which incurs unnecessary computational overhead and allows task-irrelevant background information to act as noise. Inspired by the principles of human active perception, we propose a dynamic information fusion framework designed to maximize the efficiency and robustness of VLA models. Our approach introduces a lightweight adaptive routing architecture that analyzes the current text prompt and observations from a wrist-mounted camera in real-time to predict the task-relevance of multiple camera views. By conditionally attenuating computations for views with low informational utility and selectively providing only essential visual features to the policy network, Our framework achieves computation efficiency proportional to task relevance. Furthermore, to efficiently secure large-scale annotation data for router training, we established an automated labeling pipeline utilizing Vision-Language Models (VLMs) to minimize data collection and annotation costs. Experimental results in real-world robotic manipulation scenarios demonstrate that the proposed approach achieves significant improvements in both inference efficiency and control performance compared to existing VLA models, validating the effectiveness and practicality of dynamic information fusion in resource-constrained, real-time robot control environments.
☆ Efficient Knowledge Transfer for Jump-Starting Control Policy Learning of Multirotors through Physics-Aware Neural Architectures
Efficiently training control policies for robots is a major challenge that can greatly benefit from utilizing knowledge gained from training similar systems through cross-embodiment knowledge transfer. In this work, we focus on accelerating policy training using a library-based initialization scheme that enables effective knowledge transfer across multirotor configurations. By leveraging a physics-aware neural control architecture that combines a reinforcement learning-based controller and a supervised control allocation network, we enable the reuse of previously trained policies. To this end, we utilize a policy evaluation-based similarity measure that identifies suitable policies for initialization from a library. We demonstrate that this measure correlates with the reduction in environment interactions needed to reach target performance and is therefore suited for initialization. Extensive simulation and real-world experiments confirm that our control architecture achieves state-of-the-art control performance, and that our initialization scheme saves on average up to $73.5\%$ of environment interactions (compared to training a policy from scratch) across diverse quadrotor and hexarotor designs, paving the way for efficient cross-embodiment transfer in reinforcement learning.
comment: 8 pages. Accepted to IEEE Robotics and Automation Letters
☆ Improving MLLMs in Embodied Exploration and Question Answering with Human-Inspired Memory Modeling
Deploying Multimodal Large Language Models as the brain of embodied agents remains challenging, particularly under long-horizon observations and limited context budgets. Existing memory assisted methods often rely on textual summaries, which discard rich visual and spatial details and remain brittle in non-stationary environments. In this work, we propose a non-parametric memory framework that explicitly disentangles episodic and semantic memory for embodied exploration and question answering. Our retrieval-first, reasoning-assisted paradigm recalls episodic experiences via semantic similarity and verifies them through visual reasoning, enabling robust reuse of past observations without rigid geometric alignment. In parallel, we introduce a program-style rule extraction mechanism that converts experiences into structured, reusable semantic memory, facilitating cross-environment generalization. Extensive experiments demonstrate state-of-the-art performance on embodied question answering and exploration benchmarks, yielding a 7.3% gain in LLM-Match and an 11.4% gain in LLM MatchXSPL on A-EQA, as well as +7.7% success rate and +6.8% SPL on GOAT-Bench. Analyses reveal that our episodic memory primarily improves exploration efficiency, while semantic memory strengthens complex reasoning of embodied agents.
☆ Lyapunov-Based $\mathcal{L}_2$-Stable PI-Like Control of a Four-Wheel Independently Driven and Steered Robot
In this letter, Lyapunov-based synthesis of a PI-like controller is proposed for $\mathcal{L}_2$-stable motion control of an independently driven and steered four-wheel mobile robot. An explicit, structurally verified model is used to enable systematic controller design with stability and performance guarantees suitable for real-time operation. A Lyapunov function is constructed to yield explicit bounds and $\mathcal{L}_2$ stability results, supporting feedback synthesis that reduces configuration dependent effects. The resulting control law maintains a PI-like form suitable for standard embedded implementation while preserving rigorous stability properties. Effectiveness and robustness are demonstrated experimentally on a real four-wheel mobile robot platform.
comment: SUBMITTED FOR POTENTIAL PUBLICATION IN IEEE ROBOTICS AND AUTOMATION LETTERS. PREPRINT VERSION
☆ One Agent to Guide Them All: Empowering MLLMs for Vision-and-Language Navigation via Explicit World Representation
A navigable agent needs to understand both high-level semantic instructions and precise spatial perceptions. Building navigation agents centered on Multimodal Large Language Models (MLLMs) demonstrates a promising solution due to their powerful generalization ability. However, the current tightly coupled design dramatically limits system performance. In this work, we propose a decoupled design that separates low-level spatial state estimation from high-level semantic planning. Unlike previous methods that rely on predefined, oversimplified textual maps, we introduce an interactive metric world representation that maintains rich and consistent information, allowing MLLMs to interact with and reason on it for decision-making. Furthermore, counterfactual reasoning is introduced to further elicit MLLMs' capacity, while the metric world representation ensures the physical validity of the produced actions. We conduct comprehensive experiments in both simulated and real-world environments. Our method establishes a new zero-shot state-of-the-art, achieving 48.8\% Success Rate (SR) in R2R-CE and 42.2\% in RxR-CE benchmarks. Furthermore, to validate the versatility of our metric representation, we demonstrate zero-shot sim-to-real transfer across diverse embodiments, including a wheeled TurtleBot 4 and a custom-built aerial drone. These real-world deployments verify that our decoupled framework serves as a robust, domain-invariant interface for embodied Vision-and-Language navigation.
☆ Hybrid F' and ROS2 Architecture for Vision-Based Autonomous Flight: Design and Experimental Validation
Autonomous aerospace systems require architectures that balance deterministic real-time control with advanced perception capabilities. This paper presents an integrated system combining NASA's F' flight software framework with ROS2 middleware via Protocol Buffers bridging. We evaluate the architecture through a 32.25-minute indoor quadrotor flight test using vision-based navigation. The vision system achieved 87.19 Hz position estimation with 99.90\% data continuity and 11.47 ms mean latency, validating real-time performance requirements. All 15 ground commands executed successfully with 100 % success rate, demonstrating robust F'--PX4 integration. System resource utilization remained low (15.19 % CPU, 1,244 MB RAM) with zero stale telemetry messages, confirming efficient operation on embedded platforms. Results validate the feasibility of hybrid flight-software architectures combining certification-grade determinism with flexible autonomy for autonomous aerial vehicles.
comment: Paper accepted to ICIT 2026
☆ ActionCodec: What Makes for Good Action Tokenizers
Vision-Language-Action (VLA) models leveraging the native autoregressive paradigm of Vision-Language Models (VLMs) have demonstrated superior instruction-following and training efficiency. Central to this paradigm is action tokenization, yet its design has primarily focused on reconstruction fidelity, failing to address its direct impact on VLA optimization. Consequently, the fundamental question of \textit{what makes for good action tokenizers} remains unanswered. In this paper, we bridge this gap by establishing design principles specifically from the perspective of VLA optimization. We identify a set of best practices based on information-theoretic insights, including maximized temporal token overlap, minimized vocabulary redundancy, enhanced multimodal mutual information, and token independence. Guided by these principles, we introduce \textbf{ActionCodec}, a high-performance action tokenizer that significantly enhances both training efficiency and VLA performance across diverse simulation and real-world benchmarks. Notably, on LIBERO, a SmolVLM2-2.2B fine-tuned with ActionCodec achieves a 95.5\% success rate without any robotics pre-training. With advanced architectural enhancements, this reaches 97.4\%, representing a new SOTA for VLA models without robotics pre-training. We believe our established design principles, alongside the released model, will provide a clear roadmap for the community to develop more effective action tokenizers.
☆ Fluoroscopy-Constrained Magnetic Robot Control via Zernike-Based Field Modeling and Nonlinear MPC
Magnetic actuation enables surgical robots to navigate complex anatomical pathways while reducing tissue trauma and improving surgical precision. However, clinical deployment is limited by the challenges of controlling such systems under fluoroscopic imaging, which provides low frame rate and noisy pose feedback. This paper presents a control framework that remains accurate and stable under such conditions by combining a nonlinear model predictive control (NMPC) framework that directly outputs coil currents, an analytically differentiable magnetic field model based on Zernike polynomials, and a Kalman filter to estimate the robot state. Experimental validation is conducted with two magnetic robots in a 3D-printed fluid workspace and a spine phantom replicating drug delivery in the epidural space. Results show the proposed control method remains highly accurate when feedback is downsampled to 3 Hz with added Gaussian noise (sigma = 2 mm), mimicking clinical fluoroscopy. In the spine phantom experiments, the proposed method successfully executed a drug delivery trajectory with a root mean square (RMS) position error of 1.18 mm while maintaining safe clearance from critical anatomical boundaries.
☆ A Comparison of Bayesian Prediction Techniques for Mobile Robot Trajectory Tracking
This paper presents a performance comparison of different estimation and prediction techniques applied to the problem of tracking multiple robots. The main performance criteria are the magnitude of the estimation or prediction error, the computational effort and the robustness of each method to non-Gaussian noise. Among the different techniques compared are the well known Kalman filters and their different variants (e.g. extended and unscented), and the more recent techniques relying on Sequential Monte Carlo Sampling methods, such as particle filters and Gaussian Mixture Sigma Point Particle Filter.
comment: Accepted in Robotica (Dec. 2007), vol. 26, n. 5, pp. 571-585 (c) 2008 Cambridge University Press. https://doi.org/10.1017/S0263574708004153
☆ Feasibility-aware Imitation Learning from Observation with Multimodal Feedback
Imitation learning frameworks that learn robot control policies from demonstrators' motions via hand-mounted demonstration interfaces have attracted increasing attention. However, due to differences in physical characteristics between demonstrators and robots, this approach faces two limitations: i) the demonstration data do not include robot actions, and ii) the demonstrated motions may be infeasible for robots. These limitations make policy learning difficult. To address them, we propose Feasibility-Aware Behavior Cloning from Observation (FABCO). FABCO integrates behavior cloning from observation, which complements robot actions using robot dynamics models, with feasibility estimation. In feasibility estimation, the demonstrated motions are evaluated using a robot-dynamics model, learned from the robot's execution data, to assess reproducibility under the robot's dynamics. The estimated feasibility is used for multimodal feedback and feasibility-aware policy learning to improve the demonstrator's motions and learn robust policies. Multimodal feedback provides feasibility through the demonstrator's visual and haptic senses to promote feasible demonstrated motions. Feasibility-aware policy learning reduces the influence of demonstrated motions that are infeasible for robots, enabling the learning of policies that robots can execute stably. We conducted experiments with 15 participants on two tasks and confirmed that FABCO improves imitation learning performance by more than 3.2 times compared to the case without feasibility feedback.
☆ OSCAR: An Ovipositor-Inspired Self-Propelling Capsule Robot for Colonoscopy
Self-propelling robotic capsules eliminate shaft looping of conventional colonoscopy, reducing patient discomfort. However, reliably moving within the slippery, viscoelastic environment of the colon remains a significant challenge. We present OSCAR, an ovipositor-inspired self-propelling capsule robot that translates the transport strategy of parasitic wasps into a propulsion mechanism for colonoscopy. OSCAR mechanically encodes the ovipositor-inspired motion pattern through a spring-loaded cam system that drives twelve circumferential sliders in a coordinated, phase-shifted sequence. By tuning the motion profile to maximize the retract phase relative to the advance phase, the capsule creates a controlled friction anisotropy at the interface that generates net forward thrust. We developed an analytical model incorporating a Kelvin-Voigt formulation to capture the viscoelastic stick--slip interactions between the sliders and the tissue, linking the asymmetry between advance and retract phase durations to mean thrust, and slider-reversal synchronization to thrust stability. Comprehensive force characterization experiments in ex-vivo porcine colon revealed a mean steady-state traction force of 0.85 N, closely matching the model. Furthermore, experiments confirmed that thrust generation is speed-independent and scales linearly with the phase asymmetry, in agreement with theoretical predictions, underscoring the capsule's predictable performance and scalability. In locomotion validation experiments, OSCAR demonstrated robust performance, achieving an average speed of 3.08 mm/s, a velocity sufficient to match the cecal intubation times of conventional colonoscopy. By coupling phase-encoded friction anisotropy with a predictive model, OSCAR delivers controllable thrust generation at low normal loads, enabling safer and more robust self-propelling locomotion for robotic capsule colonoscopy.
♻ ☆ Toward Context-Aware Exoskeleton Assistance: Integrating Computer Vision Payload Estimation with a User-Centric Optimization Space
Back-support exoskeletons (BSEs) mitigate musculoskeletal strain, yet their efficacy depends on precise, context-aware modulation. This paper introduces a user-centric optimization framework and a vision-based adaptive control strategy for industrial BSEs. First, we constructed a multi-metric optimization space, integrating electromyography reduction, perceived discomfort, and user preference, through baseline experiments with 12 subjects. This revealed a non-linear relationship between optimal assistance and payload. Second, we developed a predictive computer vision pipeline using a Vision Transformer (DINOv2) to estimate payloads before lifting, effectively overcoming actuation latency. Validation with 12 subjects confirmed the system's robustness, achieving over 82% estimation accuracy. Crucially, the adaptive controller reduced peak back muscle activation by up to 23% compared to static baselines while optimizing user comfort. These results validate the proposed framework, demonstrating that pre-lift environmental perception and user-centric optimization significantly enhance physical assistance and human-robot interaction in industrial settings.
♻ ☆ Learning Diffusion Policies for Robotic Manipulation of Timber Joinery under Fabrication Uncertainty
Fabrication uncertainties, such as tolerance accumulation and material imperfections, pose a significant challenge to contact-rich robotic manipulation in construction by hindering precise and robust assembly. In this paper, we investigate the performance and robustness of diffusion policy learning for contact-rich assembly at the construction scale, using a tight-fitting timber mortise and tenon joint as a case study. A two-phase experimental study is conducted: first, to evaluate baseline policy performance and applicability; second, to assess policy robustness under fabrication-induced uncertainties modeled as randomized perturbations to the mortise position. The diffusion policy is trained on teleoperated demonstrations using an industrial robotic arm conditioned on end-effector pose and force/torque feedback. The best-performing policy achieved a total average success rate of 75% under perturbations up to 10 mm, including 100% success in unperturbed cases. The results demonstrate the potential of sensory-motor diffusion policies to enable high-precision contact-rich manipulation on large-scale industrial robotic arms, reducing reliance on skilled manual intervention. This work advances robotic construction under uncertainty and provides practical insights for deploying learning-based control in real-world Architectural, Engineering, and Construction applications.
♻ ☆ Collaborative Multi-Robot Non-Prehensile Manipulation via Flow-Matching Co-Generation
Coordinating a team of robots to reposition multiple objects in cluttered environments requires reasoning jointly about where robots should establish contact, how to manipulate objects once contact is made, and how to navigate safely and efficiently at scale. Prior approaches typically fall into two extremes -- either learning the entire task or relying on privileged information and hand-designed planners -- both of which struggle to handle diverse objects in long-horizon tasks. To address these challenges, we present a unified framework for collaborative multi-robot, multi-object non-prehensile manipulation that integrates flow-matching co-generation with anonymous multi-robot motion planning. Within this framework, a generative model co-generates contact formations and manipulation trajectories from visual observations, while a novel motion planner conveys robots at scale. Crucially, the same planner also supports coordination at the object level, assigning manipulated objects to larger target structures and thereby unifying robot- and object-level reasoning within a single algorithmic framework. Experiments in challenging simulated environments demonstrate that our approach outperforms baselines in both motion planning and manipulation tasks, highlighting the benefits of generative co-design and integrated planning for scaling collaborative manipulation to complex multi-agent, multi-object settings. Visit gco-paper.github.io for code and demonstrations.
♻ ☆ FlowDrive: moderated flow matching with data balancing for trajectory planning
Learning-based planners are sensitive to the long-tailed distribution of driving data. Common maneuvers dominate datasets, while dangerous or rare scenarios are sparse. This imbalance can bias models toward the frequent cases and degrade performance on critical scenarios. To tackle this problem, we compare balancing strategies for sampling training data and find reweighting by trajectory pattern an effective approach. We then present FlowDrive, a flow-matching trajectory planner that learns a conditional rectified flow to map noise directly to trajectory distributions with few flow-matching steps. We further introduce moderated, in-the-loop guidance that injects small perturbation between flow steps to systematically increase trajectory diversity while remaining scene-consistent. On nuPlan and the interaction-focused interPlan benchmarks, FlowDrive achieves state-of-the-art results among learning-based planners and approaches methods with rule-based refinements. After adding moderated guidance and light post-processing (FlowDrive*), it achieves overall state-of-the-art performance across nearly all benchmark splits. Our code is available at https://github.com/einsteinguang/flow_drive_planner.
♻ ☆ Online Fine-Tuning of Pretrained Controllers for Autonomous Driving via Real-Time Recurrent RL
Deploying pretrained policies in real-world applications presents substantial challenges that fundamentally limit the practical applicability of learning-based control systems. When autonomous systems encounter environmental changes in system dynamics, sensor drift, or task objectives, fixed policies rapidly degrade in performance. We show that employing Real-Time Recurrent Reinforcement Learning (RTRRL), a biologically plausible algorithm for online adaptation, can effectively fine-tune a pretrained policy to improve autonomous agents' performance on driving tasks. We further show that RTRRL synergizes with a recent biologically inspired recurrent network model, the Liquid-Resistance Liquid-Capacitance RNN. We demonstrate the effectiveness of this closed-loop approach in a simulated CarRacing environment and in a real-world line-following task with a RoboRacer car equipped with an event camera.
♻ ☆ Fast and Near-Optimal Collision-Free Robot Scheduling On Paths
In this paper, we address the problem of scheduling a set of robots to complete tasks in a laboratory environment, modelled as a graph, while avoiding collisions. We analyze the dynamic programming algorithm (PA) introduced in arXiv:2402.12019 and present three baselines for comparison: an integer programming approach (IP) that always yields an optimal solution, a greedy algorithm (GA), and a simple randomized algorithm (RA). We show that for a path graph, PA, GA, and RA find solutions several orders of magnitude faster than IP (the optimal baseline), with PA returning optimal results in the vast majority of cases. Our scaled experiments comparing non-optimal algorithms show that the average schedule timespan produced by PA is less than half that of RA and GA. This outperformance is consistent across varying path lengths, task durations and distributions, number and allocations of tasks and robots, and task-to-robot ratios. This work serves two purposes. First, we present three algorithms for scheduling on line graphs, including a novel integer programming formulation for finding optimal solutions. Second, we demonstrate that PA produces near-optimal schedules that outperform all non-optimal baselines while maintaining a comparable runtime. Code is available at https://github.com/sea26-robots/code.
♻ ☆ Prescribed Performance Control of Unknown Euler-Lagrange Systems Under Input Constraints
In this paper, we present a prescribed performance control framework for trajectory tracking in Euler-Lagrange systems with unknown dynamics and prescribed input constraints. The proposed approach enforces hard funnel constraints, meaning that the prescribed performance bounds must not be violated during operation. We derive feasibility conditions that guarantee the tracking error evolves within these predefined funnels while ensuring bounded control inputs. To handle situations where the feasibility conditions are not satisfied, we introduce two approximation-free control strategies: one that actively drives the error back toward the funnel and another that prioritizes safety by preventing further deviation. The effectiveness and robustness of the proposed method are demonstrated through simulation studies and hardware experiments, highlighting its suitability for real-world robotic systems operating under strict input limits.
♻ ☆ MoIRA: Modular Instruction Routing Architecture for Multi-Task Robotics
Mixture-of-Experts (MoE) approaches have recently gained traction in robotics applications due to their ability to dynamically allocate computational resources and specialize sub-networks for distinct tasks or environmental contexts, enabling more efficient decision-making. Such systems often comprise sparsely activated experts combined under a single monolithic architecture and require a well-configured internal routing mechanism, which does not allow for selective low-level expert and router customization and requires additional training. We propose MoIRA, an architecture-agnostic modular MoE framework designed to coordinate existing experts with an external text-based router. MoIRA incorporates two zero-shot routing options: embedding-based similarity and prompt-driven language model inference. In our experiments, we choose large Vision-Language-Action models, gr00t-N1 and $π_0$, as the underlying experts, and train low-rank adapters for low-overhead inference. We evaluate MoIRA on various GR1 Humanoid tasks and LIBERO Spatial and Goal benchmarks, where it consistently outperforms generalist models and competes with other MoE pipelines. Additionally, we analyse the robustness of the proposed approach to the variations of the instructions. While relying solely on textual descriptions of tasks and experts, MoIRA demonstrates the practical viability of modular deployment with precise, low-effort routing and provides an alternative, scalable foundation for future multi-expert robotic systems.
comment: Updated to reflect the final accepted version published in Neurocomputing
♻ ☆ SafeFlowMPC: Predictive and Safe Trajectory Planning for Robot Manipulators with Learning-based Policies ICRA 2026
The emerging integration of robots into everyday life brings several major challenges. Compared to classical industrial applications, more flexibility is needed in combination with real-time reactivity. Learning-based methods can train powerful policies based on demonstrated trajectories, such that the robot generalizes a task to similar situations. However, these black-box models lack interpretability and rigorous safety guarantees. Optimization-based methods provide these guarantees but lack the required flexibility and generalization capabilities. This work proposes SafeFlowMPC, a combination of flow matching and online optimization to combine the strengths of learning and optimization. This method guarantees safety at all times and is designed to meet the demands of real-time execution by using a suboptimal model-predictive control formulation. SafeFlowMPC achieves strong performance in three real-world experiments on a KUKA 7-DoF manipulator, namely two grasping experiment and a dynamic human-robot object handover experiment. A video of the experiments is available at http://www.acin.tuwien.ac.at/42d6. The code is available at https://github.com/TU-Wien-ACIN-CDS/SafeFlowMPC.
comment: Accepted at ICRA 2026
♻ ☆ ETGL-DDPG: A Deep Deterministic Policy Gradient Algorithm for Sparse Reward Continuous Control
We consider deep deterministic policy gradient (DDPG) in the context of reinforcement learning with sparse rewards. To enhance exploration, we introduce a search procedure, \emph{$ε{t}$-greedy}, which generates exploratory options for exploring less-visited states. We prove that search using $εt$-greedy has polynomial sample complexity under mild MDP assumptions. To more efficiently use the information provided by rewarded transitions, we develop a new dual experience replay buffer framework, \emph{GDRB}, and implement \emph{longest n-step returns}. The resulting algorithm, \emph{ETGL-DDPG}, integrates all three techniques: \bm{$εt$}-greedy, \textbf{G}DRB, and \textbf{L}ongest $n$-step, into DDPG. We evaluate ETGL-DDPG on standard benchmarks and demonstrate that it outperforms DDPG, as well as other state-of-the-art methods, across all tested sparse-reward continuous environments. Ablation studies further highlight how each strategy individually enhances the performance of DDPG in this setting.
comment: We have expanded the related work section with more detailed discussions and enhanced our experiments by incorporating additional data and analysis
♻ ☆ Octopus-like Reaching Motion: A Perspective Inspired by Whipping
The stereotypical reaching motion of the octopus arm has drawn growing attention for its efficient control of a highly deformable body. Previous studies suggest that its characteristic bend propagation may share underlying principles with the dynamics of a whip. This work investigates whether whip-like passive dynamics in water can reproduce the kinematic features observed in biological reaching and their similarities and differences. Platform-based whipping tests were performed in water and air while systematically varying material stiffness and driving speed. Image-based quantification revealed that the Ecoflex Gel 2 arm driven at 150 rpm (motor speed) reproduced curvature propagation similar to that observed in octopus reaching. However, its bend-point velocity decreased monotonically rather than exhibiting the biological bell-shaped profile, confirming that the octopus reaching movement is not merely a passive whipping behavior. The absence of propagation in air further highlights the critical role of the surrounding medium in forming octopus-like reaching motion. This study provides a new perspective for understand biological reaching movement, and offers a potential platform for future hydrodynamic research.
comment: The first two listed authors contributed equally. Yiyuan Zhang is the corresponding author
♻ ☆ Satellite Autonomous Clock Fault Monitoring with Inter-Satellite Ranges Using Euclidean Distance Matrices
To address the need for robust positioning, navigation, and timing services in lunar environments, this paper proposes a novel onboard clock phase jump detection framework for satellite constellations using range measurements obtained from dual one-way inter-satellite links. Our approach leverages vertex redundantly rigid graphs to detect faults without relying on prior knowledge of satellite positions or clock biases, providing flexibility for lunar satellite networks with diverse satellite types and operators. We model satellite constellations as graphs, where satellites are vertices and inter-satellite links are edges. The proposed algorithm detects and identifies satellites with clock jumps by monitoring the singular values of the geometric-centered Euclidean distance matrix (GCEDM) of 5-clique sub-graphs. The proposed method is validated through simulations of a GPS constellation and a notional constellation around the Moon, demonstrating its effectiveness in various configurations.
comment: This manuscript was submitted to the NAVIGATION: Journal of the Institute of Navigation
Robotics 45
☆ Neurosim: A Fast Simulator for Neuromorphic Robot Perception
Neurosim is a fast, real-time, high-performance library for simulating sensors such as dynamic vision sensors, RGB cameras, depth sensors, and inertial sensors. It can also simulate agile dynamics of multi-rotor vehicles in complex and dynamic environments. Neurosim can achieve frame rates as high as ~2700 FPS on a desktop GPU. Neurosim integrates with a ZeroMQ-based communication library called Cortex to facilitate seamless integration with machine learning and robotics workflows. Cortex provides a high-throughput, low-latency message-passing system for Python and C++ applications, with native support for NumPy arrays and PyTorch tensors. This paper discusses the design philosophy behind Neurosim and Cortex. It demonstrates how they can be used to (i) train neuromorphic perception and control algorithms, e.g., using self-supervised learning on time-synchronized multi-modal data, and (ii) test real-time implementations of these algorithms in closed-loop. Neurosim and Cortex are available at https://github.com/grasp-lyrl/neurosim .
comment: 13 pages, 6 figures
☆ BPP: Long-Context Robot Imitation Learning by Focusing on Key History Frames
Many robot tasks require attending to the history of past observations. For example, finding an item in a room requires remembering which places have already been searched. However, the best-performing robot policies typically condition only on the current observation, limiting their applicability to such tasks. Naively conditioning on past observations often fails due to spurious correlations: policies latch onto incidental features of training histories that do not generalize to out-of-distribution trajectories upon deployment. We analyze why policies latch onto these spurious correlations and find that this problem stems from limited coverage over the space of possible histories during training, which grows exponentially with horizon. Existing regularization techniques provide inconsistent benefits across tasks, as they do not fundamentally address this coverage problem. Motivated by these findings, we propose Big Picture Policies (BPP), an approach that conditions on a minimal set of meaningful keyframes detected by a vision-language model. By projecting diverse rollouts onto a compact set of task-relevant events, BPP substantially reduces distribution shift between training and deployment, without sacrificing expressivity. We evaluate BPP on four challenging real-world manipulation tasks and three simulation tasks, all requiring history conditioning. BPP achieves 70% higher success rates than the best comparison on real-world evaluations.
☆ DM0: An Embodied-Native Vision-Language-Action Model towards Physical AI
Moving beyond the traditional paradigm of adapting internet-pretrained models to physical tasks, we present DM0, an Embodied-Native Vision-Language-Action (VLA) framework designed for Physical AI. Unlike approaches that treat physical grounding as a fine-tuning afterthought, DM0 unifies embodied manipulation and navigation by learning from heterogeneous data sources from the onset. Our methodology follows a comprehensive three-stage pipeline: Pretraining, Mid-Training, and Post-Training. First, we conduct large-scale unified pretraining on the Vision-Language Model (VLM) using diverse corpora--seamlessly integrating web text, autonomous driving scenarios, and embodied interaction logs-to jointly acquire semantic knowledge and physical priors. Subsequently, we build a flow-matching action expert atop the VLM. To reconcile high-level reasoning with low-level control, DM0 employs a hybrid training strategy: for embodied data, gradients from the action expert are not backpropagated to the VLM to preserve generalized representations, while the VLM remains trainable on non-embodied data. Furthermore, we introduce an Embodied Spatial Scaffolding strategy to construct spatial Chain-of-Thought (CoT) reasoning, effectively constraining the action solution space. Experiments on the RoboChallenge benchmark demonstrate that DM0 achieves state-of-the-art performance in both Specialist and Generalist settings on Table30.
comment: Authors are listed in alphabetical order. Code is available at https://github.com/Dexmal/dexbotic
☆ PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene Arrangement ICLR 2026
Automatically generating interactive 3D environments is crucial for scaling up robotic data collection in simulation. While prior work has primarily focused on 3D asset placement, it often overlooks the physical relationships between objects (e.g., contact, support, balance, and containment), which are essential for creating complex and realistic manipulation scenarios such as tabletop arrangements, shelf organization, or box packing. Compared to classical 3D layout generation, producing complex physical scenes introduces additional challenges: (a) higher object density and complexity (e.g., a small shelf may hold dozens of books), (b) richer supporting relationships and compact spatial layouts, and (c) the need to accurately model both spatial placement and physical properties. To address these challenges, we propose PhyScensis, an LLM agent-based framework powered by a physics engine, to produce physically plausible scene configurations with high complexity. Specifically, our framework consists of three main components: an LLM agent iteratively proposes assets with spatial and physical predicates; a solver, equipped with a physics engine, realizes these predicates into a 3D scene; and feedback from the solver informs the agent to refine and enrich the configuration. Moreover, our framework preserves strong controllability over fine-grained textual descriptions and numerical parameters (e.g., relative positions, scene stability), enabled through probabilistic programming for stability and a complementary heuristic that jointly regulates stability and spatial relations. Experimental results show that our method outperforms prior approaches in scene complexity, visual quality, and physical accuracy, offering a unified pipeline for generating complex physical scene layouts for robotic manipulation.
comment: ICLR 2026
☆ PAct: Part-Decomposed Single-View Articulated Object Generation
Articulated objects are central to interactive 3D applications, including embodied AI, robotics, and VR/AR, where functional part decomposition and kinematic motion are essential. Yet producing high-fidelity articulated assets remains difficult to scale because it requires reliable part decomposition and kinematic rigging. Existing approaches largely fall into two paradigms: optimization-based reconstruction or distillation, which can be accurate but often takes tens of minutes to hours per instance, and inference-time methods that rely on template or part retrieval, producing plausible results that may not match the specific structure and appearance in the input observation. We introduce a part-centric generative framework for articulated object creation that synthesizes part geometry, composition, and articulation under explicit part-aware conditioning. Our representation models an object as a set of movable parts, each encoded by latent tokens augmented with part identity and articulation cues. Conditioned on a single image, the model generates articulated 3D assets that preserve instance-level correspondence while maintaining valid part structure and motion. The resulting approach avoids per-instance optimization, enables fast feed-forward inference, and supports controllable assembly and articulation, which are important for embodied interaction. Experiments on common articulated categories (e.g., drawers and doors) show improved input consistency, part accuracy, and articulation plausibility over optimization-based and retrieval-driven baselines, while substantially reducing inference time.
comment: Technical Report(11 figures, 14 pages), Project Page: https://PAct-project.github.io
☆ Morphing of and writing with a scissor linkage mechanism
Kinematics of mechanisms is intricately coupled to their geometry and their utility often arises out of the ability to perform reproducible motion with fewer actuating degrees of freedom. In this article, we explore the assembly of scissor-units, each made of two rigid linear members connected by a pin joint. The assembly has a single degree of freedom, where actuating any single unit results in a shape change of the entire assembly. We derive expressions for the effective curvature of the unit and the trajectory of the mechanism's tip as a function of the geometric variables which we then use as the basis to program two tasks in the mechanism: shape morphing and writing. By phrasing these tasks as optimization problems and utilizing the differentiable simulation framework, we arrive at solutions that are then tested in table-top experiments. Our results show that the geometry of scissor assemblies can be leveraged for automated navigation and inspection in complex domains, in light of the optimization framework. However, we highlight that the challenges associated with rapid programming and error-free implementation in experiments without feedback still remain.
☆ Kalman Filtering Based Flight Management System Modeling for AAM Aircraft
Advanced Aerial Mobility (AAM) operations require strategic flight planning services that predict both spatial and temporal uncertainties to safely validate flight plans against hazards such as weather cells, restricted airspaces, and CNS disruption areas. Current uncertainty estimation methods for AAM vehicles rely on conservative linear models due to limited real-world performance data. This paper presents a novel Kalman Filter-based uncertainty propagation method that models AAM Flight Management System (FMS) architectures through sigmoid-blended measurement noise covariance. Unlike existing approaches with fixed uncertainty thresholds, our method continuously adapts the filter's measurement trust based on progress toward waypoints, enabling FMS correction behavior to emerge naturally. The approach scales proportionally with control inputs and is tunable to match specific aircraft characteristics or route conditions. We validate the method using real ADS-B data from general aviation aircraft divided into training and verification sets. Uncertainty propagation parameters were tuned on the training set, achieving 76% accuracy in predicting arrival times when compared against the verification dataset, demonstrating the method's effectiveness for strategic flight plan validation in AAM operations.
☆ Affordance Transfer Across Object Instances via Semantically Anchored Functional Map
Traditional learning from demonstration (LfD) generally demands a cumbersome collection of physical demonstrations, which can be time-consuming and challenging to scale. Recent advances show that robots can instead learn from human videos by extracting interaction cues without direct robot involvement. However, a fundamental challenge remains: how to generalize demonstrated interactions across different object instances that share similar functionality but vary significantly in geometry. In this work, we propose \emph{Semantic Anchored Functional Maps} (SemFM), a framework for transferring affordances across objects from a single visual demonstration. Starting from a coarse mesh reconstructed from an image, our method identifies semantically corresponding functional regions between objects, selects mutually exclusive semantic anchors, and propagates these constraints over the surface using a functional map to obtain a dense, semantically consistent correspondence. This enables demonstrated interaction regions to be transferred across geometrically diverse objects in a lightweight and interpretable manner. Experiments on synthetic object categories and real-world robotic manipulation tasks show that our approach enables accurate affordance transfer with modest computational cost, making it well-suited for practical robotic perception-to-action pipelines.
☆ Scalable Multi-Robot Path Planning via Quadratic Unconstrained Binary Optimization
Multi-Agent Path Finding (MAPF) remains a fundamental challenge in robotics, where classical centralized approaches exhibit exponential growth in joint-state complexity as the number of agents increases. This paper investigates Quadratic Unconstrained Binary Optimization (QUBO) as a structurally scalable alternative for simultaneous multi-robot path planning. This approach is a robotics-oriented QUBO formulation incorporating BFS-based logical pre-processing (achieving over 95% variable reduction), adaptive penalty design for collision and constraint enforcement, and a time-windowed decomposition strategy that enables execution within current hardware limitations. An experimental evaluation in grid environments with up to four robots demonstrated near-optimal solutions in dense scenarios and favorable scaling behavior compared to sequential classical planning. These results establish a practical and reproducible baseline for future quantum and quantum-inspired multi-robot coordinations.
comment: 21 pages, 9 figures, 1 table. Accompanying open-source implementation at https://github.com/JavideuS/Spooky
☆ Analysis of a Cuspidal 6R Robot
We present a theoretical and numerical analysis of the kinematics for the "Transpressor", a cuspidal 6R robot. It admits up to 16 inverse kinematics solutions which are described geometrically. For special target poses, we provide the solutions analytically and present a simple numerical solver for the general case. Moreover, an analytical estimate of the Jacobian determinant on a path between two solutions proves cuspidality for a class of robots similar to the transpressor.
☆ ROSA: Roundabout Optimized Speed Advisory with Multi-Agent Trajectory Prediction in Multimodal Traffic SC
We present ROSA -- Roundabout Optimized Speed Advisory -- a system that combines multi-agent trajectory prediction with coordinated speed guidance for multimodal, mixed traffic at roundabouts. Using a Transformer-based model, ROSA jointly predicts the future trajectories of vehicles and Vulnerable Road Users (VRUs) at roundabouts. Trained for single-step prediction and deployed autoregressively, it generates deterministic outputs, enabling actionable speed advisories. Incorporating motion dynamics, the model achieves high accuracy (ADE: 1.29m, FDE: 2.99m at a five-second prediction horizon), surpassing prior work. Adding route intention further improves performance (ADE: 1.10m, FDE: 2.36m), demonstrating the value of connected vehicle data. Based on predicted conflicts with VRUs and circulating vehicles, ROSA provides real-time, proactive speed advisories for approaching and entering the roundabout. Despite prediction uncertainty, ROSA significantly improves vehicle efficiency and safety, with positive effects even on perceived safety from a VRU perspective. The source code of this work is available under: github.com/urbanAIthi/ROSA.
comment: 8 pages, 1 figure, 4 tables, 2026 IEEE International Conference on Intelligent Transportation Systems (ITSC)
☆ ManeuverNet: A Soft Actor-Critic Framework for Precise Maneuvering of Double-Ackermann-Steering Robots with Optimized Reward Functions ICRA
Autonomous control of double-Ackermann-steering robots is essential in agricultural applications, where robots must execute precise and complex maneuvers within a limited space. Classical methods, such as the Timed Elastic Band (TEB) planner, can address this problem, but they rely on parameter tuning, making them highly sensitive to changes in robot configuration or environment and impractical to deploy without constant recalibration. At the same time, end-to-end deep reinforcement learning (DRL) methods often fail due to unsuitable reward functions for non-holonomic constraints, resulting in sub-optimal policies and poor generalization. To address these challenges, this paper presents ManeuverNet, a DRL framework tailored for double-Ackermann systems, combining Soft Actor-Critic with CrossQ. Furthermore, ManeuverNet introduces four specifically designed reward functions to support maneuver learning. Unlike prior work, ManeuverNet does not depend on expert data or handcrafted guidance. We extensively evaluate ManeuverNet against both state-of-the-art DRL baselines and the TEB planner. Experimental results demonstrate that our framework substantially improves maneuverability and success rates, achieving more than a 40% gain over DRL baselines. Moreover, ManeuverNet effectively mitigates the strong parameter sensitivity observed in the TEB planner. In real-world trials, ManeuverNet achieved up to a 90% increase in maneuvering trajectory efficiency, highlighting its robustness and practical applicability.
comment: 8 pages, 5, figures, Accepted for 2026 IEEE International Conference on Robotics & Automation (ICRA)
☆ Real-time Monocular 2D and 3D Perception of Endoluminal Scenes for Controlling Flexible Robotic Endoscopic Instruments
Endoluminal surgery offers a minimally invasive option for early-stage gastrointestinal and urinary tract cancers but is limited by surgical tools and a steep learning curve. Robotic systems, particularly continuum robots, provide flexible instruments that enable precise tissue resection, potentially improving outcomes. This paper presents a visual perception platform for a continuum robotic system in endoluminal surgery. Our goal is to utilize monocular endoscopic image-based perception algorithms to identify position and orientation of flexible instruments and measure their distances from tissues. We introduce 2D and 3D learning-based perception algorithms and develop a physically-realistic simulator that models flexible instruments dynamics. This simulator generates realistic endoluminal scenes, enabling control of flexible robots and substantial data collection. Using a continuum robot prototype, we conducted module and system-level evaluations. Results show that our algorithms improve control of flexible instruments, reducing manipulation time by over 70% for trajectory-following tasks and enhancing understanding of surgical scenarios, leading to robust endoluminal surgeries.
☆ Advances in Global Solvers for 3D Vision
Global solvers have emerged as a powerful paradigm for 3D vision, offering certifiable solutions to nonconvex geometric optimization problems traditionally addressed by local or heuristic methods. This survey presents the first systematic review of global solvers in geometric vision, unifying the field through a comprehensive taxonomy of three core paradigms: Branch-and-Bound (BnB), Convex Relaxation (CR), and Graduated Non-Convexity (GNC). We present their theoretical foundations, algorithmic designs, and practical enhancements for robustness and scalability, examining how each addresses the fundamental nonconvexity of geometric estimation problems. Our analysis spans ten core vision tasks, from Wahba problem to bundle adjustment, revealing the optimality-robustness-scalability trade-offs that govern solver selection. We identify critical future directions: scaling algorithms while maintaining guarantees, integrating data-driven priors with certifiable optimization, establishing standardized benchmarks, and addressing societal implications for safety-critical deployment. By consolidating theoretical foundations, practical advances, and broader impacts, this survey provides a unified perspective and roadmap toward certifiable, trustworthy perception for real-world applications. A continuously-updated literature summary and companion code tutorials are available at https://github.com/ericzzj1989/Awesome-Global-Solvers-for-3D-Vision.
comment: Comprehensive survey; 37 pages, 7 figures, 3 tables. Project page with literature tracking and code tutorials: https://github.com/ericzzj1989/Awesome-Global-Solvers-for-3D-Vision
☆ Simulation-based Learning of Electrical Cabinet Assembly Using Robot Skills
This paper presents a simulation-driven approach for automating the force-controlled assembly of electrical terminals on DIN-rails, a task traditionally hindered by high programming effort and product variability. The proposed method integrates deep reinforcement learning (DRL) with parameterizable robot skills in a physics-based simulation environment. To realistically model the snap-fit assembly process, we develop and evaluate two types of joining models: analytical models based on beam theory and rigid-body models implemented in the MuJoCo physics engine. These models enable accurate simulation of interaction forces, essential for training DRL agents. The robot skills are structured using the pitasc framework, allowing modular, reusable control strategies. Training is conducted in simulation using Soft Actor-Critic (SAC) and Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithms. Domain randomization is applied to improve robustness. The trained policies are transferred to a physical UR10e robot system without additional tuning. Experimental results demonstrate high success rates (up to 100%) in both simulation and real-world settings, even under significant positional and rotational deviations. The system generalizes well to new terminal types and positions, significantly reducing manual programming effort. This work highlights the potential of combining simulation-based learning with modular robot skills for flexible, scalable automation in small-batch manufacturing. Future work will explore hybrid learning methods, automated environment parameterization, and further refinement of joining models for design integration.
comment: 20 pages, 14 Figures
☆ Replanning Human-Robot Collaborative Tasks with Vision-Language Models via Semantic and Physical Dual-Correction
Human-Robot Collaboration (HRC) plays an important role in assembly tasks by enabling robots to plan and adjust their motions based on interactive, real-time human instructions. However, such instructions are often linguistically ambiguous and underspecified, making it difficult to generate physically feasible and cooperative robot behaviors. To address this challenge, many studies have applied Vision-Language Models (VLMs) to interpret high-level instructions and generate corresponding actions. Nevertheless, VLM-based approaches still suffer from hallucinated reasoning and an inability to anticipate physical execution failures. To address these challenges, we propose an HRC framework that augments a VLM-based reasoning with a dual-correction mechanism: an internal correction model that verifies logical consistency and task feasibility prior to action execution, and an external correction model that detects and rectifies physical failures through post-execution feedback. Simulation ablation studies demonstrate that the proposed method improves the success rate compared to baselines without correction models. Our real-world experiments in collaborative assembly tasks supported by object fixation or tool preparation by an upper body humanoid robot further confirm the framewor's effectiveness in enabling interactive replanning across different collaborative tasks in response to human instructions, validating its practical feasibility.
comment: 16 pages, 8 figures
☆ Multimodal Covariance Steering in Belief Space with Active Probing and Influence for Autonomous Driving ICRA 2026
Autonomous driving in complex traffic requires reasoning under uncertainty. Common approaches rely on prediction-based planning or risk-aware control, but these are typically treated in isolation, limiting their ability to capture the coupled nature of action and inference in interactive settings. This gap becomes especially critical in uncertain scenarios, where simply reacting to predictions can lead to unsafe maneuvers or overly conservative behavior. Our central insight is that safe interaction requires not only estimating human behavior but also shaping it when ambiguity poses risks. To this end, we introduce a hierarchical belief model that structures human behavior across coarse discrete intents and fine motion modes, updated via Bayesian inference for interpretable multi-resolution reasoning. On top of this, we develop an active probing strategy that identifies when multimodal ambiguity in human predictions may compromise safety and plans disambiguating actions that both reveal intent and gently steer human decisions toward safer outcomes. Finally, a runtime risk-evaluation layer based on Conditional Value-at-Risk (CVaR) ensures that all probing actions remain within human risk tolerance during influence. Our simulations in lane-merging and unsignaled intersection scenarios demonstrate that our approach achieves higher success rates and shorter completion times compared to existing methods. These results highlight the benefit of coupling belief inference, probing, and risk monitoring, yielding a principled and interpretable framework for planning under uncertainty.
comment: Accepted to IEEE International Conference on Robotics and Automation (ICRA 2026)
☆ TWISTED-RL: Hierarchical Skilled Agents for Knot-Tying without Human Demonstrations
Robotic knot-tying represents a fundamental challenge in robotics due to the complex interactions between deformable objects and strict topological constraints. We present TWISTED-RL, a framework that improves upon the previous state-of-the-art in demonstration-free knot-tying (TWISTED), which smartly decomposed a single knot-tying problem into manageable subproblems, each addressed by a specialized agent. Our approach replaces TWISTED's single-step inverse model that was learned via supervised learning with a multi-step Reinforcement Learning policy conditioned on abstract topological actions rather than goal states. This change allows more delicate topological state transitions while avoiding costly and ineffective data collection protocols, thus enabling better generalization across diverse knot configurations. Experimental results demonstrate that TWISTED-RL manages to solve previously unattainable knots of higher complexity, including commonly used knots such as the Figure-8 and the Overhand. Furthermore, the increase in success rates and drop in planning time establishes TWISTED-RL as the new state-of-the-art in robotic knot-tying without human demonstrations.
☆ Learning Transferability: A Two-Stage Reinforcement Learning Approach for Enhancing Quadruped Robots' Performance in U-Shaped Stair Climbing
Quadruped robots are employed in various scenarios in building construction. However, autonomous stair climbing across different indoor staircases remains a major challenge for robot dogs to complete building construction tasks. In this project, we employed a two-stage end-to-end deep reinforcement learning (RL) approach to optimize a robot's performance on U-shaped stairs. The training robot-dog modality, Unitree Go2, was first trained to climb stairs on Isaac Lab's pyramid-stair terrain, and then to climb a U-shaped indoor staircase using the learned policies. This project explores end-to-end RL methods that enable robot dogs to autonomously climb stairs. The results showed (1) the successful goal reached for robot dogs climbing U-shaped stairs with a stall penalty, and (2) the transferability from the policy trained on U-shaped stairs to deployment on straight, L-shaped, and spiral stair terrains, and transferability from other stair models to deployment on U-shaped terrain.
comment: 8 pages, 4 figures, International Conference on Computing in Civil Engineering (i3CE 2026)
☆ RoboSolver: A Multi-Agent Large Language Model Framework for Solving Robotic Arm Problems
This study proposes an intelligent multi-agent framework built on LLMs and VLMs and specifically tailored to robotics. The goal is to integrate the strengths of LLMs and VLMs with computational tools to automatically analyze and solve problems related to robotic manipulators. Our developed framework accepts both textual and visual inputs and can automatically perform forward and inverse kinematics, compute velocities and accelerations of key points, generate 3D simulations of the robot, and ultimately execute motion control within the simulated environment, all according to the user's query. To evaluate the framework, three benchmark tests were designed, each consisting of ten questions. In the first benchmark test, the framework was evaluated while connected to GPT-4o, DeepSeek-V3.2, and Claude-Sonnet-4.5, as well as their corresponding raw models. The objective was to extract the forward kinematics of robots directly from textual descriptions. The results showed that the framework integrated with GPT-4o achieved the highest accuracy, reaching 0.97 in computing the final solution, whereas the raw model alone attained an accuracy of only 0.30 for the same task. Similarly, for the other two models, the framework consistently outperformed the corresponding raw models in terms of accuracy. The second benchmark test was identical to the first, except that the input was provided in visual form. In this test, the GPT-4o LLM was used alongside the Gemini 2.5 Pro VLM. The results showed that the framework achieved an accuracy of 0.93 in obtaining the final answer, which is approximately 20% higher than that of the corresponding raw model. The third benchmark test encompassed a range of robotic tasks, including simulation, control, velocity and acceleration computation, as well as inverse kinematics and Jacobian calculation, for which the framework achieved an accuracy of 0.97.
☆ A Soft Wrist with Anisotropic and Selectable Stiffness for Robust Robot Learning in Contact-rich Manipulation
Contact-rich manipulation tasks in unstructured environments pose significant robustness challenges for robot learning, where unexpected collisions can cause damage and hinder policy acquisition. Existing soft end-effectors face fundamental limitations: they either provide a limited deformation range, lack directional stiffness control, or require complex actuation systems that compromise practicality. This study introduces CLAW (Compliant Leaf-spring Anisotropic soft Wrist), a novel soft wrist mechanism that addresses these limitations through a simple yet effective design using two orthogonal leaf springs and rotary joints with a locking mechanism. CLAW provides large 6-degree-of-freedom deformation (40mm lateral, 20mm vertical), anisotropic stiffness that is tunable across three distinct modes, while maintaining lightweight construction (330g) at low cost ($550). Experimental evaluations using imitation learning demonstrate that CLAW achieves 76% success rate in benchmark peg-insertion tasks, outperforming both the Fin Ray gripper (43%) and rigid gripper alternatives (36%). CLAW successfully handles diverse contact-rich scenarios, including precision assembly with tight tolerances and delicate object manipulation, demonstrating its potential to enable robust robot learning in contact-rich domains. Project page: https://project-page-manager.github.io/CLAW/
☆ Understanding Sensor Vulnerabilities in Industrial XR Tracking
Extended Reality (XR) systems deployed in industrial and operational settings rely on Visual--Inertial Odometry (VIO) for continuous six-degree-of-freedom pose tracking, yet these environments often involve sensing conditions that deviate from ideal assumptions. Despite this, most VIO evaluations emphasize nominal sensor behavior, leaving the effects of sustained sensor degradation under operational conditions insufficiently understood. This paper presents a controlled empirical study of VIO behavior under degraded sensing, examining faults affecting visual and inertial modalities across a range of operating regimes. Through systematic fault injection and quantitative evaluation, we observe a pronounced asymmetry in fault impact where degradations affecting visual sensing typically lead to bounded pose errors on the order of centimeters, whereas degradations affecting inertial sensing can induce substantially larger trajectory deviations, in some cases reaching hundreds to thousands of meters. These observations motivate greater emphasis on inertial reliability in the evaluation and design of XR systems for real-life industrial settings.
comment: IEEE VR XRIOS 2026 Workshop
☆ AdaptManip: Learning Adaptive Whole-Body Object Lifting and Delivery with Online Recurrent State Estimation
This paper presents Adaptive Whole-body Loco-Manipulation, AdaptManip, a fully autonomous framework for humanoid robots to perform integrated navigation, object lifting, and delivery. Unlike prior imitation learning-based approaches that rely on human demonstrations and are often brittle to disturbances, AdaptManip aims to train a robust loco-manipulation policy via reinforcement learning without human demonstrations or teleoperation data. The proposed framework consists of three coupled components: (1) a recurrent object state estimator that tracks the manipulated object in real time under limited field-of-view and occlusions; (2) a whole-body base policy for robust locomotion with residual manipulation control for stable object lifting and delivery; and (3) a LiDAR-based robot global position estimator that provides drift-robust localization. All components are trained in simulation using reinforcement learning and deployed on real hardware in a zero-shot manner. Experimental results show that AdaptManip significantly outperforms baseline methods, including imitation learning-based approaches, in adaptability and overall success rate, while accurate object state estimation improves manipulation performance even under occlusion. We further demonstrate fully autonomous real-world navigation, object lifting, and delivery on a humanoid robot.
comment: Website: https://morganbyrd03.github.io/adaptmanip/
☆ SEG-JPEG: Simple Visual Semantic Communications for Remote Operation of Automated Vehicles over Unreliable Wireless Networks
Remote Operation is touted as being key to the rapid deployment of automated vehicles. Streaming imagery to control connected vehicles remotely currently requires a reliable, high throughput network connection, which can be limited in real-world remote operation deployments relying on public network infrastructure. This paper investigates how the application of computer vision assisted semantic communication can be used to circumvent data loss and corruption associated with traditional image compression techniques. By encoding the segmentations of detected road users into colour coded highlights within low resolution greyscale imagery, the required data rate can be reduced by 50 \% compared with conventional techniques, while maintaining visual clarity. This enables a median glass-to-glass latency of below 200ms even when the network data rate is below 500kbit/s, while clearly outlining salient road users to enhance situational awareness of the remote operator. The approach is demonstrated in an area of variable 4G mobile connectivity using an automated last-mile delivery vehicle. With this technique, the results indicate that large-scale deployment of remotely operated automated vehicles could be possible even on the often constrained public 4G/5G mobile network, providing the potential to expedite the nationwide roll-out of automated vehicles.
☆ DexEvolve: Evolutionary Optimization for Robust and Diverse Dexterous Grasp Synthesis
Dexterous grasping is fundamental to robotics, yet data-driven grasp prediction heavily relies on large, diverse datasets that are costly to generate and typically limited to a narrow set of gripper morphologies. Analytical grasp synthesis can be used to scale data collection, but necessary simplifying assumptions often yield physically infeasible grasps that need to be filtered in high-fidelity simulators, significantly reducing the total number of grasps and their diversity. We propose a scalable generate-and-refine pipeline for synthesizing large-scale, diverse, and physically feasible grasps. Instead of using high-fidelity simulators solely for verification and filtering, we leverage them as an optimization stage that continuously improves grasp quality without discarding precomputed candidates. More specifically, we initialize an evolutionary search with a seed set of analytically generated, potentially suboptimal grasps. We then refine these proposals directly in a high-fidelity simulator (Isaac Sim) using an asynchronous, gradient-free evolutionary algorithm, improving stability while maintaining diversity. In addition, this refinement stage can be guided toward human preferences and/or domain-specific quality metrics without requiring a differentiable objective. We further distill the refined grasp distribution into a diffusion model for robust real-world deployment, and highlight the role of diversity for both effective training and during deployment. Experiments on a newly introduced Handles dataset and a DexGraspNet subset demonstrate that our approach achieves over 120 distinct stable grasps per object (a 1.7-6x improvement over unrefined analytical methods) while outperforming diffusion-based alternatives by 46-60\% in unique grasp coverage.
☆ Time-Archival Camera Virtualization for Sports and Visual Performances
Camera virtualization -- an emerging solution to novel view synthesis -- holds transformative potential for visual entertainment, live performances, and sports broadcasting by enabling the generation of photorealistic images from novel viewpoints using images from a limited set of calibrated multiple static physical cameras. Despite recent advances, achieving spatially and temporally coherent and photorealistic rendering of dynamic scenes with efficient time-archival capabilities, particularly in fast-paced sports and stage performances, remains challenging for existing approaches. Recent methods based on 3D Gaussian Splatting (3DGS) for dynamic scenes could offer real-time view-synthesis results. Yet, they are hindered by their dependence on accurate 3D point clouds from the structure-from-motion method and their inability to handle large, non-rigid, rapid motions of different subjects (e.g., flips, jumps, articulations, sudden player-to-player transitions). Moreover, independent motions of multiple subjects can break the Gaussian-tracking assumptions commonly used in 4DGS, ST-GS, and other dynamic splatting variants. This paper advocates reconsidering a neural volume rendering formulation for camera virtualization and efficient time-archival capabilities, making it useful for sports broadcasting and related applications. By modeling a dynamic scene as rigid transformations across multiple synchronized camera views at a given time, our method performs neural representation learning, providing enhanced visual rendering quality at test time. A key contribution of our approach is its support for time-archival, i.e., users can revisit any past temporal instance of a dynamic scene and can perform novel view synthesis, enabling retrospective rendering for replay, analysis, and archival of live events, a functionality absent in existing neural rendering approaches and novel view synthesis...
comment: Project Page: https://yunxiaozhangjack.com/tacv/; Under minor revision in Journal of Computer Vision and Image Understanding (CVIU); Special Issue: Computer Vision for Sports and Winter Sports. Outcome of a master and bachelor student project completed in Visual and Spatial AI Lab at TAMU
☆ A ROS2 Benchmarking Framework for Hierarchical Control Strategies in Mobile Robots for Mediterranean Greenhouses
Mobile robots operating in agroindustrial environments, such as Mediterranean greenhouses, are subject to challenging conditions, including uneven terrain, variable friction, payload changes, and terrain slopes, all of which significantly affect control performance and stability. Despite the increasing adoption of robotic platforms in agriculture, the lack of standardized, reproducible benchmarks impedes fair comparisons and systematic evaluations of control strategies under realistic operating conditions. This paper presents a comprehensive benchmarking framework for evaluating mobile robot controllers in greenhouse environments. The proposed framework integrates an accurate three dimensional model of the environment, a physics based simulator, and a hierarchical control architecture comprising low, mid, and high level control layers. Three benchmark categories are defined to enable modular assessment, ranging from actuator level control to full autonomous navigation. Additionally, three disturbance scenarios payload variation, terrain type, and slope are explicitly modeled to replicate real world agricultural conditions. To ensure objective and reproducible evaluation, standardized performance metrics are introduced, including the Squared Absolute Error (SAE), the Squared Control Input (SCI), and composite performance indices. Statistical analysis based on repeated trials is employed to mitigate the influence of sensor noise and environmental variability. The framework is further enhanced by a plugin based architecture that facilitates seamless integration of user defined controllers and planners. The proposed benchmark provides a robust and extensible tool for the quantitative comparison of classical, predictive, and planning based control strategies in realistic conditions, bridging the gap between simulation based analysis and real world agroindustrial applications.
comment: 53 pages
☆ Augmenting Human Balance with Generic Supernumerary Robotic Limbs
Supernumerary robotic limbs (SLs) have the potential to transform a wide range of human activities, yet their usability remains limited by key technical challenges, particularly in ensuring safety and achieving versatile control. Here, we address the critical problem of maintaining balance in the human-SLs system, a prerequisite for safe and comfortable augmentation tasks. Unlike previous approaches that developed SLs specifically for stability support, we propose a general framework for preserving balance with SLs designed for generic use. Our hierarchical three-layer architecture consists of: (i) a prediction layer that estimates human trunk and center of mass (CoM) dynamics, (ii) a planning layer that generates optimal CoM trajectories to counteract trunk movements and computes the corresponding SL control inputs, and (iii) a control layer that executes these inputs on the SL hardware. We evaluated the framework with ten participants performing forward and lateral bending tasks. The results show a clear reduction in stance instability, demonstrating the framework's effectiveness in enhancing balance. This work paves the path towards safe and versatile human-SLs interactions. [This paper has been submitted for publication to IEEE.]
♻ ☆ Event-Grounding Graph: Unified Spatio-Temporal Scene Graph from Robotic Observations
A fundamental aspect for building intelligent autonomous robots that can assist humans in their daily lives is the construction of rich environmental representations. While advances in semantic scene representations have enriched robotic scene understanding, current approaches lack a connection between spatial features and dynamic events; e.g., connecting the blue mug to the event washing a mug. In this work, we introduce the event-grounding graph (EGG), a framework grounding event interactions to spatial features of a scene. This representation allows robots to perceive, reason, and respond to complex spatio-temporal queries. Experiments using real robotic data demonstrate EGG's capability to retrieve relevant information and respond accurately to human inquiries concerning the environment and events within. Furthermore, the EGG framework's source code and evaluation dataset are released as open-source at: https://github.com/aalto-intelligent-robotics/EGG.
comment: Accepted to RA-L
♻ ☆ Foundation Models in Autonomous Driving: A Survey on Scenario Generation and Scenario Analysis
For autonomous vehicles, safe navigation in complex environments depends on handling a broad range of diverse and rare driving scenarios. Simulation- and scenario-based testing have emerged as key approaches to development and validation of autonomous driving systems. Traditional scenario generation relies on rule-based systems, knowledge-driven models, and data-driven synthesis, often producing limited diversity and unrealistic safety-critical cases. With the emergence of foundation models, which represent a new generation of pre-trained, general-purpose AI models, developers can process heterogeneous inputs (e.g., natural language, sensor data, HD maps, and control actions), enabling the synthesis and interpretation of complex driving scenarios. In this paper, we conduct a survey about the application of foundation models for scenario generation and scenario analysis in autonomous driving (as of May 2025). Our survey presents a unified taxonomy that includes large language models, vision-language models, multimodal large language models, diffusion models, and world models for the generation and analysis of autonomous driving scenarios. In addition, we review the methodologies, open-source datasets, simulation platforms, and benchmark challenges, and we examine the evaluation metrics tailored explicitly to scenario generation and analysis. Finally, the survey concludes by highlighting the open challenges and research questions, and outlining promising future research directions. All reviewed papers are listed in a continuously maintained repository, which contains supplementary materials and is available at https://github.com/TUM-AVS/FM-for-Scenario-Generation-Analysis.
comment: IEEE Open Journal of Intelligent Transportation Systems
♻ ☆ Language Movement Primitives: Grounding Language Models in Robot Motion
Enabling robots to perform novel manipulation tasks from natural language instructions remains a fundamental challenge in robotics, despite significant progress in generalized problem solving with foundational models. Large vision and language models (VLMs) are capable of processing high-dimensional input data for visual scene and language understanding, as well as decomposing tasks into a sequence of logical steps; however, they struggle to ground those steps in embodied robot motion. On the other hand, robotics foundation models output action commands, but require in-domain fine-tuning or experience before they are able to perform novel tasks successfully. At its core, there still remains the fundamental challenge of connecting abstract task reasoning with low-level motion control. To address this disconnect, we propose Language Movement Primitives (LMPs), a framework that grounds VLM reasoning in Dynamic Movement Primitive (DMP) parameterization. Our key insight is that DMPs provide a small number of interpretable parameters, and VLMs can set these parameters to specify diverse, continuous, and stable trajectories. Put another way: VLMs can reason over free-form natural language task descriptions, and semantically ground their desired motions into DMPs -- bridging the gap between high-level task reasoning and low-level position and velocity control. Building on this combination of VLMs and DMPs, we formulate our LMP pipeline for zero-shot robot manipulation that effectively completes tabletop manipulation problems by generating a sequence of DMP motions. Across 20 real-world manipulation tasks, we show that LMP achieves 80% task success as compared to 31% for the best-performing baseline. See videos at our website: https://collab.me.vt.edu/lmp
♻ ☆ 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/
♻ ☆ 3DRot: Rediscovering the Missing Primitive for RGB-Based 3D Augmentation
RGB-based 3D tasks, e.g., 3D detection, depth estimation, 3D keypoint estimation, still suffer from scarce, expensive annotations and a thin augmentation toolbox, since many image transforms, including rotations and warps, disrupt geometric consistency. While horizontal flipping and color jitter are standard, rigorous 3D rotation augmentation has surprisingly remained absent from RGB-based pipelines, largely due to the misconception that it requires scene depth or scene reconstruction. In this paper, we introduce 3DRot, a plug-and-play augmentation that rotates and mirrors images about the camera's optical center while synchronously updating RGB images, camera intrinsics, object poses, and 3D annotations to preserve projective geometry, achieving geometry-consistent rotations and reflections without relying on any scene depth. We first validate 3DRot on a classical RGB-based 3D task, monocular 3D detection. On SUN RGB-D, inserting 3DRot into a frozen DINO-X + Cube R-CNN pipeline raises $IoU_{3D}$ from 43.21 to 44.51, cuts rotation error (ROT) from 22.91$^\circ$ to 20.93$^\circ$, and boosts $mAP_{0.5}$ from 35.70 to 38.11; smaller but consistent gains appear on a cross-domain IN10 split. Beyond monocular detection, adding 3DRot on top of the standard BTS augmentation schedule further improves NYU Depth v2 from 0.1783 to 0.1685 in abs-rel (and 0.7472 to 0.7548 in $δ<1.25$), and reduces cross-dataset error on SUN RGB-D. On KITTI, applying the same camera-centric rotations in MVX-Net (LiDAR+RGB) raises moderate 3D AP from about 63.85 to 65.16 while remaining compatible with standard 3D augmentations.
♻ ☆ Deep Reinforcement Learning based Autonomous Decision-Making for Cooperative UAVs: A Search and Rescue Real World Application
This paper presents the first end-to-end framework that combines guidance, navigation, and centralised task allocation for multiple UAVs performing autonomous search-and-rescue (SAR) in GNSS-denied indoor environments. A Twin Delayed Deep Deterministic Policy Gradient controller is trained with an Artificial Potential Field (APF) reward that blends attractive and repulsive potentials with continuous control, accelerating convergence and yielding smoother, safer trajectories than distance-only baselines. Collaborative mission assignment is solved by a deep Graph Attention Network that, at each decision step, reasons over the drone-task graph to produce near-optimal allocations with negligible on-board compute. To arrest the notorious Z-drift of indoor LiDAR-SLAM, we fuse depth-camera altimetry with IMU vertical velocity in a lightweight complementary filter, giving centimetre-level altitude stability without external beacons. The resulting system was deployed on two 1m-class quad-rotors and flight-tested in a cluttered, multi-level disaster mock-up designed for the NATO-Sapience Autonomous Cooperative Drone Competition. Compared with prior DRL guidance that remains largely in simulation, our framework demonstrates an ability to navigate complex indoor environments, securing first place in the 2024 event. These results demonstrate that APF-shaped DRL and GAT-driven cooperation can translate to reliable real-world SAR operations.
comment: 22 Pages, 24 Figures
♻ ☆ BoundPlanner: A convex-set-based approach to bounded manipulator trajectory planning
Online trajectory planning enables robot manipulators to react quickly to changing environments or tasks. Many robot trajectory planners exist for known environments but are often too slow for online computations. Current methods in online trajectory planning do not find suitable trajectories in challenging scenarios that respect the limits of the robot and account for collisions. This work proposes a trajectory planning framework consisting of the novel Cartesian path planner based on convex sets, called BoundPlanner, and the online trajectory planner BoundMPC. BoundPlanner explores and maps the collision-free space using convex sets to compute a reference path with bounds. BoundMPC is extended in this work to handle convex sets for path deviations, which allows the robot to optimally follow the path within the bounds while accounting for the robot's kinematics. Collisions of the robot's kinematic chain are considered by a novel convex-set-based collision avoidance formulation independent on the number of obstacles. Simulations and experiments with a 7-DoF manipulator show the performance of the proposed planner compared to state-of-the-art methods. The source code is available at github.com/TU-Wien-ACIN-CDS/BoundPlanner and videos of the experiments can be found at www.acin.tuwien.ac.at/42d4.
comment: Published at RA-L
♻ ☆ BoundMPC: Cartesian path following with error bounds based on model predictive control in the joint space
This work introduces the BoundMPC strategy, an innovative online model-predictive path-following approach for robot manipulators. This joint-space trajectory planner allows the following of Cartesian reference paths in the end-effector's position and orientation, including via-points, within the desired asymmetric bounds of the orthogonal path error. These bounds encode the obstacle-free space and additional task-specific constraints in Cartesian space. Contrary to traditional path-following concepts, BoundMPC purposefully deviates from the Cartesian reference path in position and orientation to account for the robot's kinematics, leading to more successful task executions for Cartesian reference paths. Furthermore the simple reference path formulation is computationally efficient and allows for replanning during the robot's motion. This feature makes it possible to use this planner for dynamically changing environments and varying goals. The flexibility and performance of BoundMPC are experimentally demonstrated by five scenarios on a 7-DoF Kuka LBR iiwa 14 R820 robot. The first scenario shows the transfer of a larger object from a start to a goal pose through a confined space where the object must be tilted. The second scenario deals with grasping an object from a table where the grasping point changes during the robot's motion, and collisions with other obstacles in the scene must be avoided. The adaptability of BoundMPC is showcased in scenarios such as the opening of a drawer, the transfer of an open container, and the wiping of a table, where it effectively handles task-specific constraints. The last scenario highlights the possibility of accounting for collisions with the entire robot's kinematic chain. The code is readily available at https://github.com/thieso/boundmpc, inspiring you to explore its potential and adapt it to your specific robotic tasks.
comment: 17 pages, 20 figures
♻ ☆ Experimental Evaluation of ROS-Causal in Real-World Human-Robot Spatial Interaction Scenarios
Deploying robots in human-shared environments requires a deep understanding of how nearby agents and objects interact. Employing causal inference to model cause-and-effect relationships facilitates the prediction of human behaviours and enables the anticipation of robot interventions. However, a significant challenge arises due to the absence of implementation of existing causal discovery methods within the ROS ecosystem, the standard de-facto framework in robotics, hindering effective utilisation on real robots. To bridge this gap, in our previous work we proposed ROS-Causal, a ROS-based framework designed for onboard data collection and causal discovery in human-robot spatial interactions. In this work, we present an experimental evaluation of ROS-Causal both in simulation and on a new dataset of human-robot spatial interactions in a lab scenario, to assess its performance and effectiveness. Our analysis demonstrates the efficacy of this approach, showcasing how causal models can be extracted directly onboard by robots during data collection. The online causal models generated from the simulation are consistent with those from lab experiments. These findings can help researchers to enhance the performance of robotic systems in shared environments, firstly by studying the causal relations between variables in simulation without real people, and then facilitating the actual robot deployment in real human environments. ROS-Causal: https://lcastri.github.io/roscausal
comment: Published at 2024 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)
♻ ☆ AquaROM: shape optimization pipeline for soft swimmers using parametric reduced order models
The efficient optimization of actuated soft structures, particularly under complex nonlinear forces, remains a critical challenge in advancing robotics. Simulations of nonlinear structures, such as soft-bodied robots modeled using the finite element method (FEM), often demand substantial computational resources, especially during optimization. To address this challenge, we propose a novel optimization algorithm based on a tensorial parametric reduced order model (PROM). Our algorithm leverages dimensionality reduction and solution approximation techniques to facilitate efficient solving of nonlinear constrained optimization problems. The well-structured tensorial approach enables the use of analytical gradients within a specifically chosen reduced order basis (ROB), significantly enhancing computational efficiency. To showcase the performance of our method, we apply it to optimizing soft robotic swimmer shapes. These actuated soft robots experience hydrodynamic forces, subjecting them to both internal and external nonlinear forces, which are incorporated into our optimization process using a data-free ROB for fast and accurate computations. This approach not only reduces computational complexity but also unlocks new opportunities to optimize complex nonlinear systems in soft robotics, paving the way for more efficient design and control.
♻ ☆ 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.
♻ ☆ EigenSafe: A Spectral Framework for Learning-Based Probabilistic Safety Assessment
We present EigenSafe, an operator-theoretic framework for safety assessment of learning-enabled stochastic systems. In many robotic applications, the dynamics are inherently stochastic due to factors such as sensing noise and environmental disturbances, and it is challenging for conventional methods such as Hamilton-Jacobi reachability and control barrier functions to provide a well-calibrated safety critic that is tied to the actual safety probability. We derive a linear operator that governs the dynamic programming principle for safety probability, and find that its dominant eigenpair provides critical safety information for both individual state-action pairs and the overall closed-loop system. The proposed framework learns this dominant eigenpair, which can be used to either inform or constrain policy updates. We demonstrate that the learned eigenpair effectively facilitates safe reinforcement learning. Further, we validate its applicability in enhancing the safety of learned policies from imitation learning through robot manipulation experiments using a UR3 robotic arm in a food preparation task.
comment: Inkyu Jang and Jonghae Park contributed equally to this work. Project Webpage: https://eigen-safe.github.io/
♻ ☆ V2V-LLM: Vehicle-to-Vehicle Cooperative Autonomous Driving with Multimodal Large Language Models ICRA 2026
Current autonomous driving vehicles rely mainly on their individual sensors to understand surrounding scenes and plan for future trajectories, which can be unreliable when the sensors are malfunctioning or occluded. To address this problem, cooperative perception methods via vehicle-to-vehicle (V2V) communication have been proposed, but they have tended to focus on perception tasks like detection or tracking. How those approaches contribute to overall cooperative planning performance is still under-explored. Inspired by recent progress using Large Language Models (LLMs) to build autonomous driving systems, we propose a novel problem setting that integrates a Multimodal LLM into cooperative autonomous driving, with the proposed Vehicle-to-Vehicle Question-Answering (V2V-QA) dataset and benchmark. We also propose our baseline method Vehicle-to-Vehicle Multimodal Large Language Model (V2V-LLM), which uses an LLM to fuse perception information from multiple connected autonomous vehicles (CAVs) and answer various types of driving-related questions: grounding, notable object identification, and planning. Experimental results show that our proposed V2V-LLM can be a promising unified model architecture for performing various tasks in cooperative autonomous driving, and outperforms other baseline methods that use different fusion approaches. Our work also creates a new research direction that can improve the safety of future autonomous driving systems. The code and data will be released to the public to facilitate open-source research in this field. Our project website: https://eddyhkchiu.github.io/v2vllm.github.io/ .
comment: Accepted by ICRA 2026 (IEEE International Conference on Robotics and Automation). Project: https://eddyhkchiu.github.io/v2vllm.github.io/ Code: https://github.com/eddyhkchiu/V2V-LLM Dataset: https://huggingface.co/datasets/eddyhkchiu/V2V-GoT-QA
♻ ☆ V2V-GoT: Vehicle-to-Vehicle Cooperative Autonomous Driving with Multimodal Large Language Models and Graph-of-Thoughts ICRA 2026
Current state-of-the-art autonomous vehicles could face safety-critical situations when their local sensors are occluded by large nearby objects on the road. Vehicle-to-vehicle (V2V) cooperative autonomous driving has been proposed as a means of addressing this problem, and one recently introduced framework for cooperative autonomous driving has further adopted an approach that incorporates a Multimodal Large Language Model (MLLM) to integrate cooperative perception and planning processes. However, despite the potential benefit of applying graph-of-thoughts reasoning to the MLLM, this idea has not been considered by previous cooperative autonomous driving research. In this paper, we propose a novel graph-of-thoughts framework specifically designed for MLLM-based cooperative autonomous driving. Our graph-of-thoughts includes our proposed novel ideas of occlusion-aware perception and planning-aware prediction. We curate the V2V-GoT-QA dataset and develop the V2V-GoT model for training and testing the cooperative driving graph-of-thoughts. Our experimental results show that our method outperforms other baselines in cooperative perception, prediction, and planning tasks. Our project website: https://eddyhkchiu.github.io/v2vgot.github.io/ .
comment: Accepted by ICRA 2026 (IEEE International Conference on Robotics and Automation). Project: https://eddyhkchiu.github.io/v2vgot.github.io/ Code: https://github.com/eddyhkchiu/V2V-GoT Dataset: https://huggingface.co/datasets/eddyhkchiu/V2V-GoT-QA
♻ ☆ LapSurgie: Humanoid Robots Performing Surgery via Teleoperated Handheld Laparoscopy
Robotic laparoscopic surgery has gained increasing attention in recent years for its potential to deliver more efficient and precise minimally invasive procedures. However, adoption of surgical robotic platforms remains largely confined to high-resource medical centers, exacerbating healthcare disparities in rural and low-resource regions. To close this gap, a range of solutions has been explored, from remote mentorship to fully remote telesurgery. Yet, the practical deployment of surgical robotic systems to underserved communities remains an unsolved challenge. Humanoid systems offer a promising path toward deployability, as they can directly operate in environments designed for humans without extensive infrastructure modifications -- including operating rooms. In this work, we introduce LapSurgie, the first humanoid-robot-based laparoscopic teleoperation framework. The system leverages an inverse-mapping strategy for manual-wristed laparoscopic instruments that abides to remote center-of-motion constraints, enabling precise hand-to-tool control of off-the-shelf surgical laparoscopic tools without additional setup requirements. A control console equipped with a stereo vision system provides real-time visual feedback. Finally, a comprehensive user study across platforms demonstrates the effectiveness of the proposed framework and provides initial evidence for the feasibility of deploying humanoid robots in laparoscopic procedures.
♻ ☆ Learning-Based Planning for Improving Science Return of Earth Observation Satellites
Earth observing satellites are powerful tools for collecting scientific information about our planet, however they have limitations: they cannot easily deviate from their orbital trajectories, their sensors have a limited field of view, and pointing and operating these sensors can take a large amount of the spacecraft's resources. It is important for these satellites to optimize the data they collect and include only the most important or informative measurements. Dynamic targeting is an emerging concept in which satellite resources and data from a lookahead instrument are used to intelligently reconfigure and point a primary instrument. Simulation studies have shown that dynamic targeting increases the amount of scientific information gathered versus conventional sampling strategies. In this work, we present two different learning-based approaches to dynamic targeting, using reinforcement and imitation learning, respectively. These learning methods build on a dynamic programming solution to plan a sequence of sampling locations. We evaluate our approaches against existing heuristic methods for dynamic targeting, showing the benefits of using learning for this application. Imitation learning performs on average 10.0\% better than the best heuristic method, while reinforcement learning performs on average 13.7\% better. We also show that both learning methods can be trained effectively with small amounts of data.
comment: International Symposium on Artificial Intelligence, Robotics and Automation in Space, November 2024
♻ ☆ Active Matter as a framework for living systems-inspired Robophysics
Robophysics investigates the physical principles that govern living-like robots operating in complex, realworld environments. Despite remarkable technological advances, robots continue to face fundamental efficiency limitations. At the level of individual units, locomotion remains a challenge, while at the collective level, robot swarms struggle to achieve shared purpose, coordination, communication, and cost efficiency. This perspective article examines the key challenges faced by bio-inspired robotic collectives and highlights recent research efforts that incorporate principles from active-matter physics and biology into the modeling and design of robot swarms.
Robotics 29
☆ Exploiting Structure-from-Motion for Robust Vision-Based Map Matching for Aircraft Surface Movement
In this paper we introduce a vision-aided navigation (VAN) pipeline designed to support ground navigation of autonomous aircraft. The proposed algorithm combines the computational efficiency of indirect methods with the robustness of direct image-based techniques to enhance solution integrity. The pipeline starts by processing ground images (e.g., acquired by a taxiing aircraft) and relates them via a feature-based structure-from-motion (SfM) solution. A ground plane mosaic is then constructed via homography transforms and matched to satellite imagery using a sum of squares differences (SSD) of intensities. Experimental results reveal that drift within the SfM solution, similar to that observed in dead-reckoning systems, challenges the expected accuracy benefits of map-matching with a wide-baseline ground-plane mosaic. However, the proposed algorithm demonstrates key integrity features, such as the ability to identify registration anomalies and ambiguous matches. These characteristics of the pipeline can mitigate outlier behaviors and contribute toward a robust, certifiable solution for autonomous surface movement of aircraft.
comment: Accepted to the Proceedings of the 38th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2025). 15 pages, 13 figures
☆ Autonomous Robotic Tissue Palpation and Abnormalities Characterisation via Ergodic Exploration
We propose a novel autonomous robotic palpation framework for real-time elastic mapping during tissue exploration using a viscoelastic tissue model. The method combines force-based parameter estimation using a commercial force/torque sensor with an ergodic control strategy driven by a tailored Expected Information Density, which explicitly biases exploration toward diagnostically relevant regions by jointly considering model uncertainty, stiffness magnitude, and spatial gradients. An Extended Kalman Filter is employed to estimate viscoelastic model parameters online, while Gaussian Process Regression provides spatial modelling of the estimated elasticity, and a Heat Equation Driven Area Coverage controller enables adaptive, continuous trajectory planning. Simulations on synthetic stiffness maps demonstrate that the proposed approach achieves better reconstruction accuracy, enhanced segmentation capability, and improved robustness in detecting stiff inclusions compared to Bayesian Optimisation-based techniques. Experimental validation on a silicone phantom with embedded inclusions emulating pathological tissue regions further corroborates the potential of the method for autonomous tissue characterisation in diagnostic and screening applications.
comment: Submitted to IEEE Robotics and Automation Letters (RA-L)
☆ A Latency-Aware Framework for Visuomotor Policy Learning on Industrial Robots
Industrial robots are increasingly deployed in contact-rich construction and manufacturing tasks that involve uncertainty and long-horizon execution. While learning-based visuomotor policies offer a promising alternative to open-loop control, their deployment on industrial platforms is challenged by a large observation-execution gap caused by sensing, inference, and control latency. This gap is significantly greater than on low-latency research robots due to high-level interfaces and slower closed-loop dynamics, making execution timing a critical system-level issue. This paper presents a latency-aware framework for deploying and evaluating visuomotor policies on industrial robotic arms under realistic timing constraints. The framework integrates calibrated multimodal sensing, temporally consistent synchronization, a unified communication pipeline, and a teleoperation interface for demonstration collection. Within this framework, we introduce a latency-aware execution strategy that schedules finite-horizon, policy-predicted action sequences based on temporal feasibility, enabling asynchronous inference and execution without modifying policy architectures or training. We evaluate the framework on a contact-rich industrial assembly task while systematically varying inference latency. Using identical policies and sensing pipelines, we compare latency-aware execution with blocking and naive asynchronous baselines. Results show that latency-aware execution maintains smooth motion, compliant contact behavior, and consistent task progression across a wide range of latencies while reducing idle time and avoiding instability observed in baseline methods. These findings highlight the importance of explicitly handling latency for reliable closed-loop deployment of visuomotor policies on industrial robots.
☆ GRAIL: Goal Recognition Alignment through Imitation Learning AAMAS 2026
Understanding an agent's goals from its behavior is fundamental to aligning AI systems with human intentions. Existing goal recognition methods typically rely on an optimal goal-oriented policy representation, which may differ from the actor's true behavior and hinder the accurate recognition of their goal. To address this gap, this paper introduces Goal Recognition Alignment through Imitation Learning (GRAIL), which leverages imitation learning and inverse reinforcement learning to learn one goal-directed policy for each candidate goal directly from (potentially suboptimal) demonstration trajectories. By scoring an observed partial trajectory with each learned goal-directed policy in a single forward pass, GRAIL retains the one-shot inference capability of classical goal recognition while leveraging learned policies that can capture suboptimal and systematically biased behavior. Across the evaluated domains, GRAIL increases the F1-score by more than 0.5 under systematically biased optimal behavior, achieves gains of approximately 0.1-0.3 under suboptimal behavior, and yields improvements of up to 0.4 under noisy optimal trajectories, while remaining competitive in fully optimal settings. This work contributes toward scalable and robust models for interpreting agent goals in uncertain environments.
comment: Accepted for publication at AAMAS 2026
☆ Path Planning Optimisation for SParse, AwaRe and Cooperative Networked Aerial Robot Teams (SpArC-NARTs): Optimisation Tool and Ground Sensing Coverage Use Cases
A networked aerial robot team (NART) comprises a group of agents (e.g., unmanned aerial vehicles (UAVs), ground control stations, etc.) interconnected by wireless links. Inter-agent connectivity, even if intermittent (i.e. sparse), enables data exchanges between agents and supports cooperative behaviours in several NART missions. It can benefit online decentralised decision-making and group resilience, particularly when prior knowledge is inaccurate or incomplete. These requirements can be accounted for in the offline mission planning stages to incentivise cooperative behaviours and improve mission efficiency during the NART deployment. This paper proposes a novel path planning tool for a Sparse, Aware, and Cooperative Networked Aerial Robot Team (SpArC-NART) in exploration missions. It simultaneously considers different levels of prior information regarding the environment, limited agent energy, sensing, and communication, as well as distinct NART constitutions. The communication model takes into account the limitations of user-defined radio technology and physical phenomena. The proposed tool aims to maximise the mission goals (e.g., finding one or multiple targets, covering the full area of the environment, etc.), while cooperating with other agents to reduce agent reporting times, increase their global situational awareness (e.g., their knowledge of the environment), and facilitate mission replanning, if required. The developed cooperation mechanism leverages soft-motion constraints and dynamic rewards based on the Value of Movement and the expected communication availability between the agents at each time step. A ground sensing coverage use case was chosen to illustrate the current capabilities of this tool.
comment: 20 pages, submitted to a Journal
☆ Muscle Coactivation in the Sky: Geometry and Pareto Optimality of Energy vs. Promptness in Multirotors
In robotics and human biomechanics, the tension between energy economy and kinematic readiness is well recognized; this work brings that fundamental principle to aerial multirotors. We show that the limited torque of the motors and the nonlinear aerodynamic map from rotor speed to thrust naturally give rise to the novel concept of promptness-a metric akin to dynamic aerodynamic manipulability. By treating energy consumption as a competing objective and introducing a geometric fiber-bundle formulation, we turn redundancy resolution into a principled multi-objective program on affine fibers. The use of the diffeomorphic transformation linearizing the signed-quadratic propulsion model allows us to lay the foundations for a rigorous study of the interplay between these costs. Through an illustrative case study on 4-DoF allocation on the hexarotor, we reveal that this interplay is fiber-dependent and physically shaped by hardware inequalities. For unidirectional thrusters, the feasible fibers are compact, yielding interior allocations and a short Pareto arc, while torque demands break symmetry and separate the optima. Conversely, with reversible propellers, the null space enables antagonistic rotor co-contraction that drives promptness to hardware limits, making optimal endurance and agility fundamentally incompatible in those regimes. Ultimately, rather than relying on heuristic tuning or black box algorithms to empirically improve task execution, this framework provides a foundational understanding of why and how to achieve agility through geometry-aware control allocation, offering possible guidance for vehicle design, certification metrics, and threat-aware flight operation.
☆ Learning Part-Aware Dense 3D Feature Field for Generalizable Articulated Object Manipulation ICLR 2026
Articulated object manipulation is essential for various real-world robotic tasks, yet generalizing across diverse objects remains a major challenge. A key to generalization lies in understanding functional parts (e.g., door handles and knobs), which indicate where and how to manipulate across diverse object categories and shapes. Previous works attempted to achieve generalization by introducing foundation features, while these features are mostly 2D-based and do not specifically consider functional parts. When lifting these 2D features to geometry-profound 3D space, challenges arise, such as long runtimes, multi-view inconsistencies, and low spatial resolution with insufficient geometric information. To address these issues, we propose Part-Aware 3D Feature Field (PA3FF), a novel dense 3D feature with part awareness for generalizable articulated object manipulation. PA3FF is trained by 3D part proposals from a large-scale labeled dataset, via a contrastive learning formulation. Given point clouds as input, PA3FF predicts a continuous 3D feature field in a feedforward manner, where the distance between point features reflects the proximity of functional parts: points with similar features are more likely to belong to the same part. Building on this feature, we introduce the Part-Aware Diffusion Policy (PADP), an imitation learning framework aimed at enhancing sample efficiency and generalization for robotic manipulation. We evaluate PADP on several simulated and real-world tasks, demonstrating that PA3FF consistently outperforms a range of 2D and 3D representations in manipulation scenarios, including CLIP, DINOv2, and Grounded-SAM. Beyond imitation learning, PA3FF enables diverse downstream methods, including correspondence learning and segmentation tasks, making it a versatile foundation for robotic manipulation. Project page: https://pa3ff.github.io
comment: Accept to ICLR 2026, Project page: https://pa3ff.github.io
☆ Direction Matters: Learning Force Direction Enables Sim-to-Real Contact-Rich Manipulation
Sim-to-real transfer for contact-rich manipulation remains challenging due to the inherent discrepancy in contact dynamics. While existing methods often rely on costly real-world data or utilize blind compliance through fixed controllers, we propose a framework that leverages expert-designed controller logic for transfer. Inspired by the success of privileged supervision in kinematic tasks, we employ a human-designed finite state machine based position/force controller in simulation to provide privileged guidance. The resulting policy is trained to predict the end-effector pose, contact state, and crucially the desired contact force direction. Unlike force magnitudes, which are highly sensitive to simulation inaccuracies, force directions encode high-level task geometry and remain robust across the sim-to-real gap. At deployment, these predictions configure a force-aware admittance controller. By combining the policy's directional intent with a constant, low-cost manually tuned force magnitude, the system generates adaptive, task-aligned compliance. This tuning is lightweight, typically requiring only a single scalar per contact state. We provide theoretical analysis for stability and robustness to disturbances. Experiments on four real-world tasks, i.e., microwave opening, peg-in-hole, whiteboard wiping, and door opening, demonstrate that our approach significantly outperforms strong baselines in both success rate and robustness. Videos are available at: https://yifei-y.github.io/project-pages/DirectionMatters/.
☆ Rigidity-Based Multi-Finger Coordination for Precise In-Hand Manipulation of Force-Sensitive Objects
Precise in-hand manipulation of force-sensitive objects typically requires judicious coordinated force planning as well as accurate contact force feedback and control. Unlike multi-arm platforms with gripper end effectors, multi-fingered hands rely solely on fingertip point contacts and are not able to apply pull forces, therefore poses a more challenging problem. Furthermore, calibrated torque sensors are lacking in most commercial dexterous hands, adding to the difficulty. To address these challenges, we propose a dual-layer framework for multi-finger coordination, enabling high-precision manipulation of force-sensitive objects through joint control without tactile feedback. This approach solves coordinated contact force planning by incorporating graph rigidity and force closure constraints. By employing a force-to-position mapping, the planned force trajectory is converted to a joint trajectory. We validate the framework on a custom dexterous hand, demonstrating the capability to manipulate fragile objects-including a soft yarn, a plastic cup, and a raw egg-with high precision and safety.
comment: This paper has been accepted by IEEE Robotics and Automation Letters. The experimental video is avaialable at: https://www.youtube.com/watch?v=kcf9dVW0Dpo
☆ SemanticFeels: Semantic Labeling during In-Hand Manipulation
As robots become increasingly integrated into everyday tasks, their ability to perceive both the shape and properties of objects during in-hand manipulation becomes critical for adaptive and intelligent behavior. We present SemanticFeels, an extension of the NeuralFeels framework that integrates semantic labeling with neural implicit shape representation, from vision and touch. To illustrate its application, we focus on material classification: high-resolution Digit tactile readings are processed by a fine-tuned EfficientNet-B0 convolutional neural network (CNN) to generate local material predictions, which are then embedded into an augmented signed distance field (SDF) network that jointly predicts geometry and continuous material regions. Experimental results show that the system achieves a high correspondence between predicted and actual materials on both single- and multi-material objects, with an average matching accuracy of 79.87% across multiple manipulation trials on a multi-material object.
comment: 10 pages, 5 figures
☆ Simultaneous State Estimation and Online Model Learning in a Soft Robotic System
Operating complex real-world systems, such as soft robots, can benefit from precise predictive control schemes that require accurate state and model knowledge. This knowledge is typically not available in practical settings and must be inferred from noisy measurements. In particular, it is challenging to simultaneously estimate unknown states and learn a model online from sequentially arriving measurements. In this paper, we show how a recently proposed gray-box system identification tool enables the estimation of a soft robot's current pose while at the same time learning a bending stiffness model. For estimation and learning, we rely solely on a nominal constant-curvature robot model and measurements of the robot's base reactions (e.g., base forces). The estimation scheme -- relying on a marginalized particle filter -- allows us to conveniently interface nominal constant-curvature equations with a Gaussian Process (GP) bending stiffness model to be learned. This, in contrast to estimation via a random walk over stiffness values, enables prediction of bending stiffness and improves overall model quality. We demonstrate, using real-world soft-robot data, that the method learns a bending stiffness model online while accurately estimating the robot's pose. Notably, reduced multi-step forward-prediction errors indicate that the learned bending-stiffness GP improves overall model quality.
comment: 8 pages, 3 figures, 2 tables
☆ ProAct: A Dual-System Framework for Proactive Embodied Social Agents
Embodied social agents have recently advanced in generating synchronized speech and gestures. However, most interactive systems remain fundamentally reactive, responding only to current sensory inputs within a short temporal window. Proactive social behavior, in contrast, requires deliberation over accumulated context and intent inference, which conflicts with the strict latency budget of real-time interaction. We present \emph{ProAct}, a dual-system framework that reconciles this time-scale conflict by decoupling a low-latency \emph{Behavioral System} for streaming multimodal interaction from a slower \emph{Cognitive System} which performs long-horizon social reasoning and produces high-level proactive intentions. To translate deliberative intentions into continuous non-verbal behaviors without disrupting fluency, we introduce a streaming flow-matching model conditioned on intentions via ControlNet. This mechanism supports asynchronous intention injection, enabling seamless transitions between reactive and proactive gestures within a single motion stream. We deploy ProAct on a physical humanoid robot and evaluate both motion quality and interactive effectiveness. In real-world interaction user studies, participants and observers consistently prefer ProAct over reactive variants in perceived proactivity, social presence, and overall engagement, demonstrating the benefits of dual-system proactive control for embodied social interaction.
comment: Project Page: https://proactrobot.github.io/
☆ RoboAug: One Annotation to Hundreds of Scenes via Region-Contrastive Data Augmentation for Robotic Manipulation
Enhancing the generalization capability of robotic learning to enable robots to operate effectively in diverse, unseen scenes is a fundamental and challenging problem. Existing approaches often depend on pretraining with large-scale data collection, which is labor-intensive and time-consuming, or on semantic data augmentation techniques that necessitate an impractical assumption of flawless upstream object detection in real-world scenarios. In this work, we propose RoboAug, a novel generative data augmentation framework that significantly minimizes the reliance on large-scale pretraining and the perfect visual recognition assumption by requiring only the bounding box annotation of a single image during training. Leveraging this minimal information, RoboAug employs pre-trained generative models for precise semantic data augmentation and integrates a plug-and-play region-contrastive loss to help models focus on task-relevant regions, thereby improving generalization and boosting task success rates. We conduct extensive real-world experiments on three robots, namely UR-5e, AgileX, and Tien Kung 2.0, spanning over 35k rollouts. Empirical results demonstrate that RoboAug significantly outperforms state-of-the-art data augmentation baselines. Specifically, when evaluating generalization capabilities in unseen scenes featuring diverse combinations of backgrounds, distractors, and lighting conditions, our method achieves substantial gains over the baseline without augmentation. The success rates increase from 0.09 to 0.47 on UR-5e, from 0.16 to 0.60 on AgileX, and from 0.19 to 0.67 on Tien Kung 2.0. These results highlight the superior generalization and effectiveness of RoboAug in real-world manipulation tasks. Our project is available at https://x-roboaug.github.io/.
☆ It Takes Two to Tango: A Holistic Simulator for Joint Order Scheduling and Multi-Agent Path Finding in Robotic Warehouses
The prevailing paradigm in Robotic Mobile Fulfillment Systems (RMFS) typically treats order scheduling and multi-agent pathfinding as isolated sub-problems. We argue that this decoupling is a fundamental bottleneck, masking the critical dependencies between high-level dispatching and low-level congestion. Existing simulators fail to bridge this gap, often abstracting away heterogeneous kinematics and stochastic execution failures. We propose WareRover, a holistic simulation platform that enforces a tight coupling between OS and MAPF via a unified, closed-loop optimization interface. Unlike standard benchmarks, WareRover integrates dynamic order streams, physics-aware motion constraints, and non-nominal recovery mechanisms into a single evaluation loop. Experiments reveal that SOTA algorithms often falter under these realistic coupled constraints, demonstrating that WareRover provides a necessary and challenging testbed for robust, next-generation warehouse coordination. The project and video is available at https://hhh-x.github.io/WareRover/.
☆ WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RL
Reinforcement learning (RL) promises to unlock capabilities beyond imitation learning for Vision-Language-Action (VLA) models, but its requirement for massive real-world interaction prevents direct deployment on physical robots. Recent work attempts to use learned world models as simulators for policy optimization, yet closed-loop imagined rollouts inevitably suffer from hallucination and long-horizon error accumulation. Such errors do not merely degrade visual fidelity; they corrupt the optimization signal, encouraging policies to exploit model inaccuracies rather than genuine task progress. We propose WoVR, a reliable world-model-based reinforcement learning framework for post-training VLA policies. Instead of assuming a faithful world model, WoVR explicitly regulates how RL interacts with imperfect imagined dynamics. It improves rollout stability through a controllable action-conditioned video world model, reshapes imagined interaction to reduce effective error depth via Keyframe-Initialized Rollouts, and maintains policy-simulator alignment through World Model-Policy co-evolution. Extensive experiments on LIBERO benchmarks and real-world robotic manipulation demonstrate that WoVR enables stable long-horizon imagined rollouts and effective policy optimization, improving average LIBERO success from 39.95% to 69.2% (+29.3 points) and real-robot success from 61.7% to 91.7% (+30.0 points). These results show that learned world models can serve as practical simulators for reinforcement learning when hallucination is explicitly controlled.
comment: 21pages, 8 figures
☆ Joint Task Assistance Planning via Nested Branch and Bound (Extended Version)
We introduce and study the Joint Task Assistance Planning problem which generalizes prior work on optimizing assistance in robotic collaboration. In this setting, two robots operate over predefined roadmaps, each represented as a graph corresponding to its configuration space. One robot, the task robot, must execute a timed mission, while the other, the assistance robot, provides sensor-based support that depends on their spatial relationship. The objective is to compute a path for both robots that maximizes the total duration of assistance given. Solving this problem is challenging due to the combinatorial explosion of possible path combinations together with the temporal nature of the problem (time needs to be accounted for as well). To address this, we propose a nested branch-and-bound framework that efficiently explores the space of robot paths in a hierarchical manner. We empirically evaluate our algorithm and demonstrate a speedup of up to two orders of magnitude when compared to a baseline approach.
☆ A Comprehensive Survey on Deep Learning-Based LiDAR Super-Resolution for Autonomous Driving
LiDAR sensors are often considered essential for autonomous driving, but high-resolution sensors remain expensive while affordable low-resolution sensors produce sparse point clouds that miss critical details. LiDAR super-resolution addresses this challenge by using deep learning to enhance sparse point clouds, bridging the gap between different sensor types and enabling cross-sensor compatibility in real-world deployments. This paper presents the first comprehensive survey of LiDAR super-resolution methods for autonomous driving. Despite the importance of practical deployment, no systematic review has been conducted until now. We organize existing approaches into four categories: CNN-based architectures, model-based deep unrolling, implicit representation methods, and Transformer and Mamba-based approaches. We establish fundamental concepts including data representations, problem formulation, benchmark datasets and evaluation metrics. Current trends include the adoption of range image representation for efficient processing, extreme model compression and the development of resolution-flexible architectures. Recent research prioritizes real-time inference and cross-sensor generalization for practical deployment. We conclude by identifying open challenges and future research directions for advancing LiDAR super-resolution technology.
comment: Accepted to The IEEE Intelligent Vehicles Symposium 2026 (IEEE IV 2026)
♻ ☆ Pack it in: Packing into Partially Filled Containers Through Contact
The automation of warehouse operations is crucial for improving productivity and reducing human exposure to hazardous environments. One operation frequently performed in warehouses is bin-packing where items need to be placed into containers, either for delivery to a customer, or for temporary storage in the warehouse. Whilst prior bin-packing works have largely been focused on packing items into empty containers and have adopted collision-free strategies, it is often the case that containers will already be partially filled with items, often in suboptimal arrangements due to transportation about a warehouse. This paper presents a contact-aware packing approach that exploits purposeful interactions with previously placed objects to create free space and enable successful placement of new items. This is achieved by using a contact-based multi-object trajectory optimizer within a model predictive controller, integrated with a physics-aware perception system that estimates object poses even during inevitable occlusions, and a method that suggests physically-feasible locations to place the object inside the container.
comment: 8 pages, 5 figures
♻ ☆ ORACLE-Grasp: Zero-Shot Affordance-Aligned Robotic Grasping using Large Multimodal Models
Grasping unknown objects in unstructured environments is a critical challenge for service robots, which must operate in dynamic, real-world settings such as homes, hospitals, and warehouses. Success in these environments requires both semantic understanding and spatial reasoning. Traditional methods often rely on dense training datasets or detailed geometric modeling, which demand extensive data collection and do not generalize well to novel objects or affordances. We present ORACLE-Grasp, a zero-shot framework that leverages Large Multimodal Models (LMMs) as semantic oracles to guide affordance-aligned grasp selection, without requiring task-specific training or manual input. The system reformulates grasp prediction as a structured, iterative decision process, using a dual-prompt tool-calling strategy: the first prompt extracts high-level object semantics, while the second identifies graspable regions aligned with the object's function. To address the spatial limitations of LMMs, ORACLE-Grasp discretizes the image into candidate regions and reasons over them to produce human-like and context-sensitive grasp suggestions. A depth-based refinement step improves grasp reliability when available, and an early stopping mechanism enhances computational efficiency. We evaluate ORACLE-Grasp on a diverse set of RGB and RGB-D images featuring both everyday and AI-generated objects. The results show that our method produces physically feasible and semantically appropriate grasps that align closely with human annotations, achieving high success rates in real-world pick-up tasks. Our findings highlight the potential of LMMs for enabling flexible and generalizable grasping strategies in autonomous service robots, eliminating the need for object-specific models or extensive training.
♻ ☆ Multi-Objective Neural Network-Assisted Design Optimization of Soft Fin-Ray Fingers for Enhanced Grasping Performance
The internal structure of the Fin-Ray fingers plays a significant role in their adaptability and grasping performance. However, modeling the grasp force and deformation behavior for design purposes is challenging. When the Fin-Ray finger becomes more rigid and capable of exerting higher forces, it becomes less delicate in handling objects. The contrast between these two gives rise to a multi-objective optimization problem. We employ the finite element method to estimate the deflections and contact forces of the Fin-Ray fingers grasping cylindrical objects, generating a dataset of 120 simulations. This dataset includes three input variables: the thickness of the front and support beams, the thickness of the crossbeams, and the equal spacing between the crossbeams, which are the design variables in the optimization. This dataset is then used to construct a multilayer perceptron (MLP) with four output neurons predicting the contact force and tip displacement in two directions. The magnitudes of maximum contact force and maximum tip displacement are two optimization objectives, showing the trade-off between force and delicate manipulation. The set of solutions is found using the non-dominated sorting genetic algorithm (NSGA-II). The results of the simulations demonstrate that the proposed methodology can be used to improve the design and grasping performance of soft grippers, aiding to choose a design not only for delicate grasping but also for high-force applications.
comment: Major revision correcting errors that affected the reported results. Corresponding results were regenerated, and the manuscript was updated accordingly
♻ ☆ LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfer
A long-standing goal in robotics is a generalist policy that can be deployed zero-shot on new robot embodiments without per-embodiment adaptation. Despite large-scale multi-embodiment pre-training, existing Vision-Language-Action models (VLAs) remain tightly coupled to their training embodiments and typically require costly fine-tuning. We introduce Language-Action Pre-training (LAP), a simple recipe that represents low-level robot actions directly in natural language, aligning action supervision with the pre-trained vision-language model's input-output distribution. LAP requires no learned tokenizer, no costly annotation, and no embodiment-specific architectural design. Based on LAP, we present LAP-3B, which to the best of our knowledge is the first VLA to achieve substantial zero-shot transfer to previously unseen robot embodiments without any embodiment-specific fine-tuning. Across multiple novel robots and manipulation tasks, LAP-3B attains over 50% average zero-shot success, delivering roughly a 2x improvement over the strongest prior VLAs. We further show that LAP enables efficient adaptation and favorable scaling, while unifying action prediction and VQA in a shared language-action format that yields additional gains through co-training.
comment: Project website: https://lap-vla.github.io
♻ ☆ Generation of Uncertainty-Aware High-Level Spatial Concepts in Factorized 3D Scene Graphs via Graph Neural Networks
Enabling robots to autonomously discover high-level spatial concepts (e.g., rooms and walls) from primitive geometric observations (e.g., planar surfaces) within 3D Scene Graphs is essential for robust indoor navigation and mapping. These graphs provide a hierarchical metric-semantic representation in which such concepts are organized. To further enhance graph-SLAM performance, Factorized 3D Scene Graphs incorporate these concepts as optimization factors that constrain relative geometry and enforce global consistency. However, both stages of this process remain largely manual: concepts are typically derived using hand-crafted, concept-specific heuristics, while factors and their covariances are likewise manually designed. This reliance on manual specification limits generalization across diverse environments and scalability to new concept classes. This paper presents a novel learning-based method that infers spatial concepts online from observed vertical planes and introduces them as optimizable factors within a SLAM backend, eliminating the need to handcraft concept generation, factor design, and covariance specification. We evaluate our approach in simulated environments with complex layouts, improving room detection by 20.7% and trajectory estimation by 19.2%, and further validate it on real construction sites, where room detection improves by 5.3% and map matching accuracy by 3.8%. Results confirm that learned factors can improve their handcrafted counterparts in SLAM systems and serve as a foundation for extending this approach to new spatial concepts.
comment: Submitted to IEEE Robotics and Automation Letters (RA-L)
♻ ☆ System-Level Error Propagation and Tail-Risk Amplification in Reference-Based Robotic Navigation
Image guided robotic navigation systems often rely on reference based geometric perception pipelines, where accurate spatial mapping is established through multi stage estimation processes. In biplanar X ray guided navigation, such pipelines are widely used due to their real time capability and geometric interpretability. However, navigation reliability can be constrained by an overlooked system level failure mechanism in which installation induced structural perturbations introduced at the perception stage are progressively amplified along the perception reconstruction execution chain and dominate execution level error and tail risk behavior. This paper investigates this mechanism from a system level perspective and presents a unified error propagation modeling framework that characterizes how installation induced structural perturbations propagate and couple with pixel level observation noise through biplanar imaging, projection matrix estimation, triangulation, and coordinate mapping. Using first order analytic uncertainty propagation and Monte Carlo simulations, we analyze dominant sensitivity channels and quantify worst case error behavior beyond mean accuracy metrics. The results show that rotational installation error is a primary driver of system level error amplification, while translational misalignment of comparable magnitude plays a secondary role under typical biplanar geometries. Real biplanar X ray bench top experiments further confirm that the predicted amplification trends persist under realistic imaging conditions. These findings reveal a broader structural limitation of reference based multi stage geometric perception pipelines and provide a framework for system level reliability analysis and risk aware design in safety critical robotic navigation systems.
comment: 13 pages, 8 figures
♻ ☆ DeCo: Task Decomposition and Skill Composition for Zero-Shot Generalization in Long-Horizon 3D Manipulation
Generalizing language-conditioned multi-task imitation learning (IL) models to novel long-horizon 3D manipulation tasks is challenging. To address this, we propose DeCo (Task Decomposition and Skill Composition), a model-agnostic framework that enhances zero-shot generalization to compositional long-horizon manipulation tasks. DeCo decomposes IL demonstrations into modular atomic tasks based on gripper-object interactions, creating a dataset that enables models to learn reusable skills. At inference, DeCo uses a vision-language model (VLM) to parse high-level instructions, retrieve relevant skills, and dynamically schedule their execution. A spatially-aware skill-chaining module ensures smooth, collision-free transitions between skills. We introduce DeCoBench, a benchmark designed to evaluate compositional generalization in long-horizon manipulation tasks. DeCo improves the success rate of three IL models, RVT-2, 3DDA, and ARP, by 66.67%, 21.53%, and 57.92%, respectively, on 12 novel tasks. In real-world experiments, the DeCo-enhanced model, trained on only 6 atomic tasks, completes 9 novel tasks in zero-shot, with a 53.33% improvement over the baseline model. Project website: https://deco226.github.io.
comment: RAL 2026
♻ ☆ GelSLAM: A Real-time, High-Fidelity, and Robust 3D Tactile SLAM System
Accurately perceiving an object's pose and shape is essential for precise grasping and manipulation. Compared to common vision-based methods, tactile sensing offers advantages in precision and immunity to occlusion when tracking and reconstructing objects in contact. This makes it particularly valuable for in-hand and other high-precision manipulation tasks. In this work, we present GelSLAM, a real-time 3D SLAM system that relies solely on tactile sensing to estimate object pose over long periods and reconstruct object shapes with high fidelity. Unlike traditional point cloud-based approaches, GelSLAM uses tactile-derived surface normals and curvatures for robust tracking and loop closure. It can track object motion in real time with low error and minimal drift, and reconstruct shapes with submillimeter accuracy, even for low-texture objects such as wooden tools. GelSLAM extends tactile sensing beyond local contact to enable global, long-horizon spatial perception, and we believe it will serve as a foundation for many precise manipulation tasks involving interaction with objects in hand. The video demo, code, and dataset are available at https://joehjhuang.github.io/gelslam.
comment: 20 pages
♻ ☆ Vision-Based Early Fault Diagnosis and Self-Recovery for Strawberry Harvesting Robots
Strawberry harvesting robots faced persistent challenges such as low integration of visual perception, fruit-gripper misalignment, empty grasping/misgrasp, and strawberry slippage from the gripper due to insufficient gripping force, all of which compromised harvesting stability and efficiency in orchard environments. To overcome these issues, this paper proposed a visual fault diagnosis and self-recovery framework that integrated multi-task perception with corrective control strategies. At the core of this framework was SRR-Net, an end-to-end multi-task perception model that simultaneously performed strawberry detection, segmentation, and ripeness estimation, thereby unifying visual perception with fault diagnosis.Based on this integrated perception, a relative error compensation method based on the simultaneous target-gripper detection was designed to address positional misalignment, correcting deviations when error exceeded the tolerance threshold.To mitigate empty grasping/misgrasp and fruit-slippage faults, an early abort strategy was implemented. A micro-optical camera embedded in the end-effector provided real-time visual feedback, enabling grasp classification during the deflating stage and strawberry slip prediction during snap-off through MobileNet V3-Small classifier and a time-series LSTM classifier. Experiments demonstrated that SRR-Net maintained high perception accuracy. For detection, it achieved a precision of 0.895 and recall of 0.813 on strawberries, and 0.972/0.958 on hands. In segmentation, it yielded a precision of 0.887 and recall of 0.747 for strawberries, and 0.974/0.947 for hands. For ripeness estimation, SRR-Net attained a mean absolute error of 0.035, while simultaneously supporting multi-task perception and sustaining a competitive inference speed of 163.35 FPS.
♻ ☆ Learning Category-level Last-meter Navigation from RGB Demonstrations of a Single-instance
Achieving precise positioning of the mobile manipulator's base is essential for successful manipulation actions that follow. Most of the RGB-based navigation systems only guarantee coarse, meter-level accuracy, making them less suitable for the precise positioning phase of mobile manipulation. This gap prevents manipulation policies from operating within the distribution of their training demonstrations, resulting in frequent execution failures. We address this gap by introducing an object-centric imitation learning framework for last-meter navigation, enabling a quadruped mobile manipulator robot to achieve manipulation-ready positioning using only RGB observations from its onboard cameras. Our method conditions the navigation policy on three inputs: goal images, multi-view RGB observations from the onboard cameras, and a text prompt specifying the target object. A language-driven segmentation module and a spatial score-matrix decoder then supply explicit object grounding and relative pose reasoning. Using real-world data from a single object instance within a category, the system generalizes to unseen object instances across diverse environments with challenging lighting and background conditions. To comprehensively evaluate this, we introduce two metrics: an edge-alignment metric, which uses ground truth orientation, and an object-alignment metric, which evaluates how well the robot visually faces the target. Under these metrics, our policy achieves 73.47% success in edge-alignment and 96.94% success in object-alignment when positioning relative to unseen target objects. These results show that precise last-meter navigation can be achieved at a category-level without depth, LiDAR, or map priors, enabling a scalable pathway toward unified mobile manipulation. Project page: https://rpm-lab-umn.github.io/category-level-last-meter-nav/
♻ ☆ Instance-Guided Unsupervised Domain Adaptation for Robotic Semantic Segmentation ICRA 2026
Semantic segmentation networks, which are essential for robotic perception, often suffer from performance degradation when the visual distribution of the deployment environment differs from that of the source dataset on which they were trained. Unsupervised Domain Adaptation (UDA) addresses this challenge by adapting the network to the robot's target environment without external supervision, leveraging the large amounts of data a robot might naturally collect during long-term operation. In such settings, UDA methods can exploit multi-view consistency across the environment's map to fine-tune the model in an unsupervised fashion and mitigate domain shift. However, these approaches remain sensitive to cross-view instance-level inconsistencies. In this work, we propose a method that starts from a volumetric 3D map to generate multi-view consistent pseudo-labels. We then refine these labels using the zero-shot instance segmentation capabilities of a foundation model, enforcing instance-level coherence. The refined annotations serve as supervision for self-supervised fine-tuning, enabling the robot to adapt its perception system at deployment time. Experiments on real-world data demonstrate that our approach consistently improves performance over state-of-the-art UDA baselines based on multi-view consistency, without requiring any ground-truth labels in the target domain.
comment: Accepted for publication at ICRA 2026
♻ ☆ VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Model
The goal of this paper is to improve the performance and reliability of vision-language-action (VLA) models through iterative online interaction. Since collecting policy rollouts in the real world is expensive, we investigate whether a learned simulator-specifically, an action-conditioned video generation model-can be used to generate additional rollout data. Unfortunately, existing world models lack the physical fidelity necessary for policy improvement: they are predominantly trained on demonstration datasets that lack coverage of many different physical interactions (particularly failure cases) and struggle to accurately model small yet critical physical details in contact-rich object manipulation. We propose a simple iterative improvement algorithm that uses real-world roll-out data to improve the fidelity of the world model, which can then, in turn, be used to generate supplemental synthetic data for improving the VLA model. In our experiments on a real robot, we use this approach to improve the performance of a state-of-the-art VLA model on multiple downstream tasks. We achieve a 39.2% absolute success rate improvement over the base policy and 11.6% improvement from training with the generated synthetic rollouts. Videos can be found at this anonymous website: https://sites.google.com/view/vla-w
comment: Project Page: https://sites.google.com/view/vlaw-arxiv
Robotics 35
☆ High-fidelity 3D reconstruction for planetary exploration
Planetary exploration increasingly relies on autonomous robotic systems capable of perceiving, interpreting, and reconstructing their surroundings in the absence of global positioning or real-time communication with Earth. Rovers operating on planetary surfaces must navigate under sever environmental constraints, limited visual redundancy, and communication delays, making onboard spatial awareness and visual localization key components for mission success. Traditional techniques based on Structure-from-Motion (SfM) and Simultaneous Localization and Mapping (SLAM) provide geometric consistency but struggle to capture radiometric detail or to scale efficiently in unstructured, low-texture terrains typical of extraterrestrial environments. This work explores the integration of radiance field-based methods - specifically Neural Radiance Fields (NeRF) and Gaussian Splatting - into a unified, automated environment reconstruction pipeline for planetary robotics. Our system combines the Nerfstudio and COLMAP frameworks with a ROS2-compatible workflow capable of processing raw rover data directly from rosbag recordings. This approach enables the generation of dense, photorealistic, and metrically consistent 3D representations from minimal visual input, supporting improved perception and planning for autonomous systems operating in planetary-like conditions. The resulting pipeline established a foundation for future research in radiance field-based mapping, bridging the gap between geometric and neural representations in planetary exploration.
comment: 7 pages, 3 figures, conference paper
☆ RPGD: RANSAC-P3P Gradient Descent for Extrinsic Calibration in 3D Human Pose Estimation SC
In this paper, we propose RPGD (RANSAC-P3P Gradient Descent), a human-pose-driven extrinsic calibration framework that robustly aligns MoCap-based 3D skeletal data with monocular or multi-view RGB cameras using only natural human motion. RPGD formulates extrinsic calibration as a coarse-to-fine problem tailored to human poses, combining the global robustness of RANSAC-P3P with Gradient-Descent-based refinement. We evaluate RPGD on three large-scale public 3D HPE datasets as well as on a self-collected in-the-wild dataset. Experimental results demonstrate that RPGD consistently recovers extrinsic parameters with accuracy comparable to the provided ground truth, achieving sub-pixel MPJPE reprojection error even in challenging, noisy settings. These results indicate that RPGD provides a practical and automatic solution for reliable extrinsic calibration of large-scale 3D HPE dataset collection.
comment: Accepted at AAIML 2026. This work is co-funded by the European Union's Horizon Europe research and innovation programme under MSCA with grant agreement No 101081674
☆ UAV-SEAD: State Estimation Anomaly Dataset for UAVs
Accurate state estimation in Unmanned Aerial Vehicles (UAVs) is crucial for ensuring reliable and safe operation, as anomalies occurring during mission execution may induce discrepancies between expected and observed system behaviors, thereby compromising mission success or posing potential safety hazards. It is essential to continuously monitor and detect such conditions in order to ensure a timely response and maintain system reliability. In this work, we focus on UAV state estimation anomalies and provide a large-scale real-world UAV dataset to facilitate research aimed at improving the development of anomaly detection. Unlike existing datasets that primarily rely on injected faults into simulated data, this dataset comprises 1396 real flight logs totaling over 52 hours of flight time, collected across diverse indoor and outdoor environments using a collection of PX4-based UAVs equipped with a variety of sensor configurations. The dataset comprises both normal and anomalous flights without synthetic manipulation, making it uniquely suitable for realistic anomaly detection tasks. A structured classification is proposed that categorizes UAV state estimation anomalies into four classes: mechanical and electrical, external position, global position, and altitude anomalies. These classifications reflect collective, contextual, and outlier anomalies observed in multivariate sensor data streams, including IMU, GPS, barometer, magnetometer, distance sensors, visual odometry, and optical flow, that can be found in the PX4 logging mechanism. It is anticipated that this dataset will play a key role in the development, training, and evaluation of anomaly detection and isolation systems to address the critical gap in UAV reliability research.
☆ Modeling and Optimizing the Provisioning of Exhaustible Capabilities for Simultaneous Task Allocation and Scheduling AAMAS 2026
Deploying heterogeneous robot teams to accomplish multiple tasks over extended time horizons presents significant computational challenges for task allocation and planning. In this paper, we present a comprehensive, time-extended, offline heterogeneous multi-robot task allocation framework, TRAITS, which we believe to be the first that can cope with the provisioning of exhaustible traits under battery and temporal constraints. Specifically, we introduce a nonlinear programming-based trait distribution module that can optimize the trait-provisioning rate of coalitions to yield feasible and time-efficient solutions. TRAITS provides a more accurate feasibility assessment and estimation of task execution times and makespan by leveraging trait-provisioning rates while optimizing battery consumption -- an advantage that state-of-the-art frameworks lack. We evaluate TRAITS against two state-of-the-art frameworks, with results demonstrating its advantage in satisfying complex trait and battery requirements while remaining computationally tractable.
comment: Accepted at AAMAS 2026
☆ Enabling Option Learning in Sparse Rewards with Hindsight Experience Replay
Hierarchical Reinforcement Learning (HRL) frameworks like Option-Critic (OC) and Multi-updates Option Critic (MOC) have introduced significant advancements in learning reusable options. However, these methods underperform in multi-goal environments with sparse rewards, where actions must be linked to temporally distant outcomes. To address this limitation, we first propose MOC-HER, which integrates the Hindsight Experience Replay (HER) mechanism into the MOC framework. By relabeling goals from achieved outcomes, MOC-HER can solve sparse reward environments that are intractable for the original MOC. However, this approach is insufficient for object manipulation tasks, where the reward depends on the object reaching the goal rather than on the agent's direct interaction. This makes it extremely difficult for HRL agents to discover how to interact with these objects. To overcome this issue, we introduce Dual Objectives Hindsight Experience Replay (2HER), a novel extension that creates two sets of virtual goals. In addition to relabeling goals based on the object's final state (standard HER), 2HER also generates goals from the agent's effector positions, rewarding the agent for both interacting with the object and completing the task. Experimental results in robotic manipulation environments show that MOC-2HER achieves success rates of up to 90%, compared to less than 11% for both MOC and MOC-HER. These results highlight the effectiveness of our dual objective relabeling strategy in sparse reward, multi-goal tasks.
☆ Humanoid Hanoi: Investigating Shared Whole-Body Control for Skill-Based Box Rearrangement
We investigate a skill-based framework for humanoid box rearrangement that enables long-horizon execution by sequencing reusable skills at the task level. In our architecture, all skills execute through a shared, task-agnostic whole-body controller (WBC), providing a consistent closed-loop interface for skill composition, in contrast to non-shared designs that use separate low-level controllers per skill. We find that naively reusing the same pretrained WBC can reduce robustness over long horizons, as new skills and their compositions induce shifted state and command distributions. We address this with a simple data aggregation procedure that augments shared-WBC training with rollouts from closed-loop skill execution under domain randomization. To evaluate the approach, we introduce \emph{Humanoid Hanoi}, a long-horizon Tower-of-Hanoi box rearrangement benchmark, and report results in simulation and on the Digit V3 humanoid robot, demonstrating fully autonomous rearrangement over extended horizons and quantifying the benefits of the shared-WBC approach over non-shared baselines.
comment: 10 pages, 6 figures
☆ Push-Placement: A Hybrid Approach Integrating Prehensile and Non-Prehensile Manipulation for Object Rearrangement
Efficient tabletop rearrangement remains challenging due to collisions and the need for temporary buffering when target poses are obstructed. Prehensile pick-and-place provides precise control but often requires extra moves, whereas non-prehensile pushing can be more efficient but suffers from complex, imprecise dynamics. This paper proposes push-placement, a hybrid action primitive that uses the grasped object to displace obstructing items while being placed, thereby reducing explicit buffering. The method is integrated into a physics-in-the-loop Monte Carlo Tree Search (MCTS) planner and evaluated in the PyBullet simulator. Empirical results show push-placement reduces the manipulator travel cost by up to 11.12% versus a baseline MCTS planner and 8.56% versus dynamic stacking. These findings indicate that hybrid prehensile/non-prehensile action primitives can substantially improve efficiency in long-horizon rearrangement tasks.
comment: International Conference on Robotics and Mechatronics (ICRoM 2025)
☆ Semantic-Contact Fields for Category-Level Generalizable Tactile Tool Manipulation
Generalizing tool manipulation requires both semantic planning and precise physical control. Modern generalist robot policies, such as Vision-Language-Action (VLA) models, often lack the high-fidelity physical grounding required for contact-rich tool manipulation. Conversely, existing contact-aware policies that leverage tactile or haptic sensing are typically instance-specific and fail to generalize across diverse tool geometries. Bridging this gap requires learning unified contact representations from diverse data, yet a fundamental barrier remains: diverse real-world tactile data are prohibitive at scale, while direct zero-shot sim-to-real transfer is challenging due to the complex dynamics of nonlinear deformation of soft sensors. To address this, we propose Semantic-Contact Fields (SCFields), a unified 3D representation fusing visual semantics with dense contact estimates. We enable this via a two-stage Sim-to-Real Contact Learning Pipeline: first, we pre-train on a large simulation data set to learn general contact physics; second, we fine-tune on a small set of real data, pseudo-labeled via geometric heuristics and force optimization, to align sensor characteristics. This allows physical generalization to unseen tools. We leverage SCFields as the dense observation input for a diffusion policy to enable robust execution of contact-rich tool manipulation tasks. Experiments on scraping, crayon drawing, and peeling demonstrate robust category-level generalization, significantly outperforming vision-only and raw-tactile baselines.
☆ Gaussian Sequences with Multi-Scale Dynamics for 4D Reconstruction from Monocular Casual Videos
Understanding dynamic scenes from casual videos is critical for scalable robot learning, yet four-dimensional (4D) reconstruction under strictly monocular settings remains highly ill-posed. To address this challenge, our key insight is that real-world dynamics exhibits a multi-scale regularity from object to particle level. To this end, we design the multi-scale dynamics mechanism that factorizes complex motion fields. Within this formulation, we propose Gaussian sequences with multi-scale dynamics, a novel representation for dynamic 3D Gaussians derived through compositions of multi-level motion. This layered structure substantially alleviates ambiguity of reconstruction and promotes physically plausible dynamics. We further incorporate multi-modal priors from vision foundation models to establish complementary supervision, constraining the solution space and improving the reconstruction fidelity. Our approach enables accurate and globally consistent 4D reconstruction from monocular casual videos. Experiments of dynamic novel-view synthesis (NVS) on benchmark and real-world manipulation datasets demonstrate considerable improvements over existing methods.
☆ Ontological grounding for sound and natural robot explanations via large language models AAMAS 2026
Building effective human-robot interaction requires robots to derive conclusions from their experiences that are both logically sound and communicated in ways aligned with human expectations. This paper presents a hybrid framework that blends ontology-based reasoning with large language models (LLMs) to produce semantically grounded and natural robot explanations. Ontologies ensure logical consistency and domain grounding, while LLMs provide fluent, context-aware and adaptive language generation. The proposed method grounds data from human-robot experiences, enabling robots to reason about whether events are typical or atypical based on their properties. We integrate a state-of-the-art algorithm for retrieving and constructing static contrastive ontology-based narratives with an LLM agent that uses them to produce concise, clear, interactive explanations. The approach is validated through a laboratory study replicating an industrial collaborative task. Empirical results show significant improvements in the clarity and brevity of ontology-based narratives while preserving their semantic accuracy. Initial evaluations further demonstrate the system's ability to adapt explanations to user feedback. Overall, this work highlights the potential of ontology-LLM integration to advance explainable agency, and promote more transparent human-robot collaboration.
comment: An extended abstract of this article is accepted for presentation at AAMAS 2026: Olivares-Alarcos, A., Muhammad, A., Sanjaya, S., Lin, H. and Alenyà, G. (2026). Blending ontologies and language models to generate sound and natural robot explanations. In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems. IFAAMAS
☆ MOTIF: Learning Action Motifs for Few-shot Cross-Embodiment Transfer
While vision-language-action (VLA) models have advanced generalist robotic learning, cross-embodiment transfer remains challenging due to kinematic heterogeneity and the high cost of collecting sufficient real-world demonstrations to support fine-tuning. Existing cross-embodiment policies typically rely on shared-private architectures, which suffer from limited capacity of private parameters and lack explicit adaptation mechanisms. To address these limitations, we introduce MOTIF for efficient few-shot cross-embodiment transfer that decouples embodiment-agnostic spatiotemporal patterns, termed action motifs, from heterogeneous action data. Specifically, MOTIF first learns unified motifs via vector quantization with progress-aware alignment and embodiment adversarial constraints to ensure temporal and cross-embodiment consistency. We then design a lightweight predictor that predicts these motifs from real-time inputs to guide a flow-matching policy, fusing them with robot-specific states to enable action generation on new embodiments. Evaluations across both simulation and real-world environments validate the superiority of MOTIF, which significantly outperforms strong baselines in few-shot transfer scenarios by 6.5% in simulation and 43.7% in real-world settings. Code is available at https://github.com/buduz/MOTIF.
☆ Impact-Robust Posture Optimization for Aerial Manipulation
We present a novel method for optimizing the posture of kinematically redundant torque-controlled robots to improve robustness during impacts. A rigid impact model is used as the basis for a configuration-dependent metric that quantifies the variation between pre- and post-impact velocities. By finding configurations (postures) that minimize the aforementioned metric, spikes in the robot's state and input commands can be significantly reduced during impacts, improving safety and robustness. The problem of identifying impact-robust postures is posed as a min-max optimization of the aforementioned metric. To overcome the real-time intractability of the problem, we reformulate it as a gradient-based motion task that iteratively guides the robot towards configurations that minimize the proposed metric. This task is embedded within a task-space inverse dynamics (TSID) whole-body controller, enabling seamless integration with other control objectives. The method is applied to a kinematically redundant aerial manipulator performing repeated point contact tasks. We test our method inside a realistic physics simulator and compare it with the nominal TSID. Our method leads to a reduction (up to 51% w.r.t. standard TSID) of post-impact spikes in the robot's configuration and successfully avoids actuator saturation. Moreover, we demonstrate the importance of kinematic redundancy for impact robustness using additional numerical simulations on a quadruped and a humanoid robot, resulting in up to 45% reduction of post-impact spikes in the robot's state w.r.t. nominal TSID.
☆ The More the Merrier: Running Multiple Neuromorphic Components On-Chip for Robotic Control
It has long been realized that neuromorphic hardware offers benefits for the domain of robotics such as low energy, low latency, as well as unique methods of learning. In aiming for more complex tasks, especially those incorporating multimodal data, one hurdle continuing to prevent their realization is an inability to orchestrate multiple networks on neuromorphic hardware without resorting to off-chip process management logic. To address this, we show a first example of a pipeline for vision-based robot control in which numerous complex networks can be run entirely on hardware via the use of a spiking neural state machine for process orchestration. The pipeline is validated on the Intel Loihi 2 research chip. We show that all components can run concurrently on-chip in the milli Watt regime at latencies competitive with the state-of-the-art. An equivalent network on simulated hardware is shown to accomplish robotic arm plug insertion in simulation, and the core elements of the pipeline are additionally tested on a real robotic arm.
comment: IOP Journal of Neuromorphic Computing and Engineering, preliminary acceptance
☆ XIT: Exploration and Exploitation Informed Trees for Active Gas Distribution Mapping in Unknown Environments
Mobile robotic gas distribution mapping (GDM) provides critical situational awareness during emergency responses to hazardous gas releases. However, most systems still rely on teleoperation, limiting scalability and response speed. Autonomous active GDM is challenging in unknown and cluttered environments, because the robot must simultaneously explore traversable space, map the environment, and infer the gas distribution belief from sparse chemical measurements. We address this by formulating active GDM as a next-best-trajectory informative path planning (IPP) problem and propose XIT (Exploration-Exploitation Informed Trees), a sampling-based planner that balances exploration and exploitation by generating concurrent trajectories toward exploration-rich goals while collecting informative gas measurements en route. XIT draws batches of samples from an Upper Confidence Bound (UCB) information field derived from the current gas posterior and expands trees using a cost that trades off travel effort against gas concentration and uncertainty. To enable plume-aware exploration, we introduce the gas frontier concept, defined as unobserved regions adjacent to high gas concentrations, and propose the Wavefront Gas Frontier Detection (WGFD) algorithm for their identification. High-fidelity simulations and real-world experiments demonstrate the benefits of XIT in terms of GDM quality and efficiency. Although developed for active GDM, XIT is readily applicable to other robotic information-gathering tasks in unknown environments that face the exploration and exploitation trade-off.
☆ Improving Driver Satisfaction with a Driving Function Learning from Implicit Human Feedback -- a Test Group Study
During the use of advanced driver assistance systems, drivers frequently intervene into the active driving function and adjust the system's behavior to their personal wishes. These active driver-initiated takeovers contain feedback about deviations in the driving function's behavior from the drivers' personal preferences. This feedback should be utilized to optimize and personalize the driving function's behavior. In this work, the adjustment of the speed profile of a Predictive Longitudinal Driving Function (PLDF) on a pre-defined route is highlighted. An algorithm is introduced which iteratively adjusts the PLDF's speed profile by taking into account both the original speed profile of the PLDF and the driver demonstration. This approach allows for personalization in a traded control scenario during active use of the PLDF. The applicability of the proposed algorithm is tested in a driving simulator-based test group study with 43 participants. The study finds a significant increase in driver satisfaction and a significant reduction in the intervention frequency when using the proposed adaptive PLDF. Additionally, feedback by the participants was gathered to identify further optimization potentials of the proposed system.
☆ FC-Vision: Real-Time Visibility-Aware Replanning for Occlusion-Free Aerial Target Structure Scanning in Unknown Environments
Autonomous aerial scanning of target structures is crucial for practical applications, requiring online adaptation to unknown obstacles during flight. Existing methods largely emphasize collision avoidance and efficiency, but overlook occlusion-induced visibility degradation, severely compromising scanning quality. In this study, we propose FC-Vision, an on-the-fly visibility-aware replanning framework that proactively and safely prevents target occlusions while preserving the intended coverage and efficiency of the original plan. Our approach explicitly enforces dense surface-visibility constraints to regularize replanning behavior in real-time via an efficient two-level decomposition: occlusion-free viewpoint repair that maintains coverage with minimal deviation from the nominal scan intent, followed by segment-wise clean-sensing connection in 5-DoF space. A plug-in integration strategy is also presented to seamlessly interface FC-Vision with existing UAV scanning systems without architectural changes. Comprehensive simulation and real-world evaluations show that FC-Vision consistently improves scanning quality under unexpected occluders, delivering a maximum coverage gain of 55.32% and a 73.17% reduction in the occlusion ratio, while achieving real-time performance with a moderate increase in flight time. The source code will be made publicly available.
comment: 8 pages, 8 figures, 3 tables
☆ HybridFlow: A Two-Step Generative Policy for Robotic Manipulation
Limited by inference latency, existing robot manipulation policies lack sufficient real-time interaction capability with the environment. Although faster generation methods such as flow matching are gradually replacing diffusion methods, researchers are pursuing even faster generation suitable for interactive robot control. MeanFlow, as a one-step variant of flow matching, has shown strong potential in image generation, but its precision in action generation does not meet the stringent requirements of robotic manipulation. We therefore propose \textbf{HybridFlow}, a \textbf{3-stage method} with \textbf{2-NFE}: Global Jump in MeanFlow mode, ReNoise for distribution alignment, and Local Refine in ReFlow mode. This method balances inference speed and generation quality by leveraging the rapid advantage of MeanFlow one-step generation while ensuring action precision with minimal generation steps. Through real-world experiments, HybridFlow outperforms the 16-step Diffusion Policy by \textbf{15--25\%} in success rate while reducing inference time from 152ms to 19ms (\textbf{8$\times$ speedup}, \textbf{$\sim$52Hz}); it also achieves 70.0\% success on unseen-color OOD grasping and 66.3\% on deformable object folding. We envision HybridFlow as a practical low-latency method to enhance real-world interaction capabilities of robotic manipulation policies.
☆ Symmetry-Aware Fusion of Vision and Tactile Sensing via Bilateral Force Priors for Robotic Manipulation ICRA2026
Insertion tasks in robotic manipulation demand precise, contact-rich interactions that vision alone cannot resolve. While tactile feedback is intuitively valuable, existing studies have shown that naïve visuo-tactile fusion often fails to deliver consistent improvements. In this work, we propose a Cross-Modal Transformer (CMT) for visuo-tactile fusion that integrates wrist-camera observations with tactile signals through structured self- and cross-attention. To stabilize tactile embeddings, we further introduce a physics-informed regularization that encourages bilateral force balance, reflecting principles of human motor control. Experiments on the TacSL benchmark show that CMT with symmetry regularization achieves a 96.59% insertion success rate, surpassing naïve and gated fusion baselines and closely matching the privileged "wrist + contact force" configuration (96.09%). These results highlight two central insights: (i) tactile sensing is indispensable for precise alignment, and (ii) principled multimodal fusion, further strengthened by physics-informed regularization, unlocks complementary strengths of vision and touch, approaching privileged performance under realistic sensing.
comment: Accepted By ICRA2026
☆ A Kung Fu Athlete Bot That Can Do It All Day: Highly Dynamic, Balance-Challenging Motion Dataset and Autonomous Fall-Resilient Tracking
Current humanoid motion tracking systems can execute routine and moderately dynamic behaviors, yet significant gaps remain near hardware performance limits and algorithmic robustness boundaries. Martial arts represent an extreme case of highly dynamic human motion, characterized by rapid center-of-mass shifts, complex coordination, and abrupt posture transitions. However, datasets tailored to such high-intensity scenarios remain scarce. To address this gap, we construct KungFuAthlete, a high-dynamic martial arts motion dataset derived from professional athletes' daily training videos. The dataset includes ground and jump subsets covering representative complex motion patterns. The jump subset exhibits substantially higher joint, linear, and angular velocities compared to commonly used datasets such as LAFAN1, PHUMA, and AMASS, indicating significantly increased motion intensity and complexity. Importantly, even professional athletes may fail during highly dynamic movements. Similarly, humanoid robots are prone to instability and falls under external disturbances or execution errors. Most prior work assumes motion execution remains within safe states and lacks a unified strategy for modeling unsafe states and enabling reliable autonomous recovery. We propose a novel training paradigm that enables a single policy to jointly learn high-dynamic motion tracking and fall recovery, unifying agile execution and stabilization within one framework. This framework expands robotic capability from pure motion tracking to recovery-enabled execution, promoting more robust and autonomous humanoid performance in real-world high-dynamic scenarios.
comment: 18 pages, 8 figures,5 tables
☆ SPLIT: Sparse Incremental Learning of Error Dynamics for Control-Oriented Modeling in Autonomous Vehicles
Accurate, computationally efficient, and adaptive vehicle models are essential for autonomous vehicle control. Hybrid models that combine a nominal model with a Gaussian Process (GP)-based residual model have emerged as a promising approach. However, the GP-based residual model suffers from the curse of dimensionality, high evaluation complexity, and the inefficiency of online learning, which impede the deployment in real-time vehicle controllers. To address these challenges, we propose SPLIT, a sparse incremental learning framework for control-oriented vehicle dynamics modeling. SPLIT integrates three key innovations: (i) Model Decomposition. We decompose the vehicle model into invariant elements calibrated by experiments, and variant elements compensated by the residual model to reduce feature dimensionality. (ii) Local Incremental Learning. We define the valid region in the feature space and partition it into subregions, enabling efficient online learning from streaming data. (iii) GP Sparsification. We use bayesian committee machine to ensure scalable online evaluation. Integrated into model-based controllers, SPLIT is evaluated in aggressive simulations and real-vehicle experiments. Results demonstrate that SPLIT improves model accuracy and control performance online. Moreover, it enables rapid adaptation to vehicle dynamics deviations and exhibits robust generalization to previously unseen scenarios.
comment: 21 pages, 21 figures
☆ Hierarchical Audio-Visual-Proprioceptive Fusion for Precise Robotic Manipulation
Existing robotic manipulation methods primarily rely on visual and proprioceptive observations, which may struggle to infer contact-related interaction states in partially observable real-world environments. Acoustic cues, by contrast, naturally encode rich interaction dynamics during contact, yet remain underexploited in current multimodal fusion literature. Most multimodal fusion approaches implicitly assume homogeneous roles across modalities, and thus design flat and symmetric fusion structures. However, this assumption is ill-suited for acoustic signals, which are inherently sparse and contact-driven. To achieve precise robotic manipulation through acoustic-informed perception, we propose a hierarchical representation fusion framework that progressively integrates audio, vision, and proprioception. Our approach first conditions visual and proprioceptive representations on acoustic cues, and then explicitly models higher-order cross-modal interactions to capture complementary dependencies among modalities. The fused representation is leveraged by a diffusion-based policy to directly generate continuous robot actions from multimodal observations. The combination of end-to-end learning and hierarchical fusion structure enables the policy to exploit task-relevant acoustic information while mitigating interference from less informative modalities. The proposed method has been evaluated on real-world robotic manipulation tasks, including liquid pouring and cabinet opening. Extensive experiment results demonstrate that our approach consistently outperforms state-of-the-art multimodal fusion frameworks, particularly in scenarios where acoustic cues provide task-relevant information not readily available from visual observations alone. Furthermore, a mutual information analysis is conducted to interpret the effect of audio cues in robotic manipulation via multimodal fusion.
☆ AgentRob: From Virtual Forum Agents to Hijacked Physical Robots
Large Language Model (LLM)-powered autonomous agents have demonstrated significant capabilities in virtual environments, yet their integration with the physical world remains narrowly confined to direct control interfaces. We present AgentRob, a framework that bridges online community forums, LLM-powered agents, and physical robots through the Model Context Protocol (MCP). AgentRob enables a novel paradigm where autonomous agents participate in online forums--reading posts, extracting natural language commands, dispatching physical robot actions, and reporting results back to the community. The system comprises three layers: a Forum Layer providing asynchronous, persistent, multi-agent interaction; an Agent Layer with forum agents that poll for @mention-targeted commands; and a Robot Layer with VLM-driven controllers and Unitree Go2/G1 hardware that translate commands into robot primitives via iterative tool calling. The framework supports multiple concurrent agents with distinct identities and physical embodiments coexisting in the same forum, establishing the feasibility of forum-mediated multi-agent robot orchestration.
comment: 10 pages, 2 figures
☆ TactAlign: Human-to-Robot Policy Transfer via Tactile Alignment
Human demonstrations collected by wearable devices (e.g., tactile gloves) provide fast and dexterous supervision for policy learning, and are guided by rich, natural tactile feedback. However, a key challenge is how to transfer human-collected tactile signals to robots despite the differences in sensing modalities and embodiment. Existing human-to-robot (H2R) approaches that incorporate touch often assume identical tactile sensors, require paired data, and involve little to no embodiment gap between human demonstrator and the robots, limiting scalability and generality. We propose TactAlign, a cross-embodiment tactile alignment method that transfers human-collected tactile signals to a robot with different embodiment. TactAlign transforms human and robot tactile observations into a shared latent representation using a rectified flow, without paired datasets, manual labels, or privileged information. Our method enables low-cost latent transport guided by hand-object interaction-derived pseudo-pairs. We demonstrate that TactAlign improves H2R policy transfer across multiple contact-rich tasks (pivoting, insertion, lid closing), generalizes to unseen objects and tasks with human data (less than 5 minutes), and enables zero-shot H2R transfer on a highly dexterous tasks (light bulb screwing).
comment: Website: https://yswi.github.io/tactalign/
☆ ONRAP: Occupancy-driven Noise-Resilient Autonomous Path Planning
Dynamic path planning must remain reliable in the presence of sensing noise, uncertain localization, and incomplete semantic perception. We propose a practical, implementation-friendly planner that operates on occupancy grids and optionally incorporates occupancy-flow predictions to generate ego-centric, kinematically feasible paths that safely navigate through static and dynamic obstacles. The core is a nonlinear program in the spatial domain built on a modified bicycle model with explicit feasibility and collision-avoidance penalties. The formulation naturally handles unknown obstacle classes and heterogeneous agent motion by operating purely in occupancy space. The pipeline runs in real-time (faster than 10 Hz on average), requires minimal tuning, and interfaces cleanly with standard control stacks. We validate our approach in simulation with severe localization and perception noises, and on an F1TENTH platform, demonstrating smooth and safe maneuvering through narrow passages and rough routes. The approach provides a robust foundation for noise-resilient, prediction-aware planning, eliminating the need for handcrafted heuristics. The project website can be accessed at https://honda-research-institute.github.io/onrap/
comment: 8 pages, 9 figures - Presented at 2026 IEEE Intelligent Vehicles Symposium (IV)
☆ From Snapshot Sensing to Persistent EM World Modeling: A Generative-Space Perspective for ISAC
Electromagnetic (EM) world modeling is emerging as a foundational capability for environment-aware and embodiment-enabled wireless systems. However, most existing mmWave sensing solutions are designed for snapshot-based parameter estimation and rely on hardware-intensive architectures, making scalable and persistent world modeling difficult to achieve. This article rethinks mmWave sensing from a system-level perspective and introduces a generative-space framework, in which sensing is realized through controlled traversal of a low-dimensional excitation space spanning frequency, waveform, and physical embodiment. This perspective decouples spatial observability from rigid antenna arrays and transmit-time multiplexing, enabling flexible and scalable sensing-by-design radios. To illustrate the practicality of this framework, we present a representative realization called Multi-RF Chain Frequency-as-Aperture Clip-on Aperture Fabric (MRC-FaA-CAF), where multiple FMCW sources coordinate frequency-selective modules distributed along guided-wave backbones. This architecture enables interference-free excitation, preserves beat-frequency separability, and maintains low calibration overhead. Case studies show that generative-space-driven sensing can achieve update rates comparable to phased arrays while avoiding dense RF replication and the latency penalties of TDM-MIMO systems. Overall, this work positions generative-space-driven sensing as a practical architectural foundation for mmWave systems that move beyond snapshot sensing toward persistent EM world modeling.
comment: 7 pages, 6 figures/tables
♻ ☆ On the Surprising Effectiveness of Spectral Clipping in Learning Stable Linear and Latent-Linear Dynamical Systems
When learning stable linear dynamical systems from data, three important properties are desirable: i) predictive accuracy, ii) verifiable stability, and iii) computational efficiency. Unconstrained minimization of prediction errors leads to high accuracy and efficiency but cannot guarantee stability. Existing methods to enforce stability often preserve accuracy, but do so only at the cost of increased computation. In this work, we investigate if a seemingly-naive procedure can simultaneously offer all three desiderata. Specifically, we consider a post-hoc procedure in which we surgically manipulate the spectrum of the linear system after it was learned using unconstrained least squares. We call this approach spectral clipping (SC) as it involves eigen decomposition and subsequent reconstruction of the system matrix after any eigenvalues whose magnitude exceeds one have been clipped to one (without altering the eigenvectors). We also show that SC can be readily combined with Koopman operators to learn nonlinear dynamical systems that can generate stable predictions of nonlinear phenomena, such as those underlying complex dexterous manipulation skills involving multi-fingered robotic hands. Through comprehensive experiments involving two different applications and publicly available benchmark datasets, we show that this simple technique can efficiently learn highly-accurate predictive dynamics that are provably-stable. Notably, we find that SC can match or outperform strong baselines while being orders-of-magnitude faster. Finally, we find that SC can learn stable robot policies even when the training data includes unsuccessful or truncated demonstrations. Our code and datasets can be found at https://github.com/GT-STAR-Lab/spec_clip.
♻ ☆ Grounding Bodily Awareness in Visual Representations for Efficient Policy Learning
Learning effective visual representations for robotic manipulation remains a fundamental challenge due to the complex body dynamics involved in action execution. In this paper, we study how visual representations that carry body-relevant cues can enable efficient policy learning for downstream robotic manipulation tasks. We present $\textbf{I}$nter-token $\textbf{Con}$trast ($\textbf{ICon}$), a contrastive learning method applied to the token-level representations of Vision Transformers (ViTs). ICon enforces a separation in the feature space between agent-specific and environment-specific tokens, resulting in agent-centric visual representations that embed body-specific inductive biases. This framework can be seamlessly integrated into end-to-end policy learning by incorporating the contrastive loss as an auxiliary objective. Our experiments show that ICon not only improves policy performance across various manipulation tasks but also facilitates policy transfer across different robots. The project website: https://inter-token-contrast.github.io/icon/
comment: A preprint version
♻ ☆ ReaDy-Go: Real-to-Sim Dynamic 3D Gaussian Splatting Simulation for Environment-Specific Visual Navigation with Moving Obstacles
Visual navigation models often struggle in real-world dynamic environments due to limited robustness to the sim-to-real gap and the difficulty of training policies tailored to target deployment environments (e.g., households, restaurants, and factories). Although real-to-sim navigation simulation using 3D Gaussian Splatting (GS) can mitigate these challenges, prior GS-based works have considered only static scenes or non-photorealistic human obstacles built from simulator assets, despite the importance of safe navigation in dynamic environments. To address these issues, we propose ReaDy-Go, a novel real-to-sim simulation pipeline that synthesizes photorealistic dynamic scenarios in target environments by augmenting a reconstructed static GS scene with dynamic human GS obstacles, and trains navigation policies using the generated datasets. The pipeline provides three key contributions: (1) a dynamic GS simulator that integrates static scene GS with a human animation module, enabling the insertion of animatable human GS avatars and the synthesis of plausible human motions from 2D trajectories, (2) a navigation dataset generation framework that leverages the simulator along with a robot expert planner designed for dynamic GS representations and a human planner, and (3) robust navigation policies to both the sim-to-real gap and moving obstacles. The proposed simulator generates thousands of photorealistic navigation scenarios with animatable human GS avatars from arbitrary viewpoints. ReaDy-Go outperforms baselines across target environments in both simulation and real-world experiments, demonstrating improved navigation performance even after sim-to-real transfer and in the presence of moving obstacles. Moreover, zero-shot sim-to-real deployment in an unseen environment indicates its generalization potential. Project page: https://syeon-yoo.github.io/ready-go-site/.
comment: Project page: https://syeon-yoo.github.io/ready-go-site/
♻ ☆ VividFace: Real-Time and Realistic Facial Expression Shadowing for Humanoid Robots ICRA
Humanoid facial expression shadowing enables robots to realistically imitate human facial expressions in real time, which is critical for lifelike, facially expressive humanoid robots and affective human-robot interaction. Existing progress in humanoid facial expression imitation remains limited, often failing to achieve either real-time performance or realistic expressiveness due to offline video-based inference designs and insufficient ability to capture and transfer subtle expression details. To address these limitations, we present VividFace, a real-time and realistic facial expression shadowing system for humanoid robots. An optimized imitation framework X2CNet++ enhances expressiveness by fine-tuning the human-to-humanoid facial motion transfer module and introducing a feature-adaptation training strategy for better alignment across different image sources. Real-time shadowing is further enabled by a video-stream-compatible inference pipeline and a streamlined workflow based on asynchronous I/O for efficient communication across devices. VividFace produces vivid humanoid faces by mimicking human facial expressions within 0.05 seconds, while generalizing across diverse facial configurations. Extensive real-world demonstrations validate its practical utility. Videos are available at: https://lipzh5.github.io/VividFace/.
comment: Accepted to the 2026 IEEE International Conference on Robotics and Automation (ICRA)
♻ ☆ A Pragmatist Robot: Learning to Plan Tasks by Experiencing the Real World
Large language models (LLMs) have emerged as the dominant paradigm for robotic task planning using natural language instructions. However, trained on general internet data, LLMs are not inherently aligned with the embodiment, skill sets, and limitations of real-world robotic systems. Inspired by the emerging paradigm of verbal reinforcement learning-where LLM agents improve through self-reflection and few-shot learning without parameter updates-we introduce PragmaBot, a framework that enables robots to learn task planning through real-world experience. PragmaBot employs a vision-language model (VLM) as the robot's "brain" and "eye", allowing it to visually evaluate action outcomes and self-reflect on failures. These reflections are stored in a short-term memory (STM), enabling the robot to quickly adapt its behavior during ongoing tasks. Upon task completion, the robot summarizes the lessons learned into its long-term memory (LTM). When facing new tasks, it can leverage retrieval-augmented generation (RAG) to plan more grounded action sequences by drawing on relevant past experiences and knowledge. Experiments on four challenging robotic tasks show that STM-based self-reflection increases task success rates from 35% to 84%, with emergent intelligent object interactions. In 12 real-world scenarios (including eight previously unseen tasks), the robot effectively learns from the LTM and improves single-trial success rates from 22% to 80%, with RAG outperforming naive prompting. These results highlight the effectiveness and generalizability of PragmaBot. Project webpage: https://pragmabot.github.io/
comment: Accepted to RA-L
♻ ☆ SkillWrapper: Generative Predicate Invention for Task-level Planning
Generalizing from individual skill executions to solving long-horizon tasks remains a core challenge in building autonomous agents. A promising direction is learning high-level, symbolic abstractions of the low-level skills of the agents, enabling reasoning and planning independent of the low-level state space. Among possible high-level representations, object-centric skill abstraction with symbolic predicates has been proven to be efficient because of its compatibility with domain-independent planners. Recent advances in foundation models have made it possible to generate symbolic predicates that operate on raw sensory inputs, a process we call generative predicate invention, to facilitate downstream abstraction learning. However, it remains unclear which formal properties the learned representations must satisfy, and how they can be learned to guarantee these properties. In this paper, we address both questions by presenting a formal theory of generative predicate invention for skill abstraction, resulting in symbolic operators that can be used for provably sound and complete planning. Within this framework, we propose SkillWrapper, a method that leverages foundation models to actively collect robot data and learn human-interpretable, plannable representations of black-box skills, using only RGB image observations. Our extensive empirical evaluation in simulation and on real robots shows that SkillWrapper learns abstract representations that enable solving unseen, long-horizon tasks in the real world with black-box skills.
♻ ☆ Integrated Exploration and Sequential Manipulation on Scene Graph with LLM-based Situated Replanning ICRA 2026
In partially known environments, robots must combine exploration to gather information with task planning for efficient execution. To address this challenge, we propose EPoG, an Exploration-based sequential manipulation Planning framework on Scene Graphs. EPoG integrates a graph-based global planner with a Large Language Model (LLM)-based situated local planner, continuously updating a belief graph using observations and LLM predictions to represent known and unknown objects. Action sequences are generated by computing graph edit operations between the goal and belief graphs, ordered by temporal dependencies and movement costs. This approach seamlessly combines exploration and sequential manipulation planning. In ablation studies across 46 realistic household scenes and 5 long-horizon daily object transportation tasks, EPoG achieved a success rate of 91.3%, reducing travel distance by 36.1% on average. Furthermore, a physical mobile manipulator successfully executed complex tasks in unknown and dynamic environments, demonstrating EPoG's potential for real-world applications.
comment: 8 pages, 7 figures; accepted by ICRA 2026
♻ ☆ Adaptive Time Step Flow Matching for Autonomous Driving Motion Planning
Autonomous driving requires reasoning about interactions with surrounding traffic. A prevailing approach is large-scale imitation learning on expert driving datasets, aimed at generalizing across diverse real-world scenarios. For online trajectory generation, such methods must operate at real-time rates. Diffusion models require hundreds of denoising steps at inference, resulting in high latency. Consistency models mitigate this issue but rely on carefully tuned noise schedules to capture the multimodal action distributions common in autonomous driving. Adapting the schedule, typically requires expensive retraining. To address these limitations, we propose a framework based on conditional flow matching that jointly predicts future motions of surrounding agents and plans the ego trajectory in real time. We train a lightweight variance estimator that selects the number of inference steps online, removing the need for retraining to balance runtime and imitation learning performance. To further enhance ride quality, we introduce a trajectory post-processing step cast as a convex quadratic program, with negligible computational overhead. Trained on the Waymo Open Motion Dataset, the framework performs maneuvers such as lane changes, cruise control, and navigating unprotected left turns without requiring scenario-specific tuning. Our method maintains a 20 Hz update rate on an NVIDIA RTX 3070 GPU, making it suitable for online deployment. Compared to transformer, diffusion, and consistency model baselines, we achieve improved trajectory smoothness and better adherence to dynamic constraints. Experiment videos and code implementations can be found at https://flow-matching-self-driving.github.io/.
comment: Accepted to Intelligent Vehicles Symposium 2026
♻ ☆ Efficient Long-Horizon Vision-Language-Action Models via Static-Dynamic Disentanglement
Vision-Language-Action (VLA) models have recently emerged as a promising paradigm for generalist robotic control. Built upon vision-language model (VLM) architectures, VLAs predict actions conditioned on visual observations and language instructions, achieving strong performance and generalization across tasks. However, VLAs face two major challenges: limited long-horizon context and inefficient inference due to the quadratic attention complexity and large parameter counts. Our work is motivated by the observation that much of the visual information in a trajectory remains static across timesteps (e.g., the background). Leveraging this property, we propose SD-VLA, a framework that disentangles visual inputs into multi-level static and dynamic tokens, which enables (1) retaining a single copy of static tokens across frames to significantly reduce context length, and (2) reusing the key-value (KV) cache of static tokens through a lightweight recache gate that updates only when necessary. This design enables efficient multi-frame integration and efficient inference. In addition, we introduce a new benchmark that more effectively evaluates the long-horizon temporal dependency modeling ability of VLAs. Experimental results show that our approach outperforms baselines on this benchmark by 39.8% absolute improvement in success rate, and achieves a 3.9% gain on the SimplerEnv benchmark. Moreover, SD-VLA delivers a 2.26x inference speedup over the base VLA model on the same benchmark, enabling faster and more practical real-world deployment.
♻ ☆ VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model
Pretraining Vision-Language-Action (VLA) policies on internet-scale video is appealing, yet current latent-action objectives often learn the wrong thing: they remain anchored to pixel variation rather than action-relevant state transitions, making them vulnerable to appearance bias, nuisance motion, and information leakage. We introduce VLA-JEPA, a JEPA-style pretraining framework that sidesteps these pitfalls by design. The key idea is leakage-free state prediction: a target encoder produces latent representations from future frames, while the student pathway sees only the current observation -- future information is used solely as supervision targets, never as input. By predicting in latent space rather than pixel space, VLA-JEPA learns dynamics abstractions that are robust to camera motion and irrelevant background changes. This yields a simple two-stage recipe -- JEPA pretraining followed by action-head fine-tuning -- without the multi-stage complexity of prior latent-action pipelines. Experiments on LIBERO, LIBERO-Plus, SimplerEnv and real-world manipulation tasks show that VLA-JEPA achieves consistent gains in generalization and robustness over existing methods.
Robotics 57
☆ Imitating What Works: Simulation-Filtered Modular Policy Learning from Human Videos
The ability to learn manipulation skills by watching videos of humans has the potential to unlock a new source of highly scalable data for robot learning. Here, we tackle prehensile manipulation, in which tasks involve grasping an object before performing various post-grasp motions. Human videos offer strong signals for learning the post-grasp motions, but they are less useful for learning the prerequisite grasping behaviors, especially for robots without human-like hands. A promising way forward is to use a modular policy design, leveraging a dedicated grasp generator to produce stable grasps. However, arbitrary stable grasps are often not task-compatible, hindering the robot's ability to perform the desired downstream motion. To address this challenge, we present Perceive-Simulate-Imitate (PSI), a framework for training a modular manipulation policy using human video motion data processed by paired grasp-trajectory filtering in simulation. This simulation step extends the trajectory data with grasp suitability labels, which allows for supervised learning of task-oriented grasping capabilities. We show through real-world experiments that our framework can be used to learn precise manipulation skills efficiently without any robot data, resulting in significantly more robust performance than using a grasp generator naively.
☆ Steerable Vision-Language-Action Policies for Embodied Reasoning and Hierarchical Control
Pretrained vision-language models (VLMs) can make semantic and visual inferences across diverse settings, providing valuable common-sense priors for robotic control. However, effectively grounding this knowledge in robot behaviors remains an open challenge. Prior methods often employ a hierarchical approach where VLMs reason over high-level commands to be executed by separate low-level policies, e.g., vision-language-action models (VLAs). The interface between VLMs and VLAs is usually natural language task instructions, which fundamentally limits how much VLM reasoning can steer low-level behavior. We thus introduce Steerable Policies: VLAs trained on rich synthetic commands at various levels of abstraction, like subtasks, motions, and grounded pixel coordinates. By improving low-level controllability, Steerable Policies can unlock pretrained knowledge in VLMs, enabling improved task generalization. We demonstrate this benefit by controlling our Steerable Policies with both a learned high-level embodied reasoner and an off-the-shelf VLM prompted to reason over command abstractions via in-context learning. Across extensive real-world manipulation experiments, these two novel methods outperform prior embodied reasoning VLAs and VLM-based hierarchical baselines, including on challenging generalization and long-horizon tasks. Website: steerable-policies.github.io
☆ Human Emotion-Mediated Soft Robotic Arts: Exploring the Intersection of Human Emotions, Soft Robotics and Arts
Soft robotics has emerged as a versatile field with applications across various domains, from healthcare to industrial automation, and more recently, art and interactive installations. The inherent flexibility, adaptability, and safety of soft robots make them ideal for applications that require delicate, organic, and lifelike movement, allowing for immersive and responsive interactions. This study explores the intersection of human emotions, soft robotics, and art to establish and create new forms of human emotion-mediated soft robotic art. In this paper, we introduce two soft embodiments: a soft character and a soft flower as an art display that dynamically responds to brain signals based on alpha waves, reflecting different emotion levels. We present how human emotions can be measured as alpha waves based on brain/EEG signals, how we map the alpha waves to the dynamic movements of the two soft embodiments, and demonstrate our proposed concept using experiments. The findings of this study highlight how soft robotics can embody human emotional states, offering a new medium for insightful artistic expression and interaction, and demonstrating how art displays can be embodied.
☆ Temporally-Sampled Efficiently Adaptive State Lattices for Autonomous Ground Robot Navigation in Partially Observed Environments
Due to sensor limitations, environments that off-road mobile robots operate in are often only partially observable. As the robots move throughout the environment and towards their goal, the optimal route is continuously revised as the sensors perceive new information. In traditional autonomous navigation architectures, a regional motion planner will consume the environment map and output a trajectory for the local motion planner to use as a reference. Due to the continuous revision of the regional plan guidance as a result of changing map information, the reference trajectories which are passed down to the local planner can differ significantly across sequential planning cycles. This rapidly changing guidance can result in unsafe navigation behavior, often requiring manual safety interventions during autonomous traversals in off-road environments. To remedy this problem, we propose Temporally-Sampled Efficiently Adaptive State Lattices (TSEASL), which is a regional planner arbitration architecture that considers updated and optimized versions of previously generated trajectories against the currently generated trajectory. When tested on a Clearpath Robotics Warthog Unmanned Ground Vehicle as well as real map data collected from the Warthog, results indicate that when running TSEASL, the robot did not require manual interventions in the same locations where the robot was running the baseline planner. Additionally, higher levels of planner stability were recorded with TSEASL over the baseline. The paper concludes with a discussion of further improvements to TSEASL in order to make it more generalizable to various off-road autonomy scenarios.
comment: 12 pages, 8 figures
☆ A Data-Driven Algorithm for Model-Free Control Synthesis
Presented is an algorithm to synthesize the optimal infinite-horizon LQR feedback controller for continuous-time systems. The algorithm does not require knowledge of the system dynamics but instead uses only a finite-length sampling of arbitrary input-output data. The algorithm is based on a constrained optimization problem that enforces a necessary condition on the dynamics of the optimal value function along any trajectory. In addition to calculating the standard LQR gain matrix, a feedforward gain can be found to implement a reference tracking controller. This paper presents a theoretical justification for the method and shows several examples, including a validation test on a real scale aircraft.
☆ UniManip: General-Purpose Zero-Shot Robotic Manipulation with Agentic Operational Graph
Achieving general-purpose robotic manipulation requires robots to seamlessly bridge high-level semantic intent with low-level physical interaction in unstructured environments. However, existing approaches falter in zero-shot generalization: end-to-end Vision-Language-Action (VLA) models often lack the precision required for long-horizon tasks, while traditional hierarchical planners suffer from semantic rigidity when facing open-world variations. To address this, we present UniManip, a framework grounded in a Bi-level Agentic Operational Graph (AOG) that unifies semantic reasoning and physical grounding. By coupling a high-level Agentic Layer for task orchestration with a low-level Scene Layer for dynamic state representation, the system continuously aligns abstract planning with geometric constraints, enabling robust zero-shot execution. Unlike static pipelines, UniManip operates as a dynamic agentic loop: it actively instantiates object-centric scene graphs from unstructured perception, parameterizes these representations into collision-free trajectories via a safety-aware local planner, and exploits structured memory to autonomously diagnose and recover from execution failures. Extensive experiments validate the system's robust zero-shot capability on unseen objects and tasks, demonstrating a 22.5% and 25.0% higher success rate compared to state-of-the-art VLA and hierarchical baselines, respectively. Notably, the system enables direct zero-shot transfer from fixed-base setups to mobile manipulation without fine-tuning or reconfiguration. Our open-source project page can be found at https://henryhcliu.github.io/unimanip.
comment: 15 pages, 12 figures, 6 tables, project page: https://henryhcliu.github.io/unimanip
☆ Agentic AI for Robot Control: Flexible but still Fragile
Recent work leverages the capabilities and commonsense priors of generative models for robot control. In this paper, we present an agentic control system in which a reasoning-capable language model plans and executes tasks by selecting and invoking robot skills within an iterative planner and executor loop. We deploy the system on two physical robot platforms in two settings: (i) tabletop grasping, placement, and box insertion in indoor mobile manipulation (Mobipick) and (ii) autonomous agricultural navigation and sensing (Valdemar). Both settings involve uncertainty, partial observability, sensor noise, and ambiguous natural-language commands. The system exposes structured introspection of its planning and decision process, reacts to exogenous events via explicit event checks, and supports operator interventions that modify or redirect ongoing execution. Across both platforms, our proof-of-concept experiments reveal substantial fragility, including non-deterministic suboptimal behavior, instruction-following errors, and high sensitivity to prompt specification. At the same time, the architecture is flexible: transfer to a different robot and task domain largely required updating the system prompt (domain model, affordances, and action catalogue) and re-binding the same tool interface to the platform-specific skill API.
☆ SENSE-STEP: Learning Sim-to-Real Locomotion for a Sensory-Enabled Soft Quadruped Robot
Robust closed-loop locomotion remains challenging for soft quadruped robots due to high-dimensional dynamics, actuator hysteresis, and difficult-to-model contact interactions, while conventional proprioception provides limited information about ground contact. In this paper, we present a learning-based control framework for a pneumatically actuated soft quadruped equipped with tactile suction-cup feet, and we validate the approach experimentally on physical hardware. The control policy is trained in simulation through a staged learning process that starts from a reference gait and is progressively refined under randomized environmental conditions. The resulting controller maps proprioceptive and tactile feedback to coordinated pneumatic actuation and suction-cup commands, enabling closed-loop locomotion on flat and inclined surfaces. When deployed on the real robot, the closed-loop policy outperforms an open-loop baseline, increasing forward speed by 41% on a flat surface and by 91% on a 5-degree incline. Ablation studies further demonstrate the role of tactile force estimates and inertial feedback in stabilizing locomotion, with performance improvements of up to 56% compared to configurations without sensory feedback.
☆ How Swarms Differ: Challenges in Collective Behaviour Comparison
Collective behaviours often need to be expressed through numerical features, e.g., for classification or imitation learning. This problem is often addressed by proposing an ad-hoc feature set for a particular swarm behaviour context, usually without further consideration of the solution's resilience outside of the conceived context. Yet, the development of automatic methods to design swarm behaviours is dependent on the ability to measure quantitatively the similarity of swarm behaviours. Hence, we investigate the impact of feature sets for collective behaviours. We select swarm feature sets and similarity measures from prior swarm robotics works, which mainly considered a narrow behavioural context and assess their robustness. We demonstrate that the interplay of feature set and similarity measure makes some combinations more suitable to distinguish groups of similar behaviours. We also propose a self-organised map-based approach to identify regions of the feature space where behaviours cannot be easily distinguished.
comment: Accepted for publication in the proceeding of ANTS 2026 - 15th International Conference on Swarm Intelligence
☆ Learning Native Continuation for Action Chunking Flow Policies
Action chunking enables Vision Language Action (VLA) models to run in real time, but naive chunked execution often exhibits discontinuities at chunk boundaries. Real-Time Chunking (RTC) alleviates this issue but is external to the policy, leading to spurious multimodal switching and trajectories that are not intrinsically smooth. We propose Legato, a training-time continuation method for action-chunked flow-based VLA policies. Specifically, Legato initializes denoising from a schedule-shaped mixture of known actions and noise, exposing the model to partial action information. Moreover, Legato reshapes the learned flow dynamics to ensure that the denoising process remains consistent between training and inference under per-step guidance. Legato further uses randomized schedule condition during training to support varying inference delays and achieve controllable smoothness. Empirically, Legato produces smoother trajectories and reduces spurious multimodal switching during execution, leading to less hesitation and shorter task completion time. Extensive real-world experiments show that Legato consistently outperforms RTC across five manipulation tasks, achieving approximately 10% improvements in both trajectory smoothness and task completion time.
comment: Project page: https://lyfeng001.github.io/Legato/
☆ INHerit-SG: Incremental Hierarchical Semantic Scene Graphs with RAG-Style Retrieval
Driven by advancements in foundation models, semantic scene graphs have emerged as a prominent paradigm for high-level 3D environmental abstraction in robot navigation. However, existing approaches are fundamentally misaligned with the needs of embodied tasks. As they rely on either offline batch processing or implicit feature embeddings, the maps can hardly support interpretable human-intent reasoning in complex environments. To address these limitations, we present INHerit-SG. We redefine the map as a structured, RAG-ready knowledge base where natural-language descriptions are introduced as explicit semantic anchors to better align with human intent. An asynchronous dual-process architecture, together with a Floor-Room-Area-Object hierarchy, decouples geometric segmentation from time-consuming semantic reasoning. An event-triggered map update mechanism reorganizes the graph only when meaningful semantic events occur. This strategy enables our graph to maintain long-term consistency with relatively low computational overhead. For retrieval, we deploy multi-role Large Language Models (LLMs) to decompose queries into atomic constraints and handle logical negations, and employ a hard-to-soft filtering strategy to ensure robust reasoning. This explicit interpretability improves the success rate and reliability of complex retrievals, enabling the system to adapt to a broader spectrum of human interaction tasks. We evaluate INHerit-SG on a newly constructed dataset, HM3DSem-SQR, and in real-world environments. Experiments demonstrate that our system achieves state-of-the-art performance on complex queries, and reveal its scalability for downstream navigation tasks. Project Page: https://fangyuktung.github.io/INHeritSG.github.io/
☆ Adding internal audio sensing to internal vision enables human-like in-hand fabric recognition with soft robotic fingertips
Distinguishing the feel of smooth silk from coarse cotton is a trivial everyday task for humans. When exploring such fabrics, fingertip skin senses both spatio-temporal force patterns and texture-induced vibrations that are integrated to form a haptic representation of the explored material. It is challenging to reproduce this rich, dynamic perceptual capability in robots because tactile sensors typically cannot achieve both high spatial resolution and high temporal sampling rate. In this work, we present a system that can sense both types of haptic information, and we investigate how each type influences robotic tactile perception of fabrics. Our robotic hand's middle finger and thumb each feature a soft tactile sensor: one is the open-source Minsight sensor that uses an internal camera to measure fingertip deformation and force at 50 Hz, and the other is our new sensor Minsound that captures vibrations through an internal MEMS microphone with a bandwidth from 50 Hz to 15 kHz. Inspired by the movements humans make to evaluate fabrics, our robot actively encloses and rubs folded fabric samples between its two sensitive fingers. Our results test the influence of each sensing modality on overall classification performance, showing high utility for the audio-based sensor. Our transformer-based method achieves a maximum fabric classification accuracy of 97 % on a dataset of 20 common fabrics. Incorporating an external microphone away from Minsound increases our method's robustness in loud ambient noise conditions. To show that this audio-visual tactile sensing approach generalizes beyond the training data, we learn general representations of fabric stretchiness, thickness, and roughness.
☆ SKYSURF: A Self-learning Framework for Persistent Surveillance using Cooperative Aerial Gliders
The success of surveillance applications involving small unmanned aerial vehicles (UAVs) depends on how long the limited on-board power would persist. To cope with this challenge, alternative renewable sources of lift are sought. One promising solution is to extract energy from rising masses of buoyant air. This paper proposes a local-global behavioral management and decision-making approach for the autonomous deployment of soaring-capable UAVs. The cooperative UAVs are modeled as non-deterministic finite state-based rational agents. In addition to a mission planning module for assigning tasks and issuing dynamic navigation waypoints for a new path planning scheme, in which the concepts of visibility and prediction are applied to avoid the collisions. Moreover, a delayed learning and tuning strategy is employed optimize the gains of the path tracking controller. Rigorous comparative analyses carried out with three benchmarking baselines and 15 evolutionary algorithms highlight the adequacy of the proposed approach for maintaining the surveillance persistency (staying aloft for longer periods without landing) and maximizing the detection of targets (two times better than non-cooperative and semi-cooperative approaches) with less power consumption (almost 6% of battery consumed in six hours).
☆ SafeFlowMPC: Predictive and Safe Trajectory Planning for Robot Manipulators with Learning-based Policies ICRA 2026
The emerging integration of robots into everyday life brings several major challenges. Compared to classical industrial applications, more flexibility is needed in combination with real-time reactivity. Learning-based methods can train powerful policies based on demonstrated trajectories, such that the robot generalizes a task to similar situations. However, these black-box models lack interpretability and rigorous safety guarantees. Optimization-based methods provide these guarantees but lack the required flexibility and generalization capabilities. This work proposes SafeFlowMPC, a combination of flow matching and online optimization to combine the strengths of learning and optimization. This method guarantees safety at all times and is designed to meet the demands of real-time execution by using a suboptimal model-predictive control formulation. SafeFlowMPC achieves strong performance in three real-world experiments on a KUKA 7-DoF manipulator, namely two grasping experiment and a dynamic human-robot object handover experiment. A video of the experiments is available at http://www.acin.tuwien.ac.at/42d6. The code is available at https://github.com/TU-Wien-ACIN-CDS/SafeFlowMPC.
comment: Accepted at ICRA 2026
☆ Media Framing Moderates Risk-Benefit Perceptions and Value Tradeoffs in Human-Robot Collaboration
Public acceptance of industrial human-robot collaboration (HRC) is shaped by how risks and benefits are perceived by affected employees. Positive or negative media framing may shape and shift how individuals evaluate HRC. This study examines how message framing moderates the effects of perceived risks and perceived benefits on overall attributed value. In a pre-registered study, participants (N = 1150) were randomly assigned to read either a positively or negatively framed newspaper article in one of three industrial contexts (autonomy, employment, safety) about HRC in production. Subsequently, perceived risks, benefits, and value were measured using reliable and publicly available psychometric scales. Two multiple regressions (one per framing condition) tested for main and interaction effects. Framing influenced absolute evaluations of risk, benefits, and value. In both frames, risks and benefits significantly predicted attributed value. Under positive framing, only main effects were observed (risks: beta = -0.52; benefits: beta = 0.45). Under negative framing, both predictors had stronger main effects (risks: beta = -0.69; benefits: beta = 0.63) along with a significant negative interaction (beta = -0.32), indicating that higher perceived risk diminishes the positive effect of perceived benefits. Model fit was higher for the positive frame (R^2 = 0.715) than for the negative frame (R^2 = 0.583), indicating greater explained variance in value attributions. Framing shapes the absolute evaluation of HRC and how risks and benefits are cognitively integrated in trade-offs. Negative framing produces stronger but interdependent effects, whereas positive framing supports additive evaluations. These findings highlight the role of strategic communication in fostering acceptance of HRC and underscore the need to consider framing in future HRC research.
☆ Scaling Single Human Demonstrations for Imitation Learning using Generative Foundational Models ICRA 2026
Imitation learning is a popular paradigm to teach robots new tasks, but collecting robot demonstrations through teleoperation or kinesthetic teaching is tedious and time-consuming. In contrast, directly demonstrating a task using our human embodiment is much easier and data is available in abundance, yet transfer to the robot can be non-trivial. In this work, we propose Real2Gen to train a manipulation policy from a single human demonstration. Real2Gen extracts required information from the demonstration and transfers it to a simulation environment, where a programmable expert agent can demonstrate the task arbitrarily many times, generating an unlimited amount of data to train a flow matching policy. We evaluate Real2Gen on human demonstrations from three different real-world tasks and compare it to a recent baseline. Real2Gen shows an average increase in the success rate of 26.6% and better generalization of the trained policy due to the abundance and diversity of training data. We further deploy our purely simulation-trained policy zero-shot in the real world. We make the data, code, and trained models publicly available at real2gen.cs.uni-freiburg.de.
comment: ICRA 2026, 8 pages, 6 figures, 4 tables
☆ TRANS: Terrain-aware Reinforcement Learning for Agile Navigation of Quadruped Robots under Social Interactions
This study introduces TRANS: Terrain-aware Reinforcement learning for Agile Navigation under Social interactions, a deep reinforcement learning (DRL) framework for quadrupedal social navigation over unstructured terrains. Conventional quadrupedal navigation typically separates motion planning from locomotion control, neglecting whole-body constraints and terrain awareness. On the other hand, end-to-end methods are more integrated but require high-frequency sensing, which is often noisy and computationally costly. In addition, most existing approaches assume static environments, limiting their use in human-populated settings. To address these limitations, we propose a two-stage training framework with three DRL pipelines. (1) TRANS-Loco employs an asymmetric actor-critic (AC) model for quadrupedal locomotion, enabling traversal of uneven terrains without explicit terrain or contact observations. (2) TRANS-Nav applies a symmetric AC framework for social navigation, directly mapping transformed LiDAR data to ego-agent actions under differential-drive kinematics. (3) A unified pipeline, TRANS, integrates TRANS-Loco and TRANS-Nav, supporting terrain-aware quadrupedal navigation in uneven and socially interactive environments. Comprehensive benchmarks against locomotion and social navigation baselines demonstrate the effectiveness of TRANS. Hardware experiments further confirm its potential for sim-to-real transfer.
☆ Constrained PSO Six-Parameter Fuzzy PID Tuning Method for Balanced Optimization of Depth Tracking Performance in Underwater Vehicles
Depth control of underwater vehicles in engineering applications must simultaneously satisfy requirements for rapid tracking, low overshoot, and actuator constraints. Traditional fuzzy PID tuning often relies on empirical methods, making it difficult to achieve a stable and reproducible equilibrium solution between performance enhancement and control cost. This paper proposes a constrained particle swarm optimization (PSO) method for tuning six-parameter fuzzy PID controllers. By adjusting the benchmark PID parameters alongside the fuzzy controller's input quantization factor and output proportional gain, it achieves synergistic optimization of the overall tuning strength and dynamic response characteristics of the fuzzy PID system. To ensure engineering feasibility of the optimization results, a time-weighted absolute error integral, adjustment time, relative overshoot control energy, and saturation occupancy rate are introduced. Control energy constraints are applied to construct a constraint-driven comprehensive evaluation system, suppressing pseudo-improvements achieved solely by increasing control inputs. Simulation results demonstrate that, while maintaining consistent control energy and saturation levels, the proposed method significantly enhances deep tracking performance: the time-weighted absolute error integral decreases from 0.2631 to 0.1473, the settling time shortens from 2.301 s to 1.613 s, and the relative overshoot reduces from 0.1494 to 0.01839. Control energy varied from 7980 to 7935, satisfying the energy constraint, while saturation occupancy decreased from 0.004 to 0.003. These results validate the effectiveness and engineering significance of the proposed constrained six-parameter joint tuning strategy for depth control in underwater vehicle navigation scenarios.
☆ ALOE: Action-Level Off-Policy Evaluation for Vision-Language-Action Model Post-Training
We study how to improve large foundation vision-language-action (VLA) systems through online reinforcement learning (RL) in real-world settings. Central to this process is the value function, which provides learning signals to guide VLA learning from experience. In practice, the value function is estimated from trajectory fragments collected from different data sources, including historical policies and intermittent human interventions. Estimating the value function of current behavior quality from the mixture data is inherently an off-policy evaluation problem. However, prior work often adopts conservative on-policy estimation for stability, which avoids direct evaluation of the current high-capacity policy and limits learning effectiveness. In this paper, we propose ALOE, an action-level off-policy evaluation framework for VLA post-training. ALOE applies chunking-based temporal-difference bootstrapping to evaluate individual action sequences instead of predicting final task outcomes. This design improves effective credit assignment to critical action chunks under sparse rewards and supports stable policy improvement. We evaluate our method on three real-world manipulation tasks, including smartphone packing as a high-precision task, laundry folding as a long-horizon deformable-object task, and bimanual pick-and-place involving multi-object perception. Across all tasks, ALOE improves learning efficiency without compromising execution speed, showing that off-policy RL can be reintroduced in a reliable manner for real-world VLA post-training. Videos and additional materials are available at our project website.
☆ SignScene: Visual Sign Grounding for Mapless Navigation
Navigational signs enable humans to navigate unfamiliar environments without maps. This work studies how robots can similarly exploit signs for mapless navigation in the open world. A central challenge lies in interpreting signs: real-world signs are diverse and complex, and their abstract semantic contents need to be grounded in the local 3D scene. We formalize this as sign grounding, the problem of mapping semantic instructions on signs to corresponding scene elements and navigational actions. Recent Vision-Language Models (VLMs) offer the semantic common-sense and reasoning capabilities required for this task, but are sensitive to how spatial information is represented. We propose SignScene, a sign-centric spatial-semantic representation that captures navigation-relevant scene elements and sign information, and presents them to VLMs in a form conducive to effective reasoning. We evaluate our grounding approach on a dataset of 114 queries collected across nine diverse environment types, achieving 88% grounding accuracy and significantly outperforming baselines. Finally, we demonstrate that it enables real-world mapless navigation on a Spot robot using only signs.
comment: Under review for a conference
☆ Xiaomi-Robotics-0: An Open-Sourced Vision-Language-Action Model with Real-Time Execution
In this report, we introduce Xiaomi-Robotics-0, an advanced vision-language-action (VLA) model optimized for high performance and fast and smooth real-time execution. The key to our method lies in a carefully designed training recipe and deployment strategy. Xiaomi-Robotics-0 is first pre-trained on large-scale cross-embodiment robot trajectories and vision-language data, endowing it with broad and generalizable action-generation capabilities while avoiding catastrophic forgetting of the visual-semantic knowledge of the underlying pre-trained VLM. During post-training, we propose several techniques for training the VLA model for asynchronous execution to address the inference latency during real-robot rollouts. During deployment, we carefully align the timesteps of consecutive predicted action chunks to ensure continuous and seamless real-time rollouts. We evaluate Xiaomi-Robotics-0 extensively in simulation benchmarks and on two challenging real-robot tasks that require precise and dexterous bimanual manipulation. Results show that our method achieves state-of-the-art performance across all simulation benchmarks. Moreover, Xiaomi-Robotics-0 can roll out fast and smoothly on real robots using a consumer-grade GPU, achieving high success rates and throughput on both real-robot tasks. To facilitate future research, code and model checkpoints are open-sourced at https://xiaomi-robotics-0.github.io
comment: Project page: https://xiaomi-robotics-0.github.io
☆ PMG: Parameterized Motion Generator for Human-like Locomotion Control
Recent advances in data-driven reinforcement learning and motion tracking have substantially improved humanoid locomotion, yet critical practical challenges remain. In particular, while low-level motion tracking and trajectory-following controllers are mature, whole-body reference-guided methods are difficult to adapt to higher-level command interfaces and diverse task contexts: they require large, high-quality datasets, are brittle across speed and pose regimes, and are sensitive to robot-specific calibration. To address these limitations, we propose the Parameterized Motion Generator (PMG), a real-time motion generator grounded in an analysis of human motion structure that synthesizes reference trajectories using only a compact set of parameterized motion data together with High-dimensional control commands. Combined with an imitation-learning pipeline and an optimization-based sim-to-real motor parameter identification module, we validate the complete approach on our humanoid prototype ZERITH Z1 and show that, within a single integrated system, PMG produces natural, human-like locomotion, responds precisely to high-dimensional control inputs-including VR-based teleoperation-and enables efficient, verifiable sim-to-real transfer. Together, these results establish a practical, experimentally validated pathway toward natural and deployable humanoid control.
comment: 2026 IEEE International Conference on Robotics & Automation
☆ Dual-Granularity Contrastive Reward via Generated Episodic Guidance for Efficient Embodied RL
Designing suitable rewards poses a significant challenge in reinforcement learning (RL), especially for embodied manipulation. Trajectory success rewards are suitable for human judges or model fitting, but the sparsity severely limits RL sample efficiency. While recent methods have effectively improved RL via dense rewards, they rely heavily on high-quality human-annotated data or abundant expert supervision. To tackle these issues, this paper proposes Dual-granularity contrastive reward via generated Episodic Guidance (DEG), a novel framework to seek sample-efficient dense rewards without requiring human annotations or extensive supervision. Leveraging the prior knowledge of large video generation models, DEG only needs a small number of expert videos for domain adaptation to generate dedicated task guidance for each RL episode. Then, the proposed dual-granularity reward that balances coarse-grained exploration and fine-grained matching, will guide the agent to efficiently approximate the generated guidance video sequentially in the contrastive self-supervised latent space, and finally complete the target task. Extensive experiments on 18 diverse tasks across both simulation and real-world settings show that DEG can not only serve as an efficient exploration stimulus to help the agent quickly discover sparse success rewards, but also guide effective RL and stable policy convergence independently.
☆ Real-to-Sim for Highly Cluttered Environments via Physics-Consistent Inter-Object Reasoning
Reconstructing physically valid 3D scenes from single-view observations is a prerequisite for bridging the gap between visual perception and robotic control. However, in scenarios requiring precise contact reasoning, such as robotic manipulation in highly cluttered environments, geometric fidelity alone is insufficient. Standard perception pipelines often neglect physical constraints, resulting in invalid states, e.g., floating objects or severe inter-penetration, rendering downstream simulation unreliable. To address these limitations, we propose a novel physics-constrained Real-to-Sim pipeline that reconstructs physically consistent 3D scenes from single-view RGB-D data. Central to our approach is a differentiable optimization pipeline that explicitly models spatial dependencies via a contact graph, jointly refining object poses and physical properties through differentiable rigid-body simulation. Extensive evaluations in both simulation and real-world settings demonstrate that our reconstructed scenes achieve high physical fidelity and faithfully replicate real-world contact dynamics, enabling stable and reliable contact-rich manipulation.
comment: Project page: https://physics-constrained-real2sim.github.io
☆ RLinf-Co: 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.
comment: 9 pages, 12 figures. Supplementary material included. Submitted to RSS 2026
☆ When Environments Shift: Safe Planning with Generative Priors and Robust Conformal Prediction
Autonomous systems operate in environments that may change over time. An example is the control of a self-driving vehicle among pedestrians and human-controlled vehicles whose behavior may change based on factors such as traffic density, road visibility, and social norms. Therefore, the environment encountered during deployment rarely mirrors the environment and data encountered during training -- a phenomenon known as distribution shift -- which can undermine the safety of autonomous systems. Conformal prediction (CP) has recently been used along with data from the training environment to provide prediction regions that capture the behavior of the environment with a desired probability. When embedded within a model predictive controller (MPC), one can provide probabilistic safety guarantees, but only when the deployment and training environments coincide. Once a distribution shift occurs, these guarantees collapse. We propose a planning framework that is robust under distribution shifts by: (i) assuming that the underlying data distribution of the environment is parameterized by a nuisance parameter, i.e., an observable, interpretable quantity such as traffic density, (ii) training a conditional diffusion model that captures distribution shifts as a function of the nuisance parameter, (iii) observing the nuisance parameter online and generating cheap, synthetic data from the diffusion model for the observed nuisance parameter, and (iv) designing an MPC that embeds CP regions constructed from such synthetic data. Importantly, we account for discrepancies between the underlying data distribution and the diffusion model by using robust CP. Thus, the plans computed using robust CP enjoy probabilistic safety guarantees, in contrast with plans obtained from a single, static set of training data. We empirically demonstrate safety under diverse distribution shifts in the ORCA simulator.
☆ PISHYAR: A Socially Intelligent Smart Cane for Indoor Social Navigation and Multimodal Human-Robot Interaction for Visually Impaired People
This paper presents PISHYAR, a socially intelligent smart cane designed by our group to combine socially aware navigation with multimodal human-AI interaction to support both physical mobility and interactive assistance. The system consists of two components: (1) a social navigation framework implemented on a Raspberry Pi 5 that integrates real-time RGB-D perception using an OAK-D Lite camera, YOLOv8-based object detection, COMPOSER-based collective activity recognition, D* Lite dynamic path planning, and haptic feedback via vibration motors for tasks such as locating a vacant seat; and (2) an agentic multimodal LLM-VLM interaction framework that integrates speech recognition, vision language models, large language models, and text-to-speech, with dynamic routing between voice-only and vision-only modes to enable natural voice-based communication, scene description, and object localization from visual input. The system is evaluated through a combination of simulation-based tests, real-world field experiments, and user-centered studies. Results from simulated and real indoor environments demonstrate reliable obstacle avoidance and socially compliant navigation, achieving an overall system accuracy of approximately 80% under different social conditions. Group activity recognition further shows robust performance across diverse crowd scenarios. In addition, a preliminary exploratory user study with eight visually impaired and low-vision participants evaluates the agentic interaction framework through structured tasks and a UTAUT-based questionnaire reveals high acceptance and positive perceptions of usability, trust, and perceived sociability during our experiments. The results highlight the potential of PISHYAR as a multimodal assistive mobility aid that extends beyond navigation to provide socially interactive support for such users.
☆ Hemispherical Angular Power Mapping of Installed mmWave Radar Modules Under Realistic Deployment Constraints
Characterizing the angular radiation behavior of installed millimeter-wave (mmWave) radar modules is increasingly important in practical sensing platforms, where packaging, mounting hardware, and nearby structures can significantly alter the effective emission profile. However, once a device is embedded in its host environment, conventional chamber- and turntable-based antenna measurements are often impractical. This paper presents a hemispherical angular received-power mapping methodology for in-situ EM validation of installed mmWave modules under realistic deployment constraints. The approach samples the accessible half-space around a stationary device-under-test by placing a calibrated receiving probe at prescribed (phi, theta, r) locations using geometry-consistent positioning and quasi-static acquisition. Amplitude-only received-power is recorded using standard RF instrumentation to generate hemispherical angular power maps that capture installation-dependent radiation characteristics. Proof-of-concept measurements on a 60-GHz radar module demonstrate repeatable hemi-spherical mapping with angular trends in good agreement with full-wave simulation, supporting practical on-site characterization of embedded mmWave transmitters.
☆ Eva-Tracker: ESDF-update-free, Visibility-aware Planning with Target Reacquisition for Robust Aerial Tracking ICRA 2026
The Euclidean Signed Distance Field (ESDF) is widely used in visibility evaluation to prevent occlusions and collisions during tracking. However, frequent ESDF updates introduce considerable computational overhead. To address this issue, we propose Eva-Tracker, a visibility-aware trajectory planning framework for aerial tracking that eliminates ESDF updates and incorporates a recovery-capable path generation method for target reacquisition. First, we design a target trajectory prediction method and a visibility-aware initial path generation algorithm that maintain an appropriate observation distance, avoid occlusions, and enable rapid replanning to reacquire the target when it is lost. Then, we propose the Field of View ESDF (FoV-ESDF), a precomputed ESDF tailored to the tracker's field of view, enabling rapid visibility evaluation without requiring updates. Finally, we optimize the trajectory using differentiable FoV-ESDF-based objectives to ensure continuous visibility throughout the tracking process. Extensive simulations and real-world experiments demonstrate that our approach delivers more robust tracking results with lower computational effort than existing state-of-the-art methods. The source code is available at https://github.com/Yue-0/Eva-Tracker.
comment: Accepted by ICRA 2026
Self-Supervised JEPA-based World Models for LiDAR Occupancy Completion and Forecasting
Autonomous driving, as an agent operating in the physical world, requires the fundamental capability to build \textit{world models} that capture how the environment evolves spatiotemporally in order to support long-term planning. At the same time, scalability demands learning such models in a self-supervised manner; \textit{joint-embedding predictive architecture (JEPA)} enables learning world models via leveraging large volumes of unlabeled data without relying on expensive human annotations. In this paper, we propose \textbf{AD-LiST-JEPA}, a self-supervised world model for autonomous driving that predicts future spatiotemporal evolution from LiDAR data using a JEPA framework. We evaluate the quality of the learned representations through a downstream LiDAR-based occupancy completion and forecasting (OCF) task, which jointly assesses perception and prediction. Proof of concept experiments show better OCF performance with pretrained encoder after JEPA-based world model learning.
☆ CRAFT: Adapting VLA Models to Contact-rich Manipulation via Force-aware Curriculum Fine-tuning
Vision-Language-Action (VLA) models have shown a strong capability in enabling robots to execute general instructions, yet they struggle with contact-rich manipulation tasks, where success requires precise alignment, stable contact maintenance, and effective handling of deformable objects. A fundamental challenge arises from the imbalance between high-entropy vision and language inputs and low-entropy but critical force signals, which often leads to over-reliance on perception and unstable control. To address this, we introduce CRAFT, a force-aware curriculum fine-tuning framework that integrates a variational information bottleneck module to regulate vision and language embeddings during early training. This curriculum strategy encourages the model to prioritize force signals initially, before progressively restoring access to the full multimodal information. To enable force-aware learning, we further design a homologous leader-follower teleoperation system that collects synchronized vision, language, and force data across diverse contact-rich tasks. Real-world experiments demonstrate that CRAFT consistently improves task success, generalizes to unseen objects and novel task variations, and adapts effectively across diverse VLA architectures, enabling robust and generalizable contact-rich manipulation.
☆ Monocular Reconstruction of Neural Tactile Fields
Robots operating in the real world must plan through environments that deform, yield, and reconfigure under contact, requiring interaction-aware 3D representations that extend beyond static geometric occupancy. To address this, we introduce neural tactile fields, a novel 3D representation that maps spatial locations to the expected tactile response upon contact. Our model predicts these neural tactile fields from a single monocular RGB image -- the first method to do so. When integrated with off-the-shelf path planners, neural tactile fields enable robots to generate paths that avoid high-resistance objects while deliberately routing through low-resistance regions (e.g. foliage), rather than treating all occupied space as equally impassable. Empirically, our learning framework improves volumetric 3D reconstruction by $85.8\%$ and surface reconstruction by $26.7\%$ compared to state-of-the-art monocular 3D reconstruction methods (LRM and Direct3D).
comment: 10 pages, 8 figures
☆ Composable Model-Free RL for Navigation with Input-Affine Systems
As autonomous robots move into complex, dynamic real-world environments, they must learn to navigate safely in real time, yet anticipating all possible behaviors is infeasible. We propose a composable, model-free reinforcement learning method that learns a value function and an optimal policy for each individual environment element (e.g., goal or obstacle) and composes them online to achieve goal reaching and collision avoidance. Assuming unknown nonlinear dynamics that evolve in continuous time and are input-affine, we derive a continuous-time Hamilton-Jacobi-Bellman (HJB) equation for the value function and show that the corresponding advantage function is quadratic in the action and optimal policy. Based on this structure, we introduce a model-free actor-critic algorithm that learns policies and value functions for static or moving obstacles using gradient descent. We then compose multiple reach/avoid models via a quadratically constrained quadratic program (QCQP), yielding formal obstacle-avoidance guarantees in terms of value-function level sets, providing a model-free alternative to CLF/CBF-based controllers. Simulations demonstrate improved performance over a PPO baseline applied to a discrete-time approximation.
comment: 17 pages, 8 figures. Submitted to WAFR 2026 (under review)
☆ Gradient-Enhanced Partitioned Gaussian Processes for Real-Time Quadrotor Dynamics Modeling
We present a quadrotor dynamics Gaussian Process (GP) with gradient information that achieves real-time inference via state-space partitioning and approximation, and that includes aerodynamic effects using data from mid-fidelity potential flow simulations. While traditional GP-based approaches provide reliable Bayesian predictions with uncertainty quantification, they are computationally expensive and thus unsuitable for real-time simulations. To address this challenge, we integrate gradient information to improve accuracy and introduce a novel partitioning and approximation strategy to reduce online computational cost. In particular, for the latter, we associate a local GP with each non-overlapping region; by splitting the training data into local near and far subsets, and by using Schur complements, we show that a large part of the matrix inversions required for inference can be performed offline, enabling real-time inference at frequencies above 30 Hz on standard desktop hardware. To generate a training dataset that captures aerodynamic effects, such as rotor-rotor interactions and apparent wind direction, we use the CHARM code, which is a mid-fidelity aerodynamic solver. It is applied to the SUI Endurance quadrotor to predict force and torque, along with noise at three specified locations. The derivative information is obtained via finite differences. Experimental results demonstrate that the proposed partitioned GP with gradient conditioning achieves higher accuracy than standard partitioned GPs without gradient information, while greatly reducing computational time. This framework provides an efficient foundation for real-time aerodynamic prediction and control algorithms in complex and unsteady environments.
comment: 11 pages, 7 figures. Submitted to IEEE Transactions on Robotics (under review)
☆ AsyncVLA: An Asynchronous VLA for Fast and Robust Navigation on the Edge
Robotic foundation models achieve strong generalization by leveraging internet-scale vision-language representations, but their massive computational cost creates a fundamental bottleneck: high inference latency. In dynamic environments, this latency breaks the control loop, rendering powerful models unsafe for real-time deployment. We propose AsyncVLA, an asynchronous control framework that decouples semantic reasoning from reactive execution. Inspired by hierarchical control, AsyncVLA runs a large foundation model on a remote workstation to provide high-level guidance, while a lightweight, onboard Edge Adapter continuously refines actions at high frequency. To bridge the domain gap between these asynchronous streams, we introduce an end-to-end finetuning protocol and a trajectory re-weighting strategy that prioritizes dynamic interactions. We evaluate our approach on real-world vision-based navigation tasks with communication delays up to 6 seconds. AsyncVLA achieves a 40% higher success rate than state-of-the-art baselines, effectively bridging the gap between the semantic intelligence of large models and the reactivity required for edge robotics.
comment: 13 pages, 9 figures, 2 tables
☆ Inferring Turn-Rate-Limited Engagement Zones with Sacrificial Agents for Safe Trajectory Planning
This paper presents a learning-based framework for estimating pursuer parameters in turn-rate-limited pursuit-evasion scenarios using sacrificial agents. Each sacrificial agent follows a straight-line trajectory toward an adversary and reports whether it was intercepted or survived. These binary outcomes are related to the pursuer's parameters through a geometric reachable-region (RR) model. Two formulations are introduced: a boundary-interception case, where capture occurs at the RR boundary, and an interior-interception case, which allows capture anywhere within it. The pursuer's parameters are inferred using a gradient-based multi-start optimization with custom loss functions tailored to each case. Two trajectory-selection strategies are proposed for the sacrificial agents: a geometric heuristic that maximizes the spread of expected interception points, and a Bayesian experimental-design method that maximizes the D-score of the expected Gauss-Newton information matrix, thereby selecting trajectories that yield maximal information gain. Monte Carlo experiments demonstrate accurate parameter recovery with five to twelve sacrificial agents. The learned engagement models are then used to generate safe, time-optimal paths for high-value agents that avoid all feasible pursuer engagement regions.
comment: Submitted to the Journal of Aerospace Information Systems
☆ FlowHOI: Flow-based Semantics-Grounded Generation of Hand-Object Interactions for Dexterous Robot Manipulation
Recent vision-language-action (VLA) models can generate plausible end-effector motions, yet they often fail in long-horizon, contact-rich tasks because the underlying hand-object interaction (HOI) structure is not explicitly represented. An embodiment-agnostic interaction representation that captures this structure would make manipulation behaviors easier to validate and transfer across robots. We propose FlowHOI, a two-stage flow-matching framework that generates semantically grounded, temporally coherent HOI sequences, comprising hand poses, object poses, and hand-object contact states, conditioned on an egocentric observation, a language instruction, and a 3D Gaussian splatting (3DGS) scene reconstruction. We decouple geometry-centric grasping from semantics-centric manipulation, conditioning the latter on compact 3D scene tokens and employing a motion-text alignment loss to semantically ground the generated interactions in both the physical scene layout and the language instruction. To address the scarcity of high-fidelity HOI supervision, we introduce a reconstruction pipeline that recovers aligned hand-object trajectories and meshes from large-scale egocentric videos, yielding an HOI prior for robust generation. Across the GRAB and HOT3D benchmarks, FlowHOI achieves the highest action recognition accuracy and a 1.7$\times$ higher physics simulation success rate than the strongest diffusion-based baseline, while delivering a 40$\times$ inference speedup. We further demonstrate real-robot execution on four dexterous manipulation tasks, illustrating the feasibility of retargeting generated HOI representations to real-robot execution pipelines.
comment: Project Page: https://huajian-zeng.github.io/projects/flowhoi/
☆ Learning on the Fly: Replay-Based Continual Object Perception for Indoor Drones
Autonomous agents such as indoor drones must learn new object classes in real-time while limiting catastrophic forgetting, motivating Class-Incremental Learning (CIL). However, most unmanned aerial vehicle (UAV) datasets focus on outdoor scenes and offer limited temporally coherent indoor videos. We introduce an indoor dataset of $14,400$ frames capturing inter-drone and ground vehicle footage, annotated via a semi-automatic workflow with a $98.6\%$ first-pass labeling agreement before final manual verification. Using this dataset, we benchmark 3 replay-based CIL strategies: Experience Replay (ER), Maximally Interfered Retrieval (MIR), and Forgetting-Aware Replay (FAR), using YOLOv11-nano as a resource-efficient detector for deployment-constrained UAV platforms. Under tight memory budgets ($5-10\%$ replay), FAR performs better than the rest, achieving an average accuracy (ACC, $mAP_{50-95}$ across increments) of $82.96\%$ with $5\%$ replay. Gradient-weighted class activation mapping (Grad-CAM) analysis shows attention shifts across classes in mixed scenes, which is associated with reduced localization quality for drones. The experiments further demonstrate that replay-based continual learning can be effectively applied to edge aerial systems. Overall, this work contributes an indoor UAV video dataset with preserved temporal coherence and an evaluation of replay-based CIL under limited replay budgets. Project page: https://spacetime-vision-robotics-laboratory.github.io/learning-on-the-fly-cl
comment: Accepted at European Robotics Forum (ERF) 2026
☆ High-Fidelity, Customizable Force Sensing for the Wearable Human-Robot Interface
Mechanically characterizing the human-machine interface is essential to understanding user behavior and optimizing wearable robot performance. This interface has been challenging to sensorize due to manufacturing complexity and non-linear sensor responses. Here, we measure human limb-device interaction via fluidic innervation, creating a 3D-printed silicone pad with embedded air channels to measure forces. As forces are applied to the pad, the air channels compress, resulting in a pressure change measurable by off-the-shelf pressure transducers. We demonstrate in benchtop testing that pad pressure is highly linearly related to applied force ($R^2 = 0.998$). This is confirmed with clinical dynamometer correlations with isometric knee torque, where above-knee pressure was highly correlated with flexion torque ($R^2 = 0.95$), while below-knee pressure was highly correlated with extension torque ($R^2 = 0.75$). We build on these idealized settings to test pad performance in more unconstrained settings. We place the pad over \textit{biceps brachii} during cyclic curls and stepwise isometric holds, observing a correlation between pressure and elbow angle. Finally, we integrated the sensor into the strap of a lower-extremity robotic exoskeleton and recorded pad pressure during repeated squats with the device unpowered. Pad pressure tracked squat phase and overall task dynamics consistently. Overall, our preliminary results suggest fluidic innervation is a readily customizable sensing modality with high signal-to-noise ratio and temporal resolution for capturing human-machine mechanical interaction. In the long-term, this modality may provide an alternative real-time sensing input to control / optimize wearable robotic systems and to capture user function during device use.
comment: 6 pages, 7 figures, submitted to BioRob 2026
☆ RynnBrain: Open Embodied Foundation Models
Despite rapid progress in multimodal foundation models, embodied intelligence community still lacks a unified, physically grounded foundation model that integrates perception, reasoning, and planning within real-world spatial-temporal dynamics. We introduce RynnBrain, an open-source spatiotemporal foundation model for embodied intelligence. RynnBrain strengthens four core capabilities in a unified framework: comprehensive egocentric understanding, diverse spatiotemporal localization, physically grounded reasoning, and physics-aware planning. The RynnBrain family comprises three foundation model scales (2B, 8B, and 30B-A3B MoE) and four post-trained variants tailored for downstream embodied tasks (i.e., RynnBrain-Nav, RynnBrain-Plan, and RynnBrain-VLA) or complex spatial reasoning tasks (i.e., RynnBrain-CoP). In terms of extensive evaluations on 20 embodied benchmarks and 8 general vision understanding benchmarks, our RynnBrain foundation models largely outperform existing embodied foundation models by a significant margin. The post-trained model suite further substantiates two key potentials of the RynnBrain foundation model: (i) enabling physically grounded reasoning and planning, and (ii) serving as a strong pretrained backbone that can be efficiently adapted to diverse embodied tasks.
comment: Homepage: https://alibaba-damo-academy.github.io/RynnBrain.github.io
☆ Location as a service with a MEC architecture
In recent years, automated driving has become viable, and advanced driver assistance systems (ADAS) are now part of modern cars. These systems require highly precise positioning. In this paper, a cooperative approach to localization is presented. The GPS information from several road users is collected in a Mobile Edge Computing cloud, and the characteristics of GNSS positioning are used to provide lane-precise positioning for all participants by applying probabilistic filters and HD maps.
comment: Published and presented at 2024 International Conference on Information Networking (ICOIN)
♻ ☆ Palpation Alters Auditory Pain Expressions with Gender-Specific Variations in Robopatients
Diagnostic errors remain a major cause of preventable mortality, particularly in resource limited settings. Medical training simulators, including robopatients, help reduce such errors by replicating patient responses during procedures such as abdominal palpation. However, generating realistic multimodal feedback especially auditory pain expressions remains challenging due to the complex, nonlinear relationship between applied palpation forces and perceived pain sounds. The high dimensionality and perceptual variability of pain vocalizations further limit conventional modeling approaches. We propose a novel experimental paradigm for adaptive pain expressivity in robopatients that dynamically generates auditory pain responses to palpation forces using human in the loop machine learning. Specifically, we employ Proximal Policy Optimization (PPO), a reinforcement learning algorithm suited for continuous control, to iteratively refine pain sound generation based on real time human evaluative feedback. The system initializes randomized mappings between force inputs and sound outputs, and the learning agent progressively adjusts them to align with human perceptual preferences. Results show that the framework adapts to individual palpation behaviors and subjective sound preferences while capturing a broad range of perceived pain intensities, from mild discomfort to acute distress. We also observe perceptual saturation at lower force ranges, with gender specific thresholds in pain sound perception. This work demonstrates the feasibility of human in the loop reinforcement learning for co-optimizing haptic input and auditory pain expression in medical simulators, highlighting the potential of adaptive and immersive platforms to enhance palpation training and reduce diagnostic errors.
comment: 12 pages, 9 figures, journal
♻ ☆ Auditory-Tactile Congruence for Synthesis of Adaptive Pain Expressions in RoboPatients
Misdiagnosis can result in delayed treatment and patient harm. Robotic patient simulators (robopatients) provide a controlled framework for training and evaluating clinicians in rare and complex cases. We investigate auditory tactile congruence in the synthesis of adaptive vocal pain expressions for robopatients. The system generates pain vocalizations in response to tactile stimuli applied during abdominal palpation. Haptic input is captured through an abdominal phantom and processed using an internal palpation-to-pain mapping model that drives acoustic output. To evaluate perceptual congruence between palpation force and synthesized pain expressions, we conducted a study comprising 7,680 trials with 20 participants. Participants rated perceived pain intensity based solely on auditory feedback. We analyzed the influence of acoustic parameters on agreement between applied force and perceived pain. Results indicate that amplitude and pitch significantly affect perceptual agreement, independent of pain sound category. Increased palpation force was associated with higher agreement ratings, consistent with psychophysical scaling effects. Among the tested acoustic features, pitch exerted a stronger influence than amplitude on perceived congruence. These findings demonstrate that modulation of pitch and amplitude is critical for achieving perceptually coherent auditory tactile mappings in robotic pain simulation. The proposed framework supports the development of high fidelity robotic patient simulators and provides a platform for studying multidimensional representations of pain in embodied robotic systems.
comment: 20 pages, 8 figures, journal
♻ ☆ Sim2real Image Translation Enables Viewpoint-Robust Policies from Fixed-Camera Datasets
Vision-based policies for robot manipulation have achieved significant recent success, but are still brittle to distribution shifts such as camera viewpoint variations. Robot demonstration data is scarce and often lacks appropriate variation in camera viewpoints. Simulation offers a way to collect robot demonstrations at scale with comprehensive coverage of different viewpoints, but presents a visual sim2real challenge. To bridge this gap, we propose MANGO -- an unpaired image translation method with a novel segmentation-conditioned InfoNCE loss, a highly-regularized discriminator design, and a modified PatchNCE loss. We find that these elements are crucial for maintaining viewpoint consistency during sim2real translation. When training MANGO, we only require a small amount of fixed-camera data from the real world, but show that our method can generate diverse unseen viewpoints by translating simulated observations. In this setting, MANGO outperforms all other image translation methods we tested. In certain real-world tabletop manipulation tasks, MANGO augmentation increases shifted-view success rates by over 40 percentage points compared to policies trained without augmentation.
♻ ☆ Hallucinating 360°: Panoramic Street-View Generation via Local Scenes Diffusion and Probabilistic Prompting ICRA 2026
Panoramic perception holds significant potential for autonomous driving, enabling vehicles to acquire a comprehensive 360° surround view in a single shot. However, autonomous driving is a data-driven task. Complete panoramic data acquisition requires complex sampling systems and annotation pipelines, which are time-consuming and labor-intensive. Although existing street view generation models have demonstrated strong data regeneration capabilities, they can only learn from the fixed data distribution of existing datasets and cannot leverage stitched pinhole images as a supervisory signal. In this paper, we propose the first panoramic generation method Percep360 for autonomous driving. Percep360 enables coherent generation of panoramic data with control signals based on the stitched panoramic data. Percep360 focuses on two key aspects: coherence and controllability. Specifically, to overcome the inherent information loss caused by the pinhole sampling process, we propose the Local Scenes Diffusion Method (LSDM). LSDM reformulates the panorama generation as a spatially continuous diffusion process, bridging the gaps between different data distributions. Additionally, to achieve the controllable generation of panoramic images, we propose a Probabilistic Prompting Method (PPM). PPM dynamically selects the most relevant control cues, enabling controllable panoramic image generation. We evaluate the effectiveness of the generated images from three perspectives: image quality assessment (i.e., no-reference and with reference), controllability, and their utility in real-world Bird's Eye View (BEV) segmentation. Notably, the generated data consistently outperforms the original stitched images in no-reference quality metrics and enhances downstream perception models. The source code will be publicly available at https://github.com/FeiT-FeiTeng/Percep360.
comment: Accepted to ICRA 2026. The source code will be publicly available at https://github.com/FeiT-FeiTeng/Percep360
♻ ☆ Robust Convex Model Predictive Control with collision avoidance guarantees for robot manipulators
Industrial manipulators are normally operated in cluttered environments, making safe motion planning important. Furthermore, the presence of model-uncertainties make safe motion planning more difficult. Therefore, in practice the speed is limited in order to reduce the effect of disturbances. There is a need for control methods that can guarantee safe motions that can be executed fast. We address this need by suggesting a novel model predictive control (MPC) solution for manipulators, where our two main components are a robust tube MPC and a corridor planning algorithm to obtain collision-free motion. Our solution results in a convex MPC, which we can solve fast, making our method practically useful. We demonstrate the efficacy of our method in a simulated environment with a 6 DOF industrial robot operating in cluttered environments with uncertainties in model parameters. We outperform benchmark methods, both in terms of being able to work under higher levels of model uncertainties, while also yielding faster motion.
♻ ☆ Effective Task Planning with Missing Objects using Learning-Informed Object Search
Task planning for mobile robots often assumes full environment knowledge and so popular approaches, like planning via the PDDL, cannot plan when the locations of task-critical objects are unknown. Recent learning-driven object search approaches are effective, but operate as standalone tools and so are not straightforwardly incorporated into full task planners, which must additionally determine both what objects are necessary and when in the plan they should be sought out. To address this limitation, we develop a planning framework centered around novel model-based LIOS actions: each a policy that aims to find and retrieve a single object. High-level planning treats LIOS actions as deterministic and so -- informed by model-based calculations of the expected cost of each -- generates plans that interleave search and execution for effective, sound, and complete learning-informed task planning despite uncertainty. Our work effectively reasons about uncertainty while maintaining compatibility with existing full-knowledge solvers. In simulated ProcTHOR homes and in the real world, our approach outperforms non-learned and learned baselines on tasks including retrieval and meal prep.
♻ ☆ PA-MPPI: Perception-Aware Model Predictive Path Integral Control for Quadrotor Navigation in Unknown Environments
Quadrotor navigation in unknown environments is critical for practical missions such as search-and-rescue. Solving this problem requires addressing three key challenges: path planning in non-convex free space due to obstacles, satisfying quadrotor-specific dynamics and objectives, and exploring unknown regions to expand the map. Recently, the Model Predictive Path Integral (MPPI) method has emerged as a promising solution to the first two challenges. By leveraging sampling-based optimization, it can effectively handle non-convex free space while directly optimizing over the full quadrotor dynamics, enabling the inclusion of quadrotor-specific costs such as energy consumption. However, MPPI has been limited to tracking control that optimizes trajectories only within a small neighborhood around a reference trajectory, as it lacks the ability to explore unknown regions and plan alternative paths when blocked by large obstacles. To address this limitation, we introduce Perception-Aware MPPI (PA-MPPI). In this approach, perception-awareness is characterized by planning and adapting the trajectory online based on perception objectives. Specifically, when the goal is occluded, PA-MPPI incorporates a perception cost that biases trajectories toward those that can observe unknown regions. This expands the mapped traversable space and increases the likelihood of finding alternative paths to the goal. Through hardware experiments, we demonstrate that PA-MPPI, running at 50 Hz, performs on par with the state-of-the-art quadrotor navigation planner for unknown environments in challenging test scenarios. Furthermore, we show that PA-MPPI can serve as a safe and robust action policy for navigation foundation models, which often provide goal poses that are not directly reachable.
♻ ☆ MeCo: Enhancing LLM-Empowered Multi-Robot Collaboration via Similar Task Memoization
Multi-robot systems have been widely deployed in real-world applications, providing significant improvements in efficiency and reductions in labor costs. However, most existing multi-robot collaboration methods rely on extensive task-specific training, which limits their adaptability to new or diverse scenarios. Recent research leverages the language understanding and reasoning capabilities of large language models (LLMs) to enable more flexible collaboration without specialized training. Yet, current LLM-empowered approaches remain inefficient: when confronted with identical or similar tasks, they must replan from scratch because they omit task-level similarities. To address this limitation, we propose MeCo, a similarity-aware multi-robot collaboration framework that applies the principle of ``cache and reuse'' (a.k.a., memoization) to reduce redundant computation. Unlike simple task repetition, identifying and reusing solutions for similar but not identical tasks is far more challenging, particularly in multi-robot settings. To this end, MeCo introduces a new similarity testing method that retrieves previously solved tasks with high relevance, enabling effective plan reuse without re-invoking LLMs. Furthermore, we present MeCoBench, the first benchmark designed to evaluate performance on similar-task collaboration scenarios. Experimental results show that MeCo substantially reduces planning costs and improves success rates compared with state-of-the-art approaches.
♻ ☆ Assessing Vision-Language Models for Perception in Autonomous Underwater Robotic Software
Autonomous Underwater Robots (AURs) operate in challenging underwater environments, including low visibility and harsh water conditions. Such conditions present challenges for software engineers developing perception modules for the AUR software. To successfully carry out these tasks, deep learning has been incorporated into the AUR software to support its operations. However, the unique challenges of underwater environments pose difficulties for deep learning models, which often rely on labeled data that is scarce and noisy. This may undermine the trustworthiness of AUR software that relies on perception modules. Vision-Language Models (VLMs) offer promising solutions for AUR software as they generalize to unseen objects and remain robust in noisy conditions by inferring information from contextual cues. Despite this potential, their performance and uncertainty in underwater environments remain understudied from a software engineering perspective. Motivated by the needs of an industrial partner in assurance and risk management for maritime systems to assess the potential use of VLMs in this context, we present an empirical evaluation of VLM-based perception modules within the AUR software. We assess their ability to detect underwater trash by computing performance, uncertainty, and their relationship, to enable software engineers to select appropriate VLMs for their AUR software.
comment: 16 pages, 5 figures
♻ ☆ SteerVLA: Steering Vision-Language-Action Models in Long-Tail Driving Scenarios
A fundamental challenge in autonomous driving is the integration of high-level, semantic reasoning for long-tail events with low-level, reactive control for robust driving. While large vision-language models (VLMs) trained on web-scale data offer powerful common-sense reasoning, they lack the grounded experience necessary for safe vehicle control. We posit that an effective autonomous agent should leverage the world knowledge of VLMs to guide a steerable driving policy toward robust control in driving scenarios. To this end, we propose SteerVLA, which leverages the reasoning capabilities of VLMs to produce fine-grained language instructions that steer a vision-language-action (VLA) driving policy. Key to our method is this rich language interface between the high-level VLM and low-level VLA, which allows the high-level policy to more effectively ground its reasoning in the control outputs of the low-level policy. To provide fine-grained language supervision aligned with vehicle control, we leverage a VLM to augment existing driving data with detailed language annotations, which we find to be essential for effective reasoning and steerability. We evaluate SteerVLA on a challenging closed-loop benchmark, where it outperforms state-of-the-art methods by 4.77 points in overall driving score and by 8.04 points on a long-tail subset. The project website is available at: https://steervla.github.io/.
♻ ☆ KAN We Flow? Advancing Robotic Manipulation with 3D Flow Matching via KAN & RWKV ICRA2026
Diffusion-based visuomotor policies excel at modeling action distributions but are inference-inefficient, since recursively denoising from noise to policy requires many steps and heavy UNet backbones, which hinders deployment on resource-constrained robots. Flow matching alleviates the sampling burden by learning a one-step vector field, yet prior implementations still inherit large UNet-style architectures. In this work, we present KAN-We-Flow, a flow-matching policy that draws on recent advances in Receptance Weighted Key Value (RWKV) and Kolmogorov-Arnold Networks (KAN) from vision to build a lightweight and highly expressive backbone for 3D manipulation. Concretely, we introduce an RWKV-KAN block: an RWKV first performs efficient time/channel mixing to propagate task context, and a subsequent GroupKAN layer applies learnable spline-based, groupwise functional mappings to perform feature-wise nonlinear calibration of the action mapping on RWKV outputs. Moreover, we introduce an Action Consistency Regularization (ACR), a lightweight auxiliary loss that enforces alignment between predicted action trajectories and expert demonstrations via Euler extrapolation, providing additional supervision to stabilize training and improve policy precision. Without resorting to large UNets, our design reduces parameters by 86.8\%, maintains fast runtime, and achieves state-of-the-art success rates on Adroit, Meta-World, and DexArt benchmarks. Our project page can be viewed in \href{https://zhihaochen-2003.github.io/KAN-We-Flow.github.io/}{\textcolor{red}{link}}
comment: Accepted By ICRA2026
♻ ☆ Language-in-the-Loop Culvert Inspection on the Erie Canal
Culverts on canals such as the Erie Canal, built originally in 1825, require frequent inspections to ensure safe operation. Human inspection of culverts is challenging due to age, geometry, poor illumination, weather, and lack of easy access. We introduce VISION, an end-to-end, language-in-the-loop autonomy system that couples a web-scale vision-language model (VLM) with constrained viewpoint planning for autonomous inspection of culverts. Brief prompts to the VLM solicit open-vocabulary ROI proposals with rationales and confidences, stereo depth is fused to recover scale, and a planner -- aware of culvert constraints -- commands repositioning moves to capture targeted close-ups. Deployed on a quadruped in a culvert under the Erie Canal, VISION closes the see, decide, move, re-image loop on-board and produces high-resolution images for detailed reporting without domain-specific fine-tuning. In an external evaluation by New York Canal Corporation personnel, initial ROI proposals achieved 61.4\% agreement with subject-matter experts, and final post-re-imaging assessments reached 80\%, indicating that VISION converts tentative hypotheses into grounded, expert-aligned findings.
comment: First two authors contributed equally
♻ ☆ What Matters in Building Vision-Language-Action Models for Generalist Robots
To utilize Foundation Vision Language Models (VLMs) for robotic tasks and motion planning, the community has proposed different methods for injecting action components into VLMs and building the Vision-Language-Action models (VLAs). In this work, we disclose the key factors that significantly influence the performance of VLA on robot manipulation problems and focus on answering three essential design choices: which backbone to select, how to formulate the VLA architectures, and when to add cross-embodiment data. The obtained results convince us firmly to explain why we prefer VLA and develop a new family of VLAs, RoboVLMs, which require very few manual designs and achieve a new state-of-the-art performance in three simulation tasks and real-world experiments. Through our extensive experiments, which include over 8 VLM backbones, 4 policy architectures, and over 600 distinct designed experiments, we provide a detailed guidebook for the future design of VLAs. In addition to the study, the highly flexible RoboVLMs framework, which supports easy integrations of new VLMs and free combinations of various design choices, is made public to facilitate future research. We open-source all details, including codes, models, datasets, and toolkits, along with detailed training and evaluation recipes at: robovlms.github.io.
comment: Project page: robovlms.github.io. Added limitations and future works. Fix categorization
♻ ☆ Towards Bridging the Gap between Large-Scale Pretraining and Efficient Finetuning for Humanoid Control ICLR 2026
Reinforcement learning (RL) is widely used for humanoid control, with on-policy methods such as Proximal Policy Optimization (PPO) enabling robust training via large-scale parallel simulation and, in some cases, zero-shot deployment to real robots. However, the low sample efficiency of on-policy algorithms limits safe adaptation to new environments. Although off-policy RL and model-based RL have shown improved sample efficiency, the gap between large-scale pretraining and efficient finetuning on humanoids still exists. In this paper, we find that off-policy Soft Actor-Critic (SAC), with large-batch update and a high Update-To-Data (UTD) ratio, reliably supports large-scale pretraining of humanoid locomotion policies, achieving zero-shot deployment on real robots. For adaptation, we demonstrate that these SAC-pretrained policies can be finetuned in new environments and out-of-distribution tasks using model-based methods. Data collection in the new environment executes a deterministic policy while stochastic exploration is instead confined to a physics-informed world model. This separation mitigates the risks of random exploration during adaptation while preserving exploratory coverage for improvement. Overall, the approach couples the wall-clock efficiency of large-scale simulation during pretraining with the sample efficiency of model-based learning during fine-tuning.Code and videos: https://lift-humanoid.github.io
comment: ICLR 2026
♻ ☆ HyperDet: 3D Object Detection with Hyper 4D Radar Point Clouds
4D mmWave radar provides weather-robust, velocity-aware measurements and is more cost-effective than LiDAR. However, radar-only 3D detection still trails LiDAR-based systems because radar point clouds are sparse, irregular, and often corrupted by multipath noise, yielding weak and unstable geometry. We present HyperDet, a detector-agnostic radar-only 3D detection framework that constructs a task-aware hyper 4D radar point cloud for standard LiDAR-oriented detectors. HyperDet aggregates returns from multiple surround-view 4D radars over consecutive frames to improve coverage and density, then applies geometry-aware cross-sensor consensus validation with a lightweight self-consistency check outside overlap regions to suppress inconsistent returns. It further integrates a foreground-focused diffusion module with training-time mixed radar-LiDAR supervision to densify object structures while lifting radar attributes (e.g., Doppler, RCS); the model is distilled into a consistency model for single-step inference. On MAN TruckScenes, HyperDet consistently improves over raw radar inputs with VoxelNeXt and CenterPoint, partially narrowing the radar-LiDAR gap. These results show that input-level refinement enables radar to better leverage LiDAR-oriented detectors without architectural modifications.
comment: 9 pages, 4 figures, 6 tables
♻ ☆ InternVLA-A1: Unifying Understanding, Generation and Action for Robotic Manipulation
Prevalent Vision-Language-Action (VLA) models are typically built upon Multimodal Large Language Models (MLLMs) and demonstrate exceptional proficiency in semantic understanding, but they inherently lack the capability to deduce physical world dynamics. Consequently, recent approaches have shifted toward World Models, typically formulated via video prediction; however, these methods often suffer from a lack of semantic grounding and exhibit brittleness in the presence of video prediction errors. To synergize semantic understanding with dynamic predictive capabilities, we present InternVLA-A1. This model employs a unified Mixture-of-Transformers architecture, coordinating three experts for scene understanding, visual foresight generation, and action execution. These components interact seamlessly through a unified masked self attention mechanism. Building upon InternVL3 and Qwen3-VL, we instantiate InternVLA-A1 at 2B and 3B parameter scales. We pre-train these models on heterogeneous data sources over real-world robot data, synthetic simulation data, and human videos, covering over 692M frames. This hybrid training strategy effectively harnesses the diversity of synthetic simulation data while minimizing the sim-to-real gap. We evaluated InternVLA-A1 on 12 real-world robotic tasks and a simulation benchmark. The results show that InternVLA-A1 consistently outperforms prior leading models: compared with pi0.5, it achieves +4.4\% on static manipulation tasks and +2.6\% on the RoboTwin 2.0 simulation benchmark, and delivers a +26.7\% boost on dynamic manipulation tasks.
comment: Homepage: https://internrobotics.github.io/internvla-a1.github.io/
Robotics 70
☆ Scaling Verification Can Be More Effective than Scaling Policy Learning for Vision-Language-Action Alignment
The long-standing vision of general-purpose robots hinges on their ability to understand and act upon natural language instructions. Vision-Language-Action (VLA) models have made remarkable progress toward this goal, yet their generated actions can still misalign with the given instructions. In this paper, we investigate test-time verification as a means to shrink the "intention-action gap.'' We first characterize the test-time scaling law for embodied instruction following and demonstrate that jointly scaling the number of rephrased instructions and generated actions greatly increases test-time sample diversity, often recovering correct actions more efficiently than scaling each dimension independently. To capitalize on these scaling laws, we present CoVer, a contrastive verifier for vision-language-action alignment, and show that our architecture scales gracefully with additional computational resources and data. We then introduce "boot-time compute" and a hierarchical verification inference pipeline for VLAs. At deployment, our framework precomputes a diverse set of rephrased instructions from a Vision-Language-Model (VLM), repeatedly generates action candidates for each instruction, and then uses a verifier to select the optimal high-level prompt and low-level action chunks. Compared to scaling policy pre-training on the same data, our verification approach yields 22% gains in-distribution and 13% out-of-distribution on the SIMPLER benchmark, with a further 45% improvement in real-world experiments. On the PolaRiS benchmark, CoVer achieves 14% gains in task progress and 9% in success rate.
☆ 6G Empowering Future Robotics: A Vision for Next-Generation Autonomous Systems
The convergence of robotics and next-generation communication is a critical driver of technological advancement. As the world transitions from 5G to 6G, the foundational capabilities of wireless networks are evolving to support increasingly complex and autonomous robotic systems. This paper examines the transformative impact of 6G on enhancing key robotics functionalities. It provides a systematic mapping of IMT-2030 key performance indicators to robotic functional blocks including sensing, perception, cognition, actuation and self-learning. Building upon this mapping, we propose a high-level architectural framework integrating robotic, intelligent, and network service planes, underscoring the need for a holistic approach. As an example use case, we present a real-time, dynamic safety framework enabled by IMT-2030 capabilities for safe and efficient human-robot collaboration in shared spaces.
comment: 7 pages, 3 figures, 2 tables, submitted to IEEE magazine publication
☆ Any House Any Task: Scalable Long-Horizon Planning for Abstract Human Tasks
Open world language conditioned task planning is crucial for robots operating in large-scale household environments. While many recent works attempt to address this problem using Large Language Models (LLMs) via prompting or training, a key challenge remains scalability. Performance often degrades rapidly with increasing environment size, plan length, instruction ambiguity, and constraint complexity. In this work, we propose Any House Any Task (AHAT), a household task planner optimized for long-horizon planning in large environments given ambiguous human instructions. At its core, AHAT utilizes an LLM trained to map task instructions and textual scene graphs into grounded subgoals defined in the Planning Domain Definition Language (PDDL). These subgoals are subsequently solved to generate feasible and optimal long-horizon plans through explicit symbolic reasoning. To enhance the model's ability to decompose complex and ambiguous intentions, we introduce TGPO, a novel reinforcement learning algorithm that integrates external correction of intermediate reasoning traces into Group Relative Policy Optimization (GRPO). Experiments demonstrate that AHAT achieves significant performance gains over state-of-the-art prompting, planning, and learning methods, particularly in human-style household tasks characterized by brief instructions but requiring complex execution plans.
☆ LDA-1B: Scaling Latent Dynamics Action Model via Universal Embodied Data Ingestion
Recent robot foundation models largely rely on large-scale behavior cloning, which imitates expert actions but discards transferable dynamics knowledge embedded in heterogeneous embodied data. While the Unified World Model (UWM) formulation has the potential to leverage such diverse data, existing instantiations struggle to scale to foundation-level due to coarse data usage and fragmented datasets. We introduce LDA-1B, a robot foundation model that scales through universal embodied data ingestion by jointly learning dynamics, policy, and visual forecasting, assigning distinct roles to data of varying quality. To support this regime at scale, we assemble and standardize EI-30k, an embodied interaction dataset comprising over 30k hours of human and robot trajectories in a unified format. Scalable dynamics learning over such heterogeneous data is enabled by prediction in a structured DINO latent space, which avoids redundant pixel-space appearance modeling. Complementing this representation, LDA-1B employs a multi-modal diffusion transformer to handle asynchronous vision and action streams, enabling stable training at the 1B-parameter scale. Experiments in simulation and the real world show LDA-1B outperforms prior methods (e.g., $π_{0.5}$) by up to 21\%, 48\%, and 23\% on contact-rich, dexterous, and long-horizon tasks, respectively. Notably, LDA-1B enables data-efficient fine-tuning, gaining 10\% by leveraging 30\% low-quality trajectories typically harmful and discarded.
comment: Project Page:https://pku-epic.github.io/LDA
☆ Sub--Riemannian boundary value problems for Optimal Geometric Locomotion
We propose a geometric model for optimal shape-change-induced motions of slender locomotors, e.g., snakes slithering on sand. In these scenarios, the motion of a body in world coordinates is completely determined by the sequence of shapes it assumes. Specifically, we formulate Lagrangian least-dissipation principles as boundary value problems whose solutions are given by sub-Riemannian geodesics. Notably, our geometric model accounts not only for the energy dissipated by the body's displacement through the environment, but also for the energy dissipated by the animal's metabolism or a robot's actuators to induce shape changes such as bending and stretching, thus capturing overall locomotion efficiency. Our continuous model, together with a consistent time and space discretization, enables numerical computation of sub-Riemannian geodesics for three different types of boundary conditions, i.e., fixing initial and target body, restricting to cyclic motion, or solely prescribing body displacement and orientation. The resulting optimal deformation gaits qualitatively match observed motion trajectories of organisms such as snakes and spermatozoa, as well as known optimality results for low-dimensional systems such as Purcell's swimmers. Moreover, being geometrically less rigid than previous frameworks, our model enables new insights into locomotion mechanisms of, e.g., generalized Purcell's swimmers. The code is publicly available.
☆ 3DGSNav: Enhancing Vision-Language Model Reasoning for Object Navigation via Active 3D Gaussian Splatting
Object navigation is a core capability of embodied intelligence, enabling an agent to locate target objects in unknown environments. Recent advances in vision-language models (VLMs) have facilitated zero-shot object navigation (ZSON). However, existing methods often rely on scene abstractions that convert environments into semantic maps or textual representations, causing high-level decision making to be constrained by the accuracy of low-level perception. In this work, we present 3DGSNav, a novel ZSON framework that embeds 3D Gaussian Splatting (3DGS) as persistent memory for VLMs to enhance spatial reasoning. Through active perception, 3DGSNav incrementally constructs a 3DGS representation of the environment, enabling trajectory-guided free-viewpoint rendering of frontier-aware first-person views. Moreover, we design structured visual prompts and integrate them with Chain-of-Thought (CoT) prompting to further improve VLM reasoning. During navigation, a real-time object detector filters potential targets, while VLM-driven active viewpoint switching performs target re-verification, ensuring efficient and reliable recognition. Extensive evaluations across multiple benchmarks and real-world experiments on a quadruped robot demonstrate that our method achieves robust and competitive performance against state-of-the-art approaches.The Project Page:https://aczheng-cai.github.io/3dgsnav.github.io/
☆ Multi Graph Search for High-Dimensional Robot Motion Planning
Efficient motion planning for high-dimensional robotic systems, such as manipulators and mobile manipulators, is critical for real-time operation and reliable deployment. Although advances in planning algorithms have enhanced scalability to high-dimensional state spaces, these improvements often come at the cost of generating unpredictable, inconsistent motions or requiring excessive computational resources and memory. In this work, we introduce Multi-Graph Search (MGS), a search-based motion planning algorithm that generalizes classical unidirectional and bidirectional search to a multi-graph setting. MGS maintains and incrementally expands multiple implicit graphs over the state space, focusing exploration on high-potential regions while allowing initially disconnected subgraphs to be merged through feasible transitions as the search progresses. We prove that MGS is complete and bounded-suboptimal, and empirically demonstrate its effectiveness on a range of manipulation and mobile manipulation tasks. Demonstrations, benchmarks and code are available at https://multi-graph-search.github.io/.
comment: Submitted for Publication
☆ Pack it in: Packing into Partially Filled Containers Through Contact
The automation of warehouse operations is crucial for improving productivity and reducing human exposure to hazardous environments. One operation frequently performed in warehouses is bin-packing where items need to be placed into containers, either for delivery to a customer, or for temporary storage in the warehouse. Whilst prior bin-packing works have largely been focused on packing items into empty containers and have adopted collision-free strategies, it is often the case that containers will already be partially filled with items, often in suboptimal arrangements due to transportation about a warehouse. This paper presents a contact-aware packing approach that exploits purposeful interactions with previously placed objects to create free space and enable successful placement of new items. This is achieved by using a contact-based multi-object trajectory optimizer within a model predictive controller, integrated with a physics-aware perception system that estimates object poses even during inevitable occlusions, and a method that suggests physically-feasible locations to place the object inside the container.
comment: 8 pages, 5 figures
☆ RF-Modulated Adaptive Communication Improves Multi-Agent Robotic Exploration
Reliable coordination and efficient communication are critical challenges for multi-agent robotic exploration of environments where communication is limited. This work introduces Adaptive-RF Transmission (ART), a novel communication-aware planning algorithm that dynamically modulates transmission location based on signal strength and data payload size, enabling heterogeneous robot teams to share information efficiently without unnecessary backtracking. We further explore an extension to this approach called ART-SST, which enforces signal strength thresholds for high-fidelity data delivery. Through over 480 simulations across three cave-inspired environments, ART consistently outperforms existing strategies, including full rendezvous and minimum-signal heuristic approaches, achieving up to a 58% reduction in distance traveled and up to 52% faster exploration times compared to baseline methods. These results demonstrate that adaptive, payload-aware communication significantly improves coverage efficiency and mission speed in complex, communication-constrained environments, offering a promising foundation for future planetary exploration and search-and-rescue missions.
☆ Affordance-Graphed Task Worlds: Self-Evolving Task Generation for Scalable Embodied Learning
Training robotic policies directly in the real world is expensive and unscalable. Although generative simulation enables large-scale data synthesis, current approaches often fail to generate logically coherent long-horizon tasks and struggle with dynamic physical uncertainties due to open-loop execution. To address these challenges, we propose Affordance-Graphed Task Worlds (AGT-World), a unified framework that autonomously constructs interactive simulated environments and corresponding robot task policies based on real-world observations. Unlike methods relying on random proposals or static replication, AGT-World formalizes the task space as a structured graph, enabling the precise, hierarchical decomposition of complex goals into theoretically grounded atomic primitives. Furthermore, we introduce a Self-Evolution mechanism with hybrid feedback to autonomously refine policies, combining Vision-Language Model reasoning and geometric verification. Extensive experiments demonstrate that our method significantly outperforms in success rates and generalization, achieving a self-improving cycle of proposal, execution, and correction for scalable robot learning.
☆ VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Model
The goal of this paper is to improve the performance and reliability of vision-language-action (VLA) models through iterative online interaction. Since collecting policy rollouts in the real world is expensive, we investigate whether a learned simulator-specifically, an action-conditioned video generation model-can be used to generate additional rollout data. Unfortunately, existing world models lack the physical fidelity necessary for policy improvement: they are predominantly trained on demonstration datasets that lack coverage of many different physical interactions (particularly failure cases) and struggle to accurately model small yet critical physical details in contact-rich object manipulation. We propose a simple iterative improvement algorithm that uses real-world roll-out data to improve the fidelity of the world model, which can then, in turn, be used to generate supplemental synthetic data for improving the VLA model. In our experiments on a real robot, we use this approach to improve the performance of a state-of-the-art VLA model on multiple downstream tasks. We achieve a 39.2% absolute success rate improvement over the base policy and 11.6% improvement from training with the generated synthetic rollouts. Videos can be found at this anonymous website: https://sites.google.com/view/vla-w
comment: 13 pages
☆ HoloBrain-0 Technical Report
In this work, we introduce HoloBrain-0, a comprehensive Vision-Language-Action (VLA) framework that bridges the gap between foundation model research and reliable real-world robot deployment. The core of our system is a novel VLA architecture that explicitly incorporates robot embodiment priors, including multi-view camera parameters and kinematic descriptions (URDF), to enhance 3D spatial reasoning and support diverse embodiments. We validate this design through a scalable ``pre-train then post-train" paradigm, achieving state-of-the-art results on simulation benchmarks such as RoboTwin 2.0, LIBERO, and GenieSim, as well as strong results on challenging long-horizon real-world manipulation tasks. Notably, our efficient 0.2B-parameter variant rivals significantly larger baselines, enabling low-latency on-device deployment. To further accelerate research and practical adoption, we fully open-source the entire HoloBrain ecosystem, which includes: (1) powerful pre-trained VLA foundations; (2) post-trained checkpoints for multiple simulation suites and real-world tasks; and (3) RoboOrchard, a full-stack VLA infrastructure for data curation, model training and deployment. Together with standardized data collection protocols, this release provides the community with a complete, reproducible path toward high-performance robotic manipulation.
comment: 32 pages
☆ Multi UAVs Preflight Planning in a Shared and Dynamic Airspace AAMAS 2026
Preflight planning for large-scale Unmanned Aerial Vehicle (UAV) fleets in dynamic, shared airspace presents significant challenges, including temporal No-Fly Zones (NFZs), heterogeneous vehicle profiles, and strict delivery deadlines. While Multi-Agent Path Finding (MAPF) provides a formal framework, existing methods often lack the scalability and flexibility required for real-world Unmanned Traffic Management (UTM). We propose DTAPP-IICR: a Delivery-Time Aware Prioritized Planning method with Incremental and Iterative Conflict Resolution. Our framework first generates an initial solution by prioritizing missions based on urgency. Secondly, it computes roundtrip trajectories using SFIPP-ST, a novel 4D single-agent planner (Safe Flight Interval Path Planning with Soft and Temporal Constraints). SFIPP-ST handles heterogeneous UAVs, strictly enforces temporal NFZs, and models inter-agent conflicts as soft constraints. Subsequently, an iterative Large Neighborhood Search, guided by a geometric conflict graph, efficiently resolves any residual conflicts. A completeness-preserving directional pruning technique further accelerates the 3D search. On benchmarks with temporal NFZs, DTAPP-IICR achieves near-100% success with fleets of up to 1,000 UAVs and gains up to 50% runtime reduction from pruning, outperforming batch Enhanced Conflict-Based Search in the UTM context. Scaling successfully in realistic city-scale operations where other priority-based methods fail even at moderate deployments, DTAPP-IICR is positioned as a practical and scalable solution for preflight planning in dense, dynamic urban airspace.
comment: AAMAS 2026 accepted paper
☆ Safety Beyond the Training Data: Robust Out-of-Distribution MPC via Conformalized System Level Synthesis
We present a novel framework for robust out-of-distribution planning and control using conformal prediction (CP) and system level synthesis (SLS), addressing the challenge of ensuring safety and robustness when using learned dynamics models beyond the training data distribution. We first derive high-confidence model error bounds using weighted CP with a learned, state-control-dependent covariance model. These bounds are integrated into an SLS-based robust nonlinear model predictive control (MPC) formulation, which performs constraint tightening over the prediction horizon via volume-optimized forward reachable sets. We provide theoretical guarantees on coverage and robustness under distributional drift, and analyze the impact of data density and trajectory tube size on prediction coverage. Empirically, we demonstrate our method on nonlinear systems of increasing complexity, including a 4D car and a {12D} quadcopter, improving safety and robustness compared to fixed-bound and non-robust baselines, especially outside of the data distribution.
☆ When would Vision-Proprioception Policies Fail in Robotic Manipulation? ICLR 2026
Proprioceptive information is critical for precise servo control by providing real-time robotic states. Its collaboration with vision is highly expected to enhance performances of the manipulation policy in complex tasks. However, recent studies have reported inconsistent observations on the generalization of vision-proprioception policies. In this work, we investigate this by conducting temporally controlled experiments. We found that during task sub-phases that robot's motion transitions, which require target localization, the vision modality of the vision-proprioception policy plays a limited role. Further analysis reveals that the policy naturally gravitates toward concise proprioceptive signals that offer faster loss reduction when training, thereby dominating the optimization and suppressing the learning of the visual modality during motion-transition phases. To alleviate this, we propose the Gradient Adjustment with Phase-guidance (GAP) algorithm that adaptively modulates the optimization of proprioception, enabling dynamic collaboration within the vision-proprioception policy. Specifically, we leverage proprioception to capture robotic states and estimate the probability of each timestep in the trajectory belonging to motion-transition phases. During policy learning, we apply fine-grained adjustment that reduces the magnitude of proprioception's gradient based on estimated probabilities, leading to robust and generalizable vision-proprioception policies. The comprehensive experiments demonstrate GAP is applicable in both simulated and real-world environments, across one-arm and dual-arm setups, and compatible with both conventional and Vision-Language-Action models. We believe this work can offer valuable insights into the development of vision-proprioception policies in robotic manipulation.
comment: Accepted by ICLR 2026
☆ Adaptive-Horizon Conflict-Based Search for Closed-Loop Multi-Agent Path Finding
MAPF is a core coordination problem for large robot fleets in automated warehouses and logistics. Existing approaches are typically either open-loop planners, which generate fixed trajectories and struggle to handle disturbances, or closed-loop heuristics without reliable performance guarantees, limiting their use in safety-critical deployments. This paper presents ACCBS, a closed-loop algorithm built on a finite-horizon variant of CBS with a horizon-changing mechanism inspired by iterative deepening in MPC. ACCBS dynamically adjusts the planning horizon based on the available computational budget, and reuses a single constraint tree to enable seamless transitions between horizons. As a result, it produces high-quality feasible solutions quickly while being asymptotically optimal as the budget increases, exhibiting anytime behavior. Extensive case studies demonstrate that ACCBS combines flexibility to disturbances with strong performance guarantees, effectively bridging the gap between theoretical optimality and practical robustness for large-scale robot deployment.
☆ Decentralized Multi-Robot Obstacle Detection and Tracking in a Maritime Scenario
Autonomous aerial-surface robot teams are promising for maritime monitoring. Robust deployment requires reliable perception over reflective water and scalable coordination under limited communication. We present a decentralized multi-robot framework for detecting and tracking floating containers using multiple UAVs cooperating with an autonomous surface vessel. Each UAV performs YOLOv8 and stereo-disparity-based visual detection, then tracks targets with per-object EKFs using uncertainty-aware data association. Compact track summaries are exchanged and fused conservatively via covariance intersection, ensuring consistency under unknown correlations. An information-driven assignment module allocates targets and selects UAV hover viewpoints by trading expected uncertainty reduction against travel effort and safety separation. Simulation results in a maritime scenario demonstrate improved coverage, localization accuracy, and tracking consistency while maintaining modest communication requirements.
comment: 10 pages, 10 figures
☆ 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 severe sample inefficiency. 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 fleet expansion, 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 executable tools, the agent provides precise guidance via corrective waypoints and spatial constraints for exploration pruning. We validate our approach on two 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.
☆ Robot-DIFT: Distilling Diffusion Features for Geometrically Consistent Visuomotor Control
We hypothesize that a key bottleneck in generalizable robot manipulation is not solely data scale or policy capacity, but a structural mismatch between current visual backbones and the physical requirements of closed-loop control. While state-of-the-art vision encoders (including those used in VLAs) optimize for semantic invariance to stabilize classification, manipulation typically demands geometric sensitivity the ability to map millimeter-level pose shifts to predictable feature changes. Their discriminative objective creates a "blind spot" for fine-grained control, whereas generative diffusion models inherently encode geometric dependencies within their latent manifolds, encouraging the preservation of dense multi-scale spatial structure. However, directly deploying stochastic diffusion features for control is hindered by stochastic instability, inference latency, and representation drift during fine-tuning. To bridge this gap, we propose Robot-DIFT, a framework that decouples the source of geometric information from the process of inference via Manifold Distillation. By distilling a frozen diffusion teacher into a deterministic Spatial-Semantic Feature Pyramid Network (S2-FPN), we retain the rich geometric priors of the generative model while ensuring temporal stability, real-time execution, and robustness against drift. Pretrained on the large-scale DROID dataset, Robot-DIFT demonstrates superior geometric consistency and control performance compared to leading discriminative baselines, supporting the view that how a model learns to see dictates how well it can learn to act.
☆ General Humanoid Whole-Body Control via Pretraining and Fast Adaptation
Learning a general whole-body controller for humanoid robots remains challenging due to the diversity of motion distributions, the difficulty of fast adaptation, and the need for robust balance in high-dynamic scenarios. Existing approaches often require task-specific training or suffer from performance degradation when adapting to new motions. In this paper, we present FAST, a general humanoid whole-body control framework that enables Fast Adaptation and Stable Motion Tracking. FAST introduces Parseval-Guided Residual Policy Adaptation, which learns a lightweight delta action policy under orthogonality and KL constraints, enabling efficient adaptation to out-of-distribution motions while mitigating catastrophic forgetting. To further improve physical robustness, we propose Center-of-Mass-Aware Control, which incorporates CoM-related observations and objectives to enhance balance when tracking challenging reference motions. Extensive experiments in simulation and real-world deployment demonstrate that FAST consistently outperforms state-of-the-art baselines in robustness, adaptation efficiency, and generalization.
comment: 22 pages
☆ Data-Driven Trajectory Imputation for Vessel Mobility Analysis EDBT 2026
Modeling vessel activity at sea is critical for a wide range of applications, including route planning, transportation logistics, maritime safety, and environmental monitoring. Over the past two decades, the Automatic Identification System (AIS) has enabled real-time monitoring of hundreds of thousands of vessels, generating huge amounts of data daily. One major challenge in using AIS data is the presence of large gaps in vessel trajectories, often caused by coverage limitations or intentional transmission interruptions. These gaps can significantly degrade data quality, resulting in inaccurate or incomplete analysis. State-of-the-art imputation approaches have mainly been devised to tackle gaps in vehicle trajectories, even when the underlying road network is not considered. But the motion patterns of sailing vessels differ substantially, e.g., smooth turns, maneuvering near ports, or navigating in adverse weather conditions. In this application paper, we propose HABIT, a lightweight, configurable H3 Aggregation-Based Imputation framework for vessel Trajectories. This data-driven framework provides a valuable means to impute missing trajectory segments by extracting, analyzing, and indexing motion patterns from historical AIS data. Our empirical study over AIS data across various timeframes, densities, and vessel types reveals that HABIT produces maritime trajectory imputations performing comparably to baseline methods in terms of accuracy, while performing better in terms of latency while accounting for vessel characteristics and their motion patterns.
comment: International Conference on Extending Database Technology (EDBT 2026)
☆ Learning to Manipulate Anything: Revealing Data Scaling Laws in Bounding-Box Guided Policies
Diffusion-based policies show limited generalization in semantic manipulation, posing a key obstacle to the deployment of real-world robots. This limitation arises because relying solely on text instructions is inadequate to direct the policy's attention toward the target object in complex and dynamic environments. To solve this problem, we propose leveraging bounding-box instruction to directly specify target object, and further investigate whether data scaling laws exist in semantic manipulation tasks. Specifically, we design a handheld segmentation device with an automated annotation pipeline, Label-UMI, which enables the efficient collection of demonstration data with semantic labels. We further propose a semantic-motion-decoupled framework that integrates object detection and bounding-box guided diffusion policy to improve generalization and adaptability in semantic manipulation. Throughout extensive real-world experiments on large-scale datasets, we validate the effectiveness of the approach, and reveal a power-law relationship between generalization performance and the number of bounding-box objects. Finally, we summarize an effective data collection strategy for semantic manipulation, which can achieve 85\% success rates across four tasks on both seen and unseen objects. All datasets and code will be released to the community.
☆ Where Bits Matter in World Model Planning: A Paired Mixed-Bit Study for Efficient Spatial Reasoning
Efficient spatial reasoning requires world models that remain reliable under tight precision budgets. We study whether low-bit planning behavior is determined mostly by total bitwidth or by where bits are allocated across modules. Using DINO-WM on the Wall planning task, we run a paired-goal mixed-bit evaluation across uniform, mixed, asymmetric, and layerwise variants under two planner budgets. We observe a consistent three-regime pattern: 8-bit and 6-bit settings remain close to FP16, 3-bit settings collapse, and 4-bit settings are allocation-sensitive. In that transition region, preserving encoder precision improves planning relative to uniform quantization, and near-size asymmetric variants show the same encoder-side direction. In a later strict 22-cell replication with smaller per-cell episode count, the mixed-versus-uniform INT4 sign becomes budget-conditioned, which further highlights the sensitivity of this transition regime. These findings motivate module-aware, budget-aware quantization policies as a broader research direction for efficient spatial reasoning. Code and run artifacts are available at https://github.com/suraj-ranganath/DINO-MBQuant.
comment: Workshop submission
☆ DiffPlace: Street View Generation via Place-Controllable Diffusion Model Enhancing Place Recognition ICRA 2026
Generative models have advanced significantly in realistic image synthesis, with diffusion models excelling in quality and stability. Recent multi-view diffusion models improve 3D-aware street view generation, but they struggle to produce place-aware and background-consistent urban scenes from text, BEV maps, and object bounding boxes. This limits their effectiveness in generating realistic samples for place recognition tasks. To address these challenges, we propose DiffPlace, a novel framework that introduces a place-ID controller to enable place-controllable multi-view image generation. The place-ID controller employs linear projection, perceiver transformer, and contrastive learning to map place-ID embeddings into a fixed CLIP space, allowing the model to synthesize images with consistent background buildings while flexibly modifying foreground objects and weather conditions. Extensive experiments, including quantitative comparisons and augmented training evaluations, demonstrate that DiffPlace outperforms existing methods in both generation quality and training support for visual place recognition. Our results highlight the potential of generative models in enhancing scene-level and place-aware synthesis, providing a valuable approach for improving place recognition in autonomous driving
comment: accepted by ICRA 2026
☆ LAMP: Implicit Language Map for Robot Navigation
Recent advances in vision-language models have made zero-shot navigation feasible, enabling robots to follow natural language instructions without requiring labeling. However, existing methods that explicitly store language vectors in grid or node-based maps struggle to scale to large environments due to excessive memory requirements and limited resolution for fine-grained planning. We introduce LAMP (Language Map), a novel neural language field-based navigation framework that learns a continuous, language-driven map and directly leverages it for fine-grained path generation. Unlike prior approaches, our method encodes language features as an implicit neural field rather than storing them explicitly at every location. By combining this implicit representation with a sparse graph, LAMP supports efficient coarse path planning and then performs gradient-based optimization in the learned field to refine poses near the goal. This coarse-to-fine pipeline, language-driven, gradient-guided optimization is the first application of an implicit language map for precise path generation. This refinement is particularly effective at selecting goal regions not directly observed by leveraging semantic similarities in the learned feature space. To further enhance robustness, we adopt a Bayesian framework that models embedding uncertainty via the von Mises-Fisher distribution, thereby improving generalization to unobserved regions. To scale to large environments, LAMP employs a graph sampling strategy that prioritizes spatial coverage and embedding confidence, retaining only the most informative nodes and substantially reducing computational overhead. Our experimental results, both in NVIDIA Isaac Sim and on a real multi-floor building, demonstrate that LAMP outperforms existing explicit methods in both memory efficiency and fine-grained goal-reaching accuracy.
comment: Accepted for publication in IEEE Robotics and Automation Letters (RA-L). Project page: https://lab-of-ai-and-robotics.github.io/LAMP/
☆ JEPA-VLA: Video Predictive Embedding is Needed for VLA Models
Recent vision-language-action (VLA) models built upon pretrained vision-language models (VLMs) have achieved significant improvements in robotic manipulation. However, current VLAs still suffer from low sample efficiency and limited generalization. This paper argues that these limitations are closely tied to an overlooked component, pretrained visual representation, which offers insufficient knowledge on both aspects of environment understanding and policy prior. Through an in-depth analysis, we find that commonly used visual representations in VLAs, whether pretrained via language-image contrastive learning or image-based self-supervised learning, remain inadequate at capturing crucial, task-relevant environment information and at inducing effective policy priors, i.e., anticipatory knowledge of how the environment evolves under successful task execution. In contrast, we discover that predictive embeddings pretrained on videos, in particular V-JEPA 2, are adept at flexibly discarding unpredictable environment factors and encoding task-relevant temporal dynamics, thereby effectively compensating for key shortcomings of existing visual representations in VLAs. Building on these observations, we introduce JEPA-VLA, a simple yet effective approach that adaptively integrates predictive embeddings into existing VLAs. Our experiments demonstrate that JEPA-VLA yields substantial performance gains across a range of benchmarks, including LIBERO, LIBERO-plus, RoboTwin2.0, and real-robot tasks.
☆ HAIC: Humanoid Agile Object Interaction Control via Dynamics-Aware World Model
Humanoid robots show promise for complex whole-body tasks in unstructured environments. Although Human-Object Interaction (HOI) has advanced, most methods focus on fully actuated objects rigidly coupled to the robot, ignoring underactuated objects with independent dynamics and non-holonomic constraints. These introduce control challenges from coupling forces and occlusions. We present HAIC, a unified framework for robust interaction across diverse object dynamics without external state estimation. Our key contribution is a dynamics predictor that estimates high-order object states (velocity, acceleration) solely from proprioceptive history. These predictions are projected onto static geometric priors to form a spatially grounded dynamic occupancy map, enabling the policy to infer collision boundaries and contact affordances in blind spots. We use asymmetric fine-tuning, where a world model continuously adapts to the student policy's exploration, ensuring robust state estimation under distribution shifts. Experiments on a humanoid robot show HAIC achieves high success rates in agile tasks (skateboarding, cart pushing/pulling under various loads) by proactively compensating for inertial perturbations, and also masters multi-object long-horizon tasks like carrying a box across varied terrain by predicting the dynamics of multiple objects.
comment: Webpage: https://haic-humanoid.github.io/
☆ Counterfactual Conditional Likelihood Rewards for Multiagent Exploration
Efficient exploration is critical for multiagent systems to discover coordinated strategies, particularly in open-ended domains such as search and rescue or planetary surveying. However, when exploration is encouraged only at the individual agent level, it often leads to redundancy, as agents act without awareness of how their teammates are exploring. In this work, we introduce Counterfactual Conditional Likelihood (CCL) rewards, which score each agent's exploration by isolating its unique contribution to team exploration. Unlike prior methods that reward agents solely for the novelty of their individual observations, CCL emphasizes observations that are informative with respect to the joint exploration of the team. Experiments in continuous multiagent domains show that CCL rewards accelerate learning for domains with sparse team rewards, where most joint actions yield zero rewards, and are particularly effective in tasks that require tight coordination among agents.
comment: 9 pages, 5 figures
☆ AC-MASAC: An Attentive Curriculum Learning Framework for Heterogeneous UAV Swarm Coordination
Cooperative path planning for heterogeneous UAV swarms poses significant challenges for Multi-Agent Reinforcement Learning (MARL), particularly in handling asymmetric inter-agent dependencies and addressing the risks of sparse rewards and catastrophic forgetting during training. To address these issues, this paper proposes an attentive curriculum learning framework (AC-MASAC). The framework introduces a role-aware heterogeneous attention mechanism to explicitly model asymmetric dependencies. Moreover, a structured curriculum strategy is designed, integrating hierarchical knowledge transfer and stage-proportional experience replay to address the issues of sparse rewards and catastrophic forgetting. The proposed framework is validated on a custom multi-agent simulation platform, and the results show that our method has significant advantages over other advanced methods in terms of Success Rate, Formation Keeping Rate, and Success-weighted Mission Time. The code is available at \textcolor{red}{https://github.com/Wanhao-Liu/AC-MASAC}.
☆ GSO-SLAM: Bidirectionally Coupled Gaussian Splatting and Direct Visual Odometry
We propose GSO-SLAM, a real-time monocular dense SLAM system that leverages Gaussian scene representation. Unlike existing methods that couple tracking and mapping with a unified scene, incurring computational costs, or loosely integrate them with well-structured tracking frameworks, introducing redundancies, our method bidirectionally couples Visual Odometry (VO) and Gaussian Splatting (GS). Specifically, our approach formulates joint optimization within an Expectation-Maximization (EM) framework, enabling the simultaneous refinement of VO-derived semi-dense depth estimates and the GS representation without additional computational overhead. Moreover, we present Gaussian Splat Initialization, which utilizes image information, keyframe poses, and pixel associations from VO to produce close approximations to the final Gaussian scene, thereby eliminating the need for heuristic methods. Through extensive experiments, we validate the effectiveness of our method, showing that it not only operates in real time but also achieves state-of-the-art geometric/photometric fidelity of the reconstructed scene and tracking accuracy.
comment: 8 pages, 6 figures, RA-L accepted
☆ LLM-Driven 3D Scene Generation of Agricultural Simulation Environments
Procedural generation techniques in 3D rendering engines have revolutionized the creation of complex environments, reducing reliance on manual design. Recent approaches using Large Language Models (LLMs) for 3D scene generation show promise but often lack domain-specific reasoning, verification mechanisms, and modular design. These limitations lead to reduced control and poor scalability. This paper investigates the use of LLMs to generate agricultural synthetic simulation environments from natural language prompts, specifically to address the limitations of lacking domain-specific reasoning, verification mechanisms, and modular design. A modular multi-LLM pipeline was developed, integrating 3D asset retrieval, domain knowledge injection, and code generation for the Unreal rendering engine using its API. This results in a 3D environment with realistic planting layouts and environmental context, all based on the input prompt and the domain knowledge. To enhance accuracy and scalability, the system employs a hybrid strategy combining LLM optimization techniques such as few-shot prompting, Retrieval-Augmented Generation (RAG), finetuning, and validation. Unlike monolithic models, the modular architecture enables structured data handling, intermediate verification, and flexible expansion. The system was evaluated using structured prompts and semantic accuracy metrics. A user study assessed realism and familiarity against real-world images, while an expert comparison demonstrated significant time savings over manual scene design. The results confirm the effectiveness of multi-LLM pipelines in automating domain-specific 3D scene generation with improved reliability and precision. Future work will explore expanding the asset hierarchy, incorporating real-time generation, and adapting the pipeline to other simulation domains beyond agriculture.
comment: Accepted at IEEE Conference on Artificial Intelligence 2026
☆ Clutt3R-Seg: Sparse-view 3D Instance Segmentation for Language-grounded Grasping in Cluttered Scenes ICRA 2026
Reliable 3D instance segmentation is fundamental to language-grounded robotic manipulation. Its critical application lies in cluttered environments, where occlusions, limited viewpoints, and noisy masks degrade perception. To address these challenges, we present Clutt3R-Seg, a zero-shot pipeline for robust 3D instance segmentation for language-grounded grasping in cluttered scenes. Our key idea is to introduce a hierarchical instance tree of semantic cues. Unlike prior approaches that attempt to refine noisy masks, our method leverages them as informative cues: through cross-view grouping and conditional substitution, the tree suppresses over- and under-segmentation, yielding view-consistent masks and robust 3D instances. Each instance is enriched with open-vocabulary semantic embeddings, enabling accurate target selection from natural language instructions. To handle scene changes during multi-stage tasks, we further introduce a consistency-aware update that preserves instance correspondences from only a single post-interaction image, allowing efficient adaptation without rescanning. Clutt3R-Seg is evaluated on both synthetic and real-world datasets, and validated on a real robot. Across all settings, it consistently outperforms state-of-the-art baselines in cluttered and sparse-view scenarios. Even on the most challenging heavy-clutter sequences, Clutt3R-Seg achieves an AP@25 of 61.66, over 2.2x higher than baselines, and with only four input views it surpasses MaskClustering with eight views by more than 2x. The code is available at: https://github.com/jeonghonoh/clutt3r-seg.
comment: Accepted to ICRA 2026. 9 pages, 8 figures
☆ 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.
☆ Human-Like Gaze Behavior in Social Robots: A Deep Learning Approach Integrating Human and Non-Human Stimuli
Nonverbal behaviors, particularly gaze direction, play a crucial role in enhancing effective communication in social interactions. As social robots increasingly participate in these interactions, they must adapt their gaze based on human activities and remain receptive to all cues, whether human-generated or not, to ensure seamless and effective communication. This study aims to increase the similarity between robot and human gaze behavior across various social situations, including both human and non-human stimuli (e.g., conversations, pointing, door openings, and object drops). A key innovation in this study, is the investigation of gaze responses to non-human stimuli, a critical yet underexplored area in prior research. These scenarios, were simulated in the Unity software as a 3D animation and a 360-degree real-world video. Data on gaze directions from 41 participants were collected via virtual reality (VR) glasses. Preprocessed data, trained two neural networks-LSTM and Transformer-to build predictive models based on individuals' gaze patterns. In the animated scenario, the LSTM and Transformer models achieved prediction accuracies of 67.6% and 70.4%, respectively; In the real-world scenario, the LSTM and Transformer models achieved accuracies of 72% and 71.6%, respectively. Despite the gaze pattern differences among individuals, our models outperform existing approaches in accuracy while uniquely considering non-human stimuli, offering a significant advantage over previous literature. Furthermore, deployed on the NAO robot, the system was evaluated by 275 participants via a comprehensive questionnaire, with results demonstrating high satisfaction during interactions. This work advances social robotics by enabling robots to dynamically mimic human gaze behavior in complex social contexts.
☆ ViTaS: Visual Tactile Soft Fusion Contrastive Learning for Visuomotor Learning ICRA 2026
Tactile information plays a crucial role in human manipulation tasks and has recently garnered increasing attention in robotic manipulation. However, existing approaches mostly focus on the alignment of visual and tactile features and the integration mechanism tends to be direct concatenation. Consequently, they struggle to effectively cope with occluded scenarios due to neglecting the inherent complementary nature of both modalities and the alignment may not be exploited enough, limiting the potential of their real-world deployment. In this paper, we present ViTaS, a simple yet effective framework that incorporates both visual and tactile information to guide the behavior of an agent. We introduce Soft Fusion Contrastive Learning, an advanced version of conventional contrastive learning method and a CVAE module to utilize the alignment and complementarity within visuo-tactile representations. We demonstrate the effectiveness of our method in 12 simulated and 3 real-world environments, and our experiments show that ViTaS significantly outperforms existing baselines. Project page: https://skyrainwind.github.io/ViTaS/index.html.
comment: Published to ICRA 2026
☆ ABot-N0: Technical Report on the VLA Foundation Model for Versatile Embodied Navigation
Embodied navigation has long been fragmented by task-specific architectures. We introduce ABot-N0, a unified Vision-Language-Action (VLA) foundation model that achieves a ``Grand Unification'' across 5 core tasks: Point-Goal, Object-Goal, Instruction-Following, POI-Goal, and Person-Following. ABot-N0 utilizes a hierarchical ``Brain-Action'' architecture, pairing an LLM-based Cognitive Brain for semantic reasoning with a Flow Matching-based Action Expert for precise, continuous trajectory generation. To support large-scale learning, we developed the ABot-N0 Data Engine, curating 16.9M expert trajectories and 5.0M reasoning samples across 7,802 high-fidelity 3D scenes (10.7 $\text{km}^2$). ABot-N0 achieves new SOTA performance across 7 benchmarks, significantly outperforming specialized models. Furthermore, our Agentic Navigation System integrates a planner with hierarchical topological memory, enabling robust, long-horizon missions in dynamic real-world environments.
comment: Project Page: https://amap-cvlab.github.io/ABot-Navigation/ABot-N0/
☆ ReaDy-Go: Real-to-Sim Dynamic 3D Gaussian Splatting Simulation for Environment-Specific Visual Navigation with Moving Obstacles
Visual navigation models often struggle in real-world dynamic environments due to limited robustness to the sim-to-real gap and the difficulty of training policies tailored to target deployment environments (e.g., households, restaurants, and factories). Although real-to-sim navigation simulation using 3D Gaussian Splatting (GS) can mitigate this gap, prior works have assumed only static scenes or unrealistic dynamic obstacles, despite the importance of safe navigation in dynamic environments. To address these issues, we propose ReaDy-Go, a novel real-to-sim simulation pipeline that synthesizes photorealistic dynamic scenarios for target environments. ReaDy-Go generates photorealistic navigation datasets for dynamic environments by combining a reconstructed static GS scene with dynamic human GS obstacles, and trains policies robust to both the sim-to-real gap and moving obstacles. The pipeline consists of three components: (1) a dynamic GS simulator that integrates scene GS with a human animation module, enabling the insertion of animatable human GS avatars and the synthesis of plausible human motions from 2D trajectories, (2) navigation dataset generation for dynamic environments that leverages the simulator, a robot expert planner designed for dynamic GS representations, and a human planner, and (3) policy learning using the generated datasets. ReaDy-Go outperforms baselines across target environments in both simulation and real-world experiments, demonstrating improved navigation performance even after sim-to-real transfer and in the presence of moving obstacles. Moreover, zero-shot sim-to-real deployment in an unseen environment indicates its generalization potential. Project page: https://syeon-yoo.github.io/ready-go-site/.
comment: Project page: https://syeon-yoo.github.io/ready-go-site/
☆ HyperDet: 3D Object Detection with Hyper 4D Radar Point Clouds
4D mmWave radar provides weather-robust, velocity-aware measurements and is more cost-effective than LiDAR. However, radar-only 3D detection still trails LiDAR-based systems because radar point clouds are sparse, irregular, and often corrupted by multipath noise, yielding weak and unstable geometry. We present HyperDet, a detector-agnostic radar-only 3D detection framework that constructs a task-aware hyper 4D radar point cloud for standard LiDAR-oriented detectors. HyperDet aggregates returns from multiple surround-view 4D radars over consecutive frames to improve coverage and density, then applies geometry-aware cross-sensor consensus validation with a lightweight self-consistency check outside overlap regions to suppress inconsistent returns. It further integrates a foreground-focused diffusion module with training-time mixed radar-LiDAR supervision to densify object structures while lifting radar attributes (e.g., Doppler, RCS); the model is distilled into a consistency model for single-step inference. On MAN TruckScenes, HyperDet consistently improves over raw radar inputs with VoxelNeXt and CenterPoint, partially narrowing the radar-LiDAR gap. These results show that input-level refinement enables radar to better leverage LiDAR-oriented detectors without architectural modifications.
comment: 9 pages, 4 figures, 6 tables
☆ Effective Task Planning with Missing Objects using Learning-Informed Object Search
Task planning for mobile robots often assumes full environment knowledge and so popular approaches, like planning via the PDDL, cannot plan when the locations of task-critical objects are unknown. Recent learning-driven object search approaches are effective, but operate as standalone tools and so are not straightforwardly incorporated into full task planners, which must additionally determine both what objects are necessary and when in the plan they should be sought out. To address this limitation, we develop a planning framework centered around novel model-based LIOS actions: each a policy that aims to find and retrieve a single object. High-level planning treats LIOS actions as deterministic and so -- informed by model-based calculations of the expected cost of each -- generates plans that interleave search and execution for effective, sound, and complete learning-informed task planning despite uncertainty. Our work effectively reasons about uncertainty while maintaining compatibility with existing full-knowledge solvers. In simulated ProcTHOR homes and in the real world, our approach outperforms non-learned and learned baselines on tasks including retrieval and meal prep.
☆ EasyMimic: A Low-Cost Framework for Robot Imitation Learning from Human Videos
Robot imitation learning is often hindered by the high cost of collecting large-scale, real-world data. This challenge is especially significant for low-cost robots designed for home use, as they must be both user-friendly and affordable. To address this, we propose the EasyMimic framework, a low-cost and replicable solution that enables robots to quickly learn manipulation policies from human video demonstrations captured with standard RGB cameras. Our method first extracts 3D hand trajectories from the videos. An action alignment module then maps these trajectories to the gripper control space of a low-cost robot. To bridge the human-to-robot domain gap, we introduce a simple and user-friendly hand visual augmentation strategy. We then use a co-training method, fine-tuning a model on both the processed human data and a small amount of robot data, enabling rapid adaptation to new tasks. Experiments on the low-cost LeRobot platform demonstrate that EasyMimic achieves high performance across various manipulation tasks. It significantly reduces the reliance on expensive robot data collection, offering a practical path for bringing intelligent robots into homes. Project website: https://zt375356.github.io/EasyMimic-Project/.
comment: icra 2026
☆ An Autonomous, End-to-End, Convex-Based Framework for Close-Range Rendezvous Trajectory Design and Guidance with Hardware Testbed Validation
Autonomous satellite servicing missions must execute close-range rendezvous under stringent safety and operational constraints while remaining computationally tractable for onboard use and robust to uncertainty in sensing, actuation, and dynamics. This paper presents CORTEX (Convex Optimization for Rendezvous Trajectory Execution), an autonomous, perception-enabled, real-time trajectory design and guidance framework for close-range rendezvous. CORTEX integrates a deep-learning perception pipeline with convex-optimisation-based trajectory design and guidance, including reference regeneration and abort-to-safe-orbit logic to recover from large deviations caused by sensor faults and engine failures. CORTEX is validated in high-fidelity software simulation and hardware-in-the-loop experiments. The software pipeline (Basilisk) models high-fidelity relative dynamics, realistic thruster execution, perception, and attitude control. Hardware testing uses (i) an optical navigation testbed to assess perception-to-estimation performance and (ii) a planar air-bearing testbed to evaluate the end-to-end guidance loop under representative actuation and subsystem effects. A Monte-Carlo campaign in simulation includes initial-state uncertainty, thrust-magnitude errors, and missed-thrust events; under the strongest case investigated, CORTEX achieves terminal docking errors of $36.85 \pm 44.46$ mm in relative position and $1.25 \pm 2.26$ mm/s in relative velocity. On the planar air-bearing testbed, 18 cases are executed (10 nominal; 8 off-nominal requiring recomputation and/or abort due to simulated engine failure and sensor malfunctions), yielding terminal errors of $8.09 \pm 5.29$ mm in position and $2.23 \pm 1.72$ mm/s in velocity.
☆ Control Barrier Functions with Audio Risk Awareness for Robot Safe Navigation on Construction Sites
Construction automation increasingly requires autonomous mobile robots, yet robust autonomy remains challenging on construction sites. These environments are dynamic and often visually occluded, which complicates perception and navigation. In this context, valuable information from audio sources remains underutilized in most autonomy stacks. This work presents a control barrier function (CBF)-based safety filter that provides safety guarantees for obstacle avoidance while adapting safety margins during navigation using an audio-derived risk cue. The proposed framework augments the CBF with a lightweight, real-time jackhammer detector based on signal envelope and periodicity. Its output serves as an exogenous risk that is directly enforced in the controller by modulating the barrier function. The approach is evaluated in simulation with two CBF formulations (circular and goal-aligned elliptical) with a unicycle robot navigating a cluttered construction environment. Results show that the CBF safety filter eliminates safety violations across all trials while reaching the target in 40.2% (circular) vs. 76.5% (elliptical), as the elliptical formulation better avoids deadlock. This integration of audio perception into a CBF-based controller demonstrates a pathway toward richer multimodal safety reasoning in autonomous robots for safety-critical and dynamic environments.
☆ 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
☆ Self-Refining Vision Language Model for Robotic Failure Detection and Reasoning
Reasoning about failures is crucial for building reliable and trustworthy robotic systems. Prior approaches either treat failure reasoning as a closed-set classification problem or assume access to ample human annotations. Failures in the real world are typically subtle, combinatorial, and difficult to enumerate, whereas rich reasoning labels are expensive to acquire. We address this problem by introducing ARMOR: Adaptive Round-based Multi-task mOdel for Robotic failure detection and reasoning. We formulate detection and reasoning as a multi-task self-refinement process, where the model iteratively predicts detection outcomes and natural language reasoning conditioned on past outputs. During training, ARMOR learns from heterogeneous supervision - large-scale sparse binary labels and small-scale rich reasoning annotations - optimized via a combination of offline and online imitation learning. At inference time, ARMOR generates multiple refinement trajectories and selects the most confident prediction via a self-certainty metric. Experiments across diverse environments show that ARMOR achieves state-of-the-art performance by improving over the previous approaches by up to 30% on failure detection rate and up to 100% in reasoning measured through LLM fuzzy match score, demonstrating robustness to heterogeneous supervision and open-ended reasoning beyond predefined failure modes. We provide dditional visualizations on our website: https://sites.google.com/utexas.edu/armor
☆ Zero-Shot Adaptation to Robot Structural Damage via Natural Language-Informed Kinodynamics Modeling
High-performance autonomous mobile robots endure significant mechanical stress during in-the-wild operations, e.g., driving at high speeds or over rugged terrain. Although these platforms are engineered to withstand such conditions, mechanical degradation is inevitable. Structural damage manifests as consistent and notable changes in kinodynamic behavior compared to a healthy vehicle. Given the heterogeneous nature of structural failures, quantifying various damages to inform kinodynamics is challenging. We posit that natural language can describe and thus capture this variety of damages. Therefore, we propose Zero-shot Language Informed Kinodynamics (ZLIK), which employs self-supervised learning to ground semantic information of damage descriptions in kinodynamic behaviors to learn a forward kinodynamics model in a data-driven manner. Using the high-fidelity soft-body physics simulator BeamNG.tech, we collect data from a variety of structurally compromised vehicles. Our learned model achieves zero-shot adaptation to different damages with up to 81% reduction in kinodynamics error and generalizes across the sim-to-real and full-to-1/10$^{\text{th}}$ scale gaps.
comment: 12 Pages, 6 Figures, 5 Tables
☆ Predicting Dynamic Map States from Limited Field-of-View Sensor Data
When autonomous systems are deployed in real-world scenarios, sensors are often subject to limited field-of-view (FOV) constraints, either naturally through system design, or through unexpected occlusions or sensor failures. In conditions where a large FOV is unavailable, it is important to be able to infer information about the environment and predict the state of nearby surroundings based on available data to maintain safe and accurate operation. In this work, we explore the effectiveness of deep learning for dynamic map state prediction based on limited FOV time series data. We show that by representing dynamic sensor data in a simple single-image format that captures both spatial and temporal information, we can effectively use a wide variety of existing image-to-image learning models to predict map states with high accuracy in a diverse set of sensing scenarios.
comment: 7 pages, 4 figures
☆ LongNav-R1: Horizon-Adaptive Multi-Turn RL for Long-Horizon VLA Navigation
This paper develops LongNav-R1, an end-to-end multi-turn reinforcement learning (RL) framework designed to optimize Visual-Language-Action (VLA) models for long-horizon navigation. Unlike existing single-turn paradigm, LongNav-R1 reformulates the navigation decision process as a continuous multi-turn conversation between the VLA policy and the embodied environment. This multi-turn RL framework offers two distinct advantages: i) it enables the agent to reason about the causal effects of historical interactions and sequential future outcomes; and ii) it allows the model to learn directly from online interactions, fostering diverse trajectory generation and avoiding the behavioral rigidity often imposed by human demonstrations. Furthermore, we introduce Horizon-Adaptive Policy Optimization. This mechanism explicitly accounts for varying horizon lengths during advantage estimation, facilitating accurate temporal credit assignment over extended sequences. Consequently, the agent develops diverse navigation behaviors and resists collapse during long-horizon tasks. Experiments on object navigation benchmarks validate the framework's efficacy: With 4,000 rollout trajectories, LongNav-R1 boosts the Qwen3-VL-2B success rate from 64.3% to 73.0%. These results demonstrate superior sample efficiency and significantly outperform state-of-the-art methods. The model's generalizability and robustness are further validated by its zero-shot performance in long-horizon real-world navigation settings. All source code will be open-sourced upon publication.
comment: VLA, Navigation
☆ Schur-MI: Fast Mutual Information for Robotic Information Gathering
Mutual information (MI) is a principled and widely used objective for robotic information gathering (RIG), providing strong theoretical guarantees for sensor placement (SP) and informative path planning (IPP). However, its high computational cost, dominated by repeated log-determinant evaluations, has limited its use in real-time planning. This letter presents Schur-MI, a Gaussian process (GP) MI formulation that (i) leverages the iterative structure of RIG to precompute and reuse expensive intermediate quantities across planning steps, and (ii) uses a Schur-complement factorization to avoid large determinant computations. Together, these methods reduce the per-evaluation cost of MI from $\mathcal{O}(|\mathcal{V}|^3)$ to $\mathcal{O}(|\mathcal{A}|^3)$, where $\mathcal{V}$ and $\mathcal{A}$ denote the candidate and selected sensing locations, respectively. Experiments on real-world bathymetry datasets show that Schur-MI achieves up to a $12.7\times$ speedup over the standard MI formulation. Field trials with an autonomous surface vehicle (ASV) performing adaptive IPP further validate its practicality. By making MI computation tractable for online planning, Schur-MI helps bridge the gap between information-theoretic objectives and real-time robotic exploration.
comment: preprint
☆ ForeAct: Steering Your VLA with Efficient Visual Foresight Planning
Vision-Language-Action (VLA) models convert high-level language instructions into concrete, executable actions, a task that is especially challenging in open-world environments. We present Visual Foresight Planning (ForeAct), a general and efficient planner that guides a VLA step-by-step using imagined future observations and subtask descriptions. With an imagined future observation, the VLA can focus on visuo-motor inference rather than high-level semantic reasoning, leading to improved accuracy and generalization. Our planner comprises a highly efficient foresight image generation module that predicts a high-quality 640$\times$480 future observation from the current visual input and language instruction within only 0.33s on an H100 GPU, together with a vision-language model that reasons over the task and produces subtask descriptions for both the generator and the VLA. Importantly, state-of-the-art VLAs can integrate our planner seamlessly by simply augmenting their visual inputs, without any architectural modification. The foresight generator is pretrained on over 1 million multi-task, cross-embodiment episodes, enabling it to learn robust embodied dynamics. We evaluate our framework on a benchmark that consists of 11 diverse, multi-step real-world tasks. It achieves an average success rate of 87.4%, demonstrating a +40.9% absolute improvement over the $π_0$ baseline (46.5%) and a +30.3% absolute improvement over $π_0$ augmented with textual subtask guidance (57.1%).
☆ LatentAM: Real-Time, Large-Scale Latent Gaussian Attention Mapping via Online Dictionary Learning
We present LatentAM, an online 3D Gaussian Splatting (3DGS) mapping framework that builds scalable latent feature maps from streaming RGB-D observations for open-vocabulary robotic perception. Instead of distilling high-dimensional Vision-Language Model (VLM) embeddings using model-specific decoders, LatentAM proposes an online dictionary learning approach that is both model-agnostic and pretraining-free, enabling plug-and-play integration with different VLMs at test time. Specifically, our approach associates each Gaussian primitive with a compact query vector that can be converted into approximate VLM embeddings using an attention mechanism with a learnable dictionary. The dictionary is initialized efficiently from streaming observations and optimized online to adapt to evolving scene semantics under trust-region regularization. To scale to long trajectories and large environments, we further propose an efficient map management strategy based on voxel hashing, where optimization is restricted to an active local map on the GPU, while the global map is stored and indexed on the CPU to maintain bounded GPU memory usage. Experiments on public benchmarks and a large-scale custom dataset demonstrate that LatentAM attains significantly better feature reconstruction fidelity compared to state-of-the-art methods, while achieving near-real-time speed (12-35 FPS) on the evaluated datasets. Our project page is at: https://junwoonlee.github.io/projects/LatentAM
comment: 8 pages, 5 figures
♻ ☆ One Demo Is All It Takes: Planning Domain Derivation with LLMs from A Single Demonstration ICLR 2026
Pre-trained large language models (LLMs) show promise for robotic task planning but often struggle to guarantee correctness in long-horizon problems. Task and motion planning (TAMP) addresses this by grounding symbolic plans in low-level execution, yet it relies heavily on manually engineered planning domains. To improve long-horizon planning reliability and reduce human intervention, we present Planning Domain Derivation with LLMs (PDDLLM), a framework that automatically induces symbolic predicates and actions directly from demonstration trajectories by combining LLM reasoning with physical simulation roll-outs. Unlike prior domain-inference methods that rely on partially predefined or language descriptions of planning domains, PDDLLM constructs domains without manual domain initialization and automatically integrates them with motion planners to produce executable plans, enhancing long-horizon planning automation. Across 1,200 tasks in nine environments, PDDLLM outperforms six LLM-based planning baselines, achieving at least 20\% higher success rates, reduced token costs, and successful deployment on multiple physical robot platforms.
comment: Published as a conference paper at ICLR 2026
♻ ☆ Humanoid Manipulation Interface: Humanoid Whole-Body Manipulation from Robot-Free Demonstrations
Current approaches for humanoid whole-body manipulation, primarily relying on teleoperation or visual sim-to-real reinforcement learning, are hindered by hardware logistics and complex reward engineering. Consequently, demonstrated autonomous skills remain limited and are typically restricted to controlled environments. In this paper, we present the Humanoid Manipulation Interface (HuMI), a portable and efficient framework for learning diverse whole-body manipulation tasks across various environments. HuMI enables robot-free data collection by capturing rich whole-body motion using portable hardware. This data drives a hierarchical learning pipeline that translates human motions into dexterous and feasible humanoid skills. Extensive experiments across five whole-body tasks--including kneeling, squatting, tossing, walking, and bimanual manipulation--demonstrate that HuMI achieves a 3x increase in data collection efficiency compared to teleoperation and attains a 70% success rate in unseen environments.
comment: Website: https://humanoid-manipulation-interface.github.io
♻ ☆ Model Predictive Control via Probabilistic Inference: A Tutorial and Survey
This paper presents a tutorial and survey on probabilistic inference-based model predictive control (PI-MPC) for robotics. PI-MPC reformulates finite-horizon optimal control as inference over an optimal control distribution expressed as a Boltzmann distribution weighted by a control prior, and generates actions through variational inference. In the tutorial part, we derive this formulation and explain action generation via variational inference, highlighting Model Predictive Path Integral (MPPI) control as a representative algorithm with a closed-form sampling update. In the survey part, we organize existing PI-MPC research around key design dimensions, including prior design, multi-modality, constraint handling, scalability, hardware acceleration, and theoretical analysis. This paper provides a unified conceptual perspective on PI-MPC and a practical entry point for robotics researchers and practitioners.
comment: 39 pages, 7 figures
♻ ☆ An Anatomy-Aware Shared Control Approach for Assisted Teleoperation of Lung Ultrasound Examinations
Although fully autonomous systems still face challenges due to patients' anatomical variability, teleoperated systems appear to be more practical in current healthcare settings. This paper presents an anatomy-aware control framework for teleoperated lung ultrasound. Leveraging biomechanically accurate 3D modelling, the system applies virtual constraints on the ultrasound probe pose and provides real-time visual feedback to assist in precise probe placement tasks. A twofold evaluation, one with 5 naive operators on a single volunteer and the second with a single experienced operator on 6 volunteers, compared our method with a standard teleoperation baseline. The results of the first one characterised the accuracy of the anatomical model and the improved perceived performance by the naive operators, while the second one focused on the efficiency of the system in improving probe placement and reducing procedure time compared to traditional teleoperation. The results demonstrate that the proposed framework enhances the physician's capabilities in executing remote lung ultrasound, reducing more than 20% of execution time on 4-point acquisitions, towards faster, more objective and repeatable exams.
♻ ☆ Stability Analysis of Geometric Control for a Canonical Class of Underactuated Aerial Vehicles with Spurious Forces
Standard geometric control relies on force-moment decoupling, an assumption that breaks down in many aerial platforms due to spurious forces naturally induced by control moments. While strategies for such coupled systems have been validated experimentally, a rigorous theoretical certification of their stability is currently missing. This work fills this gap by providing the first formal stability analysis for a generic class of floating rigid bodies subject to spurious forces. We introduce a canonical model and construct a Lyapunov-based proof establishing local exponential stability of the hovering equilibrium. Crucially, the analysis explicitly addresses the structural challenges - specifically the induced non-minimum-phase behavior - that prevent the application of standard cascade arguments.
♻ ☆ TreeLoc: 6-DoF LiDAR Global Localization in Forests via Inter-Tree Geometric Matching ICRA 2026
Reliable localization is crucial for navigation in forests, where GPS is often degraded and LiDAR measurements are repetitive, occluded, and structurally complex. These conditions weaken the assumptions of traditional urban-centric localization methods, which assume that consistent features arise from unique structural patterns, necessitating forest-centric solutions to achieve robustness in these environments. To address these challenges, we propose TreeLoc, a LiDAR-based global localization framework for forests that handles place recognition and 6-DoF pose estimation. We represent scenes using tree stems and their Diameter at Breast Height (DBH), which are aligned to a common reference frame via their axes and summarized using the tree distribution histogram (TDH) for coarse matching, followed by fine matching with a 2D triangle descriptor. Finally, pose estimation is achieved through a two-step geometric verification. On diverse forest benchmarks, TreeLoc outperforms baselines, achieving precise localization. Ablation studies validate the contribution of each component. We also propose applications for long-term forest management using descriptors from a compact global tree database. TreeLoc is open-sourced for the robotics community at https://github.com/minwoo0611/TreeLoc.
comment: An 8-page paper with 7 tables and 8 figures, accepted to ICRA 2026
♻ ☆ EasyUUV: An LLM-Enhanced Universal and Lightweight Sim-to-Real Reinforcement Learning Framework for UUV Attitude Control
Despite recent advances in Unmanned Underwater Vehicle (UUV) attitude control, existing methods still struggle with generalizability, robustness to real-world disturbances, and efficient deployment. To address the above challenges, this paper presents EasyUUV, a Large Language Model (LLM)-enhanced, universal, and lightweight simulation-to-reality reinforcement learning (RL) framework for robust attitude control of UUVs. EasyUUV combines parallelized RL training with a hybrid control architecture, where a learned policy outputs high-level attitude corrections executed by an adaptive S-Surface controller. A multimodal LLM is further integrated to adaptively tune controller parameters at runtime using visual and textual feedback, enabling training-free adaptation to unmodeled dynamics. Also, we have developed a low-cost 6-DoF UUV platform and applied an RL policy trained through efficient parallelized simulation. Extensive simulation and real-world experiments validate the effectiveness and outstanding performance of EasyUUV in achieving robust and adaptive UUV attitude control across diverse underwater conditions. To facilitate reproducibility and further research, the source code, LLM prompts, and supplementary video are provided in the following repositories: Homepage: https://360zmem.github.io/easyuuv/ Video:https://youtu.be/m2yLQzxiIL
comment: 10 pages, 13 figures
♻ ☆ A Feature Extraction Pipeline for Enhancing Lightweight Neural Networks in sEMG-based Joint Torque Estimation
Robot-assisted rehabilitation offers an effective approach, wherein exoskeletons adapt to users' needs and provide personalized assistance. However, to deliver such assistance, accurate prediction of the user's joint torques is essential. In this work, we propose a feature extraction pipeline using 8-channel surface electromyography (sEMG) signals to predict elbow and shoulder joint torques. For preliminary evaluation, this pipeline was integrated into two neural network models: the Multilayer Perceptron (MLP) and the Temporal Convolutional Network (TCN). Data were collected from a single subject performing elbow and shoulder movements under three load conditions (0 kg, 1.10 kg, and 1.85 kg) using three motion-capture cameras. Reference torques were estimated from center-of-mass kinematics under the assumption of static equilibrium. Our offline analyses showed that, with our feature extraction pipeline, MLP model achieved mean RMSE of 0.963 N m, 1.403 N m, and 1.434 N m (over five seeds) for elbow, front-shoulder, and side-shoulder joints, respectively, which were comparable to the TCN performance. These results demonstrate that the proposed feature extraction pipeline enables a simple MLP to achieve performance comparable to that of a network designed explicitly for temporal dependencies. This finding is particularly relevant for applications with limited training data, a common scenario patient care.
♻ ☆ Translating Flow to Policy via Hindsight Online Imitation
Recent advances in hierarchical robot systems leverage a high-level planner to propose task plans and a low-level policy to generate robot actions. This design allows training the planner on action-free or even non-robot data sources (e.g., videos), providing transferable high-level guidance. Nevertheless, grounding these high-level plans into executable actions remains challenging, especially with the limited availability of high-quality robot data. To this end, we propose to improve the low-level policy through online interactions. Specifically, our approach collects online rollouts, retrospectively annotates the corresponding high-level goals from achieved outcomes, and aggregates these hindsight-relabeled experiences to update a goal-conditioned imitation policy. Our method, Hindsight Flow-conditioned Online Imitation (HinFlow), instantiates this idea with 2D point flows as the high-level planner. Across diverse manipulation tasks in both simulation and physical world, our method achieves more than $2\times$ performance improvement over the base policy, significantly outperforming the existing methods. Moreover, our framework enables policy acquisition from planners trained on cross-embodiment video data, demonstrating its potential for scalable and transferable robot learning.
♻ ☆ Scaling World Model for Hierarchical Manipulation Policies
Vision-Language-Action (VLA) models are promising for generalist robot manipulation but remain brittle in out-of-distribution (OOD) settings, especially with limited real-robot data. To resolve the generalization bottleneck, we introduce a hierarchical Vision-Language-Action framework \our{} that leverages the generalization of large-scale pre-trained world model for robust and generalizable VIsual Subgoal TAsk decomposition VISTA. Our hierarchical framework \our{} consists of a world model as the high-level planner and a VLA as the low-level executor. The high-level world model first divides manipulation tasks into subtask sequences with goal images, and the low-level policy follows the textual and visual guidance to generate action sequences. Compared to raw textual goal specification, these synthesized goal images provide visually and physically grounded details for low-level policies, making it feasible to generalize across unseen objects and novel scenarios. We validate both visual goal synthesis and our hierarchical VLA policies in massive out-of-distribution scenarios, and the performance of the same-structured VLA in novel scenarios could boost from 14% to 69% with the guidance generated by the world model. Results demonstrate that our method outperforms previous baselines with a clear margin, particularly in out-of-distribution scenarios. Project page: \href{https://vista-wm.github.io/}{https://vista-wm.github.io}
♻ ☆ DexterCap: An Affordable and Automated System for Capturing Dexterous Hand-Object Manipulation
Capturing fine-grained hand-object interactions is challenging due to severe self-occlusion from closely spaced fingers and the subtlety of in-hand manipulation motions. Existing optical motion capture systems rely on expensive camera setups and extensive manual post-processing, while low-cost vision-based methods often suffer from reduced accuracy and reliability under occlusion. To address these challenges, we present DexterCap, a low-cost optical capture system for dexterous in-hand manipulation. DexterCap uses dense, character-coded marker patches to achieve robust tracking under severe self-occlusion, together with an automated reconstruction pipeline that requires minimal manual effort. With DexterCap, we introduce DexterHand, a dataset of fine-grained hand-object interactions covering diverse manipulation behaviors and objects, from simple primitives to complex articulated objects such as a Rubik's Cube. We release the dataset and code to support future research on dexterous hand-object interaction. Project website: https://pku-mocca.github.io/Dextercap-Page/
comment: 12 pages, 12 figures
♻ ☆ RLinf-USER: A Unified and Extensible System for Real-World Online Policy Learning in Embodied AI
Online policy learning directly in the physical world is a promising yet challenging direction for embodied intelligence. Unlike simulation, real-world systems cannot be arbitrarily accelerated, cheaply reset, or massively replicated, which makes scalable data collection, heterogeneous deployment, and long-horizon effective training difficult. These challenges suggest that real-world policy learning is not only an algorithmic issue but fundamentally a systems problem. We present USER, a Unified and extensible SystEm for Real-world online policy learning. USER treats physical robots as first-class hardware resources alongside GPUs through a unified hardware abstraction layer, enabling automatic discovery, management, and scheduling of heterogeneous robots. To address cloud-edge communication, USER introduces an adaptive communication plane with tunneling-based networking, distributed data channels for traffic localization, and streaming-multiprocessor-aware weight synchronization to regulate GPU-side overhead. On top of this infrastructure, USER organizes learning as a fully asynchronous framework with a persistent, cache-aware buffer, enabling efficient long-horizon experiments with robust crash recovery and reuse of historical data. In addition, USER provides extensible abstractions for rewards, algorithms, and policies, supporting online imitation or reinforcement learning of CNN/MLP, generative policies, and large vision-language-action (VLA) models within a unified pipeline. Results in both simulation and the real world show that USER enables multi-robot coordination, heterogeneous manipulators, edge-cloud collaboration with large models, and long-running asynchronous training, offering a unified and extensible systems foundation for real-world online policy learning.
♻ ☆ DiSPo: Diffusion-SSM based Policy Learning for Coarse-to-Fine Action Discretization ICRA 2026
We aim to solve the problem of generating coarse-to-fine skills learning from demonstrations (LfD). To scale precision, traditional LfD approaches often rely on extensive fine-grained demonstrations with external interpolations or dynamics models with limited generalization capabilities. For memory-efficient learning and convenient granularity change, we propose a novel diffusion-state space model (SSM) based policy (DiSPo) that learns from diverse coarse skills and produces varying control scales of actions by leveraging an SSM, Mamba. Our evaluations show the adoption of Mamba and the proposed step-scaling method enable DiSPo to outperform in three coarse-to-fine benchmark tests with maximum 81% higher success rate than baselines. In addition, DiSPo improves inference efficiency by generating coarse motions in less critical regions. We finally demonstrate the scalability of actions with simulation and real-world manipulation tasks. Code and Videos are available at https://robo-dispo.github.io.
comment: Accepted by ICRA 2026
♻ ☆ Vi-TacMan: Articulated Object Manipulation via Vision and Touch ICRA 2026
Autonomous manipulation of articulated objects remains a fundamental challenge for robots in human environments. Vision-based methods can infer hidden kinematics but can yield imprecise estimates on unfamiliar objects. Tactile approaches achieve robust control through contact feedback but require accurate initialization. This suggests a natural synergy: vision for global guidance, touch for local precision. Yet no framework systematically exploits this complementarity for generalized articulated manipulation. Here we present Vi-TacMan, which uses vision to propose grasps and coarse directions that seed a tactile controller for precise execution. By incorporating surface normals as geometric priors and modeling directions via von Mises-Fisher distributions, our approach achieves significant gains over baselines (all p<0.0001). Critically, manipulation succeeds without explicit kinematic models -- the tactile controller refines coarse visual estimates through real-time contact regulation. Tests on more than 50,000 simulated and diverse real-world objects confirm robust cross-category generalization. This work establishes that coarse visual cues suffice for reliable manipulation when coupled with tactile feedback, offering a scalable paradigm for autonomous systems in unstructured environments.
comment: ICRA 2026
♻ ☆ STaR: Scalable Task-Conditioned Retrieval for Long-Horizon Multimodal Robot Memory
Mobile robots are often deployed over long durations in diverse open, dynamic scenes, including indoor setting such as warehouses and manufacturing facilities, and outdoor settings such as agricultural and roadway operations. A core challenge is to build a scalable long-horizon memory that supports an agentic workflow for planning, retrieval, and reasoning over open-ended instructions at variable granularity, while producing precise, actionable answers for navigation. We present STaR, an agentic reasoning framework that (i) constructs a task-agnostic, multimodal long-term memory that generalizes to unseen queries while preserving fine-grained environmental semantics (object attributes, spatial relations, and dynamic events), and (ii) introduces a Scalable Task Conditioned Retrieval algorithm based on the Information Bottleneck principle to extract from long-term memory a compact, non-redundant, information-rich set of candidate memories for contextual reasoning. We evaluate STaR on NaVQA (mixed indoor/outdoor campus scenes) and WH-VQA, a customized warehouse benchmark with many visually similar objects built with Isaac Sim, emphasizing contextual reasoning. Across the two datasets, STaR consistently outperforms strong baselines, achieving higher success rates and markedly lower spatial error. We further deploy STaR on a real Husky wheeled robot in both indoor and outdoor environments, demonstrating robust long horizon reasoning, scalability, and practical utility. Project Website: https://trailab.github.io/STaR-website/
♻ ☆ RM-RL: Role-Model Reinforcement Learning for Precise Robot Manipulation
Precise robot manipulation is critical for fine-grained applications such as chemical and biological experiments, where even small errors (e.g., reagent spillage) can invalidate an entire task. Existing approaches often rely on pre-collected expert demonstrations and train policies via imitation learning (IL) or offline reinforcement learning (RL). However, obtaining high-quality demonstrations for precision tasks is difficult and time-consuming, while offline RL commonly suffers from distribution shifts and low data efficiency. We introduce a Role-Model Reinforcement Learning (RM-RL) framework that unifies online and offline training in real-world environments. The key idea is a role-model strategy that automatically generates labels for online training data using approximately optimal actions, eliminating the need for human demonstrations. RM-RL reformulates policy learning as supervised training, reducing instability from distribution mismatch and improving efficiency. A hybrid training scheme further leverages online role-model data for offline reuse, enhancing data efficiency through repeated sampling. Extensive experiments show that RM-RL converges faster and more stably than existing RL methods, yielding significant gains in real-world manipulation: 53% improvement in translation accuracy and 20% in rotation accuracy. Finally, we demonstrate the successful execution of a challenging task, precisely placing a cell plate onto a shelf, highlighting the framework's effectiveness where prior methods fail.
♻ ☆ A Multi-Fidelity Control Variate Approach for Policy Gradient Estimation
Many reinforcement learning (RL) algorithms are impractical for training in operational systems or computationally expensive high-fidelity simulations, as they require large amounts of data. Meanwhile, low-fidelity simulators, e.g., reduced-order models, heuristic rewards, or learned world models, can cheaply provide useful data, even if they are too coarse for zero-shot transfer. We propose multi-fidelity policy gradients (MFPGs), a sample-efficient RL framework that mixes scarce target-environment data with a control variate formed from abundant low-fidelity simulation data to construct an unbiased, variance-reduced estimator for on-policy policy gradients. We instantiate the framework with a practical, multi-fidelity variant of the classical REINFORCE algorithm. Under standard assumptions, the MFPG estimator guarantees asymptotic convergence to locally optimal policies in the target environment and achieves faster finite-sample convergence than standard REINFORCE. We evaluate MFPG on robotics benchmark tasks with limited high-fidelity data but abundant off-dynamics, low-fidelity data. When low-fidelity data are neutral or beneficial and dynamics gaps are mild-moderate, MFPG is, among the evaluated off-dynamics RL and low-fidelity-only approaches, the only method that consistently achieves statistically significant improvements over a high-fidelity-only baseline. When low-fidelity data become harmful, MFPG exhibits the strongest robustness, whereas strong off-dynamics RL methods exploit low-fidelity data aggressively and fail much more severely. An additional experiment with anti-correlated high- and low-fidelity rewards shows MFPG can remain effective even under reward misspecification. MFPG thus offers a reliable paradigm for exploiting cheap low-fidelity data (e.g., for efficient sim-to-real transfer) while managing the trade-off between policy performance and data collection cost.
♻ ☆ Learning to Feel the Future: DreamTacVLA for Contact-Rich Manipulation
Vision-Language-Action (VLA) models have shown remarkable generalization by mapping web-scale knowledge to robotic control, yet they remain blind to physical contact. Consequently, they struggle with contact-rich manipulation tasks that require reasoning about force, texture, and slip. While some approaches incorporate low-dimensional tactile signals, they fail to capture the high-resolution dynamics essential for such interactions. To address this limitation, we introduce DreamTacVLA, a framework that grounds VLA models in contact physics by learning to feel the future. Our model adopts a hierarchical perception scheme in which high-resolution tactile images serve as micro-vision inputs coupled with wrist-camera local vision and third-person macro vision. To reconcile these multi-scale sensory streams, we first train a unified policy with a Hierarchical Spatial Alignment (HSA) loss that aligns tactile tokens with their spatial counterparts in the wrist and third-person views. To further deepen the model's understanding of fine-grained contact dynamics, we finetune the system with a tactile world model that predicts future tactile signals. To mitigate tactile data scarcity and the wear-prone nature of tactile sensors, we construct a hybrid large-scale dataset sourced from both high-fidelity digital twin and real-world experiments. By anticipating upcoming tactile states, DreamTacVLA acquires a rich model of contact physics and conditions its actions on both real observations and imagined consequences. Across contact-rich manipulation tasks, it outperforms state-of-the-art VLA baselines, achieving up to 95% success, highlighting the importance of understanding physical contact for robust, touch-aware robotic agents.
♻ ☆ Efficient Multi-Robot Motion Planning for Manifold-Constrained Manipulators by Randomized Scheduling and Informed Path Generation
Multi-robot motion planning for high degree-of-freedom manipulators in shared, constrained, and narrow spaces is a complex problem and essential for many scenarios such as construction, surgery, and more. Traditional coupled methods plan directly in the composite configuration space, which scales poorly; decoupled methods, on the other hand, plan separately for each robot but lack completeness. Hybrid methods that obtain paths from individual robots together require the enumeration of many paths before they can find valid composite solutions. This paper introduces Scheduling to Avoid Collisions (StAC), a hybrid approach that more effectively composes paths from individual robots by scheduling (adding stops and coordination motion along all paths) and generates paths that are likely to be feasible by using bidirectional feedback between the scheduler and motion planner for informed sampling. StAC uses 10 to 100 times fewer paths from the low-level planner than state-of-the-art hybrid baselines on challenging problems in manipulator cases.
♻ ☆ Transforming Policy-Car Swerving for Mitigating Stop-and-Go Traffic Waves: A Practice-Oriented Jam-Absorption Driving Strategy
Stop-and-go waves, as a major form of freeway traffic congestion, cause severe and long-lasting adverse effects, including reduced traffic efficiency, increased driving risks, and higher vehicle emissions. Amongst the highway traffic management strategies, jam-absorption driving (JAD), in which a dedicated vehicle performs "slow-in" and "fast-out" maneuvers before being captured by a stop-and-go wave, has been proposed as a potential method for preventing the propagation of such waves. However, most existing JAD strategies remain impractical mainly due to the lack of discussion regarding implementation vehicles and operational conditions. Inspired by real-world observations of police-car swerving behavior, this paper first introduces a Single-Vehicle Two-Detector Jam-Absorption Driving (SVDD-JAD) problem, and then proposes a practical JAD strategy that transforms such behavior into a maneuver capable of suppressing the propagation of an isolated stop-and-go wave. Five key parameters that significantly affect the proposed strategy, namely, JAD speed, inflow traffic speed, wave width, wave speed, and in-wave speed, are identified and systematically analyzed. Using a SUMO-based simulation as an illustrative example, we further demonstrate how these parameters can be measured in practice with two stationary roadside traffic detectors. The results show that the proposed JAD strategy successfully suppresses the propagation of a stop-and-go wave, without triggering a secondary wave. This paper is expected to take a significant step toward making JAD practical, advancing it from a theoretical concept to a feasible and implementable strategy. To promote reproducibility in the transportation domain, we have also open-sourced all the code on our GitHub repository https://github.com/gotrafficgo.
Robotics 77
☆ YOR: Your Own Mobile Manipulator for Generalizable Robotics
Recent advances in robot learning have generated significant interest in capable platforms that may eventually approach human-level competence. This interest, combined with the commoditization of actuators, has propelled growth in low-cost robotic platforms. However, the optimal form factor for mobile manipulation, especially on a budget, remains an open question. We introduce YOR, an open-source, low-cost mobile manipulator that integrates an omnidirectional base, a telescopic vertical lift, and two arms with grippers to achieve whole-body mobility and manipulation. Our design emphasizes modularity, ease of assembly using off-the-shelf components, and affordability, with a bill-of-materials cost under 10,000 USD. We demonstrate YOR's capability by completing tasks that require coordinated whole-body control, bimanual manipulation, and autonomous navigation. Overall, YOR offers competitive functionality for mobile manipulation research at a fraction of the cost of existing platforms. Project website: https://www.yourownrobot.ai/
☆ 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/
☆ Data-Efficient Hierarchical Goal-Conditioned Reinforcement Learning via Normalizing Flows
Hierarchical goal-conditioned reinforcement learning (H-GCRL) provides a powerful framework for tackling complex, long-horizon tasks by decomposing them into structured subgoals. However, its practical adoption is hindered by poor data efficiency and limited policy expressivity, especially in offline or data-scarce regimes. In this work, Normalizing flow-based hierarchical implicit Q-learning (NF-HIQL), a novel framework that replaces unimodal gaussian policies with expressive normalizing flow policies at both the high- and low-levels of the hierarchy is introduced. This design enables tractable log-likelihood computation, efficient sampling, and the ability to model rich multimodal behaviors. New theoretical guarantees are derived, including explicit KL-divergence bounds for Real-valued non-volume preserving (RealNVP) policies and PAC-style sample efficiency results, showing that NF-HIQL preserves stability while improving generalization. Empirically, NF-HIQL is evaluted across diverse long-horizon tasks in locomotion, ball-dribbling, and multi-step manipulation from OGBench. NF-HIQL consistently outperforms prior goal-conditioned and hierarchical baselines, demonstrating superior robustness under limited data and highlighting the potential of flow-based architectures for scalable, data-efficient hierarchical reinforcement learning.
comment: 9 pages, 3 figures, IEEE International Conference on Robotics and Automation 2026
☆ Min-Sum Uniform Coverage Problem by Autonomous Mobile Robots
We study the \textit{min-sum uniform coverage} problem for a swarm of $n$ mobile robots on a given finite line segment and on a circle having finite positive radius, where the circle is given as an input. The robots must coordinate their movements to reach a uniformly spaced configuration that minimizes the total distance traveled by all robots. The robots are autonomous, anonymous, identical, and homogeneous, and operate under the \textit{Look-Compute-Move} (LCM) model with \textit{non-rigid} motion controlled by a fair asynchronous scheduler. They are oblivious and silent, possessing neither persistent memory nor a means of explicit communication. In the \textbf{line-segment setting}, the \textit{min-sum uniform coverage} problem requires placing the robots at uniformly spaced points along the segment so as to minimize the total distance traveled by all robots. In the \textbf{circle setting} for this problem, the robots have to arrange themselves uniformly around the given circle to form a regular $n$-gon. There is no fixed orientation or designated starting vertex, and the goal is to minimize the total distance traveled by all the robots. We present a deterministic distributed algorithm that achieves uniform coverage in the line-segment setting with minimum total movement cost. For the circle setting, we characterize all initial configurations for which the \textit{min-sum uniform coverage} problem is deterministically unsolvable under the considered robot model. For all the other remaining configurations, we provide a deterministic distributed algorithm that achieves uniform coverage while minimizing the total distance traveled. These results characterize the deterministic solvability of min-sum coverage for oblivious robots and achieve optimal cost whenever solvable.
☆ Multi-UAV Trajectory Optimization for Bearing-Only Localization in GPS Denied Environments
Accurate localization of maritime targets by unmanned aerial vehicles (UAVs) remains challenging in GPS-denied environments. UAVs equipped with gimballed electro-optical sensors are typically used to localize targets, however, reliance on these sensors increases mechanical complexity, cost, and susceptibility to single-point failures, limiting scalability and robustness in multi-UAV operations. This work presents a new trajectory optimization framework that enables cooperative target localization using UAVs with fixed, non-gimballed cameras operating in coordination with a surface vessel. This estimation-aware optimization generates dynamically feasible trajectories that explicitly account for mission constraints, platform dynamics, and out-of-frame events. Estimation-aware trajectories outperform heuristic paths by reducing localization error by more than a factor of two, motivating their use in cooperative operations. Results further demonstrate that coordinated UAVs with fixed, non-gimballed cameras achieve localization accuracy that meets or exceeds that of single gimballed systems, while substantially lowering system complexity and cost, enabling scalability, and enhancing mission resilience.
comment: 38 pages, 7 figure, and 6 tables
☆ A receding-horizon multi-contact motion planner for legged robots in challenging environments
We present a novel receding-horizon multi-contact motion planner for legged robots in challenging scenarios, able to plan motions such as chimney climbing, navigating very narrow passages or crossing large gaps. Our approach adds new capabilities to the state of the art, including the ability to reactively re-plan in response to new information, and planning contact locations and whole-body trajectories simultaneously, simplifying the implementation and removing the need for post-processing or complex multi-stage approaches. Our method is more resistant to local minima problems than other potential field based approaches, and our quadratic-program-based posture generator returns nodes more quickly than those of existing algorithms. Rigorous statistical analysis shows that, with short planning horizons (e.g., one step ahead), our planner is faster than the state-of-the-art across all scenarios tested (between 45% and 98% faster on average, depending on the scenario), while planning less efficient motions (requiring 5% fewer to 700% more stance changes on average). In all but one scenario (Chimney Walking), longer planning horizons (e.g., four steps ahead) extended the average planning times (between 73% faster and 400% slower than the state-of-the-art) but resulted in higher quality motion plans (between 8% more and 47% fewer stance changes than the state-of-the-art).
comment: Submitted to Robotics and Autonomous Systems For supplementary video, see https://www.youtube.com/watch?v=RJp8DCmhDa4
☆ Digging for Data: Experiments in Rock Pile Characterization Using Only Proprioceptive Sensing in Excavation
Characterization of fragmented rock piles is a fundamental task in the mining and quarrying industries, where rock is fragmented by blasting, transported using wheel loaders, and then sent for further processing. This field report studies a novel method for estimating the relative particle size of fragmented rock piles from only proprioceptive data collected while digging with a wheel loader. Rather than employ exteroceptive sensors (e.g., cameras or LiDAR sensors) to estimate rock particle sizes, the studied method infers rock fragmentation from an excavator's inertial response during excavation. This paper expands on research that postulated the use of wavelet analysis to construct a unique feature that is proportional to the level of rock fragmentation. We demonstrate through extensive field experiments that the ratio of wavelet features, constructed from data obtained by excavating in different rock piles with different size distributions, approximates the ratio of the mean particle size of the two rock piles. Full-scale excavation experiments were performed with a battery electric, 18-tonne capacity, load-haul-dump (LHD) machine in representative conditions in an operating quarry. The relative particle size estimates generated with the proposed sensing methodology are compared with those obtained from both a vision-based fragmentation analysis tool and from sieving of sampled materials.
comment: Accepted for publication in the IEEE Transactions on Field Robotics
☆ RISE: Self-Improving Robot Policy with Compositional World Model
Despite the sustained scaling on model capacity and data acquisition, Vision-Language-Action (VLA) models remain brittle in contact-rich and dynamic manipulation tasks, where minor execution deviations can compound into failures. While reinforcement learning (RL) offers a principled path to robustness, on-policy RL in the physical world is constrained by safety risk, hardware cost, and environment reset. To bridge this gap, we present RISE, a scalable framework of robotic reinforcement learning via imagination. At its core is a Compositional World Model that (i) predicts multi-view future via a controllable dynamics model, and (ii) evaluates imagined outcomes with a progress value model, producing informative advantages for the policy improvement. Such compositional design allows state and value to be tailored by best-suited yet distinct architectures and objectives. These components are integrated into a closed-loop self-improving pipeline that continuously generates imaginary rollouts, estimates advantages, and updates the policy in imaginary space without costly physical interaction. Across three challenging real-world tasks, RISE yields significant improvement over prior art, with more than +35% absolute performance increase in dynamic brick sorting, +45% for backpack packing, and +35% for box closing, respectively.
comment: Project page: https://opendrivelab.com/kai0-rl/
☆ SQ-CBF: Signed Distance Functions for Numerically Stable Superquadric-Based Safety Filtering
Ensuring safe robot operation in cluttered and dynamic environments remains a fundamental challenge. While control barrier functions provide an effective framework for real-time safety filtering, their performance critically depends on the underlying geometric representation, which is often simplified, leading to either overly conservative behavior or insufficient collision coverage. Superquadrics offer an expressive way to model complex shapes using a few primitives and are increasingly used for robot safety. To integrate this representation into collision avoidance, most existing approaches directly use their implicit functions as barrier candidates. However, we identify a critical but overlooked issue in this practice: the gradients of the implicit SQ function can become severely ill-conditioned, potentially rendering the optimization infeasible and undermining reliable real-time safety filtering. To address this issue, we formulate an SQ-based safety filtering framework that uses signed distance functions as barrier candidates. Since analytical SDFs are unavailable for general SQs, we compute distances using the efficient Gilbert-Johnson-Keerthi algorithm and obtain gradients via randomized smoothing. Extensive simulation and real-world experiments demonstrate consistent collision-free manipulation in cluttered and unstructured scenes, showing robustness to challenging geometries, sensing noise, and dynamic disturbances, while improving task efficiency in teleoperation tasks. These results highlight a pathway toward safety filters that remain precise and reliable under the geometric complexity of real-world environments.
☆ ContactGaussian-WM: Learning Physics-Grounded World Model from Videos
Developing world models that understand complex physical interactions is essential for advancing robotic planning and simulation.However, existing methods often struggle to accurately model the environment under conditions of data scarcity and complex contact-rich dynamic motion.To address these challenges, we propose ContactGaussian-WM, a differentiable physics-grounded rigid-body world model capable of learning intricate physical laws directly from sparse and contact-rich video sequences.Our framework consists of two core components: (1) a unified Gaussian representation for both visual appearance and collision geometry, and (2) an end-to-end differentiable learning framework that differentiates through a closed-form physics engine to infer physical properties from sparse visual observations.Extensive simulations and real-world evaluations demonstrate that ContactGaussian-WM outperforms state-of-the-art methods in learning complex scenarios, exhibiting robust generalization capabilities.Furthermore, we showcase the practical utility of our framework in downstream applications, including data synthesis and real-time MPC.
☆ Enhancing Predictability of Multi-Tenant DNN Inference for Autonomous Vehicles' Perception
Autonomous vehicles (AVs) rely on sensors and deep neural networks (DNNs) to perceive their surrounding environment and make maneuver decisions in real time. However, achieving real-time DNN inference in the AV's perception pipeline is challenging due to the large gap between the computation requirement and the AV's limited resources. Most, if not all, of existing studies focus on optimizing the DNN inference time to achieve faster perception by compressing the DNN model with pruning and quantization. In contrast, we present a Predictable Perception system with DNNs (PP-DNN) that reduce the amount of image data to be processed while maintaining the same level of accuracy for multi-tenant DNNs by dynamically selecting critical frames and regions of interest (ROIs). PP-DNN is based on our key insight that critical frames and ROIs for AVs vary with the AV's surrounding environment. However, it is challenging to identify and use critical frames and ROIs in multi-tenant DNNs for predictable inference. Given image-frame streams, PP-DNN leverages an ROI generator to identify critical frames and ROIs based on the similarities of consecutive frames and traffic scenarios. PP-DNN then leverages a FLOPs predictor to predict multiply-accumulate operations (MACs) from the dynamic critical frames and ROIs. The ROI scheduler coordinates the processing of critical frames and ROIs with multiple DNN models. Finally, we design a detection predictor for the perception of non-critical frames. We have implemented PP-DNN in an ROS-based AV pipeline and evaluated it with the BDD100K and the nuScenes dataset. PP-DNN is observed to significantly enhance perception predictability, increasing the number of fusion frames by up to 7.3x, reducing the fusion delay by >2.6x and fusion-delay variations by >2.3x, improving detection completeness by 75.4% and the cost-effectiveness by up to 98% over the baseline.
comment: 13 pages, 12 figures
☆ Multi-Task Reinforcement Learning of Drone Aerobatics by Exploiting Geometric Symmetries
Flight control for autonomous micro aerial vehicles (MAVs) is evolving from steady flight near equilibrium points toward more aggressive aerobatic maneuvers, such as flips, rolls, and Power Loop. Although reinforcement learning (RL) has shown great potential in these tasks, conventional RL methods often suffer from low data efficiency and limited generalization. This challenge becomes more pronounced in multi-task scenarios where a single policy is required to master multiple maneuvers. In this paper, we propose a novel end-to-end multi-task reinforcement learning framework, called GEAR (Geometric Equivariant Aerobatics Reinforcement), which fully exploits the inherent SO(2) rotational symmetry in MAV dynamics and explicitly incorporates this property into the policy network architecture. By integrating an equivariant actor network, FiLM-based task modulation, and a multi-head critic, GEAR achieves both efficiency and flexibility in learning diverse aerobatic maneuvers, enabling a data-efficient, robust, and unified framework for aerobatic control. GEAR attains a 98.85\% success rate across various aerobatic tasks, significantly outperforming baseline methods. In real-world experiments, GEAR demonstrates stable execution of multiple maneuvers and the capability to combine basic motion primitives to complete complex aerobatics.
☆ Scaling World Model for Hierarchical Manipulation Policies
Vision-Language-Action (VLA) models are promising for generalist robot manipulation but remain brittle in out-of-distribution (OOD) settings, especially with limited real-robot data. To resolve the generalization bottleneck, we introduce a hierarchical Vision-Language-Action framework \our{} that leverages the generalization of large-scale pre-trained world model for robust and generalizable VIsual Subgoal TAsk decomposition VISTA. Our hierarchical framework \our{} consists of a world model as the high-level planner and a VLA as the low-level executor. The high-level world model first divides manipulation tasks into subtask sequences with goal images, and the low-level policy follows the textual and visual guidance to generate action sequences. Compared to raw textual goal specification, these synthesized goal images provide visually and physically grounded details for low-level policies, making it feasible to generalize across unseen objects and novel scenarios. We validate both visual goal synthesis and our hierarchical VLA policies in massive out-of-distribution scenarios, and the performance of the same-structured VLA in novel scenarios could boost from 14% to 69% with the guidance generated by the world model. Results demonstrate that our method outperforms previous baselines with a clear margin, particularly in out-of-distribution scenarios. Project page: \href{https://vista-wm.github.io/}{https://vista-wm.github.io}
☆ RADAR: Benchmarking Vision-Language-Action Generalization via Real-World Dynamics, Spatial-Physical Intelligence, and Autonomous Evaluation
VLA models have achieved remarkable progress in embodied intelligence; however, their evaluation remains largely confined to simulations or highly constrained real-world settings. This mismatch creates a substantial reality gap, where strong benchmark performance often masks poor generalization in diverse physical environments. We identify three systemic shortcomings in current benchmarking practices that hinder fair and reliable model comparison. (1) Existing benchmarks fail to model real-world dynamics, overlooking critical factors such as dynamic object configurations, robot initial states, lighting changes, and sensor noise. (2) Current protocols neglect spatial--physical intelligence, reducing evaluation to rote manipulation tasks that do not probe geometric reasoning. (3) The field lacks scalable fully autonomous evaluation, instead relying on simplistic 2D metrics that miss 3D spatial structure or on human-in-the-loop systems that are costly, biased, and unscalable. To address these limitations, we introduce RADAR (Real-world Autonomous Dynamics And Reasoning), a benchmark designed to systematically evaluate VLA generalization under realistic conditions. RADAR integrates three core components: (1) a principled suite of physical dynamics; (2) dedicated tasks that explicitly test spatial reasoning and physical understanding; and (3) a fully autonomous evaluation pipeline based on 3D metrics, eliminating the need for human supervision. We apply RADAR to audit multiple state-of-the-art VLA models and uncover severe fragility beneath their apparent competence. Performance drops precipitously under modest physical dynamics, with the expectation of 3D IoU declining from 0.261 to 0.068 under sensor noise. Moreover, models exhibit limited spatial reasoning capability. These findings position RADAR as a necessary bench toward reliable and generalizable real-world evaluation of VLA models.
comment: 12 pages, 11 figures, 3 tables
☆ Lie Group Variational Integrator for the Geometrically Exact Rod with Circular Cross-Section Incorporating Cross-Sectional Deformation
In this paper, we derive the continuous space-time equations of motion of a three-dimensional geometrically exact rod, or the Cosserat rod, incorporating planar cross-sectional deformation. We then adopt the Lie group variational integrator technique to obtain a discrete model of the rod incorporating both rotational motion and cross-sectional deformation as well. The resulting discrete model possesses several desirable features: it ensures volume conservation of the discrete elements by considering cross-sectional deformation through a local dilatation factor, it demonstrates the beneficial properties associated with the variational integrator technique, such as the preservation of the rotational configuration, and energy conservation with a bounded error. An exhaustive set of numerical results under various initial conditions of the rod demonstrates the efficacy of the model in replicating the physics of the system.
comment: Submitted to: Computers and Mathematics with Applications
☆ Stability Analysis of Geometric Control for a Canonical Class of Underactuated Aerial Vehicles with Spurious Forces
Standard geometric control relies on force-moment decoupling, an assumption that breaks down in many aerial platforms due to spurious forces naturally induced by control moments. While strategies for such coupled systems have been validated experimentally, a rigorous theoretical certification of their stability is currently missing. This work fills this gap by providing the first formal stability analysis for a generic class of floating rigid bodies subject to spurious forces. We introduce a canonical model and construct a Lyapunov-based proof establishing local exponential stability of the hovering equilibrium. Crucially, the analysis explicitly addresses the structural challenges - specifically the induced non-minimum-phase behavior - that prevent the application of standard cascade arguments.
☆ Developing Neural Network-Based Gaze Control Systems for Social Robots
During multi-party interactions, gaze direction is a key indicator of interest and intent, making it essential for social robots to direct their attention appropriately. Understanding the social context is crucial for robots to engage effectively, predict human intentions, and navigate interactions smoothly. This study aims to develop an empirical motion-time pattern for human gaze behavior in various social situations (e.g., entering, leaving, waving, talking, and pointing) using deep neural networks based on participants' data. We created two video clips-one for a computer screen and another for a virtual reality headset-depicting different social scenarios. Data were collected from 30 participants: 15 using an eye-tracker and 15 using an Oculus Quest 1 headset. Deep learning models, specifically Long Short-Term Memory (LSTM) and Transformers, were used to analyze and predict gaze patterns. Our models achieved 60% accuracy in predicting gaze direction in a 2D animation and 65% accuracy in a 3D animation. Then, the best model was implemented onto the Nao robot; and 36 new participants evaluated its performance. The feedback indicated overall satisfaction, with those experienced in robotics rating the models more favorably.
☆ Towards Learning a Generalizable 3D Scene Representation from 2D Observations
We introduce a Generalizable Neural Radiance Field approach for predicting 3D workspace occupancy from egocentric robot observations. Unlike prior methods operating in camera-centric coordinates, our model constructs occupancy representations in a global workspace frame, making it directly applicable to robotic manipulation. The model integrates flexible source views and generalizes to unseen object arrangements without scene-specific finetuning. We demonstrate the approach on a humanoid robot and evaluate predicted geometry against 3D sensor ground truth. Trained on 40 real scenes, our model achieves 26mm reconstruction error, including occluded regions, validating its ability to infer complete 3D occupancy beyond traditional stereo vision methods.
comment: Paper accepted at ESANN 2026
☆ Design, Development, and Use of Maya Robot as an Assistant for the Therapy/Education of Children with Cancer: a Pilot Study
This study centers around the design and implementation of the Maya Robot, a portable elephant-shaped social robot, intended to engage with children undergoing cancer treatment. Initial efforts were devoted to enhancing the robot's facial expression recognition accuracy, achieving a 98% accuracy through deep neural networks. Two subsequent preliminary exploratory experiments were designed to advance the study's objectives. The first experiment aimed to compare pain levels experienced by children during the injection process, with and without the presence of the Maya robot. Twenty-five children, aged 4 to 9, undergoing cancer treatment participated in this counterbalanced study. The paired T-test results revealed a significant reduction in perceived pain when the robot was actively present in the injection room. The second experiment sought to assess perspectives of hospitalized children and their mothers during engagement with Maya through a game. Forty participants, including 20 children aged 4 to 9 and their mothers, were involved. Post Human-Maya Interactions, UTAUT questionnaire results indicated that children experienced significantly less anxiety than their parents during the interaction and game play. Notably, children exhibited higher trust levels in both the robot and the games, presenting a statistically significant difference in trust levels compared to their parents (P-value < 0.05). This preliminary exploratory study highlights the positive impact of utilizing Maya as an assistant for therapy/education in a clinical setting, particularly benefiting children undergoing cancer treatment. The findings underscore the potential of social robots in pediatric healthcare contexts, emphasizing improved pain management and emotional well-being among young patients.
☆ Safe mobility support system using crowd mapping and avoidance route planning using VLM
Autonomous mobile robots offer promising solutions for labor shortages and increased operational efficiency. However, navigating safely and effectively in dynamic environments, particularly crowded areas, remains challenging. This paper proposes a novel framework that integrates Vision-Language Models (VLM) and Gaussian Process Regression (GPR) to generate dynamic crowd-density maps (``Abstraction Maps'') for autonomous robot navigation. Our approach utilizes VLM's capability to recognize abstract environmental concepts, such as crowd densities, and represents them probabilistically via GPR. Experimental results from real-world trials on a university campus demonstrated that robots successfully generated routes avoiding both static obstacles and dynamic crowds, enhancing navigation safety and adaptability.
☆ Biomimetic Mantaray robot toward the underwater autonomous -- Experimental verification of swimming and diving by flapping motion -
This study presents the development and experimental verification of a biomimetic manta ray robot for underwater autonomous exploration. Inspired by manta rays, the robot uses flapping motion for propulsion to minimize seabed disturbance and enhance efficiency compared to traditional screw propulsion. The robot features pectoral fins driven by servo motors and a streamlined control box to reduce fluid resistance. The control system, powered by a Raspberry Pi 3B, includes an IMU and pressure sensor for real-time monitoring and control. Experiments in a pool assessed the robot's swimming and diving capabilities. Results show stable swimming and diving motions with PD control. The robot is suitable for applications in environments like aquariums and fish nurseries, requiring minimal disturbance and efficient maneuverability. Our findings demonstrate the potential of bio-inspired robotic designs to improve ecological monitoring and underwater exploration.
☆ Semi-Supervised Cross-Domain Imitation Learning
Cross-domain imitation learning (CDIL) accelerates policy learning by transferring expert knowledge across domains, which is valuable in applications where the collection of expert data is costly. Existing methods are either supervised, relying on proxy tasks and explicit alignment, or unsupervised, aligning distributions without paired data, but often unstable. We introduce the Semi-Supervised CDIL (SS-CDIL) setting and propose the first algorithm for SS-CDIL with theoretical justification. Our method uses only offline data, including a small number of target expert demonstrations and some unlabeled imperfect trajectories. To handle domain discrepancy, we propose a novel cross-domain loss function for learning inter-domain state-action mappings and design an adaptive weight function to balance the source and target knowledge. Experiments on MuJoCo and Robosuite show consistent gains over the baselines, demonstrating that our approach achieves stable and data-efficient policy learning with minimal supervision. Our code is available at~ https://github.com/NYCU-RL-Bandits-Lab/CDIL.
comment: Published in Transactions on Machine Learning Research (TMLR)
☆ From Steering to Pedalling: Do Autonomous Driving VLMs Generalize to Cyclist-Assistive Spatial Perception and Planning?
Cyclists often encounter safety-critical situations in urban traffic, highlighting the need for assistive systems that support safe and informed decision-making. Recently, vision-language models (VLMs) have demonstrated strong performance on autonomous driving benchmarks, suggesting their potential for general traffic understanding and navigation-related reasoning. However, existing evaluations are predominantly vehicle-centric and fail to assess perception and reasoning from a cyclist-centric viewpoint. To address this gap, we introduce CyclingVQA, a diagnostic benchmark designed to probe perception, spatio-temporal understanding, and traffic-rule-to-lane reasoning from a cyclist's perspective. Evaluating 31+ recent VLMs spanning general-purpose, spatially enhanced, and autonomous-driving-specialized models, we find that current models demonstrate encouraging capabilities, while also revealing clear areas for improvement in cyclist-centric perception and reasoning, particularly in interpreting cyclist-specific traffic cues and associating signs with the correct navigational lanes. Notably, several driving-specialized models underperform strong generalist VLMs, indicating limited transfer from vehicle-centric training to cyclist-assistive scenarios. Finally, through systematic error analysis, we identify recurring failure modes to guide the development of more effective cyclist-assistive intelligent systems.
comment: Preprint
☆ From Representational Complementarity to Dual Systems: Synergizing VLM and Vision-Only Backbones for End-to-End Driving
Vision-Language-Action (VLA) driving augments end-to-end (E2E) planning with language-enabled backbones, yet it remains unclear what changes beyond the usual accuracy--cost trade-off. We revisit this question with 3--RQ analysis in RecogDrive by instantiating the system with a full VLM and vision-only backbones, all under an identical diffusion Transformer planner. RQ1: At the backbone level, the VLM can introduce additional subspaces upon the vision-only backbones. RQ2: This unique subspace leads to a different behavioral in some long-tail scenario: the VLM tends to be more aggressive whereas ViT is more conservative, and each decisively wins on about 2--3% of test scenarios; With an oracle that selects, per scenario, the better trajectory between the VLM and ViT branches, we obtain an upper bound of 93.58 PDMS. RQ3: To fully harness this observation, we propose HybridDriveVLA, which runs both ViT and VLM branches and selects between their endpoint trajectories using a learned scorer, improving PDMS to 92.10. Finally, DualDriveVLA implements a practical fast--slow policy: it runs ViT by default and invokes the VLM only when the scorer's confidence falls below a threshold; calling the VLM on 15% of scenarios achieves 91.00 PDMS while improving throughput by 3.2x. Code will be released.
comment: 22 pages (10 pages main text + 12 pages appendix), 18 figures
☆ Say, Dream, and Act: Learning Video World Models for Instruction-Driven Robot Manipulation
Robotic manipulation requires anticipating how the environment evolves in response to actions, yet most existing systems lack this predictive capability, often resulting in errors and inefficiency. While Vision-Language Models (VLMs) provide high-level guidance, they cannot explicitly forecast future states, and existing world models either predict only short horizons or produce spatially inconsistent frames. To address these challenges, we propose a framework for fast and predictive video-conditioned action. Our approach first selects and adapts a robust video generation model to ensure reliable future predictions, then applies adversarial distillation for fast, few-step video generation, and finally trains an action model that leverages both generated videos and real observations to correct spatial errors. Extensive experiments show that our method produces temporally coherent, spatially accurate video predictions that directly support precise manipulation, achieving significant improvements in embodiment consistency, spatial referring ability, and task completion over existing baselines. Codes & Models will be released.
☆ (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}.
☆ Omnidirectional Dual-Arm Aerial Manipulator with Proprioceptive Contact Localization for Landing on Slanted Roofs ICRA
Operating drones in urban environments often means they need to land on rooftops, which can have different geometries and surface irregularities. Accurately detecting roof inclination using conventional sensing methods, such as vision-based or acoustic techniques, can be unreliable, as measurement quality is strongly influenced by external factors including weather conditions and surface materials. To overcome these challenges, we propose a novel unmanned aerial manipulator morphology featuring a dual-arm aerial manipulator with an omnidirectional 3D workspace and extended reach. Building on this design, we develop a proprioceptive contact detection and contact localization strategy based on a momentum-based torque observer. This enables the UAM to infer the inclination of slanted surfaces blindly - through physical interaction - prior to touchdown. We validate the approach in flight experiments, demonstrating robust landings on surfaces with inclinations of up to 30.5 degrees and achieving an average surface inclination estimation error of 2.87 degrees over 9 experiments at different incline angles.
comment: Accepted into 2026 International Conference on Robotics and Automation (ICRA) in Vienna
☆ A Unified Experimental Architecture for Informative Path Planning: from Simulation to Deployment with GuadalPlanner
The evaluation of informative path planning algorithms for autonomous vehicles is often hindered by fragmented execution pipelines and limited transferability between simulation and real-world deployment. This paper introduces a unified architecture that decouples high-level decision-making from vehicle-specific control, enabling algorithms to be evaluated consistently across different abstraction levels without modification. The proposed architecture is realized through GuadalPlanner, which defines standardized interfaces between planning, sensing, and vehicle execution. It is an open and extensible research tool that supports discrete graph-based environments and interchangeable planning strategies, and is built upon widely adopted robotics technologies, including ROS2, MAVLink, and MQTT. Its design allows the same algorithmic logic to be deployed in fully simulated environments, software-in-the-loop configurations, and physical autonomous vehicles using an identical execution pipeline. The approach is validated through a set of experiments, including real-world deployment on an autonomous surface vehicle performing water quality monitoring with real-time sensor feedback.
☆ 3D-Printed Anisotropic Soft Magnetic Coating for Directional Rolling of a Magnetically Actuated Capsule Robot
Capsule robots are promising tools for minimally invasive diagnostics and therapy, with applications from gastrointestinal endoscopy to targeted drug delivery and biopsy sampling. Conventional magnetic capsule robots embed bulky permanent magnets at both ends, reducing the usable cavity by about 10-20 mm and limiting integration of functional modules. We propose a compact, 3D-printed soft capsule robot with a magnetic coating that replaces internal magnets, enabling locomotion via a thin, functional shell while preserving the entire interior cavity as a continuous volume for medical payloads. The compliant silicone-magnetic composite also improves swallowability, even with a slightly larger capsule size. Magnetostatic simulations and experiments confirm that programmed NSSN/SNNS pole distributions provide strong anisotropy and reliable torque generation, enabling stable bidirectional rolling, omnidirectional steering, climbing on 7.5 degree inclines, and traversal of 5 mm protrusions. Rolling motion is sustained when the magnetic field at the capsule reaches at least 0.3 mT, corresponding to an effective actuation depth of 30 mm in our setup. Future work will optimize material composition, coating thickness, and magnetic layouts to enhance force output and durability, while next-generation robotic-arm-based field generators with closed-loop feedback will address nonlinearities and expand maneuverability. Together, these advances aim to transition coating-based capsule robots toward reliable clinical deployment and broaden their applications in minimally invasive diagnostics and therapy.
comment: Submitted for review at IEEE/ASME Advanced Intelligenet Mechatronics Conference (2026)
☆ Free-Flying Crew Cooperative Robots on the ISS: A Joint Review of Astrobee, CIMON, and Int-Ball Operations
Intra-vehicular free-flying robots are anticipated to support various work in human spaceflight while working side-by-side with astronauts. Such example of robots includes NASA's Astrobee, DLR's CIMON, and JAXA's Int-Ball, which are deployed on the International Space Station. This paper presents the first joint analyses of these robot's shared experiences, co-authored by their development and operation team members. Despite the different origins and design philosophies, the development and operations of these platforms encountered various convergences. Hence, this paper presents a detailed overview of these robots, presenting their objectives, design, and onboard operations. Hence, joint lessons learned across the lifecycle are presented, from design to on-orbit operations. These lessons learned are anticipated to serve for future development and research as design recommendations.
comment: Author's version of a manuscript accepted at the 2025 International Conference on Space Robotics (iSpaRo25). (c)IEEE
☆ Assessing Vision-Language Models for Perception in Autonomous Underwater Robotic Software
Autonomous Underwater Robots (AURs) operate in challenging underwater environments, including low visibility and harsh water conditions. Such conditions present challenges for software engineers developing perception modules for the AUR software. To successfully carry out these tasks, deep learning has been incorporated into the AUR software to support its operations. However, the unique challenges of underwater environments pose difficulties for deep learning models, which often rely on labeled data that is scarce and noisy. This may undermine the trustworthiness of AUR software that relies on perception modules. Vision-Language Models (VLMs) offer promising solutions for AUR software as they generalize to unseen objects and remain robust in noisy conditions by inferring information from contextual cues. Despite this potential, their performance and uncertainty in underwater environments remain understudied from a software engineering perspective. Motivated by the needs of an industrial partner in assurance and risk management for maritime systems to assess the potential use of VLMs in this context, we present an empirical evaluation of VLM-based perception modules within the AUR software. We assess their ability to detect underwater trash by computing performance, uncertainty, and their relationship, to enable software engineers to select appropriate VLMs for their AUR software.
comment: 10 pages, 5 figures, submitted to ICST 2026
☆ Pitch Angle Control of a Magnetically Actuated Capsule Robot with Nonlinear FEA-based MPC and EKF Multisensory Fusion
Magnetically actuated capsule robots promise minimally invasive diagnosis and therapy in the gastrointestinal (GI) tract, but existing systems largely neglect control of capsule pitch, a degree of freedom critical for contact-rich interaction with inclined gastric walls. This paper presents a nonlinear, model-based framework for magnetic pitch control of an ingestible capsule robot actuated by a four-coil electromagnetic array. Angle-dependent magnetic forces and torques acting on embedded permanent magnets are characterized using three-dimensional finite-element simulations and embedded as lookup tables in a control-oriented rigid-body pitching model with rolling contact and actuator dynamics. A constrained model predictive controller (MPC) is designed to regulate pitch while respecting hardware-imposed current and slew-rate limits. Experiments on a compliant stomach-inspired surface demonstrate robust pitch reorientation from both horizontal and upright configurations, achieving about three to five times faster settling and reduced oscillatory motion than on-off control. Furthermore, an extended Kalman filter (EKF) fusing inertial sensing with intermittent visual measurements enables stable closed-loop control when the camera update rate is reduced from 30 Hz to 1 Hz, emulating clinically realistic imaging constraints. These results establish finite-element-informed MPC with sensor fusion as a scalable strategy for pitch regulation, controlled docking, and future multi-degree-of-freedom capsule locomotion.
comment: This version is submitted for review at IEEE/ASME Transactions on Mechatronics
☆ Flow-Enabled Generalization to Human Demonstrations in Few-Shot Imitation Learning ICRA 2026
Imitation Learning (IL) enables robots to learn complex skills from demonstrations without explicit task modeling, but it typically requires large amounts of demonstrations, creating significant collection costs. Prior work has investigated using flow as an intermediate representation to enable the use of human videos as a substitute, thereby reducing the amount of required robot demonstrations. However, most prior work has focused on the flow, either on the object or on specific points of the robot/hand, which cannot describe the motion of interaction. Meanwhile, relying on flow to achieve generalization to scenarios observed only in human videos remains limited, as flow alone cannot capture precise motion details. Furthermore, conditioning on scene observation to produce precise actions may cause the flow-conditioned policy to overfit to training tasks and weaken the generalization indicated by the flow. To address these gaps, we propose SFCrP, which includes a Scene Flow prediction model for Cross-embodiment learning (SFCr) and a Flow and Cropped point cloud conditioned Policy (FCrP). SFCr learns from both robot and human videos and predicts any point trajectories. FCrP follows the general flow motion and adjusts the action based on observations for precision tasks. Our method outperforms SOTA baselines across various real-world task settings, while also exhibiting strong spatial and instance generalization to scenarios seen only in human videos.
comment: Accepted to ICRA 2026
☆ Morphogenetic Assembly and Adaptive Control for Heterogeneous Modular Robots ICRA 2026
This paper presents a closed-loop automation framework for heterogeneous modular robots, covering the full pipeline from morphological construction to adaptive control. In this framework, a mobile manipulator handles heterogeneous functional modules including structural, joint, and wheeled modules to dynamically assemble diverse robot configurations and provide them with immediate locomotion capability. To address the state-space explosion in large-scale heterogeneous reconfiguration, we propose a hierarchical planner: the high-level planner uses a bidirectional heuristic search with type-penalty terms to generate module-handling sequences, while the low level planner employs A* search to compute optimal execution trajectories. This design effectively decouples discrete configuration planning from continuous motion execution. For adaptive motion generation of unknown assembled configurations, we introduce a GPU accelerated Annealing-Variance Model Predictive Path Integral (MPPI) controller. By incorporating a multi stage variance annealing strategy to balance global exploration and local convergence, the controller enables configuration-agnostic, real-time motion control. Large scale simulations show that the type-penalty term is critical for planning robustness in heterogeneous scenarios. Moreover, the greedy heuristic produces plans with lower physical execution costs than the Hungarian heuristic. The proposed annealing-variance MPPI significantly outperforms standard MPPI in both velocity tracking accuracy and control frequency, achieving real time control at 50 Hz. The framework validates the full-cycle process, including module assembly, robot merging and splitting, and dynamic motion generation.
comment: Accepted by ICRA 2026
☆ LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfer
A long-standing goal in robotics is a generalist policy that can be deployed zero-shot on new robot embodiments without per-embodiment adaptation. Despite large-scale multi-embodiment pre-training, existing Vision-Language-Action models (VLAs) remain tightly coupled to their training embodiments and typically require costly fine-tuning. We introduce Language-Action Pre-training (LAP), a simple recipe that represents low-level robot actions directly in natural language, aligning action supervision with the pre-trained vision-language model's input-output distribution. LAP requires no learned tokenizer, no costly annotation, and no embodiment-specific architectural design. Based on LAP, we present LAP-3B, which to the best of our knowledge is the first VLA to achieve substantial zero-shot transfer to previously unseen robot embodiments without any embodiment-specific fine-tuning. Across multiple novel robots and manipulation tasks, LAP-3B attains over 50% average zero-shot success, delivering roughly a 2x improvement over the strongest prior VLAs. We further show that LAP enables efficient adaptation and favorable scaling, while unifying action prediction and VQA in a shared language-action format that yields additional gains through co-training.
comment: Project website: https://lap-vla.github.io
☆ An Ontology-driven Dynamic Knowledge Base for Uninhabited Ground Vehicles
In this paper, the concept of Dynamic Contextual Mission Data (DCMD) is introduced to develop an ontology-driven dynamic knowledge base for Uninhabited Ground Vehicles (UGVs) at the tactical edge. The dynamic knowledge base with DCMD is added to the UGVs to: support enhanced situation awareness; improve autonomous decision making; and facilitate agility within complex and dynamic environments. As UGVs are heavily reliant on the a priori information added pre-mission, unexpected occurrences during a mission can cause identification ambiguities and require increased levels of user input. Updating this a priori information with contextual information can help UGVs realise their full potential. To address this, the dynamic knowledge base was designed using an ontology-driven representation, supported by near real-time information acquisition and analysis, to provide in-mission on-platform DCMD updates. This was implemented on a team of four UGVs that executed a laboratory based surveillance mission. The results showed that the ontology-driven dynamic representation of the UGV operational environment was machine actionable, producing contextual information to support a successful and timely mission, and contributed directly to the situation awareness.
comment: 10 pages, 11 figures, 2025 Australasian Conference on Robotics and Automation (ACRA 2025)
☆ ReSPEC: A Framework for Online Multispectral Sensor Reconfiguration in Dynamic Environments
Multi-sensor fusion is central to robust robotic perception, yet most existing systems operate under static sensor configurations, collecting all modalities at fixed rates and fidelity regardless of their situational utility. This rigidity wastes bandwidth, computation, and energy, and prevents systems from prioritizing sensors under challenging conditions such as poor lighting or occlusion. Recent advances in reinforcement learning (RL) and modality-aware fusion suggest the potential for adaptive perception, but prior efforts have largely focused on re-weighting features at inference time, ignoring the physical cost of sensor data collection. We introduce a framework that unifies sensing, learning, and actuation into a closed reconfiguration loop. A task-specific detection backbone extracts multispectral features (e.g. RGB, IR, mmWave, depth) and produces quantitative contribution scores for each modality. These scores are passed to an RL agent, which dynamically adjusts sensor configurations, including sampling frequency, resolution, sensing range, and etc., in real time. Less informative sensors are down-sampled or deactivated, while critical sensors are sampled at higher fidelity as environmental conditions evolve. We implement and evaluate this framework on a mobile rover, showing that adaptive control reduces GPU load by 29.3\% with only a 5.3\% accuracy drop compared to a heuristic baseline. These results highlight the potential of resource-aware adaptive sensing for embedded robotic platforms.
comment: 8 pages, 4 figures. This work has been submitted to the IEEE for possible publication
☆ Co-jump: Cooperative Jumping with Quadrupedal Robots via Multi-Agent Reinforcement Learning
While single-agent legged locomotion has witnessed remarkable progress, individual robots remain fundamentally constrained by physical actuation limits. To transcend these boundaries, we introduce Co-jump, a cooperative task where two quadrupedal robots synchronize to execute jumps far beyond their solo capabilities. We tackle the high-impulse contact dynamics of this task under a decentralized setting, achieving synchronization without explicit communication or pre-specified motion primitives. Our framework leverages Multi-Agent Proximal Policy Optimization (MAPPO) enhanced by a progressive curriculum strategy, which effectively overcomes the sparse-reward exploration challenges inherent in mechanically coupled systems. We demonstrate robust performance in simulation and successful transfer to physical hardware, executing multi-directional jumps onto platforms up to 1.5 m in height. Specifically, one of the robots achieves a foot-end elevation of 1.1 m, which represents a 144% improvement over the 0.45 m jump height of a standalone quadrupedal robot, demonstrating superior vertical performance. Notably, this precise coordination is achieved solely through proprioceptive feedback, establishing a foundation for communication-free collaborative locomotion in constrained environments.
comment: 14 pages, 7 figures
☆ Towards Long-Lived Robots: Continual Learning VLA Models via Reinforcement Fine-Tuning
Pretrained on large-scale and diverse datasets, VLA models demonstrate strong generalization and adaptability as general-purpose robotic policies. However, Supervised Fine-Tuning (SFT), which serves as the primary mechanism for adapting VLAs to downstream domains, requires substantial amounts of task-specific data and is prone to catastrophic forgetting. To address these limitations, we propose LifeLong-RFT, a simple yet effective Reinforcement Fine-Tuning (RFT) strategy for VLA models independent of online environmental feedback and pre-trained reward models. By integrating chunking-level on-policy reinforcement learning with the proposed Multi-Dimensional Process Reward (MDPR) mechanism, LifeLong-RFT quantifies the heterogeneous contributions of intermediate action chunks across three dimensions to facilitate policy optimization. Specifically, (1) the Quantized Action Consistency Reward (QACR) ensures accurate action prediction within the discrete action space; (2) the Continuous Trajectory Alignment Reward (CTAR) aligns decoded continuous action chunks with reference trajectories to ensure precise control; (3) the Format Compliance Reward (FCR) guarantees the structural validity of outputs. Comprehensive experiments across SimplerEnv, LIBERO, and real-world tasks demonstrate that LifeLong-RFT exhibits strong performance in multi-task learning. Furthermore, for continual learning on the LIBERO benchmark, our method achieves a 22% gain in average success rate over SFT, while effectively adapting to new tasks using only 20% of the training data. Overall, our method provides a promising post-training paradigm for VLAs.
☆ End-to-End LiDAR optimization for 3D point cloud registration BMVC
LiDAR sensors are a key modality for 3D perception, yet they are typically designed independently of downstream tasks such as point cloud registration. Conventional registration operates on pre-acquired datasets with fixed LiDAR configurations, leading to suboptimal data collection and significant computational overhead for sampling, noise filtering, and parameter tuning. In this work, we propose an adaptive LiDAR sensing framework that dynamically adjusts sensor parameters, jointly optimizing LiDAR acquisition and registration hyperparameters. By integrating registration feedback into the sensing loop, our approach optimally balances point density, noise, and sparsity, improving registration accuracy and efficiency. Evaluations in the CARLA simulation demonstrate that our method outperforms fixed-parameter baselines while retaining generalization abilities, highlighting the potential of adaptive LiDAR for autonomous perception and robotic applications.
comment: 36th British Machine Vision Conference 2025, {BMVC} 2025, Sheffield, UK, November 24-27, 2025. Project page: https://lvsn.github.io/e2e-lidar-registration/
☆ RadarEye: Robust Liquid Level Tracking Using mmWave Radar in Robotic Pouring ICASSP 2026
Transparent liquid manipulation in robotic pouring remains challenging for perception systems: specular/refraction effects and lighting variability degrade visual cues, undermining reliable level estimation. To address this challenge, we introduce RadarEye, a real-time mmWave radar signal processing pipeline for robust liquid level estimation and tracking during the whole pouring process. RadarEye integrates (i) a high-resolution range-angle beamforming module for liquid level sensing and (ii) a physics-informed mid-pour tracker that suppresses multipath to maintain lock on the liquid surface despite stream-induced clutter and source container reflections. The pipeline delivers sub-millisecond latency. In real-robot water-pouring experiments, RadarEye achieves a 0.35 cm median absolute height error at 0.62 ms per update, substantially outperforming vision and ultrasound baselines.
comment: To appear in IEEE ICASSP 2026
☆ LocoVLM: Grounding Vision and Language for Adapting Versatile Legged Locomotion Policies
Recent advances in legged locomotion learning are still dominated by the utilization of geometric representations of the environment, limiting the robot's capability to respond to higher-level semantics such as human instructions. To address this limitation, we propose a novel approach that integrates high-level commonsense reasoning from foundation models into the process of legged locomotion adaptation. Specifically, our method utilizes a pre-trained large language model to synthesize an instruction-grounded skill database tailored for legged robots. A pre-trained vision-language model is employed to extract high-level environmental semantics and ground them within the skill database, enabling real-time skill advisories for the robot. To facilitate versatile skill control, we train a style-conditioned policy capable of generating diverse and robust locomotion skills with high fidelity to specified styles. To the best of our knowledge, this is the first work to demonstrate real-time adaptation of legged locomotion using high-level reasoning from environmental semantics and instructions with instruction-following accuracy of up to 87% without the need for online query to on-the-cloud foundation models.
comment: Project page: https://locovlm.github.io
☆ Can We Really Learn One Representation to Optimize All Rewards?
As machine learning has moved towards leveraging large models as priors for downstream tasks, the community has debated the right form of prior for solving reinforcement learning (RL) problems. If one were to try to prefetch as much computation as possible, they would attempt to learn a prior over the policies for some yet-to-be-determined reward function. Recent work (forward-backward (FB) representation learning) has tried this, arguing that an unsupervised representation learning procedure can enable optimal control over arbitrary rewards without further fine-tuning. However, FB's training objective and learning behavior remain mysterious. In this paper, we demystify FB by clarifying when such representations can exist, what its objective optimizes, and how it converges in practice. We draw connections with rank matching, fitted Q-evaluation, and contraction mapping. Our analysis suggests a simplified unsupervised pre-training method for RL that, instead of enabling optimal control, performs one step of policy improvement. We call our proposed method $\textbf{one-step forward-backward representation learning (one-step FB)}$. Experiments in didactic settings, as well as in $10$ state-based and image-based continuous control domains, demonstrate that one-step FB converges to errors $10^5$ smaller and improves zero-shot performance by $+24\%$ on average. Our project website is available at https://chongyi-zheng.github.io/onestep-fb.
☆ Human Preference Modeling Using Visual Motion Prediction Improves Robot Skill Learning from Egocentric Human Video
We present an approach to robot learning from egocentric human videos by modeling human preferences in a reward function and optimizing robot behavior to maximize this reward. Prior work on reward learning from human videos attempts to measure the long-term value of a visual state as the temporal distance between it and the terminal state in a demonstration video. These approaches make assumptions that limit performance when learning from video. They must also transfer the learned value function across the embodiment and environment gap. Our method models human preferences by learning to predict the motion of tracked points between subsequent images and defines a reward function as the agreement between predicted and observed object motion in a robot's behavior at each step. We then use a modified Soft Actor Critic (SAC) algorithm initialized with 10 on-robot demonstrations to estimate a value function from this reward and optimize a policy that maximizes this value function, all on the robot. Our approach is capable of learning on a real robot, and we show that policies learned with our reward model match or outperform prior work across multiple tasks in both simulation and on the real robot.
comment: 15 pages, 11 figures
☆ MolmoSpaces: A Large-Scale Open Ecosystem for Robot Navigation and Manipulation
Deploying robots at scale demands robustness to the long tail of everyday situations. The countless variations in scene layout, object geometry, and task specifications that characterize real environments are vast and underrepresented in existing robot benchmarks. Measuring this level of generalization requires infrastructure at a scale and diversity that physical evaluation alone cannot provide. We introduce MolmoSpaces, a fully open ecosystem to support large-scale benchmarking of robot policies. MolmoSpaces consists of over 230k diverse indoor environments, ranging from handcrafted household scenes to procedurally generated multiroom houses, populated with 130k richly annotated object assets, including 48k manipulable objects with 42M stable grasps. Crucially, these environments are simulator-agnostic, supporting popular options such as MuJoCo, Isaac, and ManiSkill. The ecosystem supports the full spectrum of embodied tasks: static and mobile manipulation, navigation, and multiroom long-horizon tasks requiring coordinated perception, planning, and interaction across entire indoor environments. We also design MolmoSpaces-Bench, a benchmark suite of 8 tasks in which robots interact with our diverse scenes and richly annotated objects. Our experiments show MolmoSpaces-Bench exhibits strong sim-to-real correlation (R = 0.96, \r{ho} = 0.98), confirm newer and stronger zero-shot policies outperform earlier versions in our benchmarks, and identify key sensitivities to prompt phrasing, initial joint positions, and camera occlusion. Through MolmoSpaces and its open-source assets and tooling, we provide a foundation for scalable data generation, policy training, and benchmark creation for robot learning research.
☆ ExtremControl: Low-Latency Humanoid Teleoperation with Direct Extremity Control
Building a low-latency humanoid teleoperation system is essential for collecting diverse reactive and dynamic demonstrations. However, existing approaches rely on heavily pre-processed human-to-humanoid motion retargeting and position-only PD control, resulting in substantial latency that severely limits responsiveness and prevents tasks requiring rapid feedback and fast reactions. To address this problem, we propose ExtremControl, a low latency whole-body control framework that: (1) operates directly on SE(3) poses of selected rigid links, primarily humanoid extremities, to avoid full-body retargeting; (2) utilizes a Cartesian-space mapping to directly convert human motion to humanoid link targets; and (3) incorporates velocity feedforward control at low level to support highly responsive behavior under rapidly changing control interfaces. We further provide a unified theoretical formulation of ExtremControl and systematically validate its effectiveness through experiments in both simulation and real-world environments. Building on ExtremControl, we implement a low-latency humanoid teleoperation system that supports both optical motion capture and VR-based motion tracking, achieving end-to-end latency as low as 50ms and enabling highly responsive behaviors such as ping-pong ball balancing, juggling, and real-time return, thereby substantially surpassing the 200ms latency limit observed in prior work.
comment: Project website: https://owenowl.github.io/extremcontrol
☆ H-WM: Robotic Task and Motion Planning Guided by Hierarchical World Model
World models are becoming central to robotic planning and control, as they enable prediction of future state transitions. Existing approaches often emphasize video generation or natural language prediction, which are difficult to directly ground in robot actions and suffer from compounding errors over long horizons. Traditional task and motion planning relies on symbolic logic world models, such as planning domains, that are robot-executable and robust for long-horizon reasoning. However, these methods typically operate independently of visual perception, preventing synchronized symbolic and perceptual state prediction. We propose a Hierarchical World Model (H-WM) that jointly predicts logical and visual state transitions within a unified bilevel framework. H-WM combines a high-level logical world model with a low-level visual world model, integrating the robot-executable, long-horizon robustness of symbolic reasoning with perceptual grounding from visual observations. The hierarchical outputs provide stable and consistent intermediate guidance for long-horizon tasks, mitigating error accumulation and enabling robust execution across extended task sequences. To train H-WM, we introduce a robotic dataset that aligns robot motion with symbolic states, actions, and visual observations. Experiments across vision-language-action (VLA) control policies demonstrate the effectiveness and generality of the approach.
comment: 14 pages, 3 figures
☆ ABot-M0: VLA Foundation Model for Robotic Manipulation with Action Manifold Learning
Building general-purpose embodied agents across diverse hardware remains a central challenge in robotics, often framed as the ''one-brain, many-forms'' paradigm. Progress is hindered by fragmented data, inconsistent representations, and misaligned training objectives. We present ABot-M0, a framework that builds a systematic data curation pipeline while jointly optimizing model architecture and training strategies, enabling end-to-end transformation of heterogeneous raw data into unified, efficient representations. From six public datasets, we clean, standardize, and balance samples to construct UniACT-dataset, a large-scale dataset with over 6 million trajectories and 9,500 hours of data, covering diverse robot morphologies and task scenarios. Unified pre-training improves knowledge transfer and generalization across platforms and tasks, supporting general-purpose embodied intelligence. To improve action prediction efficiency and stability, we propose the Action Manifold Hypothesis: effective robot actions lie not in the full high-dimensional space but on a low-dimensional, smooth manifold governed by physical laws and task constraints. Based on this, we introduce Action Manifold Learning (AML), which uses a DiT backbone to predict clean, continuous action sequences directly. This shifts learning from denoising to projection onto feasible manifolds, improving decoding speed and policy stability. ABot-M0 supports modular perception via a dual-stream mechanism that integrates VLM semantics with geometric priors and multi-view inputs from plug-and-play 3D modules such as VGGT and Qwen-Image-Edit, enhancing spatial understanding without modifying the backbone and mitigating standard VLM limitations in 3D reasoning. Experiments show components operate independently with additive benefits. We will release all code and pipelines for reproducibility and future research.
comment: Project website: https://amap-cvlab.github.io/ABot-Manipulation/ . Code: https://github.com/amap-cvlab/ABot-Manipulation . 22 pages, 10 figures, 10 tables
☆ DD-MDN: Human Trajectory Forecasting with Diffusion-Based Dual Mixture Density Networks and Uncertainty Self-Calibration
Human Trajectory Forecasting (HTF) predicts future human movements from past trajectories and environmental context, with applications in Autonomous Driving, Smart Surveillance, and Human-Robot Interaction. While prior work has focused on accuracy, social interaction modeling, and diversity, little attention has been paid to uncertainty modeling, calibration, and forecasts from short observation periods, which are crucial for downstream tasks such as path planning and collision avoidance. We propose DD-MDN, an end-to-end probabilistic HTF model that combines high positional accuracy, calibrated uncertainty, and robustness to short observations. Using a few-shot denoising diffusion backbone and a dual mixture density network, our method learns self-calibrated residence areas and probability-ranked anchor paths, from which diverse trajectory hypotheses are derived, without predefined anchors or endpoints. Experiments on the ETH/UCY, SDD, inD, and IMPTC datasets demonstrate state-of-the-art accuracy, robustness at short observation intervals, and reliable uncertainty modeling. The code is available at: https://github.com/kav-institute/ddmdn.
♻ ☆ Occlusion-Aware Consistent Model Predictive Control for Robot Navigation in Occluded Obstacle-Dense Environments
Ensuring safety and motion consistency for robot navigation in occluded, obstacle-dense environments is a critical challenge. In this context, this study presents an occlusion-aware Consistent Model Predictive Control (CMPC) strategy. To account for the occluded obstacles, it incorporates adjustable risk regions that represent their potential future locations. Subsequently, dynamic risk boundary constraints are developed online to enhance safety. Based on these constraints, the CMPC constructs multiple locally optimal trajectory branches (each tailored to different risk regions) to strike a balance between safety and performance. A shared consensus segment is generated to ensure smooth transitions between branches without significant velocity fluctuations, preserving motion consistency. To facilitate high computational efficiency and ensure coordination across local trajectories, we use the alternating direction method of multipliers (ADMM) to decompose the CMPC into manageable sub-problems for parallel solving. The proposed strategy is validated through simulations and real-world experiments on an Ackermann-steering robot platform. The results demonstrate the effectiveness of the proposed CMPC strategy through comparisons with baseline approaches in occluded, obstacle-dense environments.
♻ ☆ Proactive Local-Minima-Free Robot Navigation: Blending Motion Prediction with Safe Control
This work addresses the challenge of safe and efficient mobile robot navigation in complex dynamic environments with concave moving obstacles. Reactive safe controllers like Control Barrier Functions (CBFs) design obstacle avoidance strategies based only on the current states of the obstacles, risking future collisions. To alleviate this problem, we use Gaussian processes to learn barrier functions online from multimodal motion predictions of obstacles generated by neural networks trained with energy-based learning. The learned barrier functions are then fed into quadratic programs using modulated CBFs (MCBFs), a local-minimum-free version of CBFs, to achieve safe and efficient navigation. The proposed framework makes two key contributions. First, it develops a prediction-to-barrier function online learning pipeline. Second, it introduces an autonomous parameter tuning algorithm that adapts MCBFs to deforming, prediction-based barrier functions. The framework is evaluated in both simulations and real-world experiments, consistently outperforming baselines and demonstrating superior safety and efficiency in crowded dynamic environments.
comment: Co-first authors: Yifan Xue and Ze Zhang; Accepted by IEEE RA-L 2026
♻ ☆ MOSAIC: Bridging the Sim-to-Real Gap in Generalist Humanoid Motion Tracking and Teleoperation with Rapid Residual Adaptation
Generalist humanoid motion trackers have recently achieved strong simulation metrics by scaling data and training, yet often remain brittle on hardware during sustained teleoperation due to interface- and dynamics-induced errors. We present MOSAIC, an open-source, full-stack system for humanoid motion tracking and whole-body teleoperation across multiple interfaces. MOSAIC first learns a teleoperation-oriented general motion tracker via RL on a multi-source motion bank with adaptive resampling and rewards that emphasize world-frame motion consistency, which is critical for mobile teleoperation. To bridge the sim-to-real interface gap without sacrificing generality, MOSAIC then performs rapid residual adaptation: an interface-specific policy is trained using minimal interface-specific data, and then distilled into the general tracker through an additive residual module, outperforming naive fine-tuning or continual learning. We validate MOSAIC with systematic ablations, out-of-distribution benchmarking, and real-robot experiments demonstrating robust offline motion replay and online long-horizon teleoperation under realistic latency and noise. Project page: baai-humanoid.github.io/MOSAIC.
comment: add project page: training codes and data are open sourced
♻ ☆ Provably Optimal Reinforcement Learning under Safety Filtering
Recent advances in reinforcement learning (RL) enable its use on increasingly complex tasks, but the lack of formal safety guarantees still limits its application in safety-critical settings. A common practical approach is to augment the RL policy with a safety filter that overrides unsafe actions to prevent failures during both training and deployment. However, safety filtering is often perceived as sacrificing performance and hindering the learning process. We show that this perceived safety-performance tradeoff is not inherent and prove, for the first time, that enforcing safety with a sufficiently permissive safety filter does not degrade asymptotic performance. We formalize RL safety with a safety-critical Markov decision process (SC-MDP), which requires categorical, rather than high-probability, avoidance of catastrophic failure states. Additionally, we define an associated filtered MDP in which all actions result in safe effects, thanks to a safety filter that is considered to be a part of the environment. Our main theorem establishes that (i) learning in the filtered MDP is safe categorically, (ii) standard RL convergence carries over to the filtered MDP, and (iii) any policy that is optimal in the filtered MDP-when executed through the same filter-achieves the same asymptotic return as the best safe policy in the SC-MDP, yielding a complete separation between safety enforcement and performance optimization. We validate the theory on Safety Gymnasium with representative tasks and constraints, observing zero violations during training and final performance matching or exceeding unfiltered baselines. Together, these results shed light on a long-standing question in safety-filtered learning and provide a simple, principled recipe for safe RL: train and deploy RL policies with the most permissive safety filter that is available.
comment: Accepted for publication in the proceedings of The International Association for Safe & Ethical AI (IASEAI) 2026; 17 pages, 3 figures
♻ ☆ PA-MPPI: Perception-Aware Model Predictive Path Integral Control for Quadrotor Navigation in Unknown Environments
Quadrotor navigation in unknown environments is critical for practical missions such as search-and-rescue. Solving this problem requires addressing three key challenges: path planning in non-convex free space due to obstacles, satisfying quadrotor-specific dynamics and objectives, and exploring unknown regions to expand the map. Recently, the Model Predictive Path Integral (MPPI) method has emerged as a promising solution to the first two challenges. By leveraging sampling-based optimization, it can effectively handle non-convex free space while directly optimizing over the full quadrotor dynamics, enabling the inclusion of quadrotor-specific costs such as energy consumption. However, MPPI has been limited to tracking control that optimizes trajectories only within a small neighborhood around a reference trajectory, as it lacks the ability to explore unknown regions and plan alternative paths when blocked by large obstacles. To address this limitation, we introduce Perception-Aware MPPI (PA-MPPI). In this approach, perception-awareness is characterized by planning and adapting the trajectory online based on perception objectives. Specifically, when the goal is occluded, PA-MPPI incorporates a perception cost that biases trajectories toward those that can observe unknown regions. This expands the mapped traversable space and increases the likelihood of finding alternative paths to the goal. Through hardware experiments, we demonstrate that PA-MPPI, running at 50 Hz, performs on par with the state-of-the-art quadrotor navigation planner for unknown environments in challenging test scenarios. Furthermore, we show that PA-MPPI can serve as a safe and robust action policy for navigation foundation models, which often provide goal poses that are not directly reachable.
♻ ☆ HuMam: Humanoid Motion Control via End-to-End Deep Reinforcement Learning with Mamba
End-to-end reinforcement learning (RL) for humanoid locomotion is appealing for its compact perception-action mapping, yet practical policies often suffer from training instability, inefficient feature fusion, and high actuation cost. We present HuMam, a state-centric end-to-end RL framework that employs a single-layer Mamba encoder to fuse robot-centric states with oriented footstep targets and a continuous phase clock. The policy outputs joint position targets tracked by a low-level PD loop and is optimized with PPO. A concise six-term reward balances contact quality, swing smoothness, foot placement, posture, and body stability while implicitly promoting energy saving. On the JVRC-1 humanoid in mc-mujoco, HuMam consistently improves learning efficiency, training stability, and overall task performance over a strong feedforward baseline, while reducing power consumption and torque peaks. To our knowledge, this is the first end-to-end humanoid RL controller that adopts Mamba as the fusion backbone, demonstrating tangible gains in efficiency, stability, and control economy.
comment: 12 pages
♻ ☆ Towards Safe Path Tracking Using the Simplex Architecture
Robot navigation in complex environments necessitates controllers that prioritize safety while remaining performant and adaptable. Traditional controllers like Regulated Pure Pursuit, Dynamic Window Approach, and Model-Predictive Path Integral, while reliable, struggle to adapt to dynamic conditions. Reinforcement Learning offers adaptability but state-wise safety guarantees remain challenging and often absent in practice. To address this, we propose a path tracking controller leveraging the Simplex architecture. It combines a Reinforcement Learning controller for adaptiveness and performance with a high-assurance controller providing safety and stability. Our main goal is to provide a safe testbed for the design and evaluation of path-planning algorithms, including machine-learning-based planners. Our contribution is twofold. We firstly discuss general stability and safety considerations for designing controllers using the Simplex architecture. Secondly, we present a Simplex-based path tracking controller. Our simulation results, supported by preliminary in-field tests, demonstrate the controller's effectiveness in maintaining safety while achieving comparable performance to state-of-the-art methods.
♻ ☆ ZebraPose: Zebra Detection and Pose Estimation using only Synthetic Data WACV 2026
Collecting and labeling large real-world wild animal datasets is impractical, costly, error-prone, and labor-intensive. For animal monitoring tasks, as detection, tracking, and pose estimation, out-of-distribution viewpoints (e.g. aerial) are also typically needed but rarely found in publicly available datasets. To solve this, existing approaches synthesize data with simplistic techniques that then necessitate strategies to bridge the synthetic-to-real gap. Therefore, real images, style constraints, complex animal models, or pre-trained networks are often leveraged. In contrast, we generate a fully synthetic dataset using a 3D photorealistic simulator and demonstrate that it can eliminate such needs for detecting and estimating 2D poses of wild zebras. Moreover, existing top-down 2D pose estimation approaches using synthetic data assume reliable detection models. However, these often fail in out-of-distribution scenarios, e.g. those that include wildlife or aerial imagery. Our method overcomes this by enabling the training of both tasks using the same synthetic dataset. Through extensive benchmarks, we show that models trained from scratch exclusively on our synthetic data generalize well to real images. We perform these using multiple real-world and synthetic datasets, pre-trained and randomly initialized backbones, and different image resolutions. Code, results, models, and data can be found athttps://zebrapose.is.tue.mpg.de/.
comment: 17 pages, 5 tables, 13 figures. Published in WACV 2026
♻ ☆ Discrete Variational Autoencoding via Policy Search
Discrete latent bottlenecks in variational autoencoders (VAEs) offer high bit efficiency and can be modeled with autoregressive discrete distributions, enabling parameter-efficient multimodal search with transformers. However, discrete random variables do not allow for exact differentiable parameterization; therefore, discrete VAEs typically rely on approximations, such as Gumbel-Softmax reparameterization or straight-through gradient estimates, or employ high-variance gradient-free methods such as REINFORCE that have had limited success on high-dimensional tasks such as image reconstruction. Inspired by popular techniques in policy search, we propose a training framework for discrete VAEs that leverages the natural gradient of a non-parametric encoder to update the parametric encoder without requiring reparameterization. Our method, combined with automatic step size adaptation and a transformer-based encoder, scales to challenging datasets such as ImageNet and outperforms both approximate reparameterization methods and quantization-based discrete autoencoders in reconstructing high-dimensional data from compact latent spaces.
♻ ☆ Crane Lowering Guidance Using a Attachable Camera Module for Driver Vision Support
Cranes have long been essential equipment for lifting and placing heavy loads in construction projects. This study focuses on the lowering phase of crane operation, the stage in which the load is moved to the desired location. During this phase, a constant challenge exists: the load obstructs the operator's view of the landing point. As a result, operators traditionally have to rely on verbal or gestural instructions from ground personnel, which significantly impacts site safety. To alleviate this constraint, the proposed system incorporates a attachable camera module designed to be attached directly to the load via a suction cup. This module houses a single-board computer, battery, and compact camera. After installation, it streams and processes images of the ground directly below the load in real time to generate installation guidance. Simultaneously, this guidance is transmitted to and monitored by a host computer. Preliminary experiments were conducted by attaching this module to a test object, confirming the feasibility of real-time image acquisition and transmission. This approach has the potential to significantly improve safety on construction sites by providing crane operators with an instant visual reference of hidden landing zones.
comment: Published in the Proceedings of ICCR 2025 (IEEE)
♻ ☆ Neural-Augmented Kelvinlet for Real-Time Soft Tissue Deformation Modeling
Accurate and efficient modeling of soft-tissue interactions is fundamental for advancing surgical simulation, surgical robotics, and model-based surgical automation. To achieve real-time latency, classical Finite Element Method (FEM) solvers are often replaced with neural approximations; however, naively training such models in a fully data-driven manner without incorporating physical priors frequently leads to poor generalization and physically implausible predictions. We present a novel physics-informed neural simulation framework that enables real-time prediction of soft-tissue deformations under complex single- and multi-grasper interactions. Our approach integrates Kelvinlet-based analytical priors with large-scale FEM data, capturing both linear and nonlinear tissue responses. This hybrid design improves predictive accuracy and physical plausibility across diverse neural architectures while maintaining the low-latency performance required for interactive applications. We validate our method on challenging surgical manipulation tasks involving standard laparoscopic grasping tools, demonstrating substantial improvements in deformation fidelity and temporal stability over existing baselines. These results establish Kelvinlet-augmented learning as a principled and computationally efficient paradigm for real-time, physics-aware soft-tissue simulation in surgical AI.
♻ ☆ First Multi-Constellation Observations of Navigation Satellite Signals in the Lunar Domain by Post-Processing L1/L5 IQ Snapshots
The use of Global Navigation Satellite Systems (GNSS) to increase spacecraft autonomy for orbit determination has gained renewed momentum following the Lunar GNSS Receiver Experiment (LuGRE), which demonstrated feasible onboard GPS and Galileo signal reception and tracking at lunar distances. This work processes in-phase and quadrature (IQ) snapshots collected by the LuGRE receiver in cis-lunar space and on the lunar surface to assess multi-frequency, multi-constellation signal availability. Signals from additional systems beyond GPS and Galileo, including RNSS and SBAS constellations, are observable and successfully acquired exclusively in the recorded IQ snapshots. These observations provide the first experimental evidence that signals from multiple constellations, including systems not supported by LuGRE realtime operations, are detectable at unprecedented distances from Earth. Useful observables can be extracted from the IQ snapshots, despite minimal sampling rates, 4-bit quantization, and short durations (200 ms-2 s), through a hybrid coherent/non-coherent acquisition stage compensating for code Doppler. These observations are exploited to tune simulation tools and to perform extended simulation campaigns, showing that the inclusion of additional constellations significantly improves availability; for a 26 dB-Hz acquisition threshold, the fraction of epochs with at least four visible satellites increases from 11% to 46% of the total epoch count. These findings indicate that BeiDou, RNSS, and SBAS signals can substantially enhance GNSS-based autonomy for lunar and cislunar missions.
comment: 13 pages, 9 figures, IEEE Transactions on Aerospace and Electronic Systems
♻ ☆ First-order friction models with bristle dynamics: lumped and distributed formulations
Dynamic models, particularly rate-dependent models, have proven effective in capturing the key phenomenological features of frictional processes, whilst also possessing important mathematical properties that facilitate the design of control and estimation algorithms. However, many rate-dependent formulations are built on empirical considerations, whereas physical derivations may offer greater interpretability. In this context, starting from fundamental physical principles, this paper introduces a novel class of first-order dynamic friction models that approximate the dynamics of a bristle element by inverting the friction characteristic. Amongst the developed models, a specific formulation closely resembling the LuGre model is derived using a simple rheological equation for the bristle element. This model is rigorously analyzed in terms of stability and passivity -- important properties that support the synthesis of observers and controllers. Furthermore, a distributed version, formulated as a hyperbolic partial differential equation (PDE), is presented, which enables the modeling of frictional processes commonly encountered in rolling contact phenomena. The tribological behavior of the proposed description is evaluated through classical experiments and validated against the response predicted by the LuGre model, revealing both notable similarities and key differences.
comment: 15 pages, 9 figures. Under review at IEEE Transactions on Control Systems Technology
♻ ☆ Lateral tracking control of all-wheel steering vehicles with intelligent tires
The accurate characterization of tire dynamics is critical for advancing control strategies in autonomous road vehicles, as tire behavior significantly influences handling and stability through the generation of forces and moments at the tire-road interface. Smart tire technologies have emerged as a promising tool for sensing key variables such as road friction, tire pressure, and wear states, and for estimating kinematic and dynamic states like vehicle speed and tire forces. However, most existing estimation and control algorithms rely on empirical correlations or machine learning approaches, which require extensive calibration and can be sensitive to variations in operating conditions. In contrast, model-based techniques, which leverage infinite-dimensional representations of tire dynamics using partial differential equations (PDEs), offer a more robust approach. This paper proposes a novel model-based, output-feedback lateral tracking control strategy for all-wheel steering vehicles that integrates distributed tire dynamics with smart tire technologies. The primary contributions include the suppression of micro-shimmy phenomena at low speeds and path-following via force control, achieved through the estimation of tire slip angles, vehicle kinematics, and lateral tire forces. The proposed controller and observer are based on formulations using ODE-PDE systems, representing rigid body dynamics and distributed tire behavior. This work marks the first rigorous control strategy for vehicular systems equipped with distributed tire representations in conjunction with smart tire technologies.
comment: 16 pages, 12 figures. Under review at IEEE Transactions on Intelligent Vehicles
♻ ☆ HOGraspFlow: Taxonomy-Aware Hand-Object Retargeting for Multi-Modal SE(3) Grasp Generation ICRA 2026
We propose Hand-Object\emph{(HO)GraspFlow}, an affordance-centric approach that retargets a single RGB with hand-object interaction (HOI) into multi-modal executable parallel jaw grasps without explicit geometric priors on target objects. Building on foundation models for hand reconstruction and vision, we synthesize $SE(3)$ grasp poses with denoising flow matching (FM), conditioned on the following three complementary cues: RGB foundation features as visual semantics, HOI contact reconstruction, and taxonomy-aware prior on grasp types. Our approach demonstrates high fidelity in grasp synthesis without explicit HOI contact input or object geometry, while maintaining strong contact and taxonomy recognition. Another controlled comparison shows that \emph{HOGraspFlow} consistently outperforms diffusion-based variants (\emph{HOGraspDiff}), achieving high distributional fidelity and more stable optimization in $SE(3)$. We demonstrate a reliable, object-agnostic grasp synthesis from human demonstrations in real-world experiments, where an average success rate of over $83\%$ is achieved. Code: https://github.com/YitianShi/HOGraspFlow
comment: Accepted to ICRA 2026
♻ ☆ 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 SEC filings and 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 work 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. Our evaluation of rule-based Nav2 navigation shows that current approaches are not economically viable: the contribution margin is -22.81/run (AMCL) and -12.87/run (GPS), resulting in no break-even point. We challenge the community to develop navigation policies that achieve economic viability on CostNav. We remain method-agnostic, evaluating success solely on the metric of cost rather than the underlying architecture. All resources are available at https://github.com/worv-ai/CostNav.
♻ ☆ WorldArena: A Unified Benchmark for Evaluating Perception and Functional Utility of Embodied World Models
While world models have emerged as a cornerstone of embodied intelligence by enabling agents to reason about environmental dynamics through action-conditioned prediction, their evaluation remains fragmented. Current evaluation of embodied world models has largely focused on perceptual fidelity (e.g., video generation quality), overlooking the functional utility of these models in downstream decision-making tasks. In this work, we introduce WorldArena, a unified benchmark designed to systematically evaluate embodied world models across both perceptual and functional dimensions. WorldArena assesses models through three dimensions: video perception quality, measured with 16 metrics across six sub-dimensions; embodied task functionality, which evaluates world models as data engines, policy evaluators, and action planners integrating with subjective human evaluation. Furthermore, we propose EWMScore, a holistic metric integrating multi-dimensional performance into a single interpretable index. Through extensive experiments on 14 representative models, we reveal a significant perception-functionality gap, showing that high visual quality does not necessarily translate into strong embodied task capability. WorldArena benchmark with the public leaderboard is released at https://world-arena.ai, providing a framework for tracking progress toward truly functional world models in embodied AI.
♻ ☆ Localized Graph-Based Neural Dynamics Models for Terrain Manipulation
Predictive models can be particularly helpful for robots to effectively manipulate terrains in construction sites and extraterrestrial surfaces. However, terrain state representations become extremely high-dimensional especially to capture fine-resolution details and when depth is unknown or unbounded. This paper introduces a learning-based approach for terrain dynamics modeling and manipulation, leveraging the Graph-based Neural Dynamics (GBND) framework to represent terrain deformation as motion of a graph of particles. Based on the principle that the moving portion of a terrain is usually localized, our approach builds a large terrain graph (potentially millions of particles) but only identifies a very small active subgraph (hundreds of particles) for predicting the outcomes of robot-terrain interaction. To minimize the size of the active subgraph we introduce a learning-based approach that identifies a small region of interest (RoI) based on the robot's control inputs and the current scene. We also introduce a novel domain boundary feature encoding that allows GBNDs to perform accurate dynamics prediction in the RoI interior while avoiding particle penetration through RoI boundaries. Our proposed method is both orders of magnitude faster than naive GBND and it achieves better overall prediction accuracy. We further evaluated our framework on excavation and shaping tasks on terrain with different granularity.
♻ ☆ OAT: Ordered Action Tokenization
Autoregressive policies offer a compelling foundation for scalable robot learning by enabling discrete abstraction, token-level reasoning, and flexible inference. However, applying autoregressive modeling to continuous robot actions requires an effective action tokenization scheme. Existing approaches either rely on analytical discretization methods that produce prohibitively long token sequences, or learned latent tokenizers that lack structure, limiting their compatibility with next-token prediction. In this work, we identify three desiderata for action tokenization - high compression, total decodability, and a left-to-right causally ordered token space - and introduce Ordered Action Tokenization (OAT), a learned action tokenizer that satisfies all three. OAT discretizes action chunks into an ordered sequence of tokens using transformer with registers, finite scalar quantization, and ordering-inducing training mechanisms. The resulting token space aligns naturally with autoregressive generation and enables prefix-based detokenization, yielding an anytime trade-off between inference cost and action fidelity. Across more than 20 tasks spanning four simulation benchmarks and real-world settings, autoregressive policies equipped with OAT consistently outperform prior tokenization schemes and diffusion-based baselines, while offering significantly greater flexibility at inference time.
♻ ☆ Self-Augmented Robot Trajectory: Efficient Imitation Learning via Safe Self-augmentation with Demonstrator-annotated Precision
Imitation learning is a promising paradigm for training robot agents; however, standard approaches typically require substantial data acquisition -- via numerous demonstrations or random exploration -- to ensure reliable performance. Although exploration reduces human effort, it lacks safety guarantees and often results in frequent collisions -- particularly in clearance-limited tasks (e.g., peg-in-hole) -- thereby, necessitating manual environmental resets and imposing additional human burden. This study proposes Self-Augmented Robot Trajectory (SART), a framework that enables policy learning from a single human demonstration, while safely expanding the dataset through autonomous augmentation. SART consists of two stages: (1) human teaching only once, where a single demonstration is provided and precision boundaries -- represented as spheres around key waypoints -- are annotated, followed by one environment reset; (2) robot self-augmentation, where the robot generates diverse, collision-free trajectories within these boundaries and reconnects to the original demonstration. This design improves the data collection efficiency by minimizing human effort while ensuring safety. Extensive evaluations in simulation and real-world manipulation tasks show that SART achieves substantially higher success rates than policies trained solely on human-collected demonstrations. Video results available at https://sites.google.com/view/sart-il .
comment: 21 pages, 10 figures, Advanced Robotics accepted 2026.02.03
♻ ☆ RoboSubtaskNet: Temporal Sub-task Segmentation for Human-to-Robot Skill Transfer in Real-World Environments
Temporally locating and classifying fine-grained sub-task segments in long, untrimmed videos is crucial to safe human-robot collaboration. Unlike generic activity recognition, collaborative manipulation requires sub-task labels that are directly robot-executable. We present RoboSubtaskNet, a multi-stage human-to-robot sub-task segmentation framework that couples attention-enhanced I3D features (RGB plus optical flow) with a modified MS-TCN employing a Fibonacci dilation schedule to capture better short-horizon transitions such as reach-pick-place. The network is trained with a composite objective comprising cross-entropy and temporal regularizers (truncated MSE and a transition-aware term) to reduce over-segmentation and to encourage valid sub-task progressions. To close the gap between vision benchmarks and control, we introduce RoboSubtask, a dataset of healthcare and industrial demonstrations annotated at the sub-task level and designed for deterministic mapping to manipulator primitives. Empirically, RoboSubtaskNet outperforms MS-TCN and MS-TCN++ on GTEA and our RoboSubtask benchmark (boundary-sensitive and sequence metrics), while remaining competitive on the long-horizon Breakfast benchmark. Specifically, RoboSubtaskNet attains F1 @ 50 = 79.5%, Edit = 88.6%, Acc = 78.9% on GTEA; F1 @ 50 = 30.4%, Edit = 52.0%, Acc = 53.5% on Breakfast; and F1 @ 50 = 94.2%, Edit = 95.6%, Acc = 92.2% on RoboSubtask. We further validate the full perception-to-execution pipeline on a 7-DoF Kinova Gen3 manipulator, achieving reliable end-to-end behavior in physical trials (overall task success approx 91.25%). These results demonstrate a practical path from sub-task level video understanding to deployed robotic manipulation in real-world settings.
♻ ☆ BagelVLA: Enhancing Long-Horizon Manipulation via Interleaved Vision-Language-Action Generation
Equipping embodied agents with the ability to reason about tasks, foresee physical outcomes, and generate precise actions is essential for general-purpose manipulation. While recent Vision-Language-Action (VLA) models have leveraged pre-trained foundation models, they typically focus on either linguistic planning or visual forecasting in isolation. These methods rarely integrate both capabilities simultaneously to guide action generation, leading to suboptimal performance in complex, long-horizon manipulation tasks. To bridge this gap, we propose BagelVLA, a unified model that integrates linguistic planning, visual forecasting, and action generation within a single framework. Initialized from a pretrained unified understanding and generative model, BagelVLA is trained to interleave textual reasoning and visual prediction directly into the action execution loop. To efficiently couple these modalities, we introduce Residual Flow Guidance (RFG), which initializes from current observation and leverages single-step denoising to extract predictive visual features, guiding action generation with minimal latency. Extensive experiments demonstrate that BagelVLA outperforms existing baselines by a significant margin on multiple simulated and real-world benchmarks, particularly in tasks requiring multi-stage reasoning.
♻ ☆ Robotic Depowdering for Additive Manufacturing Via Pose Tracking
With the rapid development of powder-based additive manufacturing, depowdering, a process of removing unfused powder that covers 3D-printed parts, has become a major bottleneck to further improve its productiveness. Traditional manual depowdering is extremely time-consuming and costly, and some prior automated systems either require pre-depowdering or lack adaptability to different 3D-printed parts. To solve these problems, we introduce a robotic system that automatically removes unfused powder from the surface of 3D-printed parts. The key component is a visual perception system, which consists of a pose-tracking module that tracks the 6D pose of powder-occluded parts in real-time, and a progress estimation module that estimates the depowdering completion percentage. The tracking module can be run efficiently on a laptop CPU at up to 60 FPS. Experiments show that our depowdering system can remove unfused powder from the surface of various 3D-printed parts without causing any damage. To the best of our knowledge, this is one of the first vision-based robotic depowdering systems that adapt to parts with various shapes without the need for pre-depowdering.
comment: Github link: https://github.com/zhenweil/Robotic-Depowdering-for-Additive-Manufacturing-Via-Pose-Tracking Video link: https://www.youtube.com/watch?v=AUIkyULAhqM
♻ ☆ Dexterous Manipulation Policies from RGB Human Videos via 3D Hand-Object Trajectory Reconstruction
Multi-finger robotic hand manipulation and grasping are challenging due to the high-dimensional action space and the difficulty of acquiring large-scale training data. Existing approaches largely rely on human teleoperation with wearable devices or specialized sensing equipment to capture hand-object interactions, which limits scalability. In this work, we propose VIDEOMANIP, a device-free framework that learns dexterous manipulation directly from RGB human videos. Leveraging recent advances in computer vision, VIDEOMANIP reconstructs explicit 3D robot-object trajectories from monocular videos by estimating human hand poses, object meshes, and retargets the reconstructed human motions to robotic hands for manipulation learning. To make the reconstructed robot data suitable for dexterous manipulation training, we introduce hand-object contact optimization with interaction-centric grasp modeling, as well as a demonstration synthesis strategy that generates diverse training trajectories from a single video, enabling generalizable policy learning without additional robot demonstrations. In simulation, the learned grasping model achieves a 70.25% success rate across 20 diverse objects using the Inspire Hand. In the real world, manipulation policies trained from RGB videos achieve an average 62.86% success rate across seven tasks using the LEAP Hand, outperforming retargeting-based methods by 15.87%. Project videos are available at videomanip.github.io.
♻ ☆ Towards Dynamic Quadrupedal Gaits: A Symmetry-Guided RL Hierarchy Enables Free Gait Transitions at Varying Speeds
Quadrupedal robots exhibit a wide range of viable gaits, but generating specific footfall sequences often requires laborious expert tuning of numerous variables, such as touch-down and lift-off events and holonomic constraints for each leg. This paper presents a unified reinforcement learning framework for generating versatile quadrupedal gaits by leveraging the intrinsic symmetries and velocity-period relationship of dynamic legged systems. We propose a symmetry-guided reward function design that incorporates temporal, morphological, and time-reversal symmetries. By focusing on preserved symmetries and natural dynamics, our approach eliminates the need for predefined trajectories, enabling smooth transitions between diverse locomotion patterns such as trotting, bounding, half-bounding, and galloping. Implemented on the Unitree Go2 robot, our method demonstrates robust performance across a range of speeds in both simulations and hardware tests, significantly improving gait adaptability without extensive reward tuning or explicit foot placement control. This work provides insights into dynamic locomotion strategies and underscores the crucial role of symmetries in robotic gait design.
comment: This work is build on reusing the main novel concept from arXiv:2403.10723. Based on the reviews we accept while submitting this work, we decided to resubmit this work as a replacement of the linked work
♻ ☆ Exploring Immersive Social-Physical Interaction with Virtual Characters through Coordinated Robotic Encountered-Type Contact
This work presents novel robot-mediated immersive experiences enabled by an encountered-type haptic display (ETHD) that introduces direct physical contact in virtual environments. We focus on social-physical interactions, a class of interaction associated with meaningful human outcomes in prior human-robot interaction (HRI) research. We explore the implementation of this interaction paradigm in immersive virtual environments through an object handover, fist bump, and high five with a virtual character. Extending this HRI paradigm into immersive environments enables the study of how physically grounded robotic contact and virtual augmentation jointly shape these novel social-physical interaction experiences. To support this investigation, we introduce ETHOS (Encountered-Type Haptics for On-demand Social interaction), an experimental platform integrating a torque-controlled manipulator and interchangeable props with a headset-mediated virtual experience. ETHOS enables co-located physical interaction through marker-based physical-virtual registration while concealing the robot behind the virtual environment, decoupling contact from visible robot embodiment. Both technical characterization, through spatial alignment and interaction latency tests, and experiential evaluation, through a 55 participant user study, were completed. Overall, the findings demonstrate the feasibility and experiential value of robot-mediated social-physical interaction in VR and motivate further development of dynamic encountered-type approaches for immersive HRI.
comment: 14 pages
♻ ☆ ETHOS: A Robotic Encountered-Type Haptic Display for Social Interaction in Virtual Reality
We present ETHOS (Encountered-Type Haptics for On-demand Social Interaction), a dynamic encountered-type haptic display (ETHD) that enables natural physical contact in virtual reality (VR) during social interactions such as handovers, fist bumps, and high-fives. The system integrates a torque-controlled robotic manipulator with interchangeable passive props (silicone hand replicas and a baton), marker-based physical-virtual registration via a ChArUco board, and a safety monitor that gates motion based on the user's head and hand pose. We introduce two control strategies: (i) a static mode that presents a stationary prop aligned with its virtual counterpart, consistent with prior ETHD baselines, and (ii) a dynamic mode that continuously updates prop position by exponentially blending an initial mid-point trajectory with real-time hand tracking, generating a unique contact point for each interaction. Bench tests show static colocation accuracy of 5.09 +/- 0.94 mm, while user interactions achieved temporal alignment with an average contact latency of 28.53 +/- 31.21 ms across all interaction and control conditions. These results demonstrate the feasibility of recreating socially meaningful haptics in VR. By incorporating essential safety and control mechanisms, ETHOS establishes a practical foundation for high-fidelity, dynamic interpersonal interactions in virtual environments.
comment: This work was combined with another paper that extends the technical characterization of our system to include an experiential evaluation through a user study. The combined work creates a unified representation of our contributions. With the combined paper to be published, this work is not longer needed
♻ ☆ LightTact: A Visual-Tactile Fingertip Sensor for Deformation-Independent Contact Sensing
Contact often occurs without macroscopic surface deformation, such as during interaction with liquids, semi-liquids, or ultra-soft materials. However, most existing tactile sensors rely on deformation to infer contact, making such light-contact interactions difficult to perceive robustly. To address this, we present LightTact, a visual-tactile fingertip sensor that makes contact directly visible via a deformation-independent principle. LightTact features an ambient-blocking optical configuration that suppresses both external light and internal illumination at non-contact regions, while transmitting only the scattered light generated at true contacts. As a result, LightTact produces high-contrast raw images in which non-contact pixels remain near-black (mean gray value < 3) and contact pixels preserve the natural appearance of the contacting surface. Built on this, LightTact achieves accurate pixel-level contact segmentation that is robust to material properties, contact force, surface appearance, and environmental lighting. We further demonstrate that LightTact unlocks new robotic manipulation behaviors that require detection of extremely light contact, including water spreading, facial-cream dipping, and soft thin-film interaction. In addition, we show that LightTact's spatially aligned visual-tactile images can be directly interpreted by vision-language models.
comment: Project website: https://linchangyi1.github.io/LightTact
Robotics 68
☆ SAGE: Scalable Agentic 3D Scene Generation for Embodied AI
Real-world data collection for embodied agents remains costly and unsafe, calling for scalable, realistic, and simulator-ready 3D environments. However, existing scene-generation systems often rely on rule-based or task-specific pipelines, yielding artifacts and physically invalid scenes. We present SAGE, an agentic framework that, given a user-specified embodied task (e.g., "pick up a bowl and place it on the table"), understands the intent and automatically generates simulation-ready environments at scale. The agent couples multiple generators for layout and object composition with critics that evaluate semantic plausibility, visual realism, and physical stability. Through iterative reasoning and adaptive tool selection, it self-refines the scenes until meeting user intent and physical validity. The resulting environments are realistic, diverse, and directly deployable in modern simulators for policy training. Policies trained purely on this data exhibit clear scaling trends and generalize to unseen objects and layouts, demonstrating the promise of simulation-driven scaling for embodied AI. Code, demos, and the SAGE-10k dataset can be found on the project page here: https://nvlabs.github.io/sage.
comment: Project Page: https://nvlabs.github.io/sage
☆ Decoupled MPPI-Based Multi-Arm Motion Planning
Recent advances in sampling-based motion planning algorithms for high DOF arms leverage GPUs to provide SOTA performance. These algorithms can be used to control multiple arms jointly, but this approach scales poorly. To address this, we extend STORM, a sampling-based model-predictive-control (MPC) motion planning algorithm, to handle multiple robots in a distributed fashion. First, we modify STORM to handle dynamic obstacles. Then, we let each arm compute its own motion plan prefix, which it shares with the other arms, which treat it as a dynamic obstacle. Finally, we add a dynamic priority scheme. The new algorithm, MR-STORM, demonstrates clear empirical advantages over SOTA algorithms when operating with both static and dynamic obstacles.
☆ Learning Agile Quadrotor Flight in the Real World
Learning-based controllers have achieved impressive performance in agile quadrotor flight but typically rely on massive training in simulation, necessitating accurate system identification for effective Sim2Real transfer. However, even with precise modeling, fixed policies remain susceptible to out-of-distribution scenarios, ranging from external aerodynamic disturbances to internal hardware degradation. To ensure safety under these evolving uncertainties, such controllers are forced to operate with conservative safety margins, inherently constraining their agility outside of controlled settings. While online adaptation offers a potential remedy, safely exploring physical limits remains a critical bottleneck due to data scarcity and safety risks. To bridge this gap, we propose a self-adaptive framework that eliminates the need for precise system identification or offline Sim2Real transfer. We introduce Adaptive Temporal Scaling (ATS) to actively explore platform physical limits, and employ online residual learning to augment a simple nominal model. {Based on the learned hybrid model, we further propose Real-world Anchored Short-horizon Backpropagation Through Time (RASH-BPTT) to achieve efficient and robust in-flight policy updates. Extensive experiments demonstrate that our quadrotor reliably executes agile maneuvers near actuator saturation limits. The system evolves a conservative base policy with a peak speed of 1.9 m/s to 7.3 m/s within approximately 100 seconds of flight time. These findings underscore that real-world adaptation serves not merely to compensate for modeling errors, but as a practical mechanism for sustained performance improvement in aggressive flight regimes.
☆ ST4VLA: Spatially Guided Training for Vision-Language-Action Models ICLR 2026
Large vision-language models (VLMs) excel at multimodal understanding but fall short when extended to embodied tasks, where instructions must be transformed into low-level motor actions. We introduce ST4VLA, a dual-system Vision-Language-Action framework that leverages Spatial Guided Training to align action learning with spatial priors in VLMs. ST4VLA includes two stages: (i) spatial grounding pre-training, which equips the VLM with transferable priors via scalable point, box, and trajectory prediction from both web-scale and robot-specific data, and (ii) spatially guided action post-training, which encourages the model to produce richer spatial priors to guide action generation via spatial prompting. This design preserves spatial grounding during policy learning and promotes consistent optimization across spatial and action objectives. Empirically, ST4VLA achieves substantial improvements over vanilla VLA, with performance increasing from 66.1 -> 84.6 on Google Robot and from 54.7 -> 73.2 on WidowX Robot, establishing new state-of-the-art results on SimplerEnv. It also demonstrates stronger generalization to unseen objects and paraphrased instructions, as well as robustness to long-horizon perturbations in real-world settings. These results highlight scalable spatially guided training as a promising direction for robust, generalizable robot learning. Source code, data and models are released at https://internrobotics.github.io/internvla-m1.github.io/
comment: Spatially Training for VLA, Accepted by ICLR 2026
☆ EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration
Human demonstrations offer rich environmental diversity and scale naturally, making them an appealing alternative to robot teleoperation. While this paradigm has advanced robot-arm manipulation, its potential for the more challenging, data-hungry problem of humanoid loco-manipulation remains largely unexplored. We present EgoHumanoid, the first framework to co-train a vision-language-action policy using abundant egocentric human demonstrations together with a limited amount of robot data, enabling humanoids to perform loco-manipulation across diverse real-world environments. To bridge the embodiment gap between humans and robots, including discrepancies in physical morphology and viewpoint, we introduce a systematic alignment pipeline spanning from hardware design to data processing. A portable system for scalable human data collection is developed, and we establish practical collection protocols to improve transferability. At the core of our human-to-humanoid alignment pipeline lies two key components. The view alignment reduces visual domain discrepancies caused by camera height and perspective variation. The action alignment maps human motions into a unified, kinematically feasible action space for humanoid control. Extensive real-world experiments demonstrate that incorporating robot-free egocentric data significantly outperforms robot-only baselines by 51\%, particularly in unseen environments. Our analysis further reveals which behaviors transfer effectively and the potential for scaling human data.
comment: Project page: https://opendrivelab.com/EgoHumanoid
☆ DexImit: Learning Bimanual Dexterous Manipulation from Monocular Human Videos
Data scarcity fundamentally limits the generalization of bimanual dexterous manipulation, as real-world data collection for dexterous hands is expensive and labor-intensive. Human manipulation videos, as a direct carrier of manipulation knowledge, offer significant potential for scaling up robot learning. However, the substantial embodiment gap between human hands and robotic dexterous hands makes direct pretraining from human videos extremely challenging. To bridge this gap and unleash the potential of large-scale human manipulation video data, we propose DexImit, an automated framework that converts monocular human manipulation videos into physically plausible robot data, without any additional information. DexImit employs a four-stage generation pipeline: (1) reconstructing hand-object interactions from arbitrary viewpoints with near-metric scale; (2) performing subtask decomposition and bimanual scheduling; (3) synthesizing robot trajectories consistent with the demonstrated interactions; (4) comprehensive data augmentation for zero-shot real-world deployment. Building on these designs, DexImit can generate large-scale robot data based on human videos, either from the Internet or video generation models. DexImit is capable of handling diverse manipulation tasks, including tool use (e.g., cutting an apple), long-horizon tasks (e.g., making a beverage), and fine-grained manipulations (e.g., stacking cups).
☆ Robo3R: Enhancing Robotic Manipulation with Accurate Feed-Forward 3D Reconstruction
3D spatial perception is fundamental to generalizable robotic manipulation, yet obtaining reliable, high-quality 3D geometry remains challenging. Depth sensors suffer from noise and material sensitivity, while existing reconstruction models lack the precision and metric consistency required for physical interaction. We introduce Robo3R, a feed-forward, manipulation-ready 3D reconstruction model that predicts accurate, metric-scale scene geometry directly from RGB images and robot states in real time. Robo3R jointly infers scale-invariant local geometry and relative camera poses, which are unified into the scene representation in the canonical robot frame via a learned global similarity transformation. To meet the precision demands of manipulation, Robo3R employs a masked point head for sharp, fine-grained point clouds, and a keypoint-based Perspective-n-Point (PnP) formulation to refine camera extrinsics and global alignment. Trained on Robo3R-4M, a curated large-scale synthetic dataset with four million high-fidelity annotated frames, Robo3R consistently outperforms state-of-the-art reconstruction methods and depth sensors. Across downstream tasks including imitation learning, sim-to-real transfer, grasp synthesis, and collision-free motion planning, we observe consistent gains in performance, suggesting the promise of this alternative 3D sensing module for robotic manipulation.
☆ VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model
Pretraining Vision-Language-Action (VLA) policies on internet-scale video is appealing, yet current latent-action objectives often learn the wrong thing: they remain anchored to pixel variation rather than action-relevant state transitions, making them vulnerable to appearance bias, nuisance motion, and information leakage. We introduce VLA-JEPA, a JEPA-style pretraining framework that sidesteps these pitfalls by design. The key idea is \emph{leakage-free state prediction}: a target encoder produces latent representations from future frames, while the student pathway sees only the current observation -- future information is used solely as supervision targets, never as input. By predicting in latent space rather than pixel space, VLA-JEPA learns dynamics abstractions that are robust to camera motion and irrelevant background changes. This yields a simple two-stage recipe -- JEPA pretraining followed by action-head fine-tuning -- without the multi-stage complexity of prior latent-action pipelines. Experiments on LIBERO, LIBERO-Plus, SimplerEnv and real-world manipulation tasks show that VLA-JEPA achieves consistent gains in generalization and robustness over existing methods.
☆ UniVTAC: A Unified Simulation Platform for Visuo-Tactile Manipulation Data Generation, Learning, and Benchmarking
Robotic manipulation has seen rapid progress with vision-language-action (VLA) policies. However, visuo-tactile perception is critical for contact-rich manipulation, as tasks such as insertion are difficult to complete robustly using vision alone. At the same time, acquiring large-scale and reliable tactile data in the physical world remains costly and challenging, and the lack of a unified evaluation platform further limits policy learning and systematic analysis. To address these challenges, we propose UniVTAC, a simulation-based visuo-tactile data synthesis platform that supports three commonly used visuo-tactile sensors and enables scalable and controllable generation of informative contact interactions. Based on this platform, we introduce the UniVTAC Encoder, a visuo-tactile encoder trained on large-scale simulation-synthesized data with designed supervisory signals, providing tactile-centric visuo-tactile representations for downstream manipulation tasks. In addition, we present the UniVTAC Benchmark, which consists of eight representative visuo-tactile manipulation tasks for evaluating tactile-driven policies. Experimental results show that integrating the UniVTAC Encoder improves average success rates by 17.1% on the UniVTAC Benchmark, while real-world robotic experiments further demonstrate a 25% improvement in task success. Our webpage is available at https://univtac.github.io/.
comment: Website: https://univtac.github.io/
☆ Humanoid Factors: Design Principles for AI Humanoids in Human Worlds
Human factors research has long focused on optimizing environments, tools, and systems to account for human performance. Yet, as humanoid robots begin to share our workplaces, homes, and public spaces, the design challenge expands. We must now consider not only factors for humans but also factors for humanoids, since both will coexist and interact within the same environments. Unlike conventional machines, humanoids introduce expectations of human-like behavior, communication, and social presence, which reshape usability, trust, and safety considerations. In this article, we introduce the concept of humanoid factors as a framework structured around four pillars - physical, cognitive, social, and ethical - that shape the development of humanoids to help them effectively coexist and collaborate with humans. This framework characterizes the overlap and divergence between human capabilities and those of general-purpose humanoids powered by AI foundation models. To demonstrate our framework's practical utility, we then apply the framework to evaluate a real-world humanoid control algorithm, illustrating how conventional task completion metrics in robotics overlook key human cognitive and interaction principles. We thus position humanoid factors as a foundational framework for designing, evaluating, and governing sustained human-humanoid coexistence.
☆ A Collision-Free Sway Damping Model Predictive Controller for Safe and Reactive Forestry Crane Navigation ICRA 2026
Forestry cranes operate in dynamic, unstructured outdoor environments where simultaneous collision avoidance and payload sway control are critical for safe navigation. Existing approaches address these challenges separately, either focusing on sway damping with predefined collision-free paths or performing collision avoidance only at the global planning level. We present the first collision-free, sway-damping model predictive controller (MPC) for a forestry crane that unifies both objectives in a single control framework. Our approach integrates LiDAR-based environment mapping directly into the MPC using online Euclidean distance fields (EDF), enabling real-time environmental adaptation. The controller simultaneously enforces collision constraints while damping payload sway, allowing it to (i) replan upon quasi-static environmental changes, (ii) maintain collision-free operation under disturbances, and (iii) provide safe stopping when no bypass exists. Experimental validation on a real forestry crane demonstrates effective sway damping and successful obstacle avoidance. A video can be found at https://youtu.be/tEXDoeLLTxA.
comment: Accepted at ICRA 2026
☆ Perception with Guarantees: Certified Pose Estimation via Reachability Analysis
Agents in cyber-physical systems are increasingly entrusted with safety-critical tasks. Ensuring safety of these agents often requires localizing the pose for subsequent actions. Pose estimates can, e.g., be obtained from various combinations of lidar sensors, cameras, and external services such as GPS. Crucially, in safety-critical domains, a rough estimate is insufficient to formally determine safety, i.e., guaranteeing safety even in the worst-case scenario, and external services might additionally not be trustworthy. We address this problem by presenting a certified pose estimation in 3D solely from a camera image and a well-known target geometry. This is realized by formally bounding the pose, which is computed by leveraging recent results from reachability analysis and formal neural network verification. Our experiments demonstrate that our approach efficiently and accurately localizes agents in both synthetic and real-world experiments.
☆ RoboSubtaskNet: Temporal Sub-task Segmentation for Human-to-Robot Skill Transfer in Real-World Environments
Temporally locating and classifying fine-grained sub-task segments in long, untrimmed videos is crucial to safe human-robot collaboration. Unlike generic activity recognition, collaborative manipulation requires sub-task labels that are directly robot-executable. We present RoboSubtaskNet, a multi-stage human-to-robot sub-task segmentation framework that couples attention-enhanced I3D features (RGB plus optical flow) with a modified MS-TCN employing a Fibonacci dilation schedule to capture better short-horizon transitions such as reach-pick-place. The network is trained with a composite objective comprising cross-entropy and temporal regularizers (truncated MSE and a transition-aware term) to reduce over-segmentation and to encourage valid sub-task progressions. To close the gap between vision benchmarks and control, we introduce RoboSubtask, a dataset of healthcare and industrial demonstrations annotated at the sub-task level and designed for deterministic mapping to manipulator primitives. Empirically, RoboSubtaskNet outperforms MS-TCN and MS-TCN++ on GTEA and our RoboSubtask benchmark (boundary-sensitive and sequence metrics), while remaining competitive on the long-horizon Breakfast benchmark. Specifically, RoboSubtaskNet attains F1 @ 50 = 79.5%, Edit = 88.6%, Acc = 78.9% on GTEA; F1 @ 50 = 30.4%, Edit = 52.0%, Acc = 53.5% on Breakfast; and F1 @ 50 = 94.2%, Edit = 95.6%, Acc = 92.2% on RoboSubtask. We further validate the full perception-to-execution pipeline on a 7-DoF Kinova Gen3 manipulator, achieving reliable end-to-end behavior in physical trials (overall task success approx 91.25%). These results demonstrate a practical path from sub-task level video understanding to deployed robotic manipulation in real-world settings.
☆ Learning Force-Regulated Manipulation with a Low-Cost Tactile-Force-Controlled Gripper
Successfully manipulating many everyday objects, such as potato chips, requires precise force regulation. Failure to modulate force can lead to task failure or irreversible damage to the objects. Humans can precisely achieve this by adapting force from tactile feedback, even within a short period of physical contact. We aim to give robots this capability. However, commercial grippers exhibit high cost or high minimum force, making them unsuitable for studying force-controlled policy learning with everyday force-sensitive objects. We introduce TF-Gripper, a low-cost (~$150) force-controlled parallel-jaw gripper that integrates tactile sensing as feedback. It has an effective force range of 0.45-45N and is compatible with different robot arms. Additionally, we designed a teleoperation device paired with TF-Gripper to record human-applied grasping forces. While standard low-frequency policies can be trained on this data, they struggle with the reactive, contact-dependent nature of force regulation. To overcome this, we propose RETAF (REactive Tactile Adaptation of Force), a framework that decouples grasping force control from arm pose prediction. RETAF regulates force at high frequency using wrist images and tactile feedback, while a base policy predicts end-effector pose and gripper open/close action. We evaluate TF-Gripper and RETAF across five real-world tasks requiring precise force regulation. Results show that compared to position control, direct force control significantly improves grasp stability and task performance. We further show that tactile feedback is essential for force regulation, and that RETAF consistently outperforms baselines and can be integrated with various base policies. We hope this work opens a path for scaling the learning of force-controlled policies in robotic manipulation. Project page: https://force-gripper.github.io .
comment: 10 pages, 11 figures
☆ A Collaborative Safety Shield for Safe and Efficient CAV Lane Changes in Congested On-Ramp Merging
Lane changing in dense traffic is a significant challenge for Connected and Autonomous Vehicles (CAVs). Existing lane change controllers primarily either ensure safety or collaboratively improve traffic efficiency, but do not consider these conflicting objectives together. To address this, we propose the Multi-Agent Safety Shield (MASS), designed using Control Barrier Functions (CBFs) to enable safe and collaborative lane changes. The MASS enables collaboration by capturing multi-agent interactions among CAVs through interaction topologies constructed as a graph using a simple algorithm. Further, a state-of-the-art Multi-Agent Reinforcement Learning (MARL) lane change controller is extended by integrating MASS to ensure safety and defining a customised reward function to prioritise efficiency improvements. As a result, we propose a lane change controller, known as MARL-MASS, and evaluate it in a congested on-ramp merging simulation. The results demonstrate that MASS enables collaborative lane changes with safety guarantees by strictly respecting the safety constraints. Moreover, the proposed custom reward function improves the stability of MARL policies trained with a safety shield. Overall, by encouraging the exploration of a collaborative lane change policy while respecting safety constraints, MARL-MASS effectively balances the trade-off between ensuring safety and improving traffic efficiency in congested traffic. The code for MARL-MASS is available with an open-source licence at https://github.com/hkbharath/MARL-MASS
comment: Accepted in IEEE IV 2026
☆ Acoustic Drone Package Delivery Detection
In recent years, the illicit use of unmanned aerial vehicles (UAVs) for deliveries in restricted area such as prisons became a significant security challenge. While numerous studies have focused on UAV detection or localization, little attention has been given to delivery events identification. This study presents the first acoustic package delivery detection algorithm using a ground-based microphone array. The proposed method estimates both the drone's propeller speed and the delivery event using solely acoustic features. A deep neural network detects the presence of a drone and estimates the propeller's rotation speed or blade passing frequency (BPF) from a mel spectrogram. The algorithm analyzes the BPFs to identify probable delivery moments based on sudden changes before and after a specific time. Results demonstrate a mean absolute error of the blade passing frequency estimator of 16 Hz when the drone is less than 150 meters away from the microphone array. The drone presence detection estimator has a accuracy of 97%. The delivery detection algorithm correctly identifies 96% of events with a false positive rate of 8%. This study shows that deliveries can be identified using acoustic signals up to a range of 100 meters.
☆ RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation ICLR 2026
Advances in large vision-language models (VLMs) have stimulated growing interest in vision-language-action (VLA) systems for robot manipulation. However, existing manipulation datasets remain costly to curate, highly embodiment-specific, and insufficient in coverage and diversity, thereby hindering the generalization of VLA models. Recent approaches attempt to mitigate these limitations via a plan-then-execute paradigm, where high-level plans (e.g., subtasks, trace) are first generated and subsequently translated into low-level actions, but they critically rely on extra intermediate supervision, which is largely absent from existing datasets. To bridge this gap, we introduce the RoboInter Manipulation Suite, a unified resource including data, benchmarks, and models of intermediate representations for manipulation. It comprises RoboInter-Tool, a lightweight GUI that enables semi-automatic annotation of diverse representations, and RoboInter-Data, a large-scale dataset containing over 230k episodes across 571 diverse scenes, which provides dense per-frame annotations over more than 10 categories of intermediate representations, substantially exceeding prior work in scale and annotation quality. Building upon this foundation, RoboInter-VQA introduces 9 spatial and 20 temporal embodied VQA categories to systematically benchmark and enhance the embodied reasoning capabilities of VLMs. Meanwhile, RoboInter-VLA offers an integrated plan-then-execute framework, supporting modular and end-to-end VLA variants that bridge high-level planning with low-level execution via intermediate supervision. In total, RoboInter establishes a practical foundation for advancing robust and generalizable robotic learning via fine-grained and diverse intermediate representations.
comment: Published to ICLR 2026, 69 pages, 40 figures
☆ Hydra-Nav: Object Navigation via Adaptive Dual-Process Reasoning
While large vision-language models (VLMs) show promise for object goal navigation, current methods still struggle with low success rates and inefficient localization of unseen objects--failures primarily attributed to weak temporal-spatial reasoning. Meanwhile, recent attempts to inject reasoning into VLM-based agents improve success rates but incur substantial computational overhead. To address both the ineffectiveness and inefficiency of existing approaches, we introduce Hydra-Nav, a unified VLM architecture that adaptively switches between a deliberative slow system for analyzing exploration history and formulating high-level plans, and a reactive fast system for efficient execution. We train Hydra-Nav through a three-stage curriculum: (i) spatial-action alignment to strengthen trajectory planning, (ii) memory-reasoning integration to enhance temporal-spatial reasoning over long-horizon exploration, and (iii) iterative rejection fine-tuning to enable selective reasoning at critical decision points. Extensive experiments demonstrate that Hydra-Nav achieves state-of-the-art performance on the HM3D, MP3D, and OVON benchmarks, outperforming the second-best methods by 11.1%, 17.4%, and 21.2%, respectively. Furthermore, we introduce SOT (Success weighted by Operation Time), a new metric to measure search efficiency across VLMs with varying reasoning intensity. Results show that adaptive reasoning significantly enhances search efficiency over fixed-frequency baselines.
☆ Instruct2Act: From Human Instruction to Actions Sequencing and Execution via Robot Action Network for Robotic Manipulation
Robots often struggle to follow free-form human instructions in real-world settings due to computational and sensing limitations. We address this gap with a lightweight, fully on-device pipeline that converts natural-language commands into reliable manipulation. Our approach has two stages: (i) the instruction to actions module (Instruct2Act), a compact BiLSTM with a multi-head-attention autoencoder that parses an instruction into an ordered sequence of atomic actions (e.g., reach, grasp, move, place); and (ii) the robot action network (RAN), which uses the dynamic adaptive trajectory radial network (DATRN) together with a vision-based environment analyzer (YOLOv8) to generate precise control trajectories for each sub-action. The entire system runs on a modest system with no cloud services. On our custom proprietary dataset, Instruct2Act attains 91.5% sub-actions prediction accuracy while retaining a small footprint. Real-robot evaluations across four tasks (pick-place, pick-pour, wipe, and pick-give) yield an overall 90% success; sub-action inference completes in < 3.8s, with end-to-end executions in 30-60s depending on task complexity. These results demonstrate that fine-grained instruction-to-action parsing, coupled with DATRN-based trajectory generation and vision-guided grounding, provides a practical path to deterministic, real-time manipulation in resource-constrained, single-camera settings.
☆ TaCo: A Benchmark for Lossless and Lossy Codecs of Heterogeneous Tactile Data
Tactile sensing is crucial for embodied intelligence, providing fine-grained perception and control in complex environments. However, efficient tactile data compression, which is essential for real-time robotic applications under strict bandwidth constraints, remains underexplored. The inherent heterogeneity and spatiotemporal complexity of tactile data further complicate this challenge. To bridge this gap, we introduce TaCo, the first comprehensive benchmark for Tactile data Codecs. TaCo evaluates 30 compression methods, including off-the-shelf compression algorithms and neural codecs, across five diverse datasets from various sensor types. We systematically assess both lossless and lossy compression schemes on four key tasks: lossless storage, human visualization, material and object classification, and dexterous robotic grasping. Notably, we pioneer the development of data-driven codecs explicitly trained on tactile data, TaCo-LL (lossless) and TaCo-L (lossy). Results have validated the superior performance of our TaCo-LL and TaCo-L. This benchmark provides a foundational framework for understanding the critical trade-offs between compression efficiency and task performance, paving the way for future advances in tactile perception.
comment: 27 pages
☆ TriPilot-FF: Coordinated Whole-Body Teleoperation with Force Feedback
Mobile manipulators broaden the operational envelope for robot manipulation. However, the whole-body teleoperation of such robots remains a problem: operators must coordinate a wheeled base and two arms while reasoning about obstacles and contact. Existing interfaces are predominantly hand-centric (e.g., VR controllers and joysticks), leaving foot-operated channels underexplored for continuous base control. We present TriPilot-FF, an open-source whole-body teleoperation system for a custom bimanual mobile manipulator that introduces a foot-operated pedal with lidar-driven pedal haptics, coupled with upper-body bimanual leader-follower teleoperation. Using only a low-cost base-mounted lidar, TriPilot-FF renders a resistive pedal cue from proximity-to-obstacle signals in the commanded direction, shaping operator commands toward collision-averse behaviour without an explicit collision-avoidance controller. The system also supports arm-side force reflection for contact awareness and provides real-time force and visual guidance of bimanual manipulability to prompt mobile base repositioning, thereby improving reach. We demonstrate the capability of TriPilot-FF to effectively ``co-pilot'' the human operator over long time-horizons and tasks requiring precise mobile base movement and coordination. Finally, we incorporate teleoperation feedback signals into an Action Chunking with Transformers (ACT) policy and demonstrate improved performance when the additional information is available. We release the pedal device design, full software stack, and conduct extensive real-world evaluations on a bimanual wheeled platform. The project page of TriPilot-FF is http://bit.ly/46H3ZJT.
☆ BagelVLA: Enhancing Long-Horizon Manipulation via Interleaved Vision-Language-Action Generation
Equipping embodied agents with the ability to reason about tasks, foresee physical outcomes, and generate precise actions is essential for general-purpose manipulation. While recent Vision-Language-Action (VLA) models have leveraged pre-trained foundation models, they typically focus on either linguistic planning or visual forecasting in isolation. These methods rarely integrate both capabilities simultaneously to guide action generation, leading to suboptimal performance in complex, long-horizon manipulation tasks. To bridge this gap, we propose BagelVLA, a unified model that integrates linguistic planning, visual forecasting, and action generation within a single framework. Initialized from a pretrained unified understanding and generative model, BagelVLA is trained to interleave textual reasoning and visual prediction directly into the action execution loop. To efficiently couple these modalities, we introduce Residual Flow Guidance (RFG), which initializes from current observation and leverages single-step denoising to extract predictive visual features, guiding action generation with minimal latency. Extensive experiments demonstrate that BagelVLA outperforms existing baselines by a significant margin on multiple simulated and real-world benchmarks, particularly in tasks requiring multi-stage reasoning.
☆ Design and Evaluation of an Assisted Programming Interface for Behavior Trees in Robotics
The possibility to create reactive robot programs faster without the need for extensively trained programmers is becoming increasingly important. So far, it has not been explored how various techniques for creating Behavior Tree (BT) program representations could be combined with complete graphical user interfaces (GUIs) to allow a human user to validate and edit trees suggested by automated methods. In this paper, we introduce BEhavior TRee GUI (BETR-GUI) for creating BTs with the help of an AI assistant that combines methods using large language models, planning, genetic programming, and Bayesian optimization with a drag-and-drop editor. A user study with 60 participants shows that by combining different assistive methods, BETR-GUI enables users to perform better at solving the robot programming tasks. The results also show that humans using the full variant of BETR-GUI perform better than the AI assistant running on its own.
☆ Diverse Skill Discovery for Quadruped Robots via Unsupervised Learning
Reinforcement learning necessitates meticulous reward shaping by specialists to elicit target behaviors, while imitation learning relies on costly task-specific data. In contrast, unsupervised skill discovery can potentially reduce these burdens by learning a diverse repertoire of useful skills driven by intrinsic motivation. However, existing methods exhibit two key limitations: they typically rely on a single policy to master a versatile repertoire of behaviors without modeling the shared structure or distinctions among them, which results in low learning efficiency; moreover, they are susceptible to reward hacking, where the reward signal increases and converges rapidly while the learned skills display insufficient actual diversity. In this work, we introduce an Orthogonal Mixture-of-Experts (OMoE) architecture that prevents diverse behaviors from collapsing into overlapping representations, enabling a single policy to master a wide spectrum of locomotion skills. In addition, we design a multi-discriminator framework in which different discriminators operate on distinct observation spaces, effectively mitigating reward hacking. We evaluated our method on the 12-DOF Unitree A1 quadruped robot, demonstrating a diverse set of locomotion skills. Our experiments demonstrate that the proposed framework boosts training efficiency and yields an 18.3\% expansion in state-space coverage compared to the baseline.
☆ NavDreamer: Video Models as Zero-Shot 3D Navigators
Previous Vision-Language-Action models face critical limitations in navigation: scarce, diverse data from labor-intensive collection and static representations that fail to capture temporal dynamics and physical laws. We propose NavDreamer, a video-based framework for 3D navigation that leverages generative video models as a universal interface between language instructions and navigation trajectories. Our main hypothesis is that video's ability to encode spatiotemporal information and physical dynamics, combined with internet-scale availability, enables strong zero-shot generalization in navigation. To mitigate the stochasticity of generative predictions, we introduce a sampling-based optimization method that utilizes a VLM for trajectory scoring and selection. An inverse dynamics model is employed to decode executable waypoints from generated video plans for navigation. To systematically evaluate this paradigm in several video model backbones, we introduce a comprehensive benchmark covering object navigation, precise navigation, spatial grounding, language control, and scene reasoning. Extensive experiments demonstrate robust generalization across novel objects and unseen environments, with ablation studies revealing that navigation's high-level decision-making nature makes it particularly suited for video-based planning.
comment: Work in the progress. 22 pages, 15 figures
☆ Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization
While Vision-Language-Action (VLA) models show strong promise for generalist robot control, it remains unclear whether -- and under what conditions -- the standard "scale data" recipe translates to robotics, where training data is inherently heterogeneous across embodiments, sensors, and action spaces. We present a systematic, controlled study of VLA scaling that revisits core training choices for pretraining across diverse robots. Using a representative VLA framework that combines a vision-language backbone with flow-matching, we ablate key design decisions under matched conditions and evaluate in extensive simulation and real-robot experiments. To improve the reliability of real-world results, we introduce a Grouped Blind Ensemble protocol that blinds operators to model identity and separates policy execution from outcome judgment, reducing experimenter bias. Our analysis targets three dimensions of VLA scaling. (1) Physical alignment: we show that a unified end-effector (EEF)-relative action representation is critical for robust cross-embodiment transfer. (2) Embodiment mixture: we find that naively pooling heterogeneous robot datasets often induces negative transfer rather than gains, underscoring the fragility of indiscriminate data scaling. (3) Training regularization: we observe that intuitive strategies, such as sensory dropout and multi-stage fine-tuning, do not consistently improve performance at scale. Together, this study challenge some common assumptions about embodied scaling and provide practical guidance for training large-scale VLA policies from diverse robotic data. Project website: https://research.beingbeyond.com/rethink_vla
☆ Fast Motion Planning for Non-Holonomic Mobile Robots via a Rectangular Corridor Representation of Structured Environments ICRA 2026
We present a complete framework for fast motion planning of non-holonomic autonomous mobile robots in highly complex but structured environments. Conventional grid-based planners struggle with scalability, while many kinematically-feasible planners impose a significant computational burden due to their search space complexity. To overcome these limitations, our approach introduces a deterministic free-space decomposition that creates a compact graph of overlapping rectangular corridors. This method enables a significant reduction in the search space, without sacrificing path resolution. The framework then performs online motion planning by finding a sequence of rectangles and generating a near-time-optimal, kinematically-feasible trajectory using an analytical planner. The result is a highly efficient solution for large-scale navigation. We validate our framework through extensive simulations and on a physical robot. The implementation is publicly available as open-source software.
comment: ICRA 2026
☆ RANT: Ant-Inspired Multi-Robot Rainforest Exploration Using Particle Filter Localisation and Virtual Pheromone Coordination
This paper presents RANT, an ant-inspired multi-robot exploration framework for noisy, uncertain environments. A team of differential-drive robots navigates a 10 x 10 m terrain, collects noisy probe measurements of a hidden richness field, and builds local probabilistic maps while the supervisor maintains a global evaluation. RANT combines particle-filter localisation, a behaviour-based controller with gradient-driven hotspot exploitation, and a lightweight no-revisit coordination mechanism based on virtual pheromone blocking. We experimentally analyse how team size, localisation fidelity, and coordination influence coverage, hotspot recall, and redundancy. Results show that particle filtering is essential for reliable hotspot engagement, coordination substantially reduces overlap, and increasing team size improves coverage but yields diminishing returns due to interference.
comment: 7 pages, IEEEtran format
☆ AutoFly: Vision-Language-Action Model for UAV Autonomous Navigation in the Wild ICLR 2026
Vision-language navigation (VLN) requires intelligent agents to navigate environments by interpreting linguistic instructions alongside visual observations, serving as a cornerstone task in Embodied AI. Current VLN research for unmanned aerial vehicles (UAVs) relies on detailed, pre-specified instructions to guide the UAV along predetermined routes. However, real-world outdoor exploration typically occurs in unknown environments where detailed navigation instructions are unavailable. Instead, only coarse-grained positional or directional guidance can be provided, requiring UAVs to autonomously navigate through continuous planning and obstacle avoidance. To bridge this gap, we propose AutoFly, an end-to-end Vision-Language-Action (VLA) model for autonomous UAV navigation. AutoFly incorporates a pseudo-depth encoder that derives depth-aware features from RGB inputs to enhance spatial reasoning, coupled with a progressive two-stage training strategy that effectively aligns visual, depth, and linguistic representations with action policies. Moreover, existing VLN datasets have fundamental limitations for real-world autonomous navigation, stemming from their heavy reliance on explicit instruction-following over autonomous decision-making and insufficient real-world data. To address these issues, we construct a novel autonomous navigation dataset that shifts the paradigm from instruction-following to autonomous behavior modeling through: (1) trajectory collection emphasizing continuous obstacle avoidance, autonomous planning, and recognition workflows; (2) comprehensive real-world data integration. Experimental results demonstrate that AutoFly achieves a 3.9% higher success rate compared to state-of-the-art VLA baselines, with consistent performance across simulated and real environments.
comment: Acceped by ICLR 2026
☆ TeleGate: Whole-Body Humanoid Teleoperation via Gated Expert Selection with Motion Prior
Real-time whole-body teleoperation is a critical method for humanoid robots to perform complex tasks in unstructured environments. However, developing a unified controller that robustly supports diverse human motions remains a significant challenge. Existing methods typically distill multiple expert policies into a single general policy, which often inevitably leads to performance degradation, particularly on highly dynamic motions. This paper presents TeleGate, a unified whole-body teleoperation framework for humanoid robots that achieves high-precision tracking across various motions while avoiding the performance loss inherent in knowledge distillation. Our key idea is to preserve the full capability of domain-specific expert policies by training a lightweight gating network, which dynamically activates experts in real-time based on proprioceptive states and reference trajectories. Furthermore, to compensate for the absence of future reference trajectories in real-time teleoperation, we introduce a VAE-based motion prior module that extracts implicit future motion intent from historical observations, enabling anticipatory control for motions requiring prediction such as jumping and standing up. We conducted empirical evaluations in simulation and also deployed our technique on the Unitree G1 humanoid robot. Using only 2.5 hours of motion capture data for training, our TeleGate achieves high-precision real-time teleoperation across diverse dynamic motions (e.g., running, fall recovery, and jumping), significantly outperforming the baseline methods in both tracking accuracy and success rate.
☆ AnyTouch 2: General Optical Tactile Representation Learning For Dynamic Tactile Perception ICLR 2026
Real-world contact-rich manipulation demands robots to perceive temporal tactile feedback, capture subtle surface deformations, and reason about object properties as well as force dynamics. Although optical tactile sensors are uniquely capable of providing such rich information, existing tactile datasets and models remain limited. These resources primarily focus on object-level attributes (e.g., material) while largely overlooking fine-grained tactile temporal dynamics during physical interactions. We consider that advancing dynamic tactile perception requires a systematic hierarchy of dynamic perception capabilities to guide both data collection and model design. To address the lack of tactile data with rich dynamic information, we present ToucHD, a large-scale hierarchical tactile dataset spanning tactile atomic actions, real-world manipulations, and touch-force paired data. Beyond scale, ToucHD establishes a comprehensive tactile dynamic data ecosystem that explicitly supports hierarchical perception capabilities from the data perspective. Building on it, we propose AnyTouch 2, a general tactile representation learning framework for diverse optical tactile sensors that unifies object-level understanding with fine-grained, force-aware dynamic perception. The framework captures both pixel-level and action-specific deformations across frames, while explicitly modeling physical force dynamics, thereby learning multi-level dynamic perception capabilities from the model perspective. We evaluate our model on benchmarks that covers static object properties and dynamic physical attributes, as well as real-world manipulation tasks spanning multiple tiers of dynamic perception capabilities-from basic object-level understanding to force-aware dexterous manipulation. Experimental results demonstrate consistent and strong performance across sensors and tasks.
comment: Accepted by ICLR 2026
☆ Preference Aligned Visuomotor Diffusion Policies for Deformable Object Manipulation
Humans naturally develop preferences for how manipulation tasks should be performed, which are often subtle, personal, and difficult to articulate. Although it is important for robots to account for these preferences to increase personalization and user satisfaction, they remain largely underexplored in robotic manipulation, particularly in the context of deformable objects like garments and fabrics. In this work, we study how to adapt pretrained visuomotor diffusion policies to reflect preferred behaviors using limited demonstrations. We introduce RKO, a novel preference-alignment method that combines the benefits of two recent frameworks: RPO and KTO. We evaluate RKO against common preference learning frameworks, including these two, as well as a baseline vanilla diffusion policy, on real-world cloth-folding tasks spanning multiple garments and preference settings. We show that preference-aligned policies (particularly RKO) achieve superior performance and sample efficiency compared to standard diffusion policy fine-tuning. These results highlight the importance and feasibility of structured preference learning for scaling personalized robot behavior in complex deformable object manipulation tasks.
☆ 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 SOFT-FLOW, 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 SOFT-FLOW 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, SOFT-FLOW achieves stable, sample-efficient adaptation where standard methods struggle.
☆ Optimal Control of Microswimmers for Trajectory Tracking Using Bayesian Optimization
Trajectory tracking for microswimmers remains a key challenge in microrobotics, where low-Reynolds-number dynamics make control design particularly complex. In this work, we formulate the trajectory tracking problem as an optimal control problem and solve it using a combination of B-spline parametrization with Bayesian optimization, allowing the treatment of high computational costs without requiring complex gradient computations. Applied to a flagellated magnetic swimmer, the proposed method reproduces a variety of target trajectories, including biologically inspired paths observed in experimental studies. We further evaluate the approach on a three-sphere swimmer model, demonstrating that it can adapt to and partially compensate for wall-induced hydrodynamic effects. The proposed optimization strategy can be applied consistently across models of different fidelity, from low-dimensional ODE-based models to high-fidelity PDE-based simulations, showing its robustness and generality. These results highlight the potential of Bayesian optimization as a versatile tool for optimal control strategies in microscale locomotion under complex fluid-structure interactions.
comment: 16 pages, 16 figures
☆ LLM-Grounded Dynamic Task Planning with Hierarchical Temporal Logic for Human-Aware Multi-Robot Collaboration
While Large Language Models (LLM) enable non-experts to specify open-world multi-robot tasks, the generated plans often lack kinematic feasibility and are not efficient, especially in long-horizon scenarios. Formal methods like Linear Temporal Logic (LTL) offer correctness and optimal guarantees, but are typically confined to static, offline settings and struggle with computational scalability. To bridge this gap, we propose a neuro-symbolic framework that grounds LLM reasoning into hierarchical LTL specifications and solves the corresponding Simultaneous Task Allocation and Planning (STAP) problem. Unlike static approaches, our system resolves stochastic environmental changes, such as moving users or updated instructions via a receding horizon planning (RHP) loop with real-time perception, which dynamically refines plans through a hierarchical state space. Extensive real-world experiments demonstrate that our approach significantly outperforms baseline methods in success rate and interaction fluency while minimizing planning latency.
☆ Sci-VLA: Agentic VLA Inference Plugin for Long-Horizon Tasks in Scientific Experiments
Robotic laboratories play a critical role in autonomous scientific discovery by enabling scalable, continuous experimental execution. Recent vision-language-action (VLA) models offer a promising foundation for robotic laboratories. However, scientific experiments typically involve long-horizon tasks composed of multiple atomic tasks, posing a fundamental challenge to existing VLA models. While VLA models fine-tuned for scientific tasks can reliably execute atomic experimental actions seen during training, they often fail to perform composite tasks formed by reordering and composing these known atomic actions. This limitation arises from a distributional mismatch between training-time atomic tasks and inference-time composite tasks, which prevents VLA models from executing necessary transitional operations between atomic tasks. To address this challenge, we propose an Agentic VLA Inference Plugin for Long-Horizon Tasks in Scientific Experiments. It introduces an LLM-based agentic inference mechanism that intervenes when executing sequential manipulation tasks. By performing explicit transition inference and generating transitional robotic action code, the proposed plugin guides VLA models through missing transitional steps, enabling reliable execution of composite scientific workflows without any additional training. This inference-only intervention makes our method computationally efficient, data-efficient, and well-suited for open-ended and long-horizon robotic laboratory tasks. We build 3D assets of scientific instruments and common scientific operating scenes within an existing simulation environment. In these scenes, we have verified that our method increases the average success rate per atomic task by 42\% during inference. Furthermore, we show that our method can be easily transferred from the simulation to real scientific laboratories.
☆ First-order friction models with bristle dynamics: lumped and distributed formulations
Dynamic models, particularly rate-dependent models, have proven effective in capturing the key phenomenological features of frictional processes, whilst also possessing important mathematical properties that facilitate the design of control and estimation algorithms. However, many rate-dependent formulations are built on empirical considerations, whereas physical derivations may offer greater interpretability. In this context, starting from fundamental physical principles, this paper introduces a novel class of first-order dynamic friction models that approximate the dynamics of a bristle element by inverting the friction characteristic. Amongst the developed models, a specific formulation closely resembling the LuGre model is derived using a simple rheological equation for the bristle element. This model is rigorously analyzed in terms of stability and passivity -- important properties that support the synthesis of observers and controllers. Furthermore, a distributed version, formulated as a hyperbolic partial differential equation (PDE), is presented, which enables the modeling of frictional processes commonly encountered in rolling contact phenomena. The tribological behavior of the proposed description is evaluated through classical experiments and validated against the response predicted by the LuGre model, revealing both notable similarities and key differences.
comment: 15 pages, 9 figures. Under review at IEEE Transactions on Control Systems Technology
☆ Lateral tracking control of all-wheel steering vehicles with intelligent tires
The accurate characterization of tire dynamics is critical for advancing control strategies in autonomous road vehicles, as tire behavior significantly influences handling and stability through the generation of forces and moments at the tire-road interface. Smart tire technologies have emerged as a promising tool for sensing key variables such as road friction, tire pressure, and wear states, and for estimating kinematic and dynamic states like vehicle speed and tire forces. However, most existing estimation and control algorithms rely on empirical correlations or machine learning approaches, which require extensive calibration and can be sensitive to variations in operating conditions. In contrast, model-based techniques, which leverage infinite-dimensional representations of tire dynamics using partial differential equations (PDEs), offer a more robust approach. This paper proposes a novel model-based, output-feedback lateral tracking control strategy for all-wheel steering vehicles that integrates distributed tire dynamics with smart tire technologies. The primary contributions include the suppression of micro-shimmy phenomena at low speeds and path-following via force control, achieved through the estimation of tire slip angles, vehicle kinematics, and lateral tire forces. The proposed controller and observer are based on formulations using ODE-PDE systems, representing rigid body dynamics and distributed tire behavior. This work marks the first rigorous control strategy for vehicular systems equipped with distributed tire representations in conjunction with smart tire technologies.
comment: 17 pages, 12 figures. Under review at IEEE Transactions on Intelligent Vehicles
☆ Finite-time Stable Pose Estimation on TSE(3) using Point Cloud and Velocity Sensors
This work presents a finite-time stable pose estimator (FTS-PE) for rigid bodies undergoing rotational and translational motion in three dimensions, using measurements from onboard sensors that provide position vectors to inertially-fixed points and body velocities. The FTS-PE is a full-state observer for the pose (position and orientation) and velocities and is obtained through a Lyapunov analysis that shows its stability in finite time and its robustness to bounded measurement noise. Further, this observer is designed directly on the state space, the tangent bundle of the Lie group of rigid body motions, SE(3), without using local coordinates or (dual) quaternion representations. Therefore, it can estimate arbitrary rigid body motions without encountering singularities or the unwinding phenomenon and be readily applied to autonomous vehicles. A version of this observer that does not need translational velocity measurements and uses only point clouds and angular velocity measurements from rate gyros, is also obtained. It is discretized using the framework of geometric mechanics for numerical and experimental implementations. The numerical simulations compare the FTS-PE with a dual-quaternion extended Kalman filter and our previously developed variational pose estimator (VPE). The experimental results are obtained using point cloud images and rate gyro measurements obtained from a Zed 2i stereo depth camera sensor. These results validate the stability and robustness of the FTS-PE.
comment: 17 pages, 8 figures, submitted to Automatica
☆ Phase-Aware Policy Learning for Skateboard Riding of Quadruped Robots via Feature-wise Linear Modulation ICRA 2026
Skateboards offer a compact and efficient means of transportation as a type of personal mobility device. However, controlling them with legged robots poses several challenges for policy learning due to perception-driven interactions and multi-modal control objectives across distinct skateboarding phases. To address these challenges, we introduce Phase-Aware Policy Learning (PAPL), a reinforcement-learning framework tailored for skateboarding with quadruped robots. PAPL leverages the cyclic nature of skateboarding by integrating phase-conditioned Feature-wise Linear Modulation layers into actor and critic networks, enabling a unified policy that captures phase-dependent behaviors while sharing robot-specific knowledge across phases. Our evaluations in simulation validate command-tracking accuracy and conduct ablation studies quantifying each component's contribution. We also compare locomotion efficiency against leg and wheel-leg baselines and show real-world transferability.
comment: Accepted at ICRA 2026. Supplementary Video: https://www.youtube.com/watch?v=bCNfdQ3RYKg. M. Yoon and J. Jeong contributed equally
☆ Certified Gradient-Based Contact-Rich Manipulation via Smoothing-Error Reachable Tubes
Gradient-based methods can efficiently optimize controllers using physical priors and differentiable simulators, but contact-rich manipulation remains challenging due to discontinuous or vanishing gradients from hybrid contact dynamics. Smoothing the dynamics yields continuous gradients, but the resulting model mismatch can cause controller failures when executed on real systems. We address this trade-off by planning with smoothed dynamics while explicitly quantifying and compensating for the induced errors, providing formal guarantees of constraint satisfaction and goal reachability on the true hybrid dynamics. Our method smooths both contact dynamics and geometry via a novel differentiable simulator based on convex optimization, which enables us to characterize the discrepancy from the true dynamics as a set-valued deviation. This deviation constrains the optimization of time-varying affine feedback policies through analytical bounds on the system's reachable set, enabling robust constraint satisfaction guarantees for the true closed-loop hybrid dynamics, while relying solely on informative gradients from the smoothed dynamics. We evaluate our method on several contact-rich tasks, including planar pushing, object rotation, and in-hand dexterous manipulation, achieving guaranteed constraint satisfaction with lower safety violation and goal error than baselines. By bridging differentiable physics with set-valued robust control, our method is the first certifiable gradient-based policy synthesis method for contact-rich manipulation.
☆ CAPER: Constrained and Procedural Reasoning for Robotic Scientific Experiments
Robotic assistance in scientific laboratories requires procedurally correct long-horizon manipulation, reliable execution under limited supervision, and robustness in low-demonstration regimes. Such conditions greatly challenge end-to-end vision-language-action (VLA) models, whose assumptions of recoverable errors and data-driven policy learning often break down in protocol-sensitive experiments. We propose CAPER, a framework for Constrained And ProcEdural Reasoning for robotic scientific experiments, which explicitly restricts where learning and reasoning occur in the planning and control pipeline. Rather than strengthening end-to-end policies, CAPER enforces a responsibility-separated structure: task-level reasoning generates procedurally valid action sequences under explicit constraints, mid-level multimodal grounding realizes subtasks without delegating spatial decision-making to large language models, and low-level control adapts to physical uncertainty via reinforcement learning with minimal demonstrations. By encoding procedural commitments through interpretable intermediate representations, CAPER prevents execution-time violations of experimental logic, improving controllability, robustness, and data efficiency. Experiments on a scientific workflow benchmark and a public long-horizon manipulation dataset demonstrate consistent improvements in success rate and procedural correctness, particularly in low-data and long-horizon settings.
☆ Disambiguating Anthropomorphism and Anthropomimesis in Human-Robot Interaction
In this preliminary work, we offer an initial disambiguation of the theoretical concepts anthropomorphism and anthropomimesis in Human-Robot Interaction (HRI) and social robotics. We define anthropomorphism as users perceiving human-like qualities in robots, and anthropomimesis as robot developers designing human-like features into robots. This contribution aims to provide a clarification and exploration of these concepts for future HRI scholarship, particularly regarding the party responsible for human-like qualities - robot perceiver for anthropomorphism, and robot designer for anthropomimesis. We provide this contribution so that researchers can build on these disambiguated theoretical concepts for future robot design and evaluation.
comment: 6 pages, 0 figures. Accepted at Human-Robot Interaction (HRI) conference (ACM/IEEE) 2026, Late-Breaking Reports
☆ Solving Geodesic Equations with Composite Bernstein Polynomials for Trajectory Planning
This work presents a trajectory planning method based on composite Bernstein polynomials for autonomous systems navigating complex environments. The method is implemented in a symbolic optimization framework that enables continuous paths and precise control over trajectory shape. Trajectories are planned over a cost surface that encodes obstacles as continuous fields rather than discrete boundaries. Regions near obstacles are assigned higher costs, naturally encouraging the trajectory to maintain a safe distance while still allowing efficient routing through constrained spaces. The use of composite Bernstein polynomials preserves continuity while enabling fine control over local curvature to satisfy geodesic constraints. The symbolic representation supports exact derivatives, improving optimization efficiency. The method applies to both two- and three-dimensional environments and is suitable for ground, aerial, underwater, and space systems. In spacecraft trajectory planning, for example, it enables the generation of continuous, dynamically feasible trajectories with high numerical efficiency, making it well suited for orbital maneuvers, rendezvous and proximity operations, cluttered gravitational environments, and planetary exploration missions with limited onboard computational resources. Demonstrations show that the approach efficiently generates smooth, collision-free paths in scenarios with multiple obstacles, maintaining clearance without extensive sampling or post-processing. The optimization incorporates three constraint types: (1) a Gaussian surface inequality enforcing minimum obstacle clearance; (2) geodesic equations guiding the path along locally efficient directions on the cost surface; and (3) boundary constraints enforcing fixed start and end conditions. The method can serve as a standalone planner or as an initializer for more complex motion planning problems.
comment: Accepted for the 2026 IEEE Aerospace Conference
☆ Confounding Robust Continuous Control via Automatic Reward Shaping AAMAS 2026
Reward shaping has been applied widely to accelerate Reinforcement Learning (RL) agents' training. However, a principled way of designing effective reward shaping functions, especially for complex continuous control problems, remains largely under-explained. In this work, we propose to automatically learn a reward shaping function for continuous control problems from offline datasets, potentially contaminated by unobserved confounding variables. Specifically, our method builds upon the recently proposed causal Bellman equation to learn a tight upper bound on the optimal state values, which is then used as the potentials in the Potential-Based Reward Shaping (PBRS) framework. Our proposed reward shaping algorithm is tested with Soft-Actor-Critic (SAC) on multiple commonly used continuous control benchmarks and exhibits strong performance guarantees under unobserved confounders. More broadly, our work marks a solid first step towards confounding robust continuous control from a causal perspective. Code for training our reward shaping functions can be found at https://github.com/mateojuliani/confounding_robust_cont_control.
comment: Mateo Juliani and Mingxuan Li contributed equally to this work; accepted in AAMAS 2026
☆ 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
☆ Adaptive Time Step Flow Matching for Autonomous Driving Motion Planning
Autonomous driving requires reasoning about interactions with surrounding traffic. A prevailing approach is large-scale imitation learning on expert driving datasets, aimed at generalizing across diverse real-world scenarios. For online trajectory generation, such methods must operate at real-time rates. Diffusion models require hundreds of denoising steps at inference, resulting in high latency. Consistency models mitigate this issue but rely on carefully tuned noise schedules to capture the multimodal action distributions common in autonomous driving. Adapting the schedule, typically requires expensive retraining. To address these limitations, we propose a framework based on conditional flow matching that jointly predicts future motions of surrounding agents and plans the ego trajectory in real time. We train a lightweight variance estimator that selects the number of inference steps online, removing the need for retraining to balance runtime and imitation learning performance. To further enhance ride quality, we introduce a trajectory post-processing step cast as a convex quadratic program, with negligible computational overhead. Trained on the Waymo Open Motion Dataset, the framework performs maneuvers such as lane changes, cruise control, and navigating unprotected left turns without requiring scenario-specific tuning. Our method maintains a 20 Hz update rate on an NVIDIA RTX 3070 GPU, making it suitable for online deployment. Compared to transformer, diffusion, and consistency model baselines, we achieve improved trajectory smoothness and better adherence to dynamic constraints. Experiment videos and code implementations can be found at https://flow-matching-self-driving.github.io/.
comment: Accepted to Intelligent Vehicles Symposium 2026
☆ Transforming Policy-Car Swerving for Mitigating Stop-and-Go Traffic Waves: A Practice-Oriented Jam-Absorption Driving Strategy
Stop-and-go waves, as a major form of freeway traffic congestion, cause severe and long-lasting adverse effects, including reduced traffic efficiency, increased driving risks, and higher vehicle emissions. Amongst the highway traffic management strategies, jam-absorption driving (JAD), in which a dedicated vehicle performs "slow-in" and "fast-out" maneuvers before being captured by a stop-and-go wave, has been proposed as a potential method for preventing the propagation of such waves. However, most existing JAD strategies remain impractical mainly due to the lack of discussion regarding implementation vehicles and operational conditions. Inspired by real-world observations of police-car swerving behavior, this paper first introduces a Single-Vehicle Two-Detector Jam-Absorption Driving (SVDD-JAD) problem, and then proposes a practical JAD strategy that transforms such behavior into a maneuver capable of suppressing the propagation of an isolated stop-and-go wave. Five key parameters that significantly affect the proposed strategy, namely, JAD speed, inflow traffic speed, wave width, wave speed, and in-wave speed, are identified and systematically analyzed. Using a SUMO-based simulation as an illustrative example, we further demonstrate how these parameters can be measured in practice with two stationary roadside traffic detectors. The results show that the proposed JAD strategy successfully suppresses the propagation of a stop-and-go wave, without triggering a secondary wave. This paper is expected to take a significant step toward making JAD practical, advancing it from a theoretical concept to a feasible and implementable strategy. To promote reproducibility in the transportation domain, we have also open-sourced all the code on our GitHub repository https://github.com/gotrafficgo.
♻ ☆ 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: ICLR 2026, Project page: https://falcon-vla.github.io/
♻ ☆ Going with the Flow: Koopman Behavioral Models as Implicit Planners for Visuo-Motor Dexterity
There has been rapid and dramatic progress in learning complex visuo-motor manipulation skills from demonstrations, thanks in part to expressive policy classes that employ diffusion- and transformer-based backbones. However, these design choices require significant data and computational resources and remain far from reliable, particularly within the context of multi-fingered dexterous manipulation. Fundamentally, they model skills as reactive mappings and rely on fixed-horizon action chunking to mitigate jitter, creating a rigid trade-off between temporal coherence and reactivity. In this work, we introduce Unified Behavioral Models (UBMs), a framework that learns to represent dexterous skills as coupled dynamical systems that capture how visual features of the environment (visual flow) and proprioceptive states of the robot (action flow) co-evolve. By capturing such behavioral dynamics, UBMs can ensure temporal coherence by construction rather than by heuristic averaging. To operationalize these models, we propose Koopman-UBM, a first instantiation of UBMs that leverages Koopman Operator theory to effectively learn a unified representation in which the joint flow of latent visual and proprioceptive features is governed by a structured linear system. We demonstrate that Koopman-UBM can be viewed as an implicit planner: given an initial condition, it computes the desired robot behavior with the resulting flow of visual features over the entire skill horizon. To enable reactivity, we introduce an online replanning strategy in which the model acts as its own runtime monitor that automatically triggers replanning when predicted and observed visual flow diverge. Across seven simulated and two real-world tasks, we demonstrate that K-UBM matches or exceeds the performance of SOTA baselines, while offering faster inference, smooth execution, robustness to occlusions, and flexible replanning.
comment: Codes can be found at https://github.com/GT-STAR-Lab/K-UBM/
♻ ☆ Toward Efficient and Robust Behavior Models for Multi-Agent Driving Simulation ICRA 2026
Scalable multi-agent driving simulation requires behavior models that are both realistic and computationally efficient. We address this by optimizing the behavior model that controls individual traffic participants. To improve efficiency, we adopt an instance-centric scene representation, where each traffic participant and map element is modeled in its own local coordinate frame. This design enables efficient, viewpoint-invariant scene encoding and allows static map tokens to be reused across simulation steps. To model interactions, we employ a query-centric symmetric context encoder with relative positional encodings between local frames. We use Adversarial Inverse Reinforcement Learning to learn the behavior model and propose an adaptive reward transformation that automatically balances robustness and realism during training. Experiments demonstrate that our approach scales efficiently with the number of tokens, significantly reducing training and inference times, while outperforming several agent-centric baselines in terms of positional accuracy and robustness.
comment: Accepted for publication by IEEE International Conference on Robotics & Automation (ICRA 2026)
♻ ☆ Model Predictive Control via Probabilistic Inference: A Tutorial and Survey
This paper provides a tutorial and a survey of probabilistic inference-based model predictive control (PI-MPC) for robotics. PI-MPC defines an optimal control distribution shaped by trajectory cost, a control prior, and a temperature parameter. In the tutorial part, we derive this view and describe action generation via variational inference, highlighting Model Predictive Path Integral (MPPI) control as a representative algorithm. In the survey part, we organize the PI-MPC literature around the principal design choices identified in the tutorial: prior distribution design, multi-modal distribution handling, constraint satisfaction, scalability to high-degree-of-freedom robots, hardware acceleration, and theoretical foundations. Overall, this paper aims to serve as a coherent entry point for researchers and practitioners interested in understanding, implementing, and extending PI-MPC.
comment: 26 pages, 7 figures
♻ ☆ karl. - A Research Vehicle for Automated and Connected Driving
As highly automated driving is transitioning from single-vehicle closed-access testing to commercial deployments of public ride-hailing in selected areas (e.g., Waymo), automated driving and connected cooperative intelligent transport systems (C-ITS) remain active fields of research. Even though simulation is omnipresent in the development and validation life cycle of automated and connected driving technology, the complex nature of public road traffic and software that masters it still requires real-world integration and testing with actual vehicles. Dedicated vehicles for research and development allow testing and validation of software and hardware components under real-world conditions early on. They also enable collecting and publishing real-world datasets that let others conduct research without vehicle access, and support early demonstration of futuristic use cases. In this paper, we present karl., our new research vehicle for automated and connected driving. Apart from major corporations, few institutions worldwide have access to their own L4-capable research vehicles, restricting their ability to carry out independent research. This paper aims to help bridge that gap by sharing the reasoning, design choices, and technical details that went into making karl. a flexible and powerful platform for research, engineering, and validation in the context of automated and connected driving. More impressions of karl. are available at https://karl.ac.
comment: 8 pages; Accepted to be published as part of the 37th Intelligent Vehicles Symposium (IV), Detroit, MI, United States, June 22-25, 2026
♻ ☆ Hoi! -- A Multimodal Dataset for Force-Grounded, Cross-View Articulated Manipulation
We present a dataset for force-grounded, cross-view articulated manipulation that couples what is seen with what is done and what is felt during real human interaction. The dataset contains 3048 sequences across 381 articulated objects in 38 environments. Each object is operated under four embodiments - (i) human hand, (ii) human hand with a wrist-mounted camera, (iii) handheld UMI gripper, and (iv) a custom Hoi! gripper - where the tool embodiment provides synchronized end-effector forces and tactile sensing. Our dataset offers a holistic view of interaction understanding from video, enabling researchers to evaluate how well methods transfer between human and robotic viewpoints, but also investigate underexplored modalities such as force sensing and prediction. Further information can be found on the Website.
♻ ☆ Toward Reliable Sim-to-Real Predictability for MoE-based Robust Quadrupedal Locomotion
Reinforcement learning has shown strong promise for quadrupedal agile locomotion, even with proprioception-only sensing. In practice, however, sim-to-real gap and reward overfitting in complex terrains can produce policies that fail to transfer, while physical validation remains risky and inefficient. To address these challenges, we introduce a unified framework encompassing a Mixture-of-Experts (MoE) locomotion policy for robust multi-terrain representation with RoboGauge, a predictive assessment suite that quantifies sim-to-real transferability. The MoE policy employs a gated set of specialist experts to decompose latent terrain and command modeling, achieving superior deployment robustness and generalization via proprioception alone. RoboGauge further provides multi-dimensional proprioception-based metrics via sim-to-sim tests over terrains, difficulty levels, and domain randomizations, enabling reliable MoE policy selection without extensive physical trials. Experiments on a Unitree Go2 demonstrate robust locomotion on unseen challenging terrains, including snow, sand, stairs, slopes, and 30 cm obstacles. In dedicated high-speed tests, the robot reaches 4 m/s and exhibits an emergent narrow-width gait associated with improved stability at high velocity.
comment: fix RoboGauge linear velocity metric, friction error; update radar fig (page 1), score tables (page 6,7); fix obs stack bug
♻ ☆ Simultaneous Calibration of Noise Covariance and Kinematics for State Estimation of Legged Robots via Bi-level Optimization
Accurate state estimation is critical for legged and aerial robots operating in dynamic, uncertain environments. A key challenge lies in specifying process and measurement noise covariances, which are typically unknown or manually tuned. In this work, we introduce a bi-level optimization framework that jointly calibrates covariance matrices and kinematic parameters in an estimator-in-the-loop manner. The upper level treats noise covariances and model parameters as optimization variables, while the lower level executes a full-information estimator. Differentiating through the estimator allows direct optimization of trajectory-level objectives, resulting in accurate and consistent state estimates. We validate our approach on quadrupedal and humanoid robots, demonstrating significantly improved estimation accuracy and uncertainty calibration compared to hand-tuned baselines. Our method unifies state estimation, sensor, and kinematics calibration into a principled, data-driven framework applicable across diverse robotic platforms.
♻ ☆ VDRive: Leveraging Reinforced VLA and Diffusion Policy for End-to-end Autonomous Driving
In autonomous driving, dynamic environment and corner cases pose significant challenges to the robustness of ego vehicle's state understanding and decision making. We introduce VDRive, a novel pipeline for end-to-end autonomous driving that explicitly models state-action mapping to address these challenges, enabling interpretable and robust decision making. By leveraging the advancement of the state understanding of the Vision Language Action Model (VLA) with generative diffusion policy-based action head, our VDRive guides the driving contextually and geometrically. Contextually, VLA predicts future observations through token generation pre-training, where the observations are represented as discrete codes by a Conditional Vector Quantized Variational Autoencoder (CVQ-VAE). Geometrically, we perform reinforcement learning fine-tuning of the VLA to predict future trajectories and actions based on current driving conditions. VLA supplies the current state tokens and predicted state tokens for the action policy head to generate hierarchical actions and trajectories. During policy training, a learned critic evaluates the actions generated by the policy and provides gradient-based feedback, forming an actor-critic framework that enables a reinforcement-based policy learning pipeline. Experiments show that our VDRive achieves state-of-the-art performance in the Bench2Drive closed-loop benchmark and nuScenes open-loop planning.
comment: WIP
♻ ☆ GenTrack2: An Improved Hybrid Approach for Multi-Object Tracking
This paper proposes a visual multi-object tracking method that jointly employs stochastic and deterministic mechanisms to ensure identifier consistency for unknown and time-varying target numbers under nonlinear dynamics. A stochastic particle filter addresses nonlinear dynamics and non-Gaussian noise, with support from particle swarm optimization (PSO) to guide particles toward state distribution modes and mitigate divergence through proposed fitness measures incorporating motion consistency, appearance similarity, and social-interaction cues with neighboring targets. Deterministic association further enforces identifier consistency via a proposed cost matrix incorporating spatial consistency between particles and current detections, detection confidences, and track penalties. Subsequently, a novel scheme is proposed for the smooth updating of target states while preserving their identities, particularly for weak tracks during interactions with other targets and prolonged occlusions. Moreover, velocity regression over past states provides trend-seed velocities, enhancing particle sampling and state updates. The proposed tracker is designed to operate flexibly for both pre-recorded videos and camera live streams, where future frames are unavailable. Experimental results confirm superior performance compared to state-of-the-art trackers. The source-code reference implementations of both the proposed method and compared-trackers are provided on GitHub: https://github.com/SDU-VelKoTek/GenTrack2
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ CoLA-Flow Policy: Temporally Coherent Imitation Learning via Continuous Latent Action Flow Matching for Robotic Manipulation
Learning long-horizon robotic manipulation requires jointly achieving expressive behavior modeling, real-time inference, and stable execution, which remains challenging for existing generative policies. Diffusion-based approaches provide strong modeling capacity but typically incur high inference latency, while flow matching enables fast one-step generation yet often leads to unstable execution when applied directly in the raw action space. We propose LG-Flow Policy, a trajectory-level imitation learning framework that performs flow matching in a continuous latent action space. By encoding action sequences into temporally regularized latent trajectories and learning an explicit latent-space flow, the proposed approach decouples global motion structure from low-level control noise, resulting in smooth and reliable long-horizon execution. LG-Flow Policy further incorporates geometry-aware point cloud conditioning and execution-time multimodal modulation, with visual cues evaluated as a representative modality in real-world settings. Experimental results in simulation and on physical robot platforms demonstrate that LG-Flow Policy achieves near single-step inference, substantially improves trajectory smoothness and task success over flow-based baselines operating in the raw action space, and remains significantly more efficient than diffusion-based policies.
comment: 8 pages, 8 figures
♻ ☆ xFLIE: Leveraging Actionable Hierarchical Scene Representations for Autonomous Semantic-Aware Inspection Missions
We present a novel architecture aimed towards incremental construction and exploitation of a hierarchical 3D scene graph representation during semantic-aware inspection missions. Inspection planning, particularly of distributed targets in previously unseen environments, presents an opportunity to exploit the semantic structure of the scene during reasoning, navigation and scene understanding. Motivated by this, we propose the 3D Layered Semantic Graph (3DLSG), a hierarchical inspection scene graph constructed in an incremental manner and organized into abstraction layers that support planning demands in real-time. To address the task of semantic-aware inspection, a mission framework, termed as Enhanced First-Look Inspect Explore (xFLIE), that tightly couples the 3DLSG with an inspection planner is proposed. We assess the performance through simulations and experimental trials, evaluating target-selection, path-planning and semantic navigation tasks over the 3DLSG model. The scenarios presented are diverse, ranging from city-scale distributed to solitary infrastructure targets in simulated worlds and subsequent outdoor and subterranean environment deployments onboard a quadrupedal robot. The proposed method successfully demonstrates incremental construction and planning over the 3DLSG representation to meet the objectives of the missions. Furthermore, the framework demonstrates successful semantic navigation tasks over the structured interface at the end of the inspection missions. Finally, we report multiple orders of magnitude reduction in path-planning time compared to conventional volumetric-map-based methods over various environment scale, demonstrating the planning efficiency and scalability of the proposed approach.
comment: 20 pages
♻ ☆ Bi-Adapt: Few-shot Bimanual Adaptation for Novel Categories of 3D Objects via Semantic Correspondence
Bimanual manipulation is imperative yet challenging for robots to execute complex tasks, requiring coordinated collaboration between two arms. However, existing methods for bimanual manipulation often rely on costly data collection and training, struggling to generalize to unseen objects in novel categories efficiently. In this paper, we present Bi-Adapt, a novel framework designed for efficient generalization for bimanual manipulation via semantic correspondence. Bi-Adapt achieves cross-category affordance mapping by leveraging the strong capability of vision foundation models. Fine-tuning with restricted data on novel categories, Bi-Adapt exhibits notable generalization to out-of-category objects in a zero-shot manner. Extensive experiments conducted in both simulation and real-world environments validate the effectiveness of our approach and demonstrate its high efficiency, achieving a high success rate on different benchmark tasks across novel categories with limited data. Project website: https://biadapt-project.github.io/
♻ ☆ Implicit State Estimation via Video Replanning
Video-based representations have gained prominence in planning and decision-making due to their ability to encode rich spatiotemporal dynamics and geometric relationships. These representations enable flexible and generalizable solutions for complex tasks such as object manipulation and navigation. However, existing video planning frameworks often struggle to adapt to failures at interaction time due to their inability to reason about uncertainties in partially observed environments. To overcome these limitations, we introduce a novel framework that integrates interaction-time data into the planning process. Our approach updates model parameters online and filters out previously failed plans during generation. This enables implicit state estimation, allowing the system to adapt dynamically without explicitly modeling unknown state variables. We evaluate our framework through extensive experiments on a new simulated manipulation benchmark, demonstrating its ability to improve replanning performance and advance the field of video-based decision-making.
♻ ☆ Symphony: A Heuristic Normalized Calibrated Advantage Actor and Critic Algorithm in application for Humanoid Robots
In our work we implicitly suggest that it is a misconception to think that humans learn fast. The learning process takes time. Babies start learning to move in the restricted fluid environment of the womb. Children are often limited by underdeveloped body. Even adults are not allowed to participate in complex competitions right away. However, with robots, when learning from scratch, we often don't have the privilege of waiting for tens of millions of steps. "Swaddling" regularization is responsible for restraining an agent in rapid but unstable development penalizing action strength in a specific way not affecting actions directly. The Symphony, Transitional-policy Deterministic Actor and Critic algorithm, is a concise combination of different ideas for possibility of training humanoid robots from scratch with Sample Efficiency, Sample Proximity and Safety of Actions in mind. It is well known that continuous increase in Gaussian noise without appropriate smoothing is harmful for motors and gearboxes. Compared to Stochastic algorithms, we set limited parametric noise and promote a reduced strength of actions, safely increasing entropy, since the actions are submerged in weaker noise. When actions require more extreme values, actions rise above the weak noise. Training becomes empirically much safer for both the environment around and the robot's mechanisms. We use Fading Replay Buffer: using a fixed formula containing the hyperbolic tangent, we adjust the batch sampling probability: the memory contains a recent memory and a long-term memory trail. Fading Replay Buffer allows us to use Temporal Advantage when we improve the current Critic Network prediction compared to the exponential moving average. Temporal Advantage allows us to update the Actor and Critic in one pass, as well as combine the Actor and Critic in one Object and implement their Losses in one line.
comment: https://github.com/SuspensionRailway/symphony
♻ ☆ RLinf-USER: A Unified and Extensible System for Real-World Online Policy Learning in Embodied AI
Online policy learning directly in the physical world is a promising yet challenging direction for embodied intelligence. Unlike simulation, real-world systems cannot be arbitrarily accelerated, cheaply reset, or massively replicated, which makes scalable data collection, heterogeneous deployment, and long-horizon effective training difficult. These challenges suggest that real-world policy learning is not only an algorithmic issue but fundamentally a systems problem. We present USER, a Unified and extensible SystEm for Real-world online policy learning. USER treats physical robots as first-class hardware resources alongside GPUs through a unified hardware abstraction layer, enabling automatic discovery, management, and scheduling of heterogeneous robots. To address cloud-edge communication, USER introduces an adaptive communication plane with tunneling-based networking, distributed data channels for traffic localization, and streaming-multiprocessor-aware weight synchronization to regulate GPU-side overhead. On top of this infrastructure, USER organizes learning as a fully asynchronous framework with a persistent, cache-aware buffer, enabling efficient long-horizon experiments with robust crash recovery and reuse of historical data. In addition, USER provides extensible abstractions for rewards, algorithms, and policies, supporting online imitation or reinforcement learning of CNN/MLP, generative policies, and large vision-language-action (VLA) models within a unified pipeline. Results in both simulation and the real world show that USER enables multi-robot coordination, heterogeneous manipulators, edge-cloud collaboration with large models, and long-running asynchronous training, offering a unified and extensible systems foundation for real-world online policy learning.
♻ ☆ LCLA: Language-Conditioned Latent Alignment for Vision-Language Navigation
We propose LCLA (Language-Conditioned Latent Alignment), a framework for vision-language navigation that learns modular perception-action interfaces by aligning sensory observations to a latent representation of an expert policy. The expert is first trained with privileged state information, inducing a latent space sufficient for control, after which its latent interface and action head are frozen. A lightweight adapter is then trained to map raw visual-language observations, via a frozen vision-language model, into the expert's latent space, reducing the problem of visuomotor learning to supervised latent alignment rather than end-to-end policy optimization. This decoupling enforces a stable contract between perception and control, enabling expert behavior to be reused across sensing modalities and environmental variations. We instantiate LCLA and evaluate it on a vision-language indoor navigation task, where aligned latent spaces yield strong in-distribution performance and robust zero-shot generalization to unseen environments, lighting conditions, and viewpoints while remaining lightweight at inference time.
♻ ☆ UFM: A Simple Path towards Unified Dense Correspondence with Flow
Dense image correspondence is central to many applications, such as visual odometry, 3D reconstruction, object association, and re-identification. Historically, dense correspondence has been tackled separately for wide-baseline scenarios and optical flow estimation, despite the common goal of matching content between two images. In this paper, we develop a Unified Flow & Matching model (UFM), which is trained on unified data for pixels that are co-visible in both source and target images. UFM uses a simple, generic transformer architecture that directly regresses the (u,v) flow. It is easier to train and more accurate for large flows compared to the typical coarse-to-fine cost volumes in prior work. UFM is 28% more accurate than state-of-the-art flow methods (Unimatch), while also having 62% less error and 6.7x faster than dense wide-baseline matchers (RoMa). UFM is the first to demonstrate that unified training can outperform specialized approaches across both domains. This result enables fast, general-purpose correspondence and opens new directions for multi-modal, long-range, and real-time correspondence tasks.
comment: Project Page: https://uniflowmatch.github.io/
♻ ☆ Sim2real Image Translation Enables Viewpoint-Robust Policies from Fixed-Camera Datasets
Vision-based policies for robot manipulation have achieved significant recent success, but are still brittle to distribution shifts such as camera viewpoint variations. Robot demonstration data is scarce and often lacks appropriate variation in camera viewpoints. Simulation offers a way to collect robot demonstrations at scale with comprehensive coverage of different viewpoints, but presents a visual sim2real challenge. To bridge this gap, we propose MANGO -- an unpaired image translation method with a novel segmentation-conditioned InfoNCE loss, a highly-regularized discriminator design, and a modified PatchNCE loss. We find that these elements are crucial for maintaining viewpoint consistency during sim2real translation. When training MANGO, we only require a small amount of fixed-camera data from the real world, but show that our method can generate diverse unseen viewpoints by translating simulated observations. In this domain, MANGO outperforms all other image translation methods we tested. Imitation-learning policies trained on data augmented by MANGO are able to achieve success rates as high as 60% on views that the non-augmented policy fails completely on.
♻ ☆ Multi-Momentum Observer Contact Estimation for Bipedal Robots
As bipedal robots become more and more popular in commercial and industrial settings, the ability to control them with a high degree of reliability is critical. To that end, this paper considers how to accurately estimate which feet are currently in contact with the ground so as to avoid improper control actions that could jeopardize the stability of the robot. Additionally, modern algorithms for estimating the position and orientation of a robot's base frame rely heavily on such contact mode estimates. Dedicated contact sensors on the feet can be used to estimate this contact mode, but these sensors are prone to noise, time delays, damage/yielding from repeated impacts with the ground, and are not available on every robot. To overcome these limitations, we propose a momentum observer based method for contact mode estimation that does not rely on such contact sensors. Often, momentum observers assume that the robot's base frame can be treated as an inertial frame. However, since many humanoids' legs represent a significant portion of the overall mass, the proposed method instead utilizes multiple simultaneous dynamic models. Each of these models assumes a different contact condition. A given contact assumption is then used to constrain the full dynamics in order to avoid assuming that either the body is an inertial frame or that a fully accurate estimate of body velocity is known. The (dis)agreement between each model's estimates and measurements is used to determine which contact mode is most likely using a Markov-style fusion method. The proposed method produces contact detection accuracy of up to 98.44% with a low noise simulation and 77.12% when utilizing data collect on the Sarcos Guardian XO robot (a hybrid humanoid/exoskeleton).
Robotics 81
☆ 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.
☆ $χ_{0}$: 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.
☆ Robustness Is a Function, Not a Number: A Factorized Comprehensive Study of OOD Robustness in Vision-Based Driving
Out of distribution (OOD) robustness in autonomous driving is often reduced to a single number, hiding what breaks a policy. We decompose environments along five axes: scene (rural/urban), season, weather, time (day/night), and agent mix; and measure performance under controlled $k$-factor perturbations ($k \in \{0,1,2,3\}$). Using closed loop control in VISTA, we benchmark FC, CNN, and ViT policies, train compact ViT heads on frozen foundation-model (FM) features, and vary ID support in scale, diversity, and temporal context. (1) ViT policies are markedly more OOD-robust than comparably sized CNN/FC, and FM features yield state-of-the-art success at a latency cost. (2) Naive temporal inputs (multi-frame) do not beat the best single-frame baseline. (3) The largest single factor drops are rural $\rightarrow$ urban and day $\rightarrow$ night ($\sim 31\%$ each); actor swaps $\sim 10\%$, moderate rain $\sim 7\%$; season shifts can be drastic, and combining a time flip with other changes further degrades performance. (4) FM-feature policies stay above $85\%$ under three simultaneous changes; non-FM single-frame policies take a large first-shift hit, and all no-FM models fall below $50\%$ by three changes. (5) Interactions are non-additive: some pairings partially offset, whereas season-time combinations are especially harmful. (6) Training on winter/snow is most robust to single-factor shifts, while a rural+summer baseline gives the best overall OOD performance. (7) Scaling traces/views improves robustness ($+11.8$ points from $5$ to $14$ traces), yet targeted exposure to hard conditions can substitute for scale. (8) Using multiple ID environments broadens coverage and strengthens weak cases (urban OOD $60.6\% \rightarrow 70.1\%$) with a small ID drop; single-ID preserves peak performance but in a narrow domain. These results yield actionable design rules for OOD-robust driving policies.
☆ Contact-Anchored Policies: Contact Conditioning Creates Strong Robot Utility Models
The prevalent paradigm in robot learning attempts to generalize across environments, embodiments, and tasks with language prompts at runtime. A fundamental tension limits this approach: language is often too abstract to guide the concrete physical understanding required for robust manipulation. In this work, we introduce Contact-Anchored Policies (CAP), which replace language conditioning with points of physical contact in space. Simultaneously, we structure CAP as a library of modular utility models rather than a monolithic generalist policy. This factorization allows us to implement a real-to-sim iteration cycle: we build EgoGym, a lightweight simulation benchmark, to rapidly identify failure modes and refine our models and datasets prior to real-world deployment. We show that by conditioning on contact and iterating via simulation, CAP generalizes to novel environments and embodiments out of the box on three fundamental manipulation skills while using only 23 hours of demonstration data, and outperforms large, state-of-the-art VLAs in zero-shot evaluations by 56%. All model checkpoints, codebase, hardware, simulation, and datasets will be open-sourced. Project page: https://cap-policy.github.io/
☆ Dexterous Manipulation Policies from RGB Human Videos via 4D Hand-Object Trajectory Reconstruction
Multi-finger robotic hand manipulation and grasping are challenging due to the high-dimensional action space and the difficulty of acquiring large-scale training data. Existing approaches largely rely on human teleoperation with wearable devices or specialized sensing equipment to capture hand-object interactions, which limits scalability. In this work, we propose VIDEOMANIP, a device-free framework that learns dexterous manipulation directly from RGB human videos. Leveraging recent advances in computer vision, VIDEOMANIP reconstructs explicit 4D robot-object trajectories from monocular videos by estimating human hand poses, object meshes, and retargets the reconstructed human motions to robotic hands for manipulation learning. To make the reconstructed robot data suitable for dexterous manipulation training, we introduce hand-object contact optimization with interaction-centric grasp modeling, as well as a demonstration synthesis strategy that generates diverse training trajectories from a single video, enabling generalizable policy learning without additional robot demonstrations. In simulation, the learned grasping model achieves a 70.25% success rate across 20 diverse objects using the Inspire Hand. In the real world, manipulation policies trained from RGB videos achieve an average 62.86% success rate across seven tasks using the LEAP Hand, outperforming retargeting-based methods by 15.87%. Project videos are available at videomanip.github.io.
☆ From Obstacles to Etiquette: Robot Social Navigation with VLM-Informed Path Selection
Navigating socially in human environments requires more than satisfying geometric constraints, as collision-free paths may still interfere with ongoing activities or conflict with social norms. Addressing this challenge calls for analyzing interactions between agents and incorporating common-sense reasoning into planning. This paper presents a social robot navigation framework that integrates geometric planning with contextual social reasoning. The system first extracts obstacles and human dynamics to generate geometrically feasible candidate paths, then leverages a fine-tuned vision-language model (VLM) to evaluate these paths, informed by contextually grounded social expectations, selecting a socially optimized path for the controller. This task-specific VLM distills social reasoning from large foundation models into a smaller and efficient model, allowing the framework to perform real-time adaptation in diverse human-robot interaction contexts. Experiments in four social navigation contexts demonstrate that our method achieves the best overall performance with the lowest personal space violation duration, the minimal pedestrian-facing time, and no social zone intrusions. Project page: https://path-etiquette.github.io
comment: Accepted to IEEE Robotics and Automation Letters (RA-L)
☆ CLUE: Crossmodal disambiguation via Language-vision Understanding with attEntion
With the increasing integration of robots into daily life, human-robot interaction has become more complex and multifaceted. A critical component of this interaction is Interactive Visual Grounding (IVG), through which robots must interpret human intentions and resolve ambiguity. Existing IVG models generally lack a mechanism to determine when to ask clarification questions, as they implicitly rely on their learned representations. CLUE addresses this gap by converting the VLM's cross-modal attention into an explicit, spatially grounded signal for deciding when to ask. We extract text to image attention maps and pass them to a lightweight CNN to detect referential ambiguity, while a LoRA fine-tuned decoder conducts the dialog and emits grounding location tokens. We train on a real-world interactive dataset for IVG, and a mixed ambiguity set for the detector. With InViG-only supervision, our model surpasses a state-of-the-art method while using parameter-efficient fine-tuning. Similarly, the ambiguity detector outperforms prior baselines. Overall, CLUE turns the internal cross-modal attention of a VLM into an explicit, spatially grounded signal for deciding when to ask. The data and code are publicly available at: mouadabrini.github.io/clue
☆ WorldArena: A Unified Benchmark for Evaluating Perception and Functional Utility of Embodied World Models
While world models have emerged as a cornerstone of embodied intelligence by enabling agents to reason about environmental dynamics through action-conditioned prediction, their evaluation remains fragmented. Current evaluation of embodied world models has largely focused on perceptual fidelity (e.g., video generation quality), overlooking the functional utility of these models in downstream decision-making tasks. In this work, we introduce WorldArena, a unified benchmark designed to systematically evaluate embodied world models across both perceptual and functional dimensions. WorldArena assesses models through three dimensions: video perception quality, measured with 16 metrics across six sub-dimensions; embodied task functionality, which evaluates world models as data engines, policy evaluators, and action planners integrating with subjective human evaluation. Furthermore, we propose EWMScore, a holistic metric integrating multi-dimensional performance into a single interpretable index. Through extensive experiments on 14 representative models, we reveal a significant perception-functionality gap, showing that high visual quality does not necessarily translate into strong embodied task capability. WorldArena benchmark with the public leaderboard is released at https://worldarena.ai, providing a framework for tracking progress toward truly functional world models in embodied AI.
☆ Reduced-order Control and Geometric Structure of Learned Lagrangian Latent Dynamics
Model-based controllers can offer strong guarantees on stability and convergence by relying on physically accurate dynamic models. However, these are rarely available for high-dimensional mechanical systems such as deformable objects or soft robots. While neural architectures can learn to approximate complex dynamics, they are either limited to low-dimensional systems or provide only limited formal control guarantees due to a lack of embedded physical structure. This paper introduces a latent control framework based on learned structure-preserving reduced-order dynamics for high-dimensional Lagrangian systems. We derive a reduced tracking law for fully actuated systems and adopt a Riemannian perspective on projection-based model-order reduction to study the resulting latent and projected closed-loop dynamics. By quantifying the sources of modeling error, we derive interpretable conditions for stability and convergence. We extend the proposed controller and analysis to underactuated systems by introducing learned actuation patterns. Experimental results on simulated and real-world systems validate our theoretical investigation and the accuracy of our controllers.
comment: 20 pages, 15 figures
☆ Modeling 3D Pedestrian-Vehicle Interactions for Vehicle-Conditioned Pose Forecasting ICRA
Accurately predicting pedestrian motion is crucial for safe and reliable autonomous driving in complex urban environments. In this work, we present a 3D vehicle-conditioned pedestrian pose forecasting framework that explicitly incorporates surrounding vehicle information. To support this, we enhance the Waymo-3DSkelMo dataset with aligned 3D vehicle bounding boxes, enabling realistic modeling of multi-agent pedestrian-vehicle interactions. We introduce a sampling scheme to categorize scenes by pedestrian and vehicle count, facilitating training across varying interaction complexities. Our proposed network adapts the TBIFormer architecture with a dedicated vehicle encoder and pedestrian-vehicle interaction cross-attention module to fuse pedestrian and vehicle features, allowing predictions to be conditioned on both historical pedestrian motion and surrounding vehicles. Extensive experiments demonstrate substantial improvements in forecasting accuracy and validate different approaches for modeling pedestrian-vehicle interactions, highlighting the importance of vehicle-aware 3D pose prediction for autonomous driving. Code is available at: https://github.com/GuangxunZhu/VehCondPose3D
comment: Accepted for IEEE International Conference on Robotics and Automation (ICRA) 2026
☆ Finite-Time Teleoperation of Euler-Lagrange Systems via Energy-Shaping
This paper proposes a family of finite-time controllers for the bilateral teleoperation of fully actuated nonlinear Euler-Lagrange systems. Based on the energy-shaping framework and under the standard assumption of passive interactions with the human and the environment, the controllers ensure that the position error and velocities globally converge to zero in the absence of time delays. In this case, the closed-loop system admits a homogeneous approximation of negative degree, and thus the control objective is achieved in finite-time. The proposed controllers are simple, continuous-time proportional-plus-damping-injection schemes, validated through both simulation and experimental results.
☆ karl. -- A Research Vehicle for Automated and Connected Driving
As highly automated driving is transitioning from single-vehicle closed-access testing to commercial deployments of public ride-hailing in selected areas (e.g., Waymo), automated driving and connected cooperative intelligent transport systems (C-ITS) remain active fields of research. Even though simulation is omnipresent in the development and validation life cycle of automated and connected driving technology, the complex nature of public road traffic and software that masters it still requires real-world integration and testing with actual vehicles. Dedicated vehicles for research and development allow testing and validation of software and hardware components under real-world conditions early on. They also enable collecting and publishing real-world datasets that let others conduct research without vehicle access, and support early demonstration of futuristic use cases. In this paper, we present karl., our new research vehicle for automated and connected driving. Apart from major corporations, few institutions worldwide have access to their own L4-capable research vehicles, restricting their ability to carry out independent research. This paper aims to help bridge that gap by sharing the reasoning, design choices, and technical details that went into making karl. a flexible and powerful platform for research, engineering, and validation in the context of automated and connected driving. More impressions of karl. are available at https://karl.ac.
comment: 8 pages; Accepted to be published as part of the 37th Intelligent Vehicles Symposium (IV), Detroit, MI, United States, June 22-25, 2026
☆ Multi-Staged Framework for Safety Analysis of Offloaded Services in Distributed Intelligent Transportation Systems
The integration of service-oriented architectures (SOA) with function offloading for distributed, intelligent transportation systems (ITS) offers the opportunity for connected autonomous vehicles (CAVs) to extend their locally available services. One major goal of offloading a subset of functions in the processing chain of a CAV to remote devices is to reduce the overall computational complexity on the CAV. The extension of using remote services, however, requires careful safety analysis, since the remotely created data are corrupted more easily, e.g., through an attacker on the remote device or by intercepting the wireless transmission. To tackle this problem, we first analyze the concept of SOA for distributed environments. From this, we derive a safety framework that validates the reliability of remote services and the data received locally. Since it is possible for the autonomous driving task to offload multiple different services, we propose a specific multi-staged framework for safety analysis dependent on the service composition of local and remote services. For efficiency reasons, we directly include the multi-staged framework for safety analysis in our service-oriented function offloading framework (SOFOF) that we have proposed in earlier work. The evaluation compares the performance of the extended framework considering computational complexity, with energy savings being a major motivation for function offloading, and its capability to detect data from corrupted remote services.
☆ A Generic Service-Oriented Function Offloading Framework for Connected Automated Vehicles
Function offloading is a promising solution to address limitations concerning computational capacity and available energy of Connected Automated Vehicles~(CAVs) or other autonomous robots by distributing computational tasks between local and remote computing devices in form of distributed services. This paper presents a generic function offloading framework that can be used to offload an arbitrary set of computational tasks with a focus on autonomous driving. To provide flexibility, the function offloading framework is designed to incorporate different offloading decision making algorithms and quality of service~(QoS) requirements that can be adjusted to different scenarios or the objectives of the CAVs. With a focus on the applicability, we propose an efficient location-based approach, where the decision whether tasks are processed locally or remotely depends on the location of the CAV. We apply the proposed framework on the use case of service-oriented trajectory planning, where we offload the trajectory planning task of CAVs to a Multi-Access Edge Computing~(MEC) server. The evaluation is conducted in both simulation and real-world application. It demonstrates the potential of the function offloading framework to guarantee the QoS for trajectory planning while improving the computational efficiency of the CAVs. Moreover, the simulation results also show the adaptability of the framework to diverse scenarios involving simultaneous offloading requests from multiple CAVs.
comment: 8 pages, 6 figures, 2 tables, published in RA-L
☆ GaussianCaR: Gaussian Splatting for Efficient Camera-Radar Fusion ICRA 2026
Robust and accurate perception of dynamic objects and map elements is crucial for autonomous vehicles performing safe navigation in complex traffic scenarios. While vision-only methods have become the de facto standard due to their technical advances, they can benefit from effective and cost-efficient fusion with radar measurements. In this work, we advance fusion methods by repurposing Gaussian Splatting as an efficient universal view transformer that bridges the view disparity gap, mapping both image pixels and radar points into a common Bird's-Eye View (BEV) representation. Our main contribution is GaussianCaR, an end-to-end network for BEV segmentation that, unlike prior BEV fusion methods, leverages Gaussian Splatting to map raw sensor information into latent features for efficient camera-radar fusion. Our architecture combines multi-scale fusion with a transformer decoder to efficiently extract BEV features. Experimental results demonstrate that our approach achieves performance on par with, or even surpassing, the state of the art on BEV segmentation tasks (57.3%, 82.9%, and 50.1% IoU for vehicles, roads, and lane dividers) on the nuScenes dataset, while maintaining a 3.2x faster inference runtime. Code and project page are available online.
comment: 8 pages, 6 figures. Accepted to IEEE ICRA 2026
☆ Mind the Gap: Learning Implicit Impedance in Visuomotor Policies via Intent-Execution Mismatch
Teleoperation inherently relies on the human operator acting as a closed-loop controller to actively compensate for hardware imperfections, including latency, mechanical friction, and lack of explicit force feedback. Standard Behavior Cloning (BC), by mimicking the robot's executed trajectory, fundamentally ignores this compensatory mechanism. In this work, we propose a Dual-State Conditioning framework that shifts the learning objective to "Intent Cloning" (master command). We posit that the Intent-Execution Mismatch, the discrepancy between master command and slave response, is not noise, but a critical signal that physically encodes implicit interaction forces and algorithmically reveals the operator's strategy for overcoming system dynamics. By predicting the master intent, our policy learns to generate a "virtual equilibrium point", effectively realizing implicit impedance control. Furthermore, by explicitly conditioning on the history of this mismatch, the model performs implicit system identification, perceiving tracking errors as external forces to close the control loop. To bridge the temporal gap caused by inference latency, we further formulate the policy as a trajectory inpainter to ensure continuous control. We validate our approach on a sensorless, low-cost bi-manual setup. Empirical results across tasks requiring contact-rich manipulation and dynamic tracking reveal a decisive gap: while standard execution-cloning fails due to the inability to overcome contact stiffness and tracking lag, our mismatch-aware approach achieves robust success. This presents a minimalist behavior cloning framework for low-cost hardware, enabling force perception and dynamic compensation without relying on explicit force sensing. Videos are available on the \href{https://xucj98.github.io/mind-the-gap-page/}{project page}.
comment: 14 pages, 9 figures, 5 tables
☆ High-Speed Vision-Based Flight in Clutter with Safety-Shielded Reinforcement Learning
Quadrotor unmanned aerial vehicles (UAVs) are increasingly deployed in complex missions that demand reliable autonomous navigation and robust obstacle avoidance. However, traditional modular pipelines often incur cumulative latency, whereas purely reinforcement learning (RL) approaches typically provide limited formal safety guarantees. To bridge this gap, we propose an end-to-end RL framework augmented with model-based safety mechanisms. We incorporate physical priors in both training and deployment. During training, we design a physics-informed reward structure that provides global navigational guidance. During deployment, we integrate a real-time safety filter that projects the policy outputs onto a provably safe set to enforce strict collision-avoidance constraints. This hybrid architecture reconciles high-speed flight with robust safety assurances. Benchmark evaluations demonstrate that our method outperforms both traditional planners and recent end-to-end obstacle avoidance approaches based on differentiable physics. Extensive experiments demonstrate strong generalization, enabling reliable high-speed navigation in dense clutter and challenging outdoor forest environments at velocities up to 7.5m/s.
☆ 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: \textit{``Mimic Intent, Not just Trajectories'' (MINT)}. We achieve this via \textit{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 \textit{Intent token} that facilitates planning and transfer, and multi-scale \textit{Execution tokens} that enable precise adaptation to environmental dynamics. Building on this hierarchy, our policy generates trajectories through \textit{next-scale autoregression}, performing progressive \textit{intent-to-execution reasoning}, thus boosting learning efficiency and generalization. Crucially, this disentanglement enables \textit{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
☆ A Precise Real-Time Force-Aware Grasping System for Robust Aerial Manipulation
Aerial manipulation requires force-aware capabilities to enable safe and effective grasping and physical interaction. Previous works often rely on heavy, expensive force sensors unsuitable for typical quadrotor platforms, or perform grasping without force feedback, risking damage to fragile objects. To address these limitations, we propose a novel force-aware grasping framework incorporating six low-cost, sensitive skin-like tactile sensors. We introduce a magnetic-based tactile sensing module that provides high-precision three-dimensional force measurements. We eliminate geomagnetic interference through a reference Hall sensor and simplify the calibration process compared to previous work. The proposed framework enables precise force-aware grasping control, allowing safe manipulation of fragile objects and real-time weight measurement of grasped items. The system is validated through comprehensive real-world experiments, including balloon grasping, dynamic load variation tests, and ablation studies, demonstrating its effectiveness in various aerial manipulation scenarios. Our approach achieves fully onboard operation without external motion capture systems, significantly enhancing the practicality of force-sensitive aerial manipulation. The supplementary video is available at: https://www.youtube.com/watch?v=mbcZkrJEf1I.
☆ MOSAIC: Bridging the Sim-to-Real Gap in Generalist Humanoid Motion Tracking and Teleoperation with Rapid Residual Adaptation
Generalist humanoid motion trackers have recently achieved strong simulation metrics by scaling data and training, yet often remain brittle on hardware during sustained teleoperation due to interface- and dynamics-induced errors. We present MOSAIC, an open-source, full-stack system for humanoid motion tracking and whole-body teleoperation across multiple interfaces. MOSAIC first learns a teleoperation-oriented general motion tracker via RL on a multi-source motion bank with adaptive resampling and rewards that emphasize world-frame motion consistency, which is critical for mobile teleoperation. To bridge the sim-to-real interface gap without sacrificing generality, MOSAIC then performs rapid residual adaptation: an interface-specific policy is trained using minimal interface-specific data, and then distilled into the general tracker through an additive residual module, outperforming naive fine-tuning or continual learning. We validate MOSAIC with systematic ablations, out-of-distribution benchmarking, and real-robot experiments demonstrating robust offline motion replay and online long-horizon teleoperation under realistic latency and noise.
☆ Head-to-Head autonomous racing at the limits of handling in the A2RL challenge
Autonomous racing presents a complex challenge involving multi-agent interactions between vehicles operating at the limit of performance and dynamics. As such, it provides a valuable research and testing environment for advancing autonomous driving technology and improving road safety. This article presents the algorithms and deployment strategies developed by the TUM Autonomous Motorsport team for the inaugural Abu Dhabi Autonomous Racing League (A2RL). We showcase how our software emulates human driving behavior, pushing the limits of vehicle handling and multi-vehicle interactions to win the A2RL. Finally, we highlight the key enablers of our success and share our most significant learnings.
comment: Submitted to Science Robotics for possible publication
☆ Constrained Sampling to Guide Universal Manipulation RL
We consider how model-based solvers can be leveraged to guide training of a universal policy to control from any feasible start state to any feasible goal in a contact-rich manipulation setting. While Reinforcement Learning (RL) has demonstrated its strength in such settings, it may struggle to sufficiently explore and discover complex manipulation strategies, especially in sparse-reward settings. Our approach is based on the idea of a lower-dimensional manifold of feasible, likely-visited states during such manipulation and to guide RL with a sampler from this manifold. We propose Sample-Guided RL, which uses model-based constraint solvers to efficiently sample feasible configurations (satisfying differentiable collision, contact, and force constraints) and leverage them to guide RL for universal (goal-conditioned) manipulation policies. We study using this data directly to bias state visitation, as well as using black-box optimization of open-loop trajectories between random configurations to impose a state bias and optionally add a behavior cloning loss. In a minimalistic double sphere manipulation setting, Sample-Guided RL discovers complex manipulation strategies and achieves high success rates in reaching any statically stable state. In a more challenging panda arm setting, our approach achieves a significant success rate over a near-zero baseline, and demonstrates a breadth of complex whole-body-contact manipulation strategies.
☆ UniPlan: Vision-Language Task Planning for Mobile Manipulation with Unified PDDL Formulation
Integration of VLM reasoning with symbolic planning has proven to be a promising approach to real-world robot task planning. Existing work like UniDomain effectively learns symbolic manipulation domains from real-world demonstrations, described in Planning Domain Definition Language (PDDL), and has successfully applied them to real-world tasks. These domains, however, are restricted to tabletop manipulation. We propose UniPlan, a vision-language task planning system for long-horizon mobile-manipulation in large-scale indoor environments, that unifies scene topology, visuals, and robot capabilities into a holistic PDDL representation. UniPlan programmatically extends learned tabletop domains from UniDomain to support navigation, door traversal, and bimanual coordination. It operates on a visual-topological map, comprising navigation landmarks anchored with scene images. Given a language instruction, UniPlan retrieves task-relevant nodes from the map and uses a VLM to ground the anchored image into task-relevant objects and their PDDL states; next, it reconnects these nodes to a compressed, densely-connected topological map, also represented in PDDL, with connectivity and costs derived from the original map; Finally, a mobile-manipulation plan is generated using off-the-shelf PDDL solvers. Evaluated on human-raised tasks in a large-scale map with real-world imagery, UniPlan significantly outperforms VLM and LLM+PDDL planning in success rate, plan quality, and computational efficiency.
☆ Characteristics, Management, and Utilization of Muscles in Musculoskeletal Humanoids: Empirical Study on Kengoro and Musashi
Various musculoskeletal humanoids have been developed so far, and numerous studies on control mechanisms have been conducted to leverage the advantages of their biomimetic bodies. However, there has not been sufficient and unified discussion on the diverse properties inherent in these musculoskeletal structures, nor on how to manage and utilize them. Therefore, this study categorizes and analyzes the characteristics of muscles, as well as their management and utilization methods, based on the various research conducted on the musculoskeletal humanoids we have developed, Kengoro and Musashi. We classify the features of the musculoskeletal structure into five properties: Redundancy, Independency, Anisotropy, Variable Moment Arm, and Nonlinear Elasticity. We then organize the diverse advantages and disadvantages of musculoskeletal humanoids that arise from the combination of these properties. In particular, we discuss body schema learning and reflex control, along with muscle grouping and body schema adaptation. Also, we describe the implementation of movements through an integrated system and discuss future challenges and prospects.
comment: Accepted to Advanced Intelligent Systems
☆ Reliability-aware Execution Gating for Near-field and Off-axis Vision-guided Robotic Alignment
Vision-guided robotic systems are increasingly deployed in precision alignment tasks that require reliable execution under near-field and off-axis configurations. While recent advances in pose estimation have significantly improved numerical accuracy, practical robotic systems still suffer from frequent execution failures even when pose estimates appear accurate. This gap suggests that pose accuracy alone is insufficient to guarantee execution-level reliability. In this paper, we reveal that such failures arise from a deterministic geometric error amplification mechanism, in which small pose estimation errors are magnified through system structure and motion execution, leading to unstable or failed alignment. Rather than modifying pose estimation algorithms, we propose a Reliability-aware Execution Gating mechanism that operates at the execution level. The proposed approach evaluates geometric consistency and configuration risk before execution, and selectively rejects or scales high-risk pose updates. We validate the proposed method on a real UR5 robotic platform performing single-step visual alignment tasks under varying camera-target distances and off-axis configurations. Experimental results demonstrate that the proposed execution gating significantly improves task success rates, reduces execution variance, and suppresses tail-risk behavior, while leaving average pose accuracy largely unchanged. Importantly, the proposed mechanism is estimator-agnostic and can be readily integrated with both classical geometry-based and learning-based pose estimation pipelines. These results highlight the importance of execution-level reliability modeling and provide a practical solution for improving robustness in near-field vision-guided robotic systems.
comment: 7 pages, 1 figure
☆ UAV-Supported Maritime Search System: Experience from Valun Bay Field Trials
This paper presents the integration of flow field reconstruction, dynamic probabilistic modeling, search control, and machine vision detection in a system for autonomous maritime search operations. Field experiments conducted in Valun Bay (Cres Island, Croatia) involved real-time drifter data acquisition, surrogate flow model fitting based on computational fluid dynamics and numerical optimization, advanced multi-UAV search control and vision sensing, as well as deep learning-based object detection. The results demonstrate that a tightly coupled approach enables reliable detection of floating targets under realistic uncertainties and complex environmental conditions, providing concrete insights for future autonomous maritime search and rescue applications.
☆ Post-Collision Trajectory Restoration for a Single-track Ackermann Vehicle using Heuristic Steering and Tractive Force Functions
Post-collision trajectory restoration is a safety-critical capability for autonomous vehicles, as impact-induced lateral motion and yaw transients can rapidly drive the vehicle away from the intended path. This paper proposes a structured heuristic recovery control law that jointly commands steering and tractive force for a generalized single-track Ackermann vehicle model. The formulation explicitly accounts for time-varying longitudinal velocity in the lateral-yaw dynamics and retains nonlinear steering-coupled interaction terms that are commonly simplified in the literature. Unlike approaches that assume constant longitudinal speed, the proposed design targets the transient post-impact regime where speed variations and nonlinear coupling significantly influence recovery. The method is evaluated in simulation on the proposed generalized single-track model and a standard 3DOF single-track reference model in MATLAB, demonstrating consistent post-collision restoration behaviour across representative initial post-impact conditions.
comment: 10 pages, 6 figures
☆ SteerVLA: Steering Vision-Language-Action Models in Long-Tail Driving Scenarios
A fundamental challenge in autonomous driving is the integration of high-level, semantic reasoning for long-tail events with low-level, reactive control for robust driving. While large vision-language models (VLMs) trained on web-scale data offer powerful common-sense reasoning, they lack the grounded experience necessary for safe vehicle control. We posit that an effective autonomous agent should leverage the world knowledge of VLMs to guide a steerable driving policy toward robust control in driving scenarios. To this end, we propose SteerVLA, which leverages the reasoning capabilities of VLMs to produce fine-grained language instructions that steer a vision-language-action (VLA) driving policy. Key to our method is this rich language interface between the high-level VLM and low-level VLA, which allows the high-level policy to more effectively ground its reasoning in the control outputs of the low-level policy. To provide fine-grained language supervision aligned with vehicle control, we leverage a VLM to augment existing driving data with detailed language annotations, which we find to be essential for effective reasoning and steerability. We evaluate SteerVLA on a challenging closed-loop benchmark, where it outperforms state-of-the-art methods by 4.77 points in overall driving score and by 8.04 points on a long-tail subset. The project website is available at: https://steervla.github.io/.
☆ Bi-Adapt: Few-shot Bimanual Adaptation for Novel Categories of 3D Objects via Semantic Correspondence
Bimanual manipulation is imperative yet challenging for robots to execute complex tasks, requiring coordinated collaboration between two arms. However, existing methods for bimanual manipulation often rely on costly data collection and training, struggling to generalize to unseen objects in novel categories efficiently. In this paper, we present Bi-Adapt, a novel framework designed for efficient generalization for bimanual manipulation via semantic correspondence. Bi-Adapt achieves cross-category affordance mapping by leveraging the strong capability of vision foundation models. Fine-tuning with restricted data on novel categories, Bi-Adapt exhibits notable generalization to out-of-category objects in a zero-shot manner. Extensive experiments conducted in both simulation and real-world environments validate the effectiveness of our approach and demonstrate its high efficiency, achieving a high success rate on different benchmark tasks across novel categories with limited data. Project website: https://biadapt-project.github.io/
☆ Decentralized Intent-Based Multi-Robot Task Planner with LLM Oracles on Hyperledger Fabric
Large language models (LLMs) have opened new opportunities for transforming natural language user intents into executable actions. This capability enables embodied AI agents to perform complex tasks, without involvement of an expert, making human-robot interaction (HRI) more convenient. However these developments raise significant security and privacy challenges such as self-preferencing, where a single LLM service provider dominates the market and uses this power to promote their own preferences. LLM oracles have been recently proposed as a mechanism to decentralize LLMs by executing multiple LLMs from different vendors and aggregating their outputs to obtain a more reliable and trustworthy final result. However, the accuracy of these approaches highly depends on the aggregation method. The current aggregation methods mostly use semantic similarity between various LLM outputs, not suitable for robotic task planning, where the temporal order of tasks is important. To fill the gap, we propose an LLM oracle with a new aggregation method for robotic task planning. In addition, we propose a decentralized multi-robot infrastructure based on Hyperledger Fabric that can host the proposed oracle. The proposed infrastructure enables users to express their natural language intent to the system, which then can be decomposed into subtasks. These subtasks require coordinating different robots from different vendors, while enforcing fine-grained access control management on the data. To evaluate our methodology, we created the SkillChain-RTD benchmark made it publicly available. Our experimental results demonstrate the feasibility of the proposed architecture, and the proposed aggregation method outperforms other aggregation methods currently in use.
☆ Graph-Loc: Robust Graph-Based LiDAR Pose Tracking with Compact Structural Map Priors under Low Observability and Occlusion
Map-based LiDAR pose tracking is essential for long-term autonomous operation, where onboard map priors need be compact for scalable storage and fast retrieval, while online observations are often partial, repetitive, and heavily occluded. We propose Graph-Loc, a graph-based localization framework that tracks the platform pose against compact structural map priors represented as a lightweight point-line graph. Such priors can be constructed from heterogeneous sources commonly available in practice, including polygon outlines vectorized from occupancy/grid maps and CAD/model/floor-plan layouts. For each incoming LiDAR scan, Graph-Loc extracts sparse point and line primitives to form an observation graph, retrieves a pose-conditioned visible subgraph via LiDAR ray simulation, and performs scan-to-map association through unbalanced optimal transport with a local graph-context regularizer. The unbalanced formulation relaxes mass conservation, improving robustness to missing, spurious, and fragmented structures under occlusion. To enhance stability in low-observability segments, we estimate information anisotropy from the refinement normal matrix and defer updates along weakly constrained directions until sufficient constraints reappear. Experiments on public benchmarks, controlled stress tests, and real-world deployments demonstrate accurate and stable tracking with KB-level priors from heterogeneous map sources, including under geometrically degenerate and sustained occlusion and in the presence of gradual scene changes.
comment: 13 pages, 8 figures, 8 tables
☆ BiManiBench: A Hierarchical Benchmark for Evaluating Bimanual Coordination of Multimodal Large Language Models
Multimodal Large Language Models (MLLMs) have significantly advanced embodied AI, and using them to benchmark robotic intelligence has become a pivotal trend. However, existing frameworks remain predominantly confined to single-arm manipulation, failing to capture the spatio-temporal coordination required for bimanual tasks like lifting a heavy pot. To address this, we introduce BiManiBench, a hierarchical benchmark evaluating MLLMs across three tiers: fundamental spatial reasoning, high-level action planning, and low-level end-effector control. Our framework isolates unique bimanual challenges, such as arm reachability and kinematic constraints, thereby distinguishing perceptual hallucinations from planning failures. Analysis of over 30 state-of-the-art models reveals that despite high-level reasoning proficiency, MLLMs struggle with dual-arm spatial grounding and control, frequently resulting in mutual interference and sequencing errors. These findings suggest the current paradigm lacks a deep understanding of mutual kinematic constraints, highlighting the need for future research to focus on inter-arm collision-avoidance and fine-grained temporal sequencing.
comment: 38 pages, 9 figures. Project page:https://bimanibench.github.io/
☆ Learning Human-Like Badminton Skills for Humanoid Robots
Realizing versatile and human-like performance in high-demand sports like badminton remains a formidable challenge for humanoid robotics. Unlike standard locomotion or static manipulation, this task demands a seamless integration of explosive whole-body coordination and precise, timing-critical interception. While recent advances have achieved lifelike motion mimicry, bridging the gap between kinematic imitation and functional, physics-aware striking without compromising stylistic naturalness is non-trivial. To address this, we propose Imitation-to-Interaction, a progressive reinforcement learning framework designed to evolve a robot from a "mimic" to a capable "striker." Our approach establishes a robust motor prior from human data, distills it into a compact, model-based state representation, and stabilizes dynamics via adversarial priors. Crucially, to overcome the sparsity of expert demonstrations, we introduce a manifold expansion strategy that generalizes discrete strike points into a dense interaction volume. We validate our framework through the mastery of diverse skills, including lifts and drop shots, in simulation. Furthermore, we demonstrate the first zero-shot sim-to-real transfer of anthropomorphic badminton skills to a humanoid robot, successfully replicating the kinetic elegance and functional precision of human athletes in the physical world.
comment: 10 pages, 4 figures
☆ Vec-QMDP: Vectorized POMDP Planning on CPUs for Real-Time Autonomous Driving
Planning under uncertainty for real-world robotics tasks, such as autonomous driving, requires reasoning in enormous high-dimensional belief spaces, rendering the problem computationally intensive. While parallelization offers scalability, existing hybrid CPU-GPU solvers face critical bottlenecks due to host-device synchronization latency and branch divergence on SIMT architectures, limiting their utility for real-time planning and hindering real-robot deployment. We present Vec-QMDP, a CPU-native parallel planner that aligns POMDP search with modern CPUs' SIMD architecture, achieving $227\times$--$1073\times$ speedup over state-of-the-art serial planners. Vec-QMDP adopts a Data-Oriented Design (DOD), refactoring scattered, pointer-based data structures into contiguous, cache-efficient memory layouts. We further introduce a hierarchical parallelism scheme: distributing sub-trees across independent CPU cores and SIMD lanes, enabling fully vectorized tree expansion and collision checking. Efficiency is maximized with the help of UCB load balancing across trees and a vectorized STR-tree for coarse-level collision checking. Evaluated on large-scale autonomous driving benchmarks, Vec-QMDP achieves state-of-the-art planning performance with millisecond-level latency, establishing CPUs as a high-performance computing platform for large-scale planning under uncertainty.
☆ Controlled Flight of an Insect-Scale Flapping-Wing Robot via Integrated Onboard Sensing and Computation
Aerial insects can effortlessly navigate dense vegetation, whereas similarly sized aerial robots typically depend on offboard sensors and computation to maintain stable flight. This disparity restricts insect-scale robots to operation within motion capture environments, substantially limiting their applicability to tasks such as search-and-rescue and precision agriculture. In this work, we present a 1.29-gram aerial robot capable of hovering and tracking trajectories with solely onboard sensing and computation. The combination of a sensor suite, estimators, and a low-level controller achieved centimeter-scale positional flight accuracy. Additionally, we developed a hierarchical controller in which a human operator provides high-level commands to direct the robot's motion. In a 30-second flight experiment conducted outside a motion capture system, the robot avoided obstacles and ultimately landed on a sunflower. This level of sensing and computational autonomy represents a significant advancement for the aerial microrobotics community, further opening opportunities to explore onboard planning and power autonomy.
comment: 22 pages, 7 figures
☆ Personalized Autonomous Driving via Optimal Control with Clearance Constraints from Questionnaires
Driving without considering the preferred separation distance from surrounding vehicles may cause discomfort for users. To address this limitation, we propose a planning framework that explicitly incorporates user preferences regarding the desired level of safe clearance from surrounding vehicles. We design a questionnaire purposefully tailored to capture user preferences relevant to our framework, while minimizing unnecessary questions. Specifically, the questionnaire considers various interaction-relevant factors, including the surrounding vehicle's size, speed, position, and maneuvers of surrounding vehicles, as well as the maneuvers of the ego vehicle. The response indicates the user-preferred clearance for the scenario defined by the question and is incorporated as constraints in the optimal control problem. However, it is impractical to account for all possible scenarios that may arise in a driving environment within a single optimal control problem, as the resulting computational complexity renders real-time implementation infeasible. To overcome this limitation, we approximate the original problem by decomposing it into multiple subproblems, each dealing with one fixed scenario. We then solve these subproblems in parallel and select one using the cost function from the original problem. To validate our work, we conduct simulations using different user responses to the questionnaire. We assess how effectively our planner reflects user preferences compared to preference-agnostic baseline planners by measuring preference alignment.
☆ Benchmarking Autonomous Vehicles: A Driver Foundation Model Framework
Autonomous vehicles (AVs) are poised to revolutionize global transportation systems. However, its widespread acceptance and market penetration remain significantly below expectations. This gap is primarily driven by persistent challenges in safety, comfort, commuting efficiency and energy economy when compared to the performance of experienced human drivers. We hypothesize that these challenges can be addressed through the development of a driver foundation model (DFM). Accordingly, we propose a framework for establishing DFMs to comprehensively benchmark AVs. Specifically, we describe a large-scale dataset collection strategy for training a DFM, discuss the core functionalities such a model should possess, and explore potential technical solutions to realize these functionalities. We further present the utility of the DFM across the operational spectrum, from defining human-centric safety envelopes to establishing benchmarks for energy economy. Overall, We aim to formalize the DFM concept and introduce a new paradigm for the systematic specification, verification and validation of AVs.
☆ ReefFlex: A Generative Design Framework for Soft Robotic Grasping of Organic and Fragile objects
Climate change, invasive species and human activities are currently damaging the world's coral reefs at unprecedented rates, threatening their vast biodiversity and fisheries, and reducing coastal protection. Solving this vast challenge requires scalable coral regeneration technologies that can breed climate-resilient species and accelerate the natural regrowth processes; actions that are impeded by the absence of safe and robust tools to handle the fragile coral. We investigate ReefFlex, a generative soft finger design methodology that explores a diverse space of soft fingers to produce a set of candidates capable of safely grasping fragile and geometrically heterogeneous coral in a cluttered environment. Our key insight is encoding heterogeneous grasping into a reduced set of motion primitives, creating a simplified, tractable multi-objective optimisation problem. To evaluate the method, we design a soft robot for reef rehabilitation, which grows and manipulates coral in onshore aquaculture facilities for future reef out-planting. We demonstrate ReefFlex increases both grasp success and grasp quality (disturbance resistance, positioning accuracy) and reduces in adverse events encountered during coral manipulation compared to reference designs. ReefFlex, offers a generalisable method to design soft end-effectors for complex handling and paves a pathway towards automation in previously unachievable domains like coral handling for restoration.
☆ DexFormer: Cross-Embodied Dexterous Manipulation via History-Conditioned Transformer
Dexterous manipulation remains one of the most challenging problems in robotics, requiring coherent control of high-DoF hands and arms under complex, contact-rich dynamics. A major barrier is embodiment variability: different dexterous hands exhibit distinct kinematics and dynamics, forcing prior methods to train separate policies or rely on shared action spaces with per-embodiment decoder heads. We present DexFormer, an end-to-end, dynamics-aware cross-embodiment policy built on a modified transformer backbone that conditions on historical observations. By using temporal context to infer morphology and dynamics on the fly, DexFormer adapts to diverse hand configurations and produces embodiment-appropriate control actions. Trained over a variety of procedurally generated dexterous-hand assets, DexFormer acquires a generalizable manipulation prior and exhibits strong zero-shot transfer to Leap Hand, Allegro Hand, and Rapid Hand. Our results show that a single policy can generalize across heterogeneous hand embodiments, establishing a scalable foundation for cross-embodiment dexterous manipulation. Project website: https://davidlxu.github.io/DexFormer-web/.
☆ Informative Object-centric Next Best View for Object-aware 3D Gaussian Splatting in Cluttered Scenes ICRA 2026
In cluttered scenes with inevitable occlusions and incomplete observations, selecting informative viewpoints is essential for building a reliable representation. In this context, 3D Gaussian Splatting (3DGS) offers a distinct advantage, as it can explicitly guide the selection of subsequent viewpoints and then refine the representation with new observations. However, existing approaches rely solely on geometric cues, neglect manipulation-relevant semantics, and tend to prioritize exploitation over exploration. To tackle these limitations, we introduce an instance-aware Next Best View (NBV) policy that prioritizes underexplored regions by leveraging object features. Specifically, our object-aware 3DGS distills instancelevel information into one-hot object vectors, which are used to compute confidence-weighted information gain that guides the identification of regions associated with erroneous and uncertain Gaussians. Furthermore, our method can be easily adapted to an object-centric NBV, which focuses view selection on a target object, thereby improving reconstruction robustness to object placement. Experiments demonstrate that our NBV policy reduces depth error by up to 77.14% on the synthetic dataset and 34.10% on the real-world GraspNet dataset compared to baselines. Moreover, compared to targeting the entire scene, performing NBV on a specific object yields an additional reduction of 25.60% in depth error for that object. We further validate the effectiveness of our approach through real-world robotic manipulation tasks.
comment: 9 pages, 8 figures, 4 tables, accepted to ICRA 2026
☆ Aerial Manipulation with Contact-Aware Onboard Perception and Hybrid Control ICRA 2026
Aerial manipulation (AM) promises to move Unmanned Aerial Vehicles (UAVs) beyond passive inspection to contact-rich tasks such as grasping, assembly, and in-situ maintenance. Most prior AM demonstrations rely on external motion capture (MoCap) and emphasize position control for coarse interactions, limiting deployability. We present a fully onboard perception-control pipeline for contact-rich AM that achieves accurate motion tracking and regulated contact wrenches without MoCap. The main components are (1) an augmented visual-inertial odometry (VIO) estimator with contact-consistency factors that activate only during interaction, tightening uncertainty around the contact frame and reducing drift, and (2) image-based visual servoing (IBVS) to mitigate perception-control coupling, together with a hybrid force-motion controller that regulates contact wrenches and lateral motion for stable contact. Experiments show that our approach closes the perception-to-wrench loop using only onboard sensing, yielding an velocity estimation improvement of 66.01% at contact, reliable target approach, and stable force holding-pointing toward deployable, in-the-wild aerial manipulation.
comment: 9 pages, 7 figures. Accepted by ICRA 2026
☆ STEP: Warm-Started Visuomotor Policies with Spatiotemporal Consistency Prediction
Diffusion policies have recently emerged as a powerful paradigm for visuomotor control in robotic manipulation due to their ability to model the distribution of action sequences and capture multimodality. However, iterative denoising leads to substantial inference latency, limiting control frequency in real-time closed-loop systems. Existing acceleration methods either reduce sampling steps, bypass diffusion through direct prediction, or reuse past actions, but often struggle to jointly preserve action quality and achieve consistently low latency. In this work, we propose STEP, a lightweight spatiotemporal consistency prediction mechanism to construct high-quality warm-start actions that are both distributionally close to the target action and temporally consistent, without compromising the generative capability of the original diffusion policy. Then, we propose a velocity-aware perturbation injection mechanism that adaptively modulates actuation excitation based on temporal action variation to prevent execution stall especially for real-world tasks. We further provide a theoretical analysis showing that the proposed prediction induces a locally contractive mapping, ensuring convergence of action errors during diffusion refinement. We conduct extensive evaluations on nine simulated benchmarks and two real-world tasks. Notably, STEP with 2 steps can achieve an average 21.6% and 27.5% higher success rate than BRIDGER and DDIM on the RoboMimic benchmark and real-world tasks, respectively. These results demonstrate that STEP consistently advances the Pareto frontier of inference latency and success rate over existing methods.
comment: 13 pages, 9 figures
☆ Chamelion: Reliable Change Detection for Long-Term LiDAR Mapping in Transient Environments
Online change detection is crucial for mobile robots to efficiently navigate through dynamic environments. Detecting changes in transient settings, such as active construction sites or frequently reconfigured indoor spaces, is particularly challenging due to frequent occlusions and spatiotemporal variations. Existing approaches often struggle to detect changes and fail to update the map across different observations. To address these limitations, we propose a dual-head network designed for online change detection and long-term map maintenance. A key difficulty in this task is the collection and alignment of real-world data, as manually registering structural differences over time is both labor-intensive and often impractical. To overcome this, we develop a data augmentation strategy that synthesizes structural changes by importing elements from different scenes, enabling effective model training without the need for extensive ground-truth annotations. Experiments conducted at real-world construction sites and in indoor office environments demonstrate that our approach generalizes well across diverse scenarios, achieving efficient and accurate map updates.\resubmit{Our source code and additional material are available at: https://chamelion-pages.github.io/.
comment: 8 pages, IEEE Robot. Automat. Lett. (RA-L) 2026
Self-Supervised Bootstrapping of Action-Predictive Embodied Reasoning
Embodied Chain-of-Thought (CoT) reasoning has significantly enhanced Vision-Language-Action (VLA) models, yet current methods rely on rigid templates to specify reasoning primitives (e.g., objects in the scene, high-level plans, structural affordances). These templates can force policies to process irrelevant information that distracts from critical action-prediction signals. This creates a bottleneck: without successful policies, we cannot verify reasoning quality; without quality reasoning, we cannot build robust policies. We introduce R&B-EnCoRe, which enables models to bootstrap embodied reasoning from internet-scale knowledge through self-supervised refinement. By treating reasoning as a latent variable within importance-weighted variational inference, models can generate and distill a refined reasoning training dataset of embodiment-specific strategies without external rewards, verifiers, or human annotation. We validate R&B-EnCoRe across manipulation (Franka Panda in simulation, WidowX in hardware), legged navigation (bipedal, wheeled, bicycle, quadruped), and autonomous driving embodiments using various VLA architectures with 1B, 4B, 7B, and 30B parameters. Our approach achieves 28% gains in manipulation success, 101% improvement in navigation scores, and 21% reduction in collision-rate metric over models that indiscriminately reason about all available primitives. R&B-EnCoRe enables models to distill reasoning that is predictive of successful control, bypassing manual annotation engineering while grounding internet-scale knowledge in physical execution.
☆ Data-centric Design of Learning-based Surgical Gaze Perception Models in Multi-Task Simulation
In robot-assisted minimally invasive surgery (RMIS), reduced haptic feedback and depth cues increase reliance on expert visual perception, motivating gaze-guided training and learning-based surgical perception models. However, operative expert gaze is costly to collect, and it remains unclear how the source of gaze supervision, both expertise level (intermediate vs. novice) and perceptual modality (active execution vs. passive viewing), shapes what attention models learn. We introduce a paired active-passive, multi-task surgical gaze dataset collected on the da Vinci SimNow simulator across four drills. Active gaze was recorded during task execution using a VR headset with eye tracking, and the corresponding videos were reused as stimuli to collect passive gaze from observers, enabling controlled same-video comparisons. We quantify skill- and modality-dependent differences in gaze organization and evaluate the substitutability of passive gaze for operative supervision using fixation density overlap analyses and single-frame saliency modeling. Across settings, MSI-Net produced stable, interpretable predictions, whereas SalGAN was unstable and often poorly aligned with human fixations. Models trained on passive gaze recovered a substantial portion of intermediate active attention, but with predictable degradation, and transfer was asymmetric between active and passive targets. Notably, novice passive labels approximated intermediate-passive targets with limited loss on higher-quality demonstrations, suggesting a practical path for scalable, crowd-sourced gaze supervision in surgical coaching and perception modeling.
comment: 8 pages, conference submission pre-print
☆ STaR: Scalable Task-Conditioned Retrieval for Long-Horizon Multimodal Robot Memory
Mobile robots are often deployed over long durations in diverse open, dynamic scenes, including indoor setting such as warehouses and manufacturing facilities, and outdoor settings such as agricultural and roadway operations. A core challenge is to build a scalable long-horizon memory that supports an agentic workflow for planning, retrieval, and reasoning over open-ended instructions at variable granularity, while producing precise, actionable answers for navigation. We present STaR, an agentic reasoning framework that (i) constructs a task-agnostic, multimodal long-term memory that generalizes to unseen queries while preserving fine-grained environmental semantics (object attributes, spatial relations, and dynamic events), and (ii) introduces a Scalable TaskConditioned Retrieval algorithm based on the Information Bottleneck principle to extract from long-term memory a compact, non-redundant, information-rich set of candidate memories for contextual reasoning. We evaluate STaR on NaVQA (mixed indoor/outdoor campus scenes) and WH-VQA, a customized warehouse benchmark with many visually similar objects built with Isaac Sim, emphasizing contextual reasoning. Across the two datasets, STaR consistently outperforms strong baselines, achieving higher success rates and markedly lower spatial error. We further deploy STaR on a real Husky wheeled robot in both indoor and outdoor environments, demonstrating robust longhorizon reasoning, scalability, and practical utility.
☆ From Legible to Inscrutable Trajectories: (Il)legible Motion Planning Accounting for Multiple Observers
In cooperative environments, such as in factories or assistive scenarios, it is important for a robot to communicate its intentions to observers, who could be either other humans or robots. A legible trajectory allows an observer to quickly and accurately predict an agent's intention. In adversarial environments, such as in military operations or games, it is important for a robot to not communicate its intentions to observers. An illegible trajectory leads an observer to incorrectly predict the agent's intention or delays when an observer is able to make a correct prediction about the agent's intention. However, in some environments there are multiple observers, each of whom may be able to see only part of the environment, and each of whom may have different motives. In this work, we introduce the Mixed-Motive Limited-Observability Legible Motion Planning (MMLO-LMP) problem, which requires a motion planner to generate a trajectory that is legible to observers with positive motives and illegible to observers with negative motives while also considering the visibility limitations of each observer. We highlight multiple strategies an agent can take while still achieving the problem objective. We also present DUBIOUS, a trajectory optimizer that solves MMLO-LMP. Our results show that DUBIOUS can generate trajectories that balance legibility with the motives and limited visibility regions of the observers. Future work includes many variations of MMLO-LMP, including moving observers and observer teaming.
comment: 17 pages, 5 figures
☆ Risk-Aware Obstacle Avoidance Algorithm for Real-Time Applications
Robust navigation in changing marine environments requires autonomous systems capable of perceiving, reasoning, and acting under uncertainty. This study introduces a hybrid risk-aware navigation architecture that integrates probabilistic modeling of obstacles along the vehicle path with smooth trajectory optimization for autonomous surface vessels. The system constructs probabilistic risk maps that capture both obstacle proximity and the behavior of dynamic objects. A risk-biased Rapidly Exploring Random Tree (RRT) planner leverages these maps to generate collision-free paths, which are subsequently refined using B-spline algorithms to ensure trajectory continuity. Three distinct RRT* rewiring modes are implemented based on the cost function: minimizing the path length, minimizing risk, and optimizing a combination of the path length and total risk. The framework is evaluated in experimental scenarios containing both static and dynamic obstacles. The results demonstrate the system's ability to navigate safely, maintain smooth trajectories, and dynamically adapt to changing environmental risks. Compared with conventional LIDAR or vision-only navigation approaches, the proposed method shows improvements in operational safety and autonomy, establishing it as a promising solution for risk-aware autonomous vehicle missions in uncertain and dynamic environments.
☆ Elements of Robot Morphology: Supporting Designers in Robot Form Exploration
Robot morphology, the form, shape, and structure of robots, is a key design space in human-robot interaction (HRI), shaping how robots function, express themselves, and interact with people. Yet, despite its importance, little is known about how design frameworks can guide systematic form exploration. To address this gap, we introduce Elements of Robot Morphology, a framework that identifies five fundamental elements: perception, articulation, end effectors, locomotion, and structure. Derived from an analysis of existing robots, the framework supports structured exploration of diverse robot forms. To operationalize the framework, we developed Morphology Exploration Blocks (MEB), a set of tangible blocks that enable hands-on, collaborative experimentation with robot morphologies. We evaluate the framework and toolkit through a case study and design workshops, showing how they support analysis, ideation, reflection, and collaborative robot design.
comment: 10 pages, 5 figures, Proceedings of the 21st ACM/IEEE International Conference on Human-Robot Interaction (HRI '26)
☆ SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes
Simulation has become a key tool for training and evaluating home robots at scale, yet existing environments fail to capture the diversity and physical complexity of real indoor spaces. Current scene synthesis methods produce sparsely furnished rooms that lack the dense clutter, articulated furniture, and physical properties essential for robotic manipulation. We introduce SceneSmith, a hierarchical agentic framework that generates simulation-ready indoor environments from natural language prompts. SceneSmith constructs scenes through successive stages$\unicode{x2013}$from architectural layout to furniture placement to small object population$\unicode{x2013}$each implemented as an interaction among VLM agents: designer, critic, and orchestrator. The framework tightly integrates asset generation through text-to-3D synthesis for static objects, dataset retrieval for articulated objects, and physical property estimation. SceneSmith generates 3-6x more objects than prior methods, with <2% inter-object collisions and 96% of objects remaining stable under physics simulation. In a user study with 205 participants, it achieves 92% average realism and 91% average prompt faithfulness win rates against baselines. We further demonstrate that these environments can be used in an end-to-end pipeline for automatic robot policy evaluation.
comment: Project page: https://scenesmith.github.io/
☆ Agile asymmetric multi-legged locomotion: contact planning via geometric mechanics and spin model duality
Legged robot research is presently focused on bipedal or quadrupedal robots, despite capabilities to build robots with many more legs to potentially improve locomotion performance. This imbalance is not necessarily due to hardware limitations, but rather to the absence of principled control frameworks that explain when and how additional legs improve locomotion performance. In multi-legged systems, coordinating many simultaneous contacts introduces a severe curse of dimensionality that challenges existing modeling and control approaches. As an alternative, multi-legged robots are typically controlled using low-dimensional gaits originally developed for bipeds or quadrupeds. These strategies fail to exploit the new symmetries and control opportunities that emerge in higher-dimensional systems. In this work, we develop a principled framework for discovering new control structures in multi-legged locomotion. We use geometric mechanics to reduce contact-rich locomotion planning to a graph optimization problem, and propose a spin model duality framework from statistical mechanics to exploit symmetry breaking and guide optimal gait reorganization. Using this approach, we identify an asymmetric locomotion strategy for a hexapod robot that achieves a forward speed of 0.61 body lengths per cycle (a 50% improvement over conventional gaits). The resulting asymmetry appears at both the control and hardware levels. At the control level, the body orientation oscillates asymmetrically between fast clockwise and slow counterclockwise turning phases for forward locomotion. At the hardware level, two legs on the same side remain unactuated and can be replaced with rigid parts without degrading performance. Numerical simulations and robophysical experiments validate the framework and reveal novel locomotion behaviors that emerge from symmetry reforming in high-dimensional embodied systems.
comment: 10 pages, 7 figures. arXiv admin note: text overlap with arXiv:2302.03019
☆ Legs Over Arms: On the Predictive Value of Lower-Body Pose for Human Trajectory Prediction from Egocentric Robot Perception ICRA 2026
Predicting human trajectory is crucial for social robot navigation in crowded environments. While most existing approaches treat human as point mass, we present a study on multi-agent trajectory prediction that leverages different human skeletal features for improved forecast accuracy. In particular, we systematically evaluate the predictive utility of 2D and 3D skeletal keypoints and derived biomechanical cues as additional inputs. Through a comprehensive study on the JRDB dataset and another new dataset for social navigation with 360-degree panoramic videos, we find that focusing on lower-body 3D keypoints yields a 13% reduction in Average Displacement Error and augmenting 3D keypoint inputs with corresponding biomechanical cues provides a further 1-4% improvement. Notably, the performance gain persists when using 2D keypoint inputs extracted from equirectangular panoramic images, indicating that monocular surround vision can capture informative cues for motion forecasting. Our finding that robots can forecast human movement efficiently by watching their legs provides actionable insights for designing sensing capabilities for social robot navigation.
comment: Accepted to IEEE ICRA 2026
♻ ☆ Delay-Aware Reinforcement Learning for Highway On-Ramp Merging under Stochastic Communication Latency
Delayed and partially observable state information poses significant challenges for reinforcement learning (RL)-based control in real-world autonomous driving. In highway on-ramp merging, a roadside unit (RSU) can sense nearby traffic, perform edge perception, and transmit state estimates to the ego vehicle over vehicle-to-infrastructure (V2I) links. With recent advancements in intelligent transportation infrastructure and edge computing, such RSU-assisted perception is increasingly realistic and already deployed in modern connected roadway systems. However, edge processing time and wireless transmission can introduce stochastic V2I communication delays, violating the Markov assumption and substantially degrading control performance. In this work, we propose DAROM, a Delay-Aware Reinforcement Learning framework for On-ramp Merging that is robust to stochastic delays. We model the problem as a random delay Markov decision process (RDMDP) and develop a unified RL agent for joint longitudinal and lateral control. To recover a Markovian representation under delayed observations, we introduce a Delay-Aware Encoder that conditions on delayed observations, masked action histories, and observed delay magnitude to infer the current latent state. We further integrate a physics-based safety controller to reduce collision risk during merging. Experiments in the Simulation of Urban MObility (SUMO) simulator using real-world traffic data from the Next Generation Simulation (NGSIM) dataset demonstrate that DAROM consistently outperforms standard RL baselines across traffic densities. In particular, the gated recurrent unit (GRU)-based encoder achieves over 99% success in high-density traffic with random V2I delays of up to 2.0 seconds.
♻ ☆ 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
♻ ☆ Multi-Player, Multi-Strategy Quantum Game Model for Interaction-Aware Decision-Making in Automated Driving
Although significant progress has been made in decision-making for automated driving, challenges remain for deployment in the real world. One challenge lies in addressing interaction-awareness. Most existing approaches oversimplify interactions between the ego vehicle and surrounding agents, and often neglect interactions among the agents themselves. A common solution is to model these interactions using classical game theory. However, its formulation assumes rational players, whereas human behavior is frequently uncertain or irrational. To address these challenges, we propose the Quantum Game Decision-Making (QGDM) model, a novel framework that combines classical game theory with quantum mechanics principles (such as superposition, entanglement, and interference) to tackle multi-player, multi-strategy decision-making problems. To the best of our knowledge, this is one of the first studies to apply quantum game theory to decision-making for automated driving. QGDM runs in real time on a standard computer, without requiring quantum hardware. We evaluate QGDM in simulation across various scenarios, including roundabouts, merging, and highways, and compare its performance with multiple baseline methods. Results show that QGDM significantly improves success rates and reduces collision rates compared to classical approaches, particularly in scenarios with high interaction.
♻ ☆ Through the Perspective of LiDAR: A Feature-Enriched and Uncertainty-Aware Annotation Pipeline for Terrestrial Point Cloud Segmentation
Accurate semantic segmentation of terrestrial laser scanning (TLS) point clouds is limited by costly manual annotation. We propose a semi-automated, uncertainty-aware pipeline that integrates spherical projection, feature enrichment, ensemble learning, and targeted annotation to reduce labeling effort, while sustaining high accuracy. Our approach projects 3D points to a 2D spherical grid, enriches pixels with multi-source features, and trains an ensemble of segmentation networks to produce pseudo-labels and uncertainty maps, the latter guiding annotation of ambiguous regions. The 2D outputs are back-projected to 3D, yielding densely annotated point clouds supported by a three-tier visualization suite (2D feature maps, 3D colorized point clouds, and compact virtual spheres) for rapid triage and reviewer guidance. Using this pipeline, we build Mangrove3D, a semantic segmentation TLS dataset for mangrove forests. We further evaluate data efficiency and feature importance to address two key questions: (1) how much annotated data are needed and (2) which features matter most. Results show that performance saturates after ~12 annotated scans, geometric features contribute the most, and compact nine-channel stacks capture nearly all discriminative power, with the mean Intersection over Union (mIoU) plateauing at around 0.76. Finally, we confirm the generalization of our feature-enrichment strategy through cross-dataset tests on ForestSemantic and Semantic3D. Our contributions include: (i) a robust, uncertainty-aware TLS annotation pipeline with visualization tools; (ii) the Mangrove3D dataset; and (iii) empirical guidance on data efficiency and feature importance, thus enabling scalable, high-quality segmentation of TLS point clouds for ecological monitoring and beyond. The dataset and processing scripts are publicly available at https://fz-rit.github.io/through-the-lidars-eye/.
comment: 40 pages (28 main text), 20 figures, 4 supplementary materials; links to 3D point animations are included in the last table
♻ ☆ Aligning Microscopic Vehicle and Macroscopic Traffic Statistics: Reconstructing Driving Behavior from Partial Data
A driving algorithm that aligns with good human driving practices, or at the very least collaborates effectively with human drivers, is crucial for developing safe and efficient autonomous vehicles. In practice, two main approaches are commonly adopted: (i) supervised or imitation learning, which requires comprehensive naturalistic driving data capturing all states that influence a vehicle's decisions and corresponding actions, and (ii) reinforcement learning (RL), where the simulated driving environment either matches or is intentionally more challenging than real-world conditions. Both methods depend on high-quality observations of real-world driving behavior, which are often difficult and costly to obtain. State-of-the-art sensors on individual vehicles can gather microscopic data, but they lack context about the surrounding conditions. Conversely, roadside sensors can capture traffic flow and other macroscopic characteristics, but they cannot associate this information with individual vehicles on a microscopic level. Motivated by this complementarity, we propose a framework that reconstructs unobserved microscopic states from macroscopic observations, using microscopic data to anchor observed vehicle behaviors, and learns a shared policy whose behavior is microscopically consistent with the partially observed trajectories and actions and macroscopically aligned with target traffic statistics when deployed population-wide. Such constrained and regularized policies promote realistic flow patterns and safe coordination with human drivers at scale.
♻ ☆ A Review of Online Diffusion Policy RL Algorithms for Scalable Robotic Control
Diffusion policies have emerged as a powerful approach for robotic control, demonstrating superior expressiveness in modeling multimodal action distributions compared to conventional policy networks. However, their integration with online reinforcement learning remains challenging due to fundamental incompatibilities between diffusion model training objectives and standard RL policy improvement mechanisms. This paper presents the first comprehensive review and empirical analysis of current Online Diffusion Policy Reinforcement Learning (Online DPRL) algorithms for scalable robotic control systems. We propose a novel taxonomy that categorizes existing approaches into four distinct families--Action-Gradient, Q-Weighting, Proximity-Based, and Backpropagation Through Time (BPTT) methods--based on their policy improvement mechanisms. Through extensive experiments on a unified NVIDIA Isaac Lab benchmark encompassing 12 diverse robotic tasks, we systematically evaluate representative algorithms across five critical dimensions: task diversity, parallelization capability, diffusion step scalability, cross-embodiment generalization, and environmental robustness. Our analysis identifies key findings regarding the fundamental trade-offs inherent in each algorithmic family, particularly concerning sample efficiency and scalability. Furthermore, we reveal critical computational and algorithmic bottlenecks that currently limit the practical deployment of online DPRL. Based on these findings, we provide concrete guidelines for algorithm selection tailored to specific operational constraints and outline promising future research directions to advance the field toward more general and scalable robotic learning systems.
♻ ☆ CoBEVMoE: Heterogeneity-aware Feature Fusion with Dynamic Mixture-of-Experts for Collaborative Perception ICRA 2026
Collaborative perception aims to extend sensing coverage and improve perception accuracy by sharing information among multiple agents. However, due to differences in viewpoints and spatial positions, agents often acquire heterogeneous observations. Existing intermediate fusion methods primarily focus on aligning similar features, often overlooking the perceptual diversity among agents. To address this limitation, we propose CoBEVMoE, a novel collaborative perception framework that operates in the Bird's Eye View (BEV) space and incorporates a Dynamic Mixture-of-Experts (DMoE) architecture. In DMoE, each expert is dynamically generated based on the input features of a specific agent, enabling it to extract distinctive and reliable cues while attending to shared semantics. This design allows the fusion process to explicitly model both feature similarity and heterogeneity across agents. Furthermore, we introduce a Dynamic Expert Metric Loss (DEML) to enhance inter-expert diversity and improve the discriminability of the fused representation. Extensive experiments on the OPV2V and DAIR-V2X-C datasets demonstrate that CoBEVMoE achieves state-of-the-art performance. Specifically, it improves the IoU for Camera-based BEV segmentation by +1.5% on OPV2V and the AP@0.5 for LiDAR-based 3D object detection by +3.0% on DAIR-V2X-C, verifying the effectiveness of expert-based heterogeneous feature modeling in multi-agent collaborative perception. The source code will be made publicly available at https://github.com/godk0509/CoBEVMoE.
comment: Accepted to ICRA 2026. The source code will be made publicly available at https://github.com/godk0509/CoBEVMoE
♻ ☆ SpecPrune-VLA: Accelerating Vision-Language-Action Models via Action-Aware Self-Speculative Pruning
Pruning is a typical acceleration technique for compute-bound models by removing computation on unimportant values. Recently, it has been applied to accelerate Vision-Language-Action (VLA) model inference. However, existing acceleration methods focus on local information from the current action step and ignore the global context, leading to >20% success rate drop and limited speedup in some scenarios. In this paper, we point out spatial-temporal consistency in VLA tasks: input images in consecutive steps exhibit high similarity, and propose the key insight that token selection should combine local information with global context of the model. Based on this, we propose SpecPrune-VLA, a training-free, two-level pruning method with heuristic control. (1) Action-level static pruning. We leverage global history and local attention to statically reduce visual tokens per action. (2) Layer-level dynamic pruning. We prune tokens adaptively per layer based on layer-wise importance. (3) Lightweight action-aware controller: We classify actions as coarse- or fine-grained by the speed of the end effector and adjust pruning aggressiveness accordingly. Extensive experiments show that SpecPrune-VLA achieves up to 1.57$\times$ speedup in LIBERO simulation and 1.70$\times$ on real-world tasks, with negligible success rate degradation.
♻ ☆ SafeLink: Safety-Critical Control Under Dynamic and Irregular Unsafe Regions
Control barrier functions (CBFs) provide a theoretical foundation for safety-critical control in robotic systems. However, most existing methods rely on explicit analytical expressions of unsafe state regions, which are often impractical for irregular and dynamic unsafe regions. This paper introduces SafeLink, a novel CBF construction method based on cost-sensitive incremental random vector functional-link (RVFL) neural networks. By designing a valid cost function, SafeLink assigns different sensitivities to safe and unsafe state points, thereby eliminating false negatives in classification of unsafe state points. Under the constructed CBF, theoretical guarantees are established regarding system safety and the Lipschitz continuity of the control inputs. Furthermore, incremental update theorems are provided, enabling precise real-time adaptation to changes in unsafe regions. An analytical expression for the gradient of SafeLink is also derived to facilitate control input computation. The proposed method is validated on the endpoint position control task of a nonlinear two-link manipulator. Experimental results demonstrate that the method effectively learns the unsafe regions and rapidly adapts as these regions change, achieving computational speeds significantly faster than baseline methods while ensuring the system safely reaches its target position.
comment: 12 pages, 7 figures
♻ ☆ Addressing the Waypoint-Action Gap in End-to-End Autonomous Driving via Vehicle Motion Models
End-to-End Autonomous Driving (E2E-AD) systems are typically grouped by the nature of their outputs: (i) waypoint-based models that predict a future trajectory, and (ii) action-based models that directly output throttle, steer and brake. Most recent benchmark protocols and training pipelines are waypoint-based, which makes action-based policies harder to train and compare, slowing their progress. To bridge this waypoint-action gap, we propose a novel, differentiable vehicle-model framework that rolls out predicted action sequences to their corresponding ego-frame waypoint trajectories while supervising in waypoint space. Our approach enables action-based architectures to be trained and evaluated, for the first time, within waypoint-based benchmarks without modifying the underlying evaluation protocol. We extensively evaluate our framework across multiple challenging benchmarks and observe consistent improvements over the baselines. In particular, on NAVSIM \texttt{navhard} our approach achieves state-of-the-art performance. Our code will be made publicly available upon acceptance.
comment: 8 pages, 3 figures
♻ ☆ EgoFSD: Ego-Centric Fully Sparse Paradigm with Uncertainty Denoising and Iterative Refinement for Efficient End-to-End Self-Driving ICRA2026
Current End-to-End Autonomous Driving (E2E-AD) methods resort to unifying modular designs for various tasks (e.g. perception, prediction and planning). Although optimized with a fully differentiable framework in a planning-oriented manner, existing end-to-end driving systems lacking ego-centric designs still suffer from unsatisfactory performance and inferior efficiency, due to rasterized scene representation learning and redundant information transmission. In this paper, we propose an ego-centric fully sparse paradigm, named EgoFSD, for end-to-end self-driving. Specifically, EgoFSD consists of sparse perception, hierarchical interaction and iterative motion planner. The sparse perception module performs detection and online mapping based on sparse representation of the driving scene. The hierarchical interaction module aims to select the Closest In-Path Vehicle / Stationary (CIPV / CIPS) from coarse to fine, benefiting from an additional geometric prior. As for the iterative motion planner, both selected interactive agents and ego-vehicle are considered for joint motion prediction, where the output multi-modal ego-trajectories are optimized in an iterative fashion. In addition, position-level motion diffusion and trajectory-level planning denoising are introduced for uncertainty modeling, thereby enhancing the training stability and convergence speed. Extensive experiments are conducted on nuScenes and Bench2Drive datasets, which significantly reduces the average L2 error by 59% and collision rate by 92% than UniAD while achieves 6.9x faster running efficiency.
comment: Accepted to ICRA2026
♻ ☆ A Survey of Behavior Foundation Model: Next-Generation Whole-Body Control System of Humanoid Robots
Humanoid robots are drawing significant attention as versatile platforms for complex motor control, human-robot interaction, and general-purpose physical intelligence. However, achieving efficient whole-body control (WBC) in humanoids remains a fundamental challenge due to sophisticated dynamics, underactuation, and diverse task requirements. While learning-based controllers have shown promise for complex tasks, their reliance on labor-intensive and costly retraining for new scenarios limits real-world applicability. To address these limitations, behavior(al) foundation models (BFMs) have emerged as a new paradigm that leverages large-scale pre-training to learn reusable primitive skills and broad behavioral priors, enabling zero-shot or rapid adaptation to a wide range of downstream tasks. In this paper, we present a comprehensive overview of BFMs for humanoid WBC, tracing their development across diverse pre-training pipelines. Furthermore, we discuss real-world applications, current limitations, urgent challenges, and future opportunities, positioning BFMs as a key approach toward scalable and general-purpose humanoid intelligence. Finally, we provide a curated and regularly updated collection of BFM papers and projects to facilitate more subsequent research, which is available at https://github.com/yuanmingqi/awesome-bfm-papers.
comment: 19 pages, 10 figures
♻ ☆ Simultaneous Tactile-Visual Perception for Learning Multimodal Robot Manipulation
Robotic manipulation requires both rich multimodal perception and effective learning frameworks to handle complex real-world tasks. See-through-skin (STS) sensors, which combine tactile and visual perception, offer promising sensing capabilities, while modern imitation learning provides powerful tools for policy acquisition. However, existing STS designs lack simultaneous multimodal perception and suffer from unreliable tactile tracking. Furthermore, integrating these rich multimodal signals into learning-based manipulation pipelines remains an open challenge. We introduce TacThru, an STS sensor enabling simultaneous visual perception and robust tactile signal extraction, and TacThru-UMI, an imitation learning framework that leverages these multimodal signals for manipulation. Our sensor features a fully transparent elastomer, persistent illumination, novel keyline markers, and efficient tracking, while our learning system integrates these signals through a Transformer-based Diffusion Policy. Experiments on five challenging real-world tasks show that TacThru-UMI achieves an average success rate of 85.5%, significantly outperforming the baselines of tactile policy(66.3%) and vision-only policy (55.4%). The system excels in critical scenarios, including contact detection with thin and soft objects and precision manipulation requiring multimodal coordination. This work demonstrates that combining simultaneous multimodal perception with modern learning frameworks enables more precise, adaptable robotic manipulation.
♻ ☆ Multi-Robot Data-Free Continual Communicative Learning (CCL) from Black-Box Visual Place Recognition Models
In emerging multi-robot societies, heterogeneous agents must continually extract and integrate local knowledge from one another through communication, even when their internal models are completely opaque. Existing approaches to continual or collaborative learning for visual place recognition (VPR) largely assume white-box access to model parameters or shared training datasets, which is unrealistic when robots encounter unknown peers in the wild. This paper introduces \emph{Continual Communicative Learning (CCL)}, a data-free multi-robot framework in which a traveler robot (student) continually improves its VPR capability by communicating with black-box teacher models via a constrained query--response channel. We repurpose Membership Inference Attacks (MIA), originally developed as privacy attacks on machine learning models, as a constructive communication primitive to reconstruct pseudo-training sets from black-box VPR teachers without accessing their parameters or raw data. To overcome the intrinsic communication bottleneck caused by the low sampling efficiency of black-box MIA, we propose a prior-based query strategy that leverages the student's own VPR prior to focus queries on informative regions of the embedding space, thereby reducing the knowledge transfer (KT) cost. Experimental results on a standard multi-session VPR benchmark demonstrate that the proposed CCL framework yields substantial performance gains for low-performing robots under modest communication budgets, highlighting CCL as a promising building block for scalable and fault-tolerant multi-robot systems. Furthermore, we propose a Distributed Statistic Integration (DSI) framework that theoretically eliminates catastrophic forgetting by efficiently aggregating sufficient statistics from black-box VPR models while maintaining data privacy and reducing communication overhead to a sample-invariant constant complexity.
comment: 6 pages, 4 figures, technical report
♻ ☆ OpenGVL -- Benchmarking Visual Temporal Progress for Data Curation
Data scarcity remains one of the most limiting factors in driving progress in robotics. However, the amount of available robotics data in the wild is growing exponentially, creating new opportunities for large-scale data utilization. Reliable temporal task completion prediction could help automatically annotate and curate this data at scale. The Generative Value Learning (GVL) approach was recently proposed, leveraging the knowledge embedded in vision-language models (VLMs) to predict task progress from visual observations. Building upon GVL, we propose OpenGVL, a comprehensive benchmark for estimating task progress across diverse challenging manipulation tasks involving both robotic and human embodiments. We evaluate the capabilities of publicly available open-source foundation models, showing that open-source model families significantly underperform closed-source counterparts, achieving only approximately $70\%$ of their performance on temporal progress prediction tasks. Furthermore, we demonstrate how OpenGVL can serve as a practical tool for automated data curation and filtering, enabling efficient quality assessment of large-scale robotics datasets. We release the benchmark along with the complete codebase at \href{github.com/budzianowski/opengvl}{OpenGVL}.
comment: Workshop on Making Sense of Data in Robotics: Composition, Curation, and Interpretability at Scale at CoRL 2025
♻ ☆ Nimbus: A Unified Embodied Synthetic Data Generation Framework
Scaling data volume and diversity is critical for generalizing embodied intelligence. While synthetic data generation offers a scalable alternative to expensive physical data acquisition, existing pipelines remain fragmented and task-specific. This isolation leads to significant engineering inefficiency and system instability, failing to support the sustained, high-throughput data generation required for foundation model training. To address these challenges, we present Nimbus, a unified synthetic data generation framework designed to integrate heterogeneous navigation and manipulation pipelines. Nimbus introduces a modular four-layer architecture featuring a decoupled execution model that separates trajectory planning, rendering, and storage into asynchronous stages. By implementing dynamic pipeline scheduling, global load balancing, distributed fault tolerance, and backend-specific rendering optimizations, the system maximizes resource utilization across CPU, GPU, and I/O resources. Our evaluation demonstrates that Nimbus achieves a 2-3X improvement in end-to-end throughput compared to unoptimized baselines and ensuring robust, long-term operation in large-scale distributed environments. This framework serves as the production backbone for the InternData suite, enabling seamless cross-domain data synthesis.
♻ ☆ MomaGraph: State-Aware Unified Scene Graphs with Vision-Language Model for Embodied Task Planning
Mobile manipulators in households must both navigate and manipulate. This requires a compact, semantically rich scene representation that captures where objects are, how they function, and which parts are actionable. Scene graphs are a natural choice, yet prior work often separates spatial and functional relations, treats scenes as static snapshots without object states or temporal updates, and overlooks information most relevant for accomplishing the current task. To address these limitations, we introduce MomaGraph, a unified scene representation for embodied agents that integrates spatial-functional relationships and part-level interactive elements. However, advancing such a representation requires both suitable data and rigorous evaluation, which have been largely missing. We thus contribute MomaGraph-Scenes, the first large-scale dataset of richly annotated, task-driven scene graphs in household environments, along with MomaGraph-Bench, a systematic evaluation suite spanning six reasoning capabilities from high-level planning to fine-grained scene understanding. Built upon this foundation, we further develop MomaGraph-R1, a 7B vision-language model trained with reinforcement learning on MomaGraph-Scenes. MomaGraph-R1 predicts task-oriented scene graphs and serves as a zero-shot task planner under a Graph-then-Plan framework. Extensive experiments demonstrate that our model achieves state-of-the-art results among open-source models, reaching 71.6% accuracy on the benchmark (+11.4% over the best baseline), while generalizing across public benchmarks and transferring effectively to real-robot experiments.
comment: 25 pages, 10 figures. Project page:https://hybridrobotics.github.io/MomaGraph/
♻ ☆ AgenticLab: A Real-World Robot Agent Platform that Can See, Think, and Act
Recent advances in large vision-language models (VLMs) have demonstrated generalizable open-vocabulary perception and reasoning, yet their real-robot manipulation capability remains unclear for long-horizon, closed-loop execution in unstructured, in-the-wild environments. Prior VLM-based manipulation pipelines are difficult to compare across different research groups' setups, and many evaluations rely on simulation, privileged state, or specially designed setups. We present AgenticLab, a model-agnostic robot agent platform and benchmark for open-world manipulation. AgenticLab provides a closed-loop agent pipeline for perception, task decomposition, online verification, and replanning. Using AgenticLab, we benchmark state-of-the-art VLM-based agents on real-robot tasks in unstructured environments. Our benchmark reveals several failure modes that offline vision-language tests (e.g., VQA and static image understanding) fail to capture, including breakdowns in multi-step grounding consistency, object grounding under occlusion and scene changes, and insufficient spatial reasoning for reliable manipulation. We will release the full hardware and software stack to support reproducible evaluation and accelerate research on general-purpose robot agents.
comment: Added appendix
♻ ☆ Learning-based Adaptive Control of Quadruped Robots for Active Stabilization on Moving Platforms IROS 2024
A quadruped robot faces balancing challenges on a six-degrees-of-freedom moving platform, like subways, buses, airplanes, and yachts, due to independent platform motions and resultant diverse inertia forces on the robot. To alleviate these challenges, we present the Learning-based Active Stabilization on Moving Platforms (\textit{LAS-MP}), featuring a self-balancing policy and system state estimators. The policy adaptively adjusts the robot's posture in response to the platform's motion. The estimators infer robot and platform states based on proprioceptive sensor data. For a systematic training scheme across various platform motions, we introduce platform trajectory generation and scheduling methods. Our evaluation demonstrates superior balancing performance across multiple metrics compared to three baselines. Furthermore, we conduct a detailed analysis of the \textit{LAS-MP}, including ablation studies and evaluation of the estimators, to validate the effectiveness of each component.
comment: Accepted at IROS 2024. Project Page: https://sgvr.kaist.ac.kr/~msyoon/papers/IROS24/
♻ ☆ Enhancing Navigation Efficiency of Quadruped Robots via Leveraging Personal Transportation Platforms ICRA 2025
Quadruped robots face limitations in long-range navigation efficiency due to their reliance on legs. To ameliorate the limitations, we introduce a Reinforcement Learning-based Active Transporter Riding method (\textit{RL-ATR}), inspired by humans' utilization of personal transporters, including Segways. The \textit{RL-ATR} features a transporter riding policy and two state estimators. The policy devises adequate maneuvering strategies according to transporter-specific control dynamics, while the estimators resolve sensor ambiguities in non-inertial frames by inferring unobservable robot and transporter states. Comprehensive evaluations in simulation validate proficient command tracking abilities across various transporter-robot models and reduced energy consumption compared to legged locomotion. Moreover, we conduct ablation studies to quantify individual component contributions within the \textit{RL-ATR}. This riding ability could broaden the locomotion modalities of quadruped robots, potentially expanding the operational range and efficiency.
comment: Accepted at ICRA 2025. Project page: https://sgvr.kaist.ac.kr/~msyoon/papers/ICRA25/
♻ ☆ Learning-based Initialization of Trajectory Optimization for Path-following Problems of Redundant Manipulators ICRA 2023
Trajectory optimization (TO) is an efficient tool to generate a redundant manipulator's joint trajectory following a 6-dimensional Cartesian path. The optimization performance largely depends on the quality of initial trajectories. However, the selection of a high-quality initial trajectory is non-trivial and requires a considerable time budget due to the extremely large space of the solution trajectories and the lack of prior knowledge about task constraints in configuration space. To alleviate the issue, we present a learning-based initial trajectory generation method that generates high-quality initial trajectories in a short time budget by adopting example-guided reinforcement learning. In addition, we suggest a null-space projected imitation reward to consider null-space constraints by efficiently learning kinematically feasible motion captured in expert demonstrations. Our statistical evaluation in simulation shows the improved optimality, efficiency, and applicability of TO when we plug in our method's output, compared with three other baselines. We also show the performance improvement and feasibility via real-world experiments with a seven-degree-of-freedom manipulator.
comment: Accepted at ICRA 2023. Project page: https://sgvr.kaist.ac.kr/~msyoon/papers/ICRA23_RLITG/
♻ ☆ DREAM: Domain-aware Reasoning for Efficient Autonomous Underwater Monitoring ICRA 2026
The ocean is warming and acidifying, increasing the risk of mass mortality events for temperature-sensitive shellfish such as oysters. This motivates the development of long-term monitoring systems. However, human labor is costly and long-duration underwater work is highly hazardous, thus favoring robotic solutions as a safer and more efficient option. To enable underwater robots to make real-time, environment-aware decisions without human intervention, we must equip them with an intelligent "brain." This highlights the need for persistent,wide-area, and low-cost benthic monitoring. To this end, we present DREAM, a Vision Language Model (VLM)-guided autonomy framework for long-term underwater exploration and habitat monitoring. The results show that our framework is highly efficient in finding and exploring target objects (e.g., oysters, shipwrecks) without prior location information. In the oyster-monitoring task, our framework takes 31.5% less time than the previous baseline with the same amount of oysters. Compared to the vanilla VLM, it uses 23% fewer steps while covering 8.88% more oysters. In shipwreck scenes, our framework successfully explores and maps the wreck without collisions, requiring 27.5% fewer steps than the vanilla model and achieving 100% coverage, while the vanilla model achieves 60.23% average coverage in our shipwreck environments.
comment: In Proceeding of ICRA 2026
♻ ☆ Modelling Pedestrian Behaviour in Autonomous Vehicle Encounters Using Naturalistic Dataset
Understanding how pedestrians adjust their movement when interacting with autonomous vehicles (AVs) is essential for improving safety in mixed traffic. This study examines micro-level pedestrian behaviour during midblock encounters in the NuScenes dataset using a hybrid discrete choice-machine learning framework based on the Residual Logit (ResLogit) model. The model incorporates temporal, spatial, kinematic, and perceptual indicators. These include relative speed, visual looming, remaining distance, and directional collision risk proximity (CRP) measures. Results suggest that some of these variables may meaningfully influence movement adjustments, although predictive performance remains moderate. Marginal effects and elasticities indicate strong directional asymmetries in risk perception, with frontal and rear CRP showing opposite influences. The remaining distance exhibits a possible mid-crossing threshold. Relative speed cues appear to have a comparatively less effect. These patterns may reflect multiple behavioural tendencies driven by both risk perception and movement efficiency.
♻ ☆ Affective and Conversational Predictors of Re-Engagement in Human-Robot Interactions -- A Student-Centered Study with A Humanoid Social Robot
Humanoid social robots are increasingly present in daily life, making sustained user engagement a critical factor for their effectiveness and acceptance. While prior work has often examined affective evaluations or anthropomorphic design, less is known about the relative influence of dynamic conversational qualities and perceived robot characteristics in determining a user's intention to re-engage with Large Language Model (LLM)-driven social robots. In this study, 68 participants interacted in open-ended conversations with the Nadine humanoid social robot, completing pre- and post-interaction surveys to assess changes in robot perception, conversational quality, and intention to re-engage. The results showed that verbal interaction significantly improved the robot's perceived characteristics, with statistically significant increases in pleasantness ($p<.0001$) and approachability ($p<.0001$), and a reduction in creepiness ($p<.001$). However, these affective changes were not strong and unique predictors of users' intention to re-engage in a multiple regression model. Instead, participants' perceptions of the interestingness ($β=0.60$, $p<.001$) and naturalness ($β=0.31$, $p=0.015$) of the robot's conversation emerged as the most significant and robust independent predictors of intention to re-engage. Overall, the results highlight that conversational quality, specifically perceived interestingness and naturalness, is the dominant driver of re-engagement, indicating that LLM-driven robot design should prioritize engaging, natural dialogue over affective impression management or anthropomorphic cues.
comment: 27 pages, 7 figures
♻ ☆ CHIP: Adaptive Compliance for Humanoid Control through Hindsight Perturbation
Recent progress in humanoid robots has unlocked agile locomotion skills, including backflipping, running, and crawling. Yet it remains challenging for a humanoid robot to perform forceful manipulation tasks such as moving objects, wiping, and pushing a cart. We propose adaptive Compliance Humanoid control through hIsight Perturbation (CHIP), a plug-and-play module that enables controllable end-effector stiffness while preserving agile tracking of dynamic reference motions. CHIP is easy to implement and requires neither data augmentation nor additional reward tuning. We show that a generalist motion-tracking controller trained with CHIP can perform a diverse set of forceful manipulation tasks that require different end-effector compliance, such as multi-robot collaboration, wiping, box delivery, and door opening.
comment: The first two authors contributed equally. Project page: https://nvlabs.github.io/CHIP/
♻ ☆ VIMD: Monocular Visual-Inertial Motion and Depth Estimation
Accurate and efficient dense metric depth estimation is crucial for 3D visual perception in robotics and XR. In this paper, we develop a monocular visual-inertial motion and depth (VIMD) learning framework to estimate dense metric depth by leveraging accurate and efficient MSCKF-based monocular visual-inertial motion tracking. At the core the proposed VIMD is to exploit multi-view information to iteratively refine per-pixel scale, instead of globally fitting an invariant affine model as in the prior work. The VIMD framework is highly modular, making it compatible with a variety of existing depth estimation backbones. We conduct extensive evaluations on the TartanAir and VOID datasets and demonstrate its zero-shot generalization capabilities on the AR Table dataset. Our results show that VIMD achieves exceptional accuracy and robustness, even with extremely sparse points as few as 10-20 metric depth points per image. This makes the proposed VIMD a practical solution for deployment in resource constrained settings, while its robust performance and strong generalization capabilities offer significant potential across a wide range of scenarios.
♻ ☆ Symmetry-Guided Memory Augmentation for Efficient Locomotion Learning
Training reinforcement learning (RL) policies for legged locomotion often requires extensive environment interactions, which are costly and time-consuming. We propose Symmetry-Guided Memory Augmentation (SGMA), a framework that improves training efficiency by combining structured experience augmentation with memory-based context inference. Our method leverages robot and task symmetries to generate additional, physically consistent training experiences without requiring extra interactions. To avoid the pitfalls of naive augmentation, we extend these transformations to the policy's memory states, enabling the agent to retain task-relevant context and adapt its behavior accordingly. We evaluate the approach on quadruped and humanoid robots in simulation, as well as on a real quadruped platform. Across diverse locomotion tasks involving joint failures and payload variations, our method achieves efficient policy training while maintaining robust performance, demonstrating a practical route toward data-efficient RL for legged robots.
♻ ☆ What Do You Need for Compositional Generalization in Diffusion Planning?
In policy learning, stitching and compositional generalization refer to the extent to which the policy is able to piece together sub-trajectories of data it is trained on to generate new and diverse behaviours. While stitching has been identified as a significant strength of offline reinforcement learning, recent generative behavioural cloning (BC) methods have also shown proficiency at stitching. However, the main factors behind this are poorly understood, hindering the development of new algorithms that can reliably stitch by design. Focusing on diffusion planners trained via generative behavioural cloning, and without resorting to dynamic programming or TD-learning, we find three properties are key enablers for composition: shift equivariance, local receptive fields, and inference choices. We use these properties to explain architecture, data, and inference choices in existing generative BC methods based on diffusion planning including replanning frequency, data augmentation, and data scaling. Our experiments show that while local receptive fields are more important than shift equivariance in creating a diffusion planner capable of composition, both are crucial. Using findings from our experiments, we develop a new architecture for diffusion planners called Eq-Net, that is simple, produces diverse trajectories competitive with more computationally expensive methods such as replanning or scaling data, and can be guided to enable generalization in goal-conditioned settings. We show that Eq-Net exhibits significant compositional generalization in a variety of navigation and manipulation tasks designed to test planning diversity.
comment: 8 Pages
♻ ☆ TURBO: Utility-Aware Bandwidth Allocation for Cloud-Augmented Autonomous Control
Autonomous driving system progress has been driven by improvements in machine learning models, whose computational demands now exceed what edge devices alone can provide. The cloud offers abundant compute, but the network has long been treated as an unreliable bottleneck rather than a co-equal part of the autonomous vehicle control loop. We argue that this separation is no longer tenable: safety-critical autonomy requires co-design of control, models, and network resource allocation itself. We introduce TURBO, a cloud-augmented control framework that addresses this challenge, formulating bandwidth allocation and control pipeline configuration across both the car and cloud as a joint optimization problem. TURBO maximizes benefit to the car while guaranteeing safety in the face of highly variable network conditions. We implement TURBO and evaluate it in both simulation and real-world deployment, showing it can improve average accuracy by up to 15.6%pt over existing on-vehicle-only pipelines. Our code is made available at www.github.com/NetSys/turbo.
comment: 34 pages, 13 figures
Robotics 31
☆ Adding More Value Than Work: Practical Guidelines for Integrating Robots into Intercultural Competence Learning
While social robots have demonstrated effectiveness in supporting students' intercultural competence development, it is unclear how they can effectively be adopted for integrated use in K-12 schools. We conducted two phases of design workshops with teachers, where they co-designed robot-mediated intercultural activities while considering student needs and school integration concerns. Using thematic analysis, we identify appropriate scenarios and roles for classroom robots, explore how robots could complement rather than replace teachers, and consider how to address ethical and compliance considerations. Our findings provide practical design guidelines for the HRI community to develop social robots that can effectively support intercultural education in K-12 schools.
☆ From Ellipsoids to Midair Control of Dynamic Hitches
The ability to dynamically manipulate interaction between cables, carried by pairs of aerial vehicles attached to the ends of each cable, can greatly improve the versatility and agility of cable-assisted aerial manipulation. 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 is key to mastering the inter-cable manipulation in 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.
☆ Picasso: Holistic Scene Reconstruction with Physics-Constrained Sampling
In the presence of occlusions and measurement noise, geometrically accurate scene reconstructions -- which fit the sensor data -- can still be physically incorrect. For instance, when estimating the poses and shapes of objects in the scene and importing the resulting estimates into a simulator, small errors might translate to implausible configurations including object interpenetration or unstable equilibrium. This makes it difficult to predict the dynamic behavior of the scene using a digital twin, an important step in simulation-based planning and control of contact-rich behaviors. In this paper, we posit that object pose and shape estimation requires reasoning holistically over the scene (instead of reasoning about each object in isolation), accounting for object interactions and physical plausibility. Towards this goal, our first contribution is Picasso, a physics-constrained reconstruction pipeline that builds multi-object scene reconstructions by considering geometry, non-penetration, and physics. Picasso relies on a fast rejection sampling method that reasons over multi-object interactions, leveraging an inferred object contact graph to guide samples. Second, we propose the Picasso dataset, a collection of 10 contact-rich real-world scenes with ground truth annotations, as well as a metric to quantify physical plausibility, which we open-source as part of our benchmark. Finally, we provide an extensive evaluation of Picasso on our newly introduced dataset and on the YCB-V dataset, and show it largely outperforms the state of the art while providing reconstructions that are both physically plausible and more aligned with human intuition.
comment: 15 pages
☆ ForecastOcc: Vision-based Semantic Occupancy Forecasting
Autonomous driving requires forecasting both geometry and semantics over time to effectively reason about future environment states. Existing vision-based occupancy forecasting methods focus on motion-related categories such as static and dynamic objects, while semantic information remains largely absent. Recent semantic occupancy forecasting approaches address this gap but rely on past occupancy predictions obtained from separate networks. This makes current methods sensitive to error accumulation and prevents learning spatio-temporal features directly from images. In this work, we present ForecastOcc, the first framework for vision-based semantic occupancy forecasting that jointly predicts future occupancy states and semantic categories. Our framework yields semantic occupancy forecasts for multiple horizons directly from past camera images, without relying on externally estimated maps. We evaluate ForecastOcc in two complementary settings: multi-view forecasting on the Occ3D-nuScenes dataset and monocular forecasting on SemanticKITTI, where we establish the first benchmark for this task. We introduce the first baselines by adapting two 2D forecasting modules within our framework. Importantly, we propose a novel architecture that incorporates a temporal cross-attention forecasting module, a 2D-to-3D view transformer, a 3D encoder for occupancy prediction, and a semantic occupancy head for voxel-level forecasts across multiple horizons. Extensive experiments on both datasets show that ForecastOcc consistently outperforms baselines, yielding semantically rich, future-aware predictions that capture scene dynamics and semantics critical for autonomous driving.
☆ Analyzing the Impact of Simulation Fidelity on the Evaluation of Autonomous Driving Motion Control
Simulation is crucial in the development of autonomous driving software. In particular, assessing control algorithms requires an accurate vehicle dynamics simulation. However, recent publications use models with varying levels of detail. This disparity makes it difficult to compare individual control algorithms. Therefore, this paper aims to investigate the influence of the fidelity of vehicle dynamics modeling on the closed-loop behavior of trajectory-following controllers. For this purpose, we introduce a comprehensive Autoware-compatible vehicle model. By simplifying this, we derive models with varying fidelity. Evaluating over 550 simulation runs allows us to quantify each model's approximation quality compared to real-world data. Furthermore, we investigate whether the influence of model simplifications changes with varying margins to the acceleration limit of the vehicle. From this, we deduce to which degree a vehicle model can be simplified to evaluate control algorithms depending on the specific application. The real-world data used to validate the simulation environment originate from the Indy Autonomous Challenge race at the Autodromo Nazionale di Monza in June 2023. They show the fastest fully autonomous lap of TUM Autonomous Motorsport, with vehicle speeds reaching 267 kph and lateral accelerations of up to 15 mps2.
comment: Accepted for publication at the IEEE IV 2024
☆ Integrating Specialized and Generic Agent Motion Prediction with Dynamic Occupancy Grid Maps
Accurate prediction of driving scene is a challenging task due to uncertainty in sensor data, the complex behaviors of agents, and the possibility of multiple feasible futures. Existing prediction methods using occupancy grid maps primarily focus on agent-agnostic scene predictions, while agent-specific predictions provide specialized behavior insights with the help of semantic information. However, both paradigms face distinct limitations: agent-agnostic models struggle to capture the behavioral complexities of dynamic actors, whereas agent-specific approaches fail to generalize to poorly perceived or unrecognized agents; combining both enables robust and safer motion forecasting. To address this, we propose a unified framework by leveraging Dynamic Occupancy Grid Maps within a streamlined temporal decoding pipeline to simultaneously predict future occupancy state grids, vehicle grids, and scene flow grids. Relying on a lightweight spatiotemporal backbone, our approach is centered on a tailored, interdependent loss function that captures inter-grid dependencies and enables diverse future predictions. By using occupancy state information to enforce flow-guided transitions, the loss function acts as a regularizer that directs occupancy evolution while accounting for obstacles and occlusions. Consequently, the model not only predicts the specific behaviors of vehicle agents, but also identifies other dynamic entities and anticipates their evolution within the complex scene. Evaluations on real-world nuScenes and Woven Planet datasets demonstrate superior prediction performances for dynamic vehicles and generic dynamic scene elements compared to baseline methods.
comment: Updated version with major revisions; currently under the second round of review at IEEE Transactions on Intelligent Vehicles
☆ Feasibility-Guided Planning over Multi-Specialized Locomotion Policies ICRA 2026
Planning over unstructured terrain presents a significant challenge in the field of legged robotics. Although recent works in reinforcement learning have yielded various locomotion strategies, planning over multiple experts remains a complex issue. Existing approaches encounter several constraints: traditional planners are unable to integrate skill-specific policies, whereas hierarchical learning frameworks often lose interpretability and require retraining whenever new policies are added. In this paper, we propose a feasibility-guided planning framework that successfully incorporates multiple terrain-specific policies. Each policy is paired with a Feasibility-Net, which learned to predict feasibility tensors based on the local elevation maps and task vectors. This integration allows classical planning algorithms to derive optimal paths. Through both simulated and real-world experiments, we demonstrate that our method efficiently generates reliable plans across diverse and challenging terrains, while consistently aligning with the capabilities of the underlying policies.
comment: ICRA 2026
☆ Optimized Human-Robot Co-Dispatch Planning for Petro-Site Surveillance under Varying Criticalities
Securing petroleum infrastructure requires balancing autonomous system efficiency with human judgment for threat escalation, a challenge unaddressed by classical facility location models assuming homogeneous resources. This paper formulates the Human-Robot Co-Dispatch Facility Location Problem (HRCD-FLP), a capacitated facility location variant incorporating tiered infrastructure criticality, human-robot supervision ratio constraints, and minimum utilization requirements. We evaluate command center selection across three technology maturity scenarios. Results show transitioning from conservative (1:3 human-robot supervision) to future autonomous operations (1:10) yields significant cost reduction while maintaining complete critical infrastructure coverage. For small problems, exact methods dominate in both cost and computation time; for larger problems, the proposed heuristic achieves feasible solutions in under 3 minutes with approximately 14% optimality gap where comparison is possible. From systems perspective, our work demonstrate that optimized planning for human-robot teaming is key to achieve both cost-effective and mission-reliable deployments.
☆ Multi-Agent Route Planning as a QUBO Problem
Multi-Agent Route Planning considers selecting vehicles, each associated with a single predefined route, such that the spatial coverage of a road network is increased while redundant overlaps are limited. This paper gives a formal problem definition, proves NP-hardness by reduction from the Weighted Set Packing problem, and derives a Quadratic Unconstrained Binary Optimization formulation whose coefficients directly encode unique coverage rewards and pairwise overlap penalties. A single penalty parameter controls the coverage-overlap trade-off. We distinguish between a soft regime, which supports multi-objective exploration, and a hard regime, in which the penalty is strong enough to effectively enforce near-disjoint routes. We describe a practical pipeline for generating city instances, constructing candidate routes, building the QUBO matrix, and solving it with an exact mixed-integer solver (Gurobi), simulated annealing, and D-Wave hybrid quantum annealing. Experiments on Barcelona instances with up to 10 000 vehicles reveal a clear coverage-overlap knee and show that Pareto-optimal solutions are mainly obtained under the hard-penalty regime, while D-Wave hybrid solvers and Gurobi achieve essentially identical objective values with only minor differences in runtime as problem size grows.
☆ Incremental Mapping with Measurement Synchronization & Compression
Modern autonomous vehicles and robots utilize versatile sensors for localization and mapping. The fidelity of these maps is paramount, as an accurate environmental representation is a prerequisite for stable and precise localization. Factor graphs provide a powerful approach for sensor fusion, enabling the estimation of the maximum a posteriori solution. However, the discrete nature of graph-based representations, combined with asynchronous sensor measurements, complicates consistent state estimation. The design of an optimal factor graph topology remains an open challenge, especially in multi-sensor systems with asynchronous data. Conventional approaches rely on a rigid graph structure, which becomes inefficient with sensors of disparate rates. Although preintegration techniques can mitigate this for high-rate sensors, their applicability is limited. To address this problem, this work introduces a novel approach that incrementally constructs connected factor graphs, ensuring the incorporation of all available sensor data by choosing the optimal graph topology based on the external evaluation criteria. The proposed methodology facilitates graph compression, reducing the number of nodes (optimized variables) by ~30% on average while maintaining map quality at a level comparable to conventional approaches.
comment: 8 pages, 4 figures, 1 table
☆ Research on a Camera Position Measurement Method based on a Parallel Perspective Error Transfer Model
Camera pose estimation from sparse correspondences is a fundamental problem in geometric computer vision and remains particularly challenging in near-field scenarios, where strong perspective effects and heterogeneous measurement noise can significantly degrade the stability of analytic PnP solutions. In this paper, we present a geometric error propagation framework for camera pose estimation based on a parallel perspective approximation. By explicitly modeling how image measurement errors propagate through perspective geometry, we derive an error transfer model that characterizes the relationship between feature point distribution, camera depth, and pose estimation uncertainty. Building on this analysis, we develop a pose estimation method that leverages parallel perspective initialization and error-aware weighting within a Gauss-Newton optimization scheme, leading to improved robustness in proximity operations. Extensive experiments on both synthetic data and real-world images, covering diverse conditions such as strong illumination, surgical lighting, and underwater low-light environments, demonstrate that the proposed approach achieves accuracy and robustness comparable to state-of-the-art analytic and iterative PnP methods, while maintaining high computational efficiency. These results highlight the importance of explicit geometric error modeling for reliable camera pose estimation in challenging near-field settings.
comment: 32 pages, 19 figures
☆ System-Level Error Propagation and Tail-Risk Amplification in Reference-Based Robotic Navigation
Image guided robotic navigation systems often rely on reference based geometric perception pipelines, where accurate spatial mapping is established through multi stage estimation processes. In biplanar X ray guided navigation, such pipelines are widely used due to their real time capability and geometric interpretability. However, navigation reliability can be constrained by an overlooked system level failure mechanism in which installation induced structural perturbations introduced at the perception stage are progressively amplified along the perception reconstruction execution chain and dominate execution level error and tail risk behavior. This paper investigates this mechanism from a system level perspective and presents a unified error propagation modeling framework that characterizes how installation induced structural perturbations propagate and couple with pixel level observation noise through biplanar imaging, projection matrix estimation, triangulation, and coordinate mapping. Using first order analytic uncertainty propagation and Monte Carlo simulations, we analyze dominant sensitivity channels and quantify worst case error behavior beyond mean accuracy metrics. The results show that rotational installation error is a primary driver of system level error amplification, while translational misalignment of comparable magnitude plays a secondary role under typical biplanar geometries. Real biplanar X ray bench top experiments further confirm that the predicted amplification trends persist under realistic imaging conditions. These findings reveal a broader structural limitation of reference based multi stage geometric perception pipelines and provide a framework for system level reliability analysis and risk aware design in safety critical robotic navigation systems.
comment: 13 pages, 8 figures
☆ Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoning
Current Vision-Language-Action (VLA) models rely on fixed computational depth, expending the same amount of compute on simple adjustments and complex multi-step manipulation. While Chain-of-Thought (CoT) prompting enables variable computation, it scales memory linearly and is ill-suited for continuous action spaces. We introduce Recurrent-Depth VLA (RD-VLA), an architecture that achieves computational adaptivity via latent iterative refinement rather than explicit token generation. RD-VLA employs a recurrent, weight-tied action head that supports arbitrary inference depth with a constant memory footprint. The model is trained using truncated backpropagation through time (TBPTT) to efficiently supervise the refinement process. At inference, RD-VLA dynamically allocates compute using an adaptive stopping criterion based on latent convergence. Experiments on challenging manipulation tasks show that recurrent depth is critical: tasks that fail entirely (0 percent success) with single-iteration inference exceed 90 percent success with four iterations, while simpler tasks saturate rapidly. RD-VLA provides a scalable path to test-time compute in robotics, replacing token-based reasoning with latent reasoning to achieve constant memory usage and up to 80x inference speedup over prior reasoning-based VLA models. Project page: https://rd-vla.github.io/
comment: 11 Pages, Project page:https://rd-vla.github.io/
☆ RLinf-USER: A Unified and Extensible System for Real-World Online Policy Learning in Embodied AI
Online policy learning directly in the physical world is a promising yet challenging direction for embodied intelligence. Unlike simulation, real-world systems cannot be arbitrarily accelerated, cheaply reset, or massively replicated, which makes scalable data collection, heterogeneous deployment, and long-horizon effective training difficult. These challenges suggest that real-world policy learning is not only an algorithmic issue but fundamentally a systems problem. We present USER, a Unified and extensible SystEm for Real-world online policy learning. USER treats physical robots as first-class hardware resources alongside GPUs through a unified hardware abstraction layer, enabling automatic discovery, management, and scheduling of heterogeneous robots. To address cloud-edge communication, USER introduces an adaptive communication plane with tunneling-based networking, distributed data channels for traffic localization, and streaming-multiprocessor-aware weight synchronization to regulate GPU-side overhead. On top of this infrastructure, USER organizes learning as a fully asynchronous framework with a persistent, cache-aware buffer, enabling efficient long-horizon experiments with robust crash recovery and reuse of historical data. In addition, USER provides extensible abstractions for rewards, algorithms, and policies, supporting online imitation or reinforcement learning of CNN/MLP, generative policies, and large vision-language-action (VLA) models within a unified pipeline. Results in both simulation and the real world show that USER enables multi-robot coordination, heterogeneous manipulators, edge-cloud collaboration with large models, and long-running asynchronous training, offering a unified and extensible systems foundation for real-world online policy learning.
☆ CoLF: Learning Consistent Leader-Follower Policies for Vision-Language-Guided Multi-Robot Cooperative Transport
In this study, we address vision-language-guided multi-robot cooperative transport, where each robot grounds natural-language instructions from onboard camera observations. A key challenge in this decentralized setting is perceptual misalignment across robots, where viewpoint differences and language ambiguity can yield inconsistent interpretations and degrade cooperative transport. To mitigate this problem, we adopt a dependent leader-follower design, where one robot serves as the leader and the other as the follower. Although such a leader-follower structure appears straightforward, learning with independent and symmetric agents often yields symmetric or unstable behaviors without explicit inductive biases. To address this challenge, we propose Consistent Leader-Follower (CoLF), a multi-agent reinforcement learning (MARL) framework for stable leader-follower role differentiation. CoLF consists of two key components: (1) an asymmetric policy design that induces leader-follower role differentiation, and (2) a mutual-information-based training objective that maximizes a variational lower bound, encouraging the follower to predict the leader's action from its local observation. The leader and follower policies are jointly optimized under the centralized training and decentralized execution (CTDE) framework to balance task execution and consistent cooperative behaviors. We validate CoLF in both simulation and real-robot experiments using two quadruped robots. The demonstration video is available at https://sites.google.com/view/colf/.
comment: 9 pages, 5 figures
☆ Global Symmetry and Orthogonal Transformations from Geometrical Moment $n$-tuples
Detecting symmetry is crucial for effective object grasping for several reasons. Recognizing symmetrical features or axes within an object helps in developing efficient grasp strategies, as grasping along these axes typically results in a more stable and balanced grip, thereby facilitating successful manipulation. This paper employs geometrical moments to identify symmetries and estimate orthogonal transformations, including rotations and mirror transformations, for objects centered at the frame origin. It provides distinctive metrics for detecting symmetries and estimating orthogonal transformations, encompassing rotations, reflections, and their combinations. A comprehensive methodology is developed to obtain these functions in n-dimensional space, specifically moment \( n \)-tuples. Extensive validation tests are conducted on both 2D and 3D objects to ensure the robustness and reliability of the proposed approach. The proposed method is also compared to state-of-the-art work using iterative optimization for detecting multiple planes of symmetry. The results indicate that combining our method with the iterative one yields satisfactory outcomes in terms of the number of symmetry planes detected and computation time.
♻ ☆ Virtual Community: An Open World for Humans, Robots, and Society
The rapid progress in AI and Robotics may lead to a profound societal transformation, as humans and robots begin to coexist within shared communities, introducing both opportunities and challenges. To explore this future, we present Virtual Community-an open-world platform for humans, robots, and society-built on a universal physics engine and grounded in real-world 3D scenes. With Virtual Community, we aim to enable the study of embodied social intelligence at scale. To support these, Virtual Community features: 1) An open-source multi-agent physics simulator that supports robots, humans, and their interactions within a society; 2) A large-scale, real-world aligned community generation pipeline, including vast outdoor space, diverse indoor scenes, and a community of grounded agents with rich characters and appearances. Leveraging Virtual Community, we propose two novel challenges. The Community Planning Challenge evaluates multi-agent reasoning and planning ability in open-world settings, such as cooperating to help agents with daily activities and efficiently connecting other agents. The Community Robot Challenge requires multiple heterogeneous robots to collaborate in solving complex open-world tasks. We evaluate various baselines on these tasks and demonstrate the challenges in both high-level open-world task planning and low-level cooperation controls. We hope that Virtual Community will unlock further study of human-robot coexistence within open-world environments.
comment: website https://virtual-community-ai.github.io/
♻ ☆ Collision Risk Estimation via Loss Prediction in End-to-End Autonomous Driving
Collision risk estimation and avoidance play central roles in the safety of autonomous driving (AD) systems. Recently emerged end-to-end AD systems gain collision avoidance ability by minimizing losses to penalize planning trajectories that are too close to other objects. Despite a significant collision rate during testing, most end-to-end planners do not explicitly quantify the collision risk in their outputs. To address this, we introduce RiskMonitor, an efficient plug-and-play module that interprets planning and motion tokens from state-of-the-art end-to-end planners to estimate collision risk. Inspired by loss prediction based uncertainty quantification, RiskMonitor predicts whether the collision loss -- commonly adopted to train end-to-end planners -- is positive along planned waypoints, framing collision risk estimation as a binary classification task. We evaluate RiskMonitor on the real-world nuScenes dataset (open-loop) and the neural-rendering based simulator, NeuroNCAP (closed-loop). Our token-driven method outperforms prediction-driven approaches, including deterministic rules, Gaussian mixture models, and Monte Carlo Dropout. When integrated with a simple braking policy, RiskMonitor improves collision avoidance ability by $66.5\%$ in a closed-loop test on safety-critical scenarios. These results demonstrate that monitoring collision risk using plan and motion tokens enhances the safety of end-to-end AD without retraining it.
♻ ☆ Neural-Augmented Kelvinlet for Real-Time Soft Tissue Deformation Modeling
Accurate and efficient modeling of soft-tissue interactions is fundamental for advancing surgical simulation, surgical robotics, and model-based surgical automation. To achieve real-time latency, classical Finite Element Method (FEM) solvers are often replaced with neural approximations; however, naively training such models in a fully data-driven manner without incorporating physical priors frequently leads to poor generalization and physically implausible predictions. We present a novel physics-informed neural simulation framework that enables real-time prediction of soft-tissue deformations under complex single- and multi-grasper interactions. Our approach integrates Kelvinlet-based analytical priors with large-scale FEM data, capturing both linear and nonlinear tissue responses. This hybrid design improves predictive accuracy and physical plausibility across diverse neural architectures while maintaining the low-latency performance required for interactive applications. We validate our method on challenging surgical manipulation tasks involving standard laparoscopic grasping tools, demonstrating substantial improvements in deformation fidelity and temporal stability over existing baselines. These results establish Kelvinlet-augmented learning as a principled and computationally efficient paradigm for real-time, physics-aware soft-tissue simulation in surgical AI.
♻ ☆ RAP: 3D Rasterization Augmented End-to-End Planning
Imitation learning for end-to-end driving trains policies only on expert demonstrations. Once deployed in a closed loop, such policies lack recovery data: small mistakes cannot be corrected and quickly compound into failures. A promising direction is to generate alternative viewpoints and trajectories beyond the logged path. Prior work explores photorealistic digital twins via neural rendering or game engines, but these methods are prohibitively slow and costly, and thus mainly used for evaluation. In this work, we argue that photorealism is unnecessary for training end-to-end planners. What matters is semantic fidelity and scalability: driving depends on geometry and dynamics, not textures or lighting. Motivated by this, we propose 3D Rasterization, which replaces costly rendering with lightweight rasterization of annotated primitives, enabling augmentations such as counterfactual recovery maneuvers and cross-agent view synthesis. To transfer these synthetic views effectively to real-world deployment, we introduce a Raster-to-Real feature-space alignment that bridges the sim-to-real gap. Together, these components form Rasterization Augmented Planning (RAP), a scalable data augmentation pipeline for planning. RAP achieves state-of-the-art closed-loop robustness and long-tail generalization, ranking first on four major benchmarks: NAVSIM v1/v2, Waymo Open Dataset Vision-based E2E Driving, and Bench2Drive. Our results show that lightweight rasterization with feature alignment suffices to scale E2E training, offering a practical alternative to photorealistic rendering. Project page: https://alan-lanfeng.github.io/RAP/.
♻ ☆ Agentic Vehicles for Human-Centered Mobility
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 integrate agentic AI systems to reason, adapt, and interact within complex environments. It synthesizes recent advances in agentic systems and suggests how AgVs can complement and even reshape conventional autonomy to ensure mobility services are aligned with user and societal needs. The paper concludes by outlining key challenges in the development and governance of AgVs and their potential role in shaping future agentic transportation systems.
♻ ☆ An Adaptive Inspection Planning Approach Towards Routine Monitoring in Uncertain Environments ICRA 2026
In this work, we present a hierarchical framework designed to support robotic inspection under environment uncertainty. By leveraging a known environment model, existing methods plan and safely track inspection routes to visit points of interest. However, discrepancies between the model and actual site conditions, caused by either natural or human activities, can alter the surface morphology or introduce path obstructions. To address this challenge, the proposed framework divides the inspection task into: (a) generating the initial global view-plan for region of interests based on a historical map and (b) local view replanning to adapt to the current morphology of the inspection scene. The proposed hierarchy preserves global coverage objectives while enabling reactive adaptation to the local surface morphology. This enables the local autonomy to remain robust against environment uncertainty and complete the inspection tasks. We validate the approach through deployments in real-world subterranean mines using quadrupedal robot. A supplementary media highlighting the proposed method can be found here https://youtu.be/6TxK8S_83Lw.
comment: Accepted to ICRA 2026
♻ ☆ Omni-LIVO: Robust RGB-Colored Multi-Camera Visual-Inertial-LiDAR Odometry via Photometric Migration and ESIKF Fusion
Wide field-of-view (FoV) LiDAR sensors provide dense geometry across large environments, but existing LiDAR-inertial-visual odometry (LIVO) systems generally rely on a single camera, limiting their ability to fully exploit LiDAR-derived depth for photometric alignment and scene colorization. We present Omni-LIVO, a tightly coupled multi-camera LIVO system that leverages multi-view observations to comprehensively utilize LiDAR geometric information across extended spatial regions. Omni-LIVO introduces a Cross-View direct alignment strategy that maintains photometric consistency across non-overlapping views, and extends the Error-State Iterated Kalman Filter (ESIKF) with multi-view updates and adaptive covariance. The system is evaluated on public benchmarks and our custom dataset, showing improved accuracy and robustness over state-of-the-art LIVO, LIO, and visual-inertial SLAM baselines. Code and dataset will be released upon publication.
comment: Accepted by IEEE Robotics and Automation Letters (RA-L). Early Access version available. This version supersedes all previous versions and is the official accepted manuscript for citation
♻ ☆ Gust Estimation and Rejection with a Disturbance Observer for Proprioceptive Underwater Soft Morphing Wings ICRA
Unmanned underwater vehicles are increasingly employed for maintenance and surveying tasks at sea, but their operation in shallow waters is often hindered by hydrodynamic disturbances such as waves, currents, and turbulence. These unsteady flows can induce rapid changes in direction and speed, compromising vehicle stability and manoeuvrability. Marine organisms contend with such conditions by combining proprioceptive feedback with flexible fins and tails to reject disturbances. Inspired by this strategy, we propose soft morphing wings endowed with proprioceptive sensing to mitigate environmental perturbations. The wing's continuous deformation provides a natural means to infer dynamic disturbances: sudden changes in camber directly reflect variations in the oncoming flow. By interpreting this proprioceptive signal, a disturbance observer can reconstruct flow parameters in real time. To enable this, we develop and experimentally validate a dynamic model of a hydraulically actuated soft wing with controllable camber. We then show that curvature-based sensing allows accurate estimation of disturbances in the angle of attack. Finally, we demonstrate that a controller leveraging these proprioceptive estimates can reject disturbances in the lift response of the soft wing. By combining proprioceptive sensing with a disturbance observer, this technique mirrors biological strategies and provides a pathway for soft underwater vehicles to maintain stability in hazardous environments.
comment: 2026 IEEE International Conference on Robotics & Automation (ICRA)
♻ ☆ Sim-to-Real Dynamic Object Manipulation on Conveyor Systems via Optimization Path Shaping
Realizing generalizable dynamic object manipulation on conveyor systems is important for enhancing manufacturing efficiency, as it eliminates specialized engineering for different scenarios. To this end, imitation learning emerges as a promising paradigm, leveraging expert demonstrations to teach a policy manipulation skills. Although the generalization of an imitation learning policy can be improved by increasing demonstrations, demonstration collection is labor-intensive. Besides, public dynamic object manipulation data is scarce. In this work, we address this data scarcity problem via generating demonstrations in a simulator. A significant challenge of using simulated data lies in the appearance gap between simulated and real-world observations. To tackle this challenge, we propose Geometry-Enhanced Model (GEM), which employs our designed appearance noise annealing strategy to shape the policy optimization path, thereby prioritizing the geometry information in observations. Extensive experiments in simulated and real-world tasks demonstrate that GEM can generalize across environment backgrounds, robot embodiments, motion dynamics, and object geometries. Notably, GEM is deployed in a real canteen for tableware collection. Without test-scene data, GEM achieves a success rate of over 97% across more than 10,000 operations.
♻ ☆ Safe Exploration via Policy Priors
Safe exploration is a key requirement for reinforcement learning (RL) agents to learn and adapt online, beyond controlled (e.g. simulated) environments. In this work, we tackle this challenge by utilizing suboptimal yet conservative policies (e.g., obtained from offline data or simulators) as priors. Our approach, SOOPER, uses probabilistic dynamics models to optimistically explore, yet pessimistically fall back to the conservative policy prior if needed. We prove that SOOPER guarantees safety throughout learning, and establish convergence to an optimal policy by bounding its cumulative regret. Extensive experiments on key safe RL benchmarks and real-world hardware demonstrate that SOOPER is scalable, outperforms the state-of-the-art and validate our theoretical guarantees in practice.
♻ ☆ SPOT: Spatio-Temporal Obstacle-free Trajectory Planning for UAVs in an Unknown Dynamic Environment ICRA 2026
We address the problem of reactive motion planning for quadrotors operating in unknown environments with dynamic obstacles. Our approach leverages a 4-dimensional spatio-temporal planner, integrated with vision-based Safe Flight Corridor (SFC) generation and trajectory optimization. Unlike prior methods that rely on map fusion, our framework is mapless, enabling collision avoidance directly from perception while reducing computational overhead. Dynamic obstacles are detected and tracked using a vision-based object segmentation and tracking pipeline, allowing robust classification of static versus dynamic elements in the scene. To further enhance robustness, we introduce a backup planning module that reactively avoids dynamic obstacles when no direct path to the goal is available, mitigating the risk of collisions during deadlock situations. We validate our method extensively in both simulation and real-world hardware experiments, and benchmark it against state-of-the-art approaches, showing significant advantages for reactive UAV navigation in dynamic, unknown environments.
comment: Accepted for publication at ICRA 2026
♻ ☆ Knowledge-Centric Metacognitive Learning
Interactions are central to intelligent reasoning and learning abilities, with the interpretation of abstract knowledge guiding meaningful interaction with objects in the environment. While humans readily adapt to novel situations by leveraging abstract knowledge acquired over time, artificial intelligence systems lack principled mechanisms for incorporating abstract knowledge into learning, leading to fundamental challenges in the emergence of intelligent and adaptive behavior. To address this gap, we introduce knowledge-centric metacognitive learning based on three key principles: natural abstractions, knowledge-guided interactions through interpretation, and the composition of interactions for problem solving. Knowledge learning facilitates the acquisition of abstract knowledge and the association of interactions with knowledge, while object interactions guided by abstract knowledge enable the learning of transferable interaction concepts, abstract reasoning, and generalization. This metacognitive mechanism provides a principled approach for integrating knowledge into reinforcement learning and offers a promising pathway toward intelligent and adaptive behavior in artificial intelligence, robotics, and autonomous systems.
♻ ☆ 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 SEC filings and 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 work 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. Our evaluation of rule-based Nav2 navigation shows that current approaches are not economically viable: the contribution margin is -22.81/run (AMCL) and -12.87/run (GPS), resulting in no break-even point. We challenge the community to develop navigation policies that achieve economic viability on CostNav. We remain method-agnostic, evaluating success solely on the metric of cost rather than the underlying architecture. All resources are available at https://github.com/worv-ai/CostNav.
♻ ☆ Residual Vector Quantization For Communication-Efficient Multi-Agent Perception ICASSP 2026
Multi-agent collaborative perception (CP) improves scene understanding by sharing information across connected agents such as autonomous vehicles, unmanned aerial vehicles, and robots. Communication bandwidth, however, constrains scalability. We present ReVQom, a learned feature codec that preserves spatial identity while compressing intermediate features. ReVQom is an end-to-end method that compresses feature dimensions via a simple bottleneck network followed by multi-stage residual vector quantization (RVQ). This allows only per-pixel code indices to be transmitted, reducing payloads from 8192 bits per pixel (bpp) of uncompressed 32-bit float features to 6-30 bpp per agent with minimal accuracy loss. On DAIR-V2X real-world CP dataset, ReVQom achieves 273x compression at 30 bpp to 1365x compression at 6 bpp. At 18 bpp (455x), ReVQom matches or outperforms raw-feature CP, and at 6-12 bpp it enables ultra-low-bandwidth operation with graceful degradation. ReVQom allows efficient and accurate multi-agent collaborative perception with a step toward practical V2X deployment.
comment: Accepted at ICASSP 2026. 5 pages
♻ ☆ Constructing the Umwelt: Cognitive Planning through Belief-Intent Co-Evolution
This paper challenges a prevailing epistemological assumption in End-to-End Autonomous Driving: that high-performance planning necessitates high-fidelity world reconstruction. Inspired by cognitive science, we propose the Mental Bayesian Causal World Model (MBCWM) and instantiate it as the Tokenized Intent World Model (TIWM), a novel cognitive computing architecture. Its core philosophy posits that intelligence emerges not from pixel-level objective fidelity, but from the Cognitive Consistency between the agent's internal intentional world and physical reality. By synthesizing von Uexküll's $\textit{Umwelt}$ theory, the neural assembly hypothesis, and the triple causal model (integrating symbolic deduction, probabilistic induction, and force dynamics) into an end-to-end embodied planning system, we demonstrate the feasibility of this paradigm on the nuPlan benchmark. Experimental results in open-loop validation confirm that our Belief-Intent Co-Evolution mechanism effectively enhances planning performance. Crucially, in closed-loop simulations, the system exhibits emergent human-like cognitive behaviors, including map affordance understanding, free exploration, and self-recovery strategies. We identify Cognitive Consistency as the core learning mechanism: during long-term training, belief (state understanding) and intent (future prediction) spontaneously form a self-organizing equilibrium through implicit computational replay, achieving semantic alignment between internal representations and physical world affordances. TIWM offers a neuro-symbolic, cognition-first alternative to reconstruction-based planners, establishing a new direction: planning as active understanding, not passive reaction.
comment: 12 pages, 8 figures. A paradigm shift from reconstructing the world to understanding it: planning through Belief-Intent Co-Evolution
Robotics 32
☆ Vision and language: Novel Representations and Artificial intelligence for Driving Scene Safety Assessment and Autonomous Vehicle Planning
Vision-language models (VLMs) have recently emerged as powerful representation learning systems that align visual observations with natural language concepts, offering new opportunities for semantic reasoning in safety-critical autonomous driving. This paper investigates how vision-language representations support driving scene safety assessment and decision-making when integrated into perception, prediction, and planning pipelines. We study three complementary system-level use cases. First, we introduce a lightweight, category-agnostic hazard screening approach leveraging CLIP-based image-text similarity to produce a low-latency semantic hazard signal. This enables robust detection of diverse and out-of-distribution road hazards without explicit object detection or visual question answering. Second, we examine the integration of scene-level vision-language embeddings into a transformer-based trajectory planning framework using the Waymo Open Dataset. Our results show that naively conditioning planners on global embeddings does not improve trajectory accuracy, highlighting the importance of representation-task alignment and motivating the development of task-informed extraction methods for safety-critical planning. Third, we investigate natural language as an explicit behavioral constraint on motion planning using the doScenes dataset. In this setting, passenger-style instructions grounded in visual scene elements suppress rare but severe planning failures and improve safety-aligned behavior in ambiguous scenarios. Taken together, these findings demonstrate that vision-language representations hold significant promise for autonomous driving safety when used to express semantic risk, intent, and behavioral constraints. Realizing this potential is fundamentally an engineering problem requiring careful system design and structured grounding rather than direct feature injection.
☆ Affine Transformable Unmanned Ground Vehicle
This paper develops the proof of concept for a novel affine transformable unmanned ground vehicle (ATUGV) with the capability of safe and aggressive deformation while carrying multiple payloads. The ATUGV is a multi-body system with mobile robots that can be used to power the ATUGV morphable motion, powered cells to enclose the mobile robots, unpowered cells to contain payloads, and a deformable structure to integrate cells through bars and joints. The objective is that all powered and unpowered cells motion can safely track a desired affine transformation, where an affine transformation can be decomposed into translation, rigid body rotation, and deformation. To this end, the paper first uses a deep neural network to structure cell interconnection in such a way that every cell can freely move over the deformation plane, and the entire structure can reconfigurably deform to track a desired affine transformation. Then, the mobile robots, contained by the powered cells and stepper motors, regulating the connections of the powered and unpowered cells, design the proper controls so that all cells safely track the desired affine transformation. The functionality of the proposed ATUGV is validated through hardware experimentation and simulation.
☆ Looking and Listening Inside and Outside: Multimodal Artificial Intelligence Systems for Driver Safety Assessment and Intelligent Vehicle Decision-Making
The looking-in-looking-out (LILO) framework has enabled intelligent vehicle applications that understand both the outside scene and the driver state to improve safety outcomes, with examples in smart airbag deployment, takeover time prediction in autonomous control transitions, and driver attention monitoring. In this research, we propose an augmentation to this framework, making a case for the audio modality as an additional source of information to understand the driver, and in the evolving autonomy landscape, also the passengers and those outside the vehicle. We expand LILO by incorporating audio signals, forming the looking-and-listening inside-and-outside (L-LIO) framework to enhance driver state assessment and environment understanding through multimodal sensor fusion. We evaluate three example cases where audio enhances vehicle safety: supervised learning on driver speech audio to classify potential impairment states (e.g., intoxication), collection and analysis of passenger natural language instructions (e.g., "turn after that red building") to motivate how spoken language can interface with planning systems through audio-aligned instruction data, and limitations of vision-only systems where audio may disambiguate the guidance and gestures of external agents. Datasets include custom-collected in-vehicle and external audio samples in real-world environments. Pilot findings show that audio yields safety-relevant insights, particularly in nuanced or context-rich scenarios where sound is critical to safe decision-making or visual signals alone are insufficient. Challenges include ambient noise interference, privacy considerations, and robustness across human subjects, motivating further work on reliability in dynamic real-world contexts. L-LIO augments driver and scene understanding through multimodal fusion of audio and visual sensing, offering new paths for safety intervention.
☆ LCLA: Language-Conditioned Latent Alignment for Vision-Language Navigation
We propose LCLA (Language-Conditioned Latent Alignment), a framework for vision-language navigation that learns modular perception-action interfaces by aligning sensory observations to a latent representation of an expert policy. The expert is first trained with privileged state information, inducing a latent space sufficient for control, after which its latent interface and action head are frozen. A lightweight adapter is then trained to map raw visual-language observations, via a frozen vision-language model, into the expert's latent space, reducing the problem of visuomotor learning to supervised latent alignment rather than end-to-end policy optimization. This decoupling enforces a stable contract between perception and control, enabling expert behavior to be reused across sensing modalities and environmental variations. We instantiate LCLA and evaluate it on a vision-language indoor navigation task, where aligned latent spaces yield strong in-distribution performance and robust zero-shot generalization to unseen environments, lighting conditions, and viewpoints while remaining lightweight at inference time.
☆ "Meet My Sidekick!": Effects of Separate Identities and Control of a Single Robot in HRI
The presentation of a robot's capability and identity directly influences a human collaborator's perception and implicit trust in the robot. Unlike humans, a physical robot can simultaneously present different identities and have them reside and control different parts of the robot. This paper presents a novel study that investigates how users perceive a robot where different robot control domains (head and gripper) are presented as independent robots. We conducted a mixed design study where participants experienced one of three presentations: a single robot, two agents with shared full control (co-embodiment), or two agents with split control across robot control domains (split-embodiment). Participants underwent three distinct tasks -- a mundane data entry task where the robot provides motivational support, an individual sorting task with isolated robot failures, and a collaborative arrangement task where the robot causes a failure that directly affects the human participant. Participants perceived the robot as residing in the different control domains and were able to associate robot failure with different identities. This work signals how future robots can leverage different embodiment configurations to obtain the benefit of multiple robots within a single body.
☆ Differentiate-and-Inject: Enhancing VLAs via Functional Differentiation Induced by In-Parameter Structural Reasoning
As robots are expected to perform increasingly diverse tasks, they must understand not only low-level actions but also the higher-level structure that determines how a task should unfold. Existing vision-language-action (VLA) models struggle with this form of task-level reasoning. They either depend on prompt-based in-context decomposition, which is unstable and sensitive to linguistic variations, or end-to-end long-horizon training, which requires large-scale demonstrations and entangles task-level reasoning with low-level control. We present in-parameter structured task reasoning (iSTAR), a framework for enhancing VLA models via functional differentiation induced by in-parameter structural reasoning. Instead of treating VLAs as monolithic policies, iSTAR embeds task-level semantic structure directly into model parameters, enabling differentiated task-level inference without external planners or handcrafted prompt inputs. This injected structure takes the form of implicit dynamic scene-graph knowledge that captures object relations, subtask semantics, and task-level dependencies in parameter space. Across diverse manipulation benchmarks, iSTAR achieves more reliable task decompositions and higher success rates than both in-context and end-to-end VLA baselines, demonstrating the effectiveness of parameter-space structural reasoning for functional differentiation and improved generalization across task variations.
☆ VividFace: Real-Time and Realistic Facial Expression Shadowing for Humanoid Robots ICRA
Humanoid facial expression shadowing enables robots to realistically imitate human facial expressions in real time, which is critical for lifelike, facially expressive humanoid robots and affective human-robot interaction. Existing progress in humanoid facial expression imitation remains limited, often failing to achieve either real-time performance or realistic expressiveness due to offline video-based inference designs and insufficient ability to capture and transfer subtle expression details. To address these limitations, we present VividFace, a real-time and realistic facial expression shadowing system for humanoid robots. An optimized imitation framework X2CNet++ enhances expressiveness by fine-tuning the human-to-humanoid facial motion transfer module and introducing a feature-adaptation training strategy for better alignment across different image sources. Real-time shadowing is further enabled by a video-stream-compatible inference pipeline and a streamlined workflow based on asynchronous I/O for efficient communication across devices. VividFace produces vivid humanoid faces by mimicking human facial expressions within 0.05 seconds, while generalizing across diverse facial configurations. Extensive real-world demonstrations validate its practical utility. Videos are available at: https://lipzh5.github.io/VividFace/.
comment: Accepted to the 2026 IEEE International Conference on Robotics and Automation (ICRA)
☆ TextOp: Real-time Interactive Text-Driven Humanoid Robot Motion Generation and Control
Recent advances in humanoid whole-body motion tracking have enabled the execution of diverse and highly coordinated motions on real hardware. However, existing controllers are commonly driven either by predefined motion trajectories, which offer limited flexibility when user intent changes, or by continuous human teleoperation, which requires constant human involvement and limits autonomy. This work addresses the problem of how to drive a universal humanoid controller in a real-time and interactive manner. We present TextOp, a real-time text-driven humanoid motion generation and control framework that supports streaming language commands and on-the-fly instruction modification during execution. TextOp adopts a two-level architecture in which a high-level autoregressive motion diffusion model continuously generates short-horizon kinematic trajectories conditioned on the current text input, while a low-level motion tracking policy executes these trajectories on a physical humanoid robot. By bridging interactive motion generation with robust whole-body control, TextOp unlocks free-form intent expression and enables smooth transitions across multiple challenging behaviors such as dancing and jumping, within a single continuous motion execution. Extensive real-robot experiments and offline evaluations demonstrate instant responsiveness, smooth whole-body motion, and precise control. The project page and the open-source code are available at https://text-op.github.io/
comment: Project Page: https://text-op.github.io/
☆ Bridging Speech, Emotion, and Motion: a VLM-based Multimodal Edge-deployable Framework for Humanoid Robots
Effective human-robot interaction requires emotionally rich multimodal expressions, yet most humanoid robots lack coordinated speech, facial expressions, and gestures. Meanwhile, real-world deployment demands on-device solutions that can operate autonomously without continuous cloud connectivity. To bridging \underline{\textit{S}}peech, \underline{\textit{E}}motion, and \underline{\textit{M}}otion, we present \textit{SeM$^2$}, a Vision Language Model-based framework that orchestrates emotionally coherent multimodal interactions through three key components: a multimodal perception module capturing user contextual cues, a Chain-of-Thought reasoning for response planning, and a novel Semantic-Sequence Aligning Mechanism (SSAM) that ensures precise temporal coordination between verbal content and physical expressions. We implement both cloud-based and \underline{\textit{e}}dge-deployed versions (\textit{SeM$^2_e$}), with the latter knowledge distilled to operate efficiently on edge hardware while maintaining 95\% of the relative performance. Comprehensive evaluations demonstrate that our approach significantly outperforms unimodal baselines in naturalness, emotional clarity, and modal coherence, advancing socially expressive humanoid robotics for diverse real-world environments.
☆ Going with the Flow: Koopman Behavioral Models as Implicit Planners for Visuo-Motor Dexterity
There has been rapid and dramatic progress in robots' ability to learn complex visuo-motor manipulation skills from demonstrations, thanks in part to expressive policy classes that employ diffusion- and transformer-based backbones. However, these design choices require significant data and computational resources and remain far from reliable, particularly within the context of multi-fingered dexterous manipulation. Fundamentally, they model skills as reactive mappings and rely on fixed-horizon action chunking to mitigate jitter, creating a rigid trade-off between temporal coherence and reactivity. In this work, we introduce Unified Behavioral Models (UBMs), a framework that learns to represent dexterous skills as coupled dynamical systems that capture how visual features of the environment (visual flow) and proprioceptive states of the robot (action flow) co-evolve. By capturing such behavioral dynamics, UBMs can ensure temporal coherence by construction rather than by heuristic averaging. To operationalize these models, we propose Koopman-UBM, a first instantiation of UBMs that leverages Koopman Operator theory to effectively learn a unified representation in which the joint flow of latent visual and proprioceptive features is governed by a structured linear system. We demonstrate that Koopman-UBM can be viewed as an implicit planner: given an initial condition, it analytically computes the desired robot behavior while simultaneously ''imagining'' the resulting flow of visual features over the entire skill horizon. To enable reactivity and adaptation, we introduce an online replanning strategy in which the model acts as its own runtime monitor that automatically triggers replanning when predicted and observed visual flow diverge beyond a threshold. Across seven simulated tasks and two real-world tasks, we demonstrate that K-UBM matches or exceeds the performance of state-of-the-art baselines, while offering considerably faster inference, smooth execution, robustness to occlusions, and flexible replanning.
☆ Haptically Experienced Animacy Facilitates Emotion Regulation: A Theory-Driven Investigation
Emotion regulation (ER) is essential to mental well-being but often difficult to access, especially in high-intensity moments or for individuals with clinical vulnerabilities. While existing technology-based ER tools offer value, they typically rely on self-reflection (e.g., emotion tracking, journaling) or co-regulation through verbal modalities (reminders, text-based conversational tools), which may not be accessible or effective when most needed. The biological role of the touch modality makes it an intriguing alternate pathway, but empirical evidence is limited and under-theorized. Building on our prior theoretical framework describing how a comforting haptic co-regulating adjunct (CHORA) can support ER, we developed a zoomorphic robot CHORA with looped biomimetic breathing and heartbeat behaviors. We evaluated its effects in a mixed-methods in-lab study (N=30), providing physiological, self-report, custom questionnaire, and retrospective interview data. Our findings demonstrate the regulatory effects of haptically experienced animacy, corroborate prior work, and validate CHORA's {theoretically grounded} potential to facilitate four ER strategies.
☆ Trace-Focused Diffusion Policy for Multi-Modal Action Disambiguation in Long-Horizon Robotic Manipulation
Generative model-based policies have shown strong performance in imitation-based robotic manipulation by learning action distributions from demonstrations. However, in long-horizon tasks, visually similar observations often recur across execution stages while requiring distinct actions, which leads to ambiguous predictions when policies are conditioned only on instantaneous observations, termed multi-modal action ambiguity (MA2). To address this challenge, we propose the Trace-Focused Diffusion Policy (TF-DP), a simple yet effective diffusion-based framework that explicitly conditions action generation on the robot's execution history. TF-DP represents historical motion as an explicit execution trace and projects it into the visual observation space, providing stage-aware context when current observations alone are insufficient. In addition, the induced trace-focused field emphasizes task-relevant regions associated with historical motion, improving robustness to background visual disturbances. We evaluate TF-DP on real-world robotic manipulation tasks exhibiting pronounced multi-modal action ambiguity and visually cluttered conditions. Experimental results show that TF-DP improves temporal consistency and robustness, outperforming the vanilla diffusion policy by 80.56 percent on tasks with multi-modal action ambiguity and by 86.11 percent under visual disturbances, while maintaining inference efficiency with only a 6.4 percent runtime increase. These results demonstrate that execution-trace conditioning offers a scalable and principled approach for robust long-horizon robotic manipulation within a single policy.
☆ UEREBot: Learning Safe Quadrupedal Locomotion under Unstructured Environments and High-Speed Dynamic Obstacles
Quadruped robots are increasingly deployed in unstructured environments. Safe locomotion in these settings requires long-horizon goal progress, passability over uneven terrain and static constraints, and collision avoidance against high-speed dynamic obstacles. A single system cannot fully satisfy all three objectives simultaneously: planning-based decisions can be too slow, while purely reactive decisions can sacrifice goal progress and passability. To resolve this conflict, we propose UEREBot (Unstructured-Environment Reflexive Evasion Robot), a hierarchical framework that separates slow planning from instantaneous reflexive evasion and coordinates them during execution. UEREBot formulates the task as a constrained optimal control problem blueprint. It adopts a spatial--temporal planner that provides reference guidance toward the goal and threat signals. It then uses a threat-aware handoff to fuse navigation and reflex actions into a nominal command, and a control barrier function shield as a final execution safeguard. We evaluate UEREBot in Isaac Lab simulation and deploy it on a Unitree Go2 quadruped equipped with onboard perception. Across diverse environments with complex static structure and high-speed dynamic obstacles, UEREBot achieves higher avoidance success and more stable locomotion while maintaining goal progress than representative baselines, demonstrating improved safety--progress trade-offs.
☆ Seeing Roads Through Words: A Language-Guided Framework for RGB-T Driving Scene Segmentation
Robust semantic segmentation of road scenes under adverse illumination, lighting, and shadow conditions remain a core challenge for autonomous driving applications. RGB-Thermal fusion is a standard approach, yet existing methods apply static fusion strategies uniformly across all conditions, allowing modality-specific noise to propagate throughout the network. Hence, we propose CLARITY that dynamically adapts its fusion strategy to the detected scene condition. Guided by vision-language model (VLM) priors, the network learns to modulate each modality's contribution based on the illumination state while leveraging object embeddings for segmentation, rather than applying a fixed fusion policy. We further introduce two mechanisms, i.e., one which preserves valid dark-object semantics that prior noise-suppression methods incorrectly discard, and a hierarchical decoder that enforces structural consistency across scales to sharpen boundaries on thin objects. Experiments on the MFNet dataset demonstrate that CLARITY establishes a new state-of-the-art (SOTA), achieving 62.3% mIoU and 77.5% mAcc.
☆ Scalable Dexterous Robot Learning with AR-based Remote Human-Robot Interactions
This paper focuses on the scalable robot learning for manipulation in the dexterous robot arm-hand systems, where the remote human-robot interactions via augmented reality (AR) are established to collect the expert demonstration data for improving efficiency. In such a system, we present a unified framework to address the general manipulation task problem. Specifically, the proposed method consists of two phases: i) In the first phase for pretraining, the policy is created in a behavior cloning (BC) manner, through leveraging the learning data from our AR-based remote human-robot interaction system; ii) In the second phase, a contrastive learning empowered reinforcement learning (RL) method is developed to obtain more efficient and robust policy than the BC, and thus a projection head is designed to accelerate the learning progress. An event-driven augmented reward is adopted for enhancing the safety. To validate the proposed method, both the physics simulations via PyBullet and real-world experiments are carried out. The results demonstrate that compared to the classic proximal policy optimization and soft actor-critic policies, our method not only significantly speeds up the inference, but also achieves much better performance in terms of the success rate for fulfilling the manipulation tasks. By conducting the ablation study, it is confirmed that the proposed RL with contrastive learning overcomes policy collapse. Supplementary demonstrations are available at https://cyberyyc.github.io/.
☆ RAPiD: Real-time Deterministic Trajectory Planning via Diffusion Behavior Priors for Safe and Efficient Autonomous Driving
Diffusion-based trajectory planners have demonstrated strong capability for modeling the multimodal nature of human driving behavior, but their reliance on iterative stochastic sampling poses critical challenges for real-time, safety-critical deployment. In this work, we present RAPiD, a deterministic policy extraction framework that distills a pretrained diffusion-based planner into an efficient policy while eliminating diffusion sampling. Using score-regularized policy optimization, we leverage the score function of a pre-trained diffusion planner as a behavior prior to regularize policy learning. To promote safety and passenger comfort, the policy is optimized using a critic trained to imitate a predictive driver controller, providing dense, safety-focused supervision beyond conventional imitation learning. Evaluations demonstrate that RAPiD achieves competitive performance on closed-loop nuPlan scenarios with an 8x speedup over diffusion baselines, while achieving state-of-the-art generalization among learning-based planners on the interPlan benchmark. The official website of this work is: https://github.com/ruturajreddy/RAPiD.
☆ Why Look at It at All?: Vision-Free Multifingered Blind Grasping Using Uniaxial Fingertip Force Sensing
Grasping under limited sensing remains a fundamental challenge for real-world robotic manipulation, as vision and high-resolution tactile sensors often introduce cost, fragility, and integration complexity. This work demonstrates that reliable multifingered grasping can be achieved under extremely minimal sensing by relying solely on uniaxial fingertip force feedback and joint proprioception, without vision or multi-axis/tactile sensing. To enable such blind grasping, we employ an efficient teacher-student training pipeline in which a reinforcement-learned teacher exploits privileged simulation-only observations to generate demonstrations for distilling a transformer-based student policy operating under partial observation. The student policy is trained to act using only sensing modalities available at real-world deployment. We validate the proposed approach on real hardware across 18 objects, including both in-distribution and out-of-distribution cases, achieving a 98.3~$\%$ overall grasp success rate. These results demonstrate strong robustness and generalization beyond the simulation training distribution, while significantly reducing sensing requirements for real-world grasping systems.
comment: Submitted to Journal (under review)
☆ Action-to-Action Flow Matching
Diffusion-based policies have recently achieved remarkable success in robotics by formulating action prediction as a conditional denoising process. However, the standard practice of sampling from random Gaussian noise often requires multiple iterative steps to produce clean actions, leading to high inference latency that incurs a major bottleneck for real-time control. In this paper, we challenge the necessity of uninformed noise sampling and propose Action-to-Action flow matching (A2A), a novel policy paradigm that shifts from random sampling to initialization informed by the previous action. Unlike existing methods that treat proprioceptive action feedback as static conditions, A2A leverages historical proprioceptive sequences, embedding them into a high-dimensional latent space as the starting point for action generation. This design bypasses costly iterative denoising while effectively capturing the robot's physical dynamics and temporal continuity. Extensive experiments demonstrate that A2A exhibits high training efficiency, fast inference speed, and improved generalization. Notably, A2A enables high-quality action generation in as few as a single inference step (0.56 ms latency), and exhibits superior robustness to visual perturbations and enhanced generalization to unseen configurations. Lastly, we also extend A2A to video generation, demonstrating its broader versatility in temporal modeling. Project site: https://lorenzo-0-0.github.io/A2A_Flow_Matching.
comment: 18 pages, 18 figures
♻ ☆ Parallel Simulation of Contact and Actuation for Soft Growing Robots
Soft growing robots, commonly referred to as vine robots, have demonstrated remarkable ability to interact safely and robustly with unstructured and dynamic environments. It is therefore natural to exploit contact with the environment for planning and design optimization tasks. Previous research has focused on planning under contact for passively deforming robots with pre-formed bends. However, adding active steering to these soft growing robots is necessary for successful navigation in more complex environments. To this end, we develop a unified modeling framework that integrates vine robot growth, bending, actuation, and obstacle contact. We extend the beam moment model to include the effects of actuation on kinematics under growth and then use these models to develop a fast parallel simulation framework. We validate our model and simulator with real robot experiments. To showcase the capabilities of our framework, we apply our model in a design optimization task to find designs for vine robots navigating through cluttered environments, identifying designs that minimize the number of required actuators by exploiting environmental contacts. We show the robustness of the designs to environmental and manufacturing uncertainties. Finally, we fabricate an optimized design and successfully deploy it in an obstacle-rich environment.
comment: 37 pages, 12 figures, 2 tables. To appear in Soft Robotics
♻ ☆ GrndCtrl: Grounding World Models via Self-Supervised Reward Alignment
Recent advances in video world modeling have enabled large-scale generative models to simulate embodied environments with high visual fidelity, providing strong priors for prediction, planning, and control. Yet, despite their realism, these models often lack geometric grounding, limiting their use in navigation tasks that require spatial coherence and stability. We introduce Reinforcement Learning with World Grounding (RLWG), a self-supervised post-training framework that aligns pretrained world models with a physically verifiable structure through geometric and perceptual rewards. Analogous to reinforcement learning from verifiable feedback (RLVR) in language models, RLWG can use multiple rewards that measure pose cycle-consistency, depth reprojection, and temporal coherence. We instantiate this framework with GrndCtrl, a reward-aligned adaptation method based on Group Relative Policy Optimization (GRPO), yielding world models that maintain stable trajectories, consistent geometry, and reliable rollouts for embodied navigation. Like post-training alignment in large language models, GrndCtrl leverages verifiable rewards to bridge generative pretraining and grounded behavior, achieving superior spatial coherence and navigation stability over supervised fine-tuning in outdoor environments.
♻ ☆ 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
♻ ☆ Optimizing Automated Picking Systems in Warehouse Robots Using Machine Learning
With the rapid growth of global e-commerce, the demand for automation in the logistics industry is increasing. This study focuses on automated picking systems in warehouses, utilizing deep learning and reinforcement learning technologies to enhance picking efficiency and accuracy while reducing system failure rates. Through empirical analysis, we demonstrate the effectiveness of these technologies in improving robot picking performance and adaptability to complex environments. The results show that the integrated machine learning model significantly outperforms traditional methods, effectively addressing the challenges of peak order processing, reducing operational errors, and improving overall logistics efficiency. Additionally, by analyzing environmental factors, this study further optimizes system design to ensure efficient and stable operation under variable conditions. This research not only provides innovative solutions for logistics automation but also offers a theoretical and empirical foundation for future technological development and application.
comment: Published in IEEE Xplore
♻ ☆ RLinf-VLA: A Unified and Efficient Framework for Reinforcement Learning of Vision-Language-Action Models
Recent advances in vision-language-action (VLA) models have motivated the extension of their capabilities to embodied settings, where reinforcement learning (RL) offers a principled way to optimize task success through interaction. However, existing methods remain fragmented, lacking both a unified platform for fair comparison across architectures and algorithms and an efficient system design for scalable training. To address these challenges, we introduce RLinf-VLA, a unified and efficient framework for scalable RL training of VLA models. RLinf-VLA achieves unification by providing a unified interface that standardizes the integration of diverse VLA architectures, multiple RL algorithms, and heterogeneous simulators, enabling extensibility. To ensure efficiency, the system adopts a flexible resource allocation architecture for rendering, inference, and training workloads in RL pipelines. In particular, for GPU-parallelized simulators, RLinf-VLA introduces a hybrid fine-grained pipeline allocation strategy, yielding a 1.61x-1.88x training speedup. Using this unified system, models trained with RLinf-VLA demonstrate consistent performance improvements of approximately 20-85% across multiple simulation benchmarks, including LIBERO, ManiSkill, and RoboTwin. Furthermore, we distill a set of training practices for effective RL-based VLA training. We position RLinf-VLA as a foundational system to enable efficient, unified, and reproducible research in embodied intelligence.
comment: This is the technical report of the RLinf Team, focusing on the algorithm side. For the system-level design, please refer to arXiv:2509.15965 . The open-sourced code link: https://github.com/RLinf/RLinf
♻ ☆ StreamVLA: Breaking the Reason-Act Cycle via Completion-State Gating
Long-horizon robotic manipulation requires bridging the gap between high-level planning (System 2) and low-level control (System 1). Current Vision-Language-Action (VLA) models often entangle these processes, performing redundant multimodal reasoning at every timestep, which leads to high latency and goal instability. To address this, we present StreamVLA, a dual-system architecture that unifies textual task decomposition, visual goal imagination, and continuous action generation within a single parameter-efficient backbone. We introduce a "Lock-and-Gated" mechanism to intelligently modulate computation: only when a sub-task transition is detected, the model triggers slow thinking to generate a textual instruction and imagines the specific visual completion state, rather than generic future frames. Crucially, this completion state serves as a time-invariant goal anchor, making the policy robust to execution speed variations. During steady execution, these high-level intents are locked to condition a Flow Matching action head, allowing the model to bypass expensive autoregressive decoding for 72% of timesteps. This hierarchical abstraction ensures sub-goal focus while significantly reducing inference latency. Extensive evaluations demonstrate that StreamVLA achieves state-of-the-art performance, with a 98.5% success rate on the LIBERO benchmark and robust recovery in real-world interference scenarios, achieving a 48% reduction in latency compared to full-reasoning baselines.
♻ ☆ SPICE-HL3: Single-Photon, Inertial, and Stereo Camera dataset for Exploration of High-Latitude Lunar Landscapes
Exploring high-latitude lunar regions presents an extremely challenging visual environment for robots. The low sunlight elevation angle and minimal light scattering result in a visual field dominated by a high dynamic range featuring long, dynamic shadows. Reproducing these conditions on Earth requires sophisticated simulators and specialized facilities. We introduce a unique dataset recorded at the LunaLab from the SnT - University of Luxembourg, an indoor test facility designed to replicate the optical characteristics of multiple lunar latitudes. Our dataset includes images, inertial measurements, and wheel odometry data from robots navigating seven distinct trajectories under multiple illumination scenarios, simulating high-latitude lunar conditions from dawn to nighttime with and without the aid of headlights, resulting in 88 distinct sequences containing a total of 1.3M images. Data was captured using a stereo RGB-inertial sensor, a monocular monochrome camera, and for the first time, a novel single-photon avalanche diode (SPAD) camera. We recorded both static and dynamic image sequences, with robots navigating at slow (5 cm/s) and fast (50 cm/s) speeds. All data is calibrated, synchronized, and timestamped, providing a valuable resource for validating perception tasks from vision-based autonomous navigation to scientific imaging for future lunar missions targeting high-latitude regions or those intended for robots operating across perceptually degraded environments. The dataset and all supplementary material can be accessed from and found at https://github.com/spaceuma/spice-hl3.
comment: 12 pages, 7 figures, dataset
♻ ☆ 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
♻ ☆ Collaborative-Online-Learning-Enabled Distributionally Robust Motion Control for Multi-Robot Systems
This paper develops a novel COllaborative-Online-Learning (COOL)-enabled motion control framework for multi-robot systems to avoid collision amid randomly moving obstacles whose motion distributions are partially observable through decentralized data streams. To address the notable challenge of data acquisition due to occlusion, a COOL approach based on the Dirichlet process mixture model is proposed to efficiently extract motion distribution information by exchanging among robots selected learning structures. By leveraging the fine-grained local-moment information learned through COOL, a data-stream-driven ambiguity set for obstacle motion is constructed. We then introduce a novel ambiguity set propagation method, which theoretically admits the derivation of the ambiguity sets for obstacle positions over the entire prediction horizon by utilizing obstacle current positions and the ambiguity set for obstacle motion. Additionally, we develop a compression scheme with its safety guarantee to automatically adjust the complexity and granularity of the ambiguity set by aggregating basic ambiguity sets that are close in a measure space, thereby striking an attractive trade-off between control performance and computation time. Then the probabilistic collision-free trajectories are generated through distributionally robust optimization problems. The distributionally robust obstacle avoidance constraints based on the compressed ambiguity set are equivalently reformulated by deriving separating hyperplanes through tractable semi-definite programming. Finally, we establish the probabilistic collision avoidance guarantee and the long-term tracking performance guarantee for the proposed framework. The numerical simulations are used to demonstrate the efficacy and superiority of the proposed approach compared with state-of-the-art methods.
comment: 26 pages
♻ ☆ RoboPaint: From Human Demonstration to Any Robot and Any View
Acquiring large-scale, high-fidelity robot demonstration data remains a critical bottleneck for scaling Vision-Language-Action (VLA) models in dexterous manipulation. We propose a Real-Sim-Real data collection and data editing pipeline that transforms human demonstrations into robot-executable, environment-specific training data without direct robot teleoperation. Standardized data collection rooms are built to capture multimodal human demonstrations (synchronized 3 RGB-D videos, 11 RGB videos, 29-DoF glove joint angles, and 14-channel tactile signals). Based on these human demonstrations, we introduce a tactile-aware retargeting method that maps human hand states to robot dex-hand states via geometry and force-guided optimization. Then the retargeted robot trajectories are rendered in a photorealistic Isaac Sim environment to build robot training data. Real world experiments have demonstrated: (1) The retargeted dex-hand trajectories achieve an 84\% success rate across 10 diverse object manipulation tasks. (2) VLA policies (Pi0.5) trained exclusively on our generated data achieve 80\% average success rate on three representative tasks, i.e., pick-and-place, pushing and pouring. To conclude, robot training data can be efficiently "painted" from human demonstrations using our real-sim-real data pipeline. We offer a scalable, cost-effective alternative to teleoperation with minimal performance loss for complex dexterous manipulation.
comment: 17 pages
♻ ☆ CTBC: Contact-Triggered Blind Climbing for Wheeled Bipedal Robots with Instruction Learning and Reinforcement Learning
In recent years, wheeled bipedal robots have garnered significant attention due to their exceptional mobility on flat terrain. However, while stair climbing has been achieved in prior studies, these existing methods often suffer from a severe lack of versatility, making them difficult to adapt to varying hardware specifications or diverse complex terrains. To overcome these limitations, we propose a generalized Contact-Triggered Blind Climbing (CTBC) framework. Upon detecting wheel-obstacle contact, the framework triggers a leg-lifting motion integrated with a strongly-guided feedforward trajectory. This allows the robot to rapidly acquire agile climbing skills, significantly enhancing its capability to traverse unstructured environments. Distinct from previous approaches, CTBC demonstrates superior robustness and adaptability, having been validated across multiple wheeled bipedal platforms with different wheel radii and tire materials. Real-world experiments demonstrate that, relying solely on proprioceptive feedback, the proposed framework enables robots to achieve reliable and continuous climbing over obstacles well beyond their wheel radius.
♻ ☆ A Software-Only Post-Processor for Indexed Rotary Machining on GRBL-Based CNCs
Affordable desktop CNC routers are common in education, prototyping, and makerspaces, but most lack a rotary axis, limiting fabrication of rotationally symmetric or multi-sided parts. Existing solutions often require hardware retrofits, alternative controllers, or commercial CAM software, raising cost and complexity. This work presents a software-only framework for indexed rotary machining on GRBL-based CNCs. A custom post-processor converts planar toolpaths into discrete rotary steps, executed through a browser-based interface. While not equivalent to continuous 4-axis machining, the method enables practical rotary-axis fabrication using only standard, off-the-shelf mechanics, without firmware modification. By reducing technical and financial barriers, the framework expands access to multi-axis machining in classrooms, makerspaces, and small workshops, supporting hands-on learning and rapid prototyping.
comment: 13 pages, 10 figures, Python post-processor source code and web interface included
♻ ☆ GRAND: Guidance, Rebalancing, and Assignment for Networked Dispatch in Multi-Agent Path Finding
Large robot fleets are now common in warehouses and other logistics settings, where small control gains translate into large operational impacts. In this article, we address task scheduling for lifelong Multi-Agent Pickup-and-Delivery (MAPD) and propose a hybrid method that couples learning-based global guidance with lightweight optimization. A graph neural network policy trained via reinforcement learning outputs a desired distribution of free agents over an aggregated warehouse graph. This signal is converted into region-to-region rebalancing through a minimum-cost flow, and finalized by small, local assignment problems, preserving accuracy while keeping per-step latency within a 1 s compute budget. On congested warehouse benchmarks from the League of Robot Runners (LoRR) with up to 500 agents, our approach improves throughput by up to 10% over the 2024 winning scheduler while maintaining real-time execution. The results indicate that coupling graph-structured learned guidance with tractable solvers reduces congestion and yields a practical, scalable blueprint for high-throughput scheduling in large fleets.
♻ ☆ Efficient Policy Optimization in Robust Constrained MDPs with Iteration Complexity Guarantees
Constrained decision-making is essential for designing safe policies in real-world control systems, yet simulated environments often fail to capture real-world adversities. We consider the problem of learning a policy that will maximize the cumulative reward while satisfying a constraint, even when there is a mismatch between the real model and an accessible simulator/nominal model. In particular, we consider the robust constrained Markov decision problem (RCMDP) where an agent needs to maximize the reward and satisfy the constraint against the worst possible stochastic model under the uncertainty set centered around an unknown nominal model. Primal-dual methods, effective for standard constrained MDP (CMDP), are not applicable here because of the lack of the strong duality property. Further, one cannot apply the standard robust value-iteration based approach on the composite value function either as the worst case models may be different for the reward value function and the constraint value function. We propose a novel technique that effectively minimizes the constraint value function--to satisfy the constraints; on the other hand, when all the constraints are satisfied, it can simply maximize the robust reward value function. We prove that such an algorithm finds a policy with at most $ε$ sub-optimality and feasible policy after $O(ε^{-2})$ iterations. In contrast to the state-of-the-art method, we do not need to employ a binary search, thus, we reduce the computation time by at least 4x for smaller value of discount factor ($γ$) and by at least 6x for larger value of $γ$.
Robotics 22
☆ aerial-autonomy-stack -- a Faster-than-real-time, Autopilot-agnostic, ROS2 Framework to Simulate and Deploy Perception-based Drones
Unmanned aerial vehicles are rapidly transforming multiple applications, from agricultural and infrastructure monitoring to logistics and defense. Introducing greater autonomy to these systems can simultaneously make them more effective as well as reliable. Thus, the ability to rapidly engineer and deploy autonomous aerial systems has become of strategic importance. In the 2010s, a combination of high-performance compute, data, and open-source software led to the current deep learning and AI boom, unlocking decades of prior theoretical work. Robotics is on the cusp of a similar transformation. However, physical AI faces unique hurdles, often combined under the umbrella term "simulation-to-reality gap". These span from modeling shortcomings to the complexity of vertically integrating the highly heterogeneous hardware and software systems typically found in field robots. To address the latter, we introduce aerial-autonomy-stack, an open-source, end-to-end framework designed to streamline the pipeline from (GPU-accelerated) perception to (flight controller-based) action. Our stack allows the development of aerial autonomy using ROS2 and provides a common interface for two of the most popular autopilots: PX4 and ArduPilot. We show that it supports over 20x faster-than-real-time, end-to-end simulation of a complete development and deployment stack -- including edge compute and networking -- significantly compressing the build-test-release cycle of perception-based autonomy.
☆ Realistic Synthetic Household Data Generation at Scale AAAI 2026
Advancements in foundation models have catalyzed research in Embodied AI to develop interactive agents capable of environmental reasoning and interaction. Developing such agents requires diverse, large-scale datasets. Prior frameworks generate synthetic data for long-term human-robot interactions but fail to model the bidirectional influence between human behavior and household environments. Our proposed generative framework creates household datasets at scale through loosely coupled generation of long-term human-robot interactions and environments. Human personas influence environment generation, while environment schematics and semantics shape human-robot interactions. The generated 3D data includes rich static context such as object and environment semantics, and temporal context capturing human and agent behaviors over extended periods. Our flexible tool allows users to define dataset characteristics via natural language prompts, enabling configuration of environment and human activity data through natural language specifications. The tool creates variations of user-defined configurations, enabling scalable data generation. We validate our framework through statistical evaluation using multi-modal embeddings and key metrics: cosine similarity, mutual information gain, intervention analysis, and iterative improvement validation. Statistical comparisons show good alignment with real-world datasets (HOMER) with cosine similarity (0.60), while synthetic datasets (Wang et al.) show moderate alignment (0.27). Intervention analysis across age, organization, and sleep pattern changes shows statistically significant effects (p < 0.001) with large effect sizes (Cohen's d = 0.51-1.12), confirming bidirectional coupling translates persona traits into measurable environmental and behavioral differences. These contributions enable development and testing of household smart devices at scale.
comment: Accepted at Agentic AI Benchmarks and Applications for Enterprise Tasks workshop at AAAI 2026
☆ Cerebellar-Inspired Residual Control for Fault Recovery: From Inference-Time Adaptation to Structural Consolidation
Robotic policies deployed in real-world environments often encounter post-training faults, where retraining, exploration, or system identification are impractical. We introduce an inference-time, cerebellar-inspired residual control framework that augments a frozen reinforcement learning policy with online corrective actions, enabling fault recovery without modifying base policy parameters. The framework instantiates core cerebellar principles, including high-dimensional pattern separation via fixed feature expansion, parallel microzone-style residual pathways, and local error-driven plasticity with excitatory and inhibitory eligibility traces operating at distinct time scales. These mechanisms enable fast, localized correction under post-training disturbances while avoiding destabilizing global policy updates. A conservative, performance-driven meta-adaptation regulates residual authority and plasticity, preserving nominal behavior and suppressing unnecessary intervention. Experiments on MuJoCo benchmarks under actuator, dynamic, and environmental perturbations show improvements of up to $+66\%$ on \texttt{HalfCheetah-v5} and $+53\%$ on \texttt{Humanoid-v5} under moderate faults, with graceful degradation under severe shifts and complementary robustness from consolidating persistent residual corrections into policy parameters.
☆ Continuum Robot Localization using Distributed Time-of-Flight Sensors
Localization and mapping of an environment are crucial tasks for any robot operating in unstructured environments. Time-of-flight (ToF) sensors (e.g.,~lidar) have proven useful in mobile robotics, where high-resolution sensors can be used for simultaneous localization and mapping. In soft and continuum robotics, however, these high-resolution sensors are too large for practical use. This, combined with the deformable nature of such robots, has resulted in continuum robot (CR) localization and mapping in unstructured environments being a largely untouched area. In this work, we present a localization technique for CRs that relies on small, low-resolution ToF sensors distributed along the length of the robot. By fusing measurement information with a robot shape prior, we show that accurate localization is possible despite each sensor experiencing frequent degenerate scenarios. We achieve an average localization error of 2.5cm in position and 7.2° in rotation across all experimental conditions with a 53cm long robot. We demonstrate that the results are repeated across multiple environments, in both simulation and real-world experiments, and study robustness in the estimation to deviations in the prior map.
☆ A compliant ankle-actuated compass walker with triggering timing control
Passive dynamic walkers are widely adopted as a mathematical model to represent biped walking. The stable locomotion of these models is limited to tilted surfaces, requiring gravitational energy. Various techniques, such as actuation through the ankle and hip joints, have been proposed to extend the applicability of these models to level ground and rough terrain with improved locomotion efficiency. However, most of these techniques rely on impulsive energy injection schemes and torsional springs, which are quite challenging to implement in a physical platform. Here, a new model is proposed, named triggering controlled ankle actuated compass gait (TC-AACG), which allows non-instantaneous compliant ankle pushoff. The proposed technique can be implemented in physical platforms via series elastic actuators (SEAs). Our systematic examination shows that the proposed approach extends the locomotion capabilities of a biped model compared to impulsive ankle pushoff approach. We provide extensive simulation analysis investigating the locomotion speed, mechanical cost of transport, and basin of attraction of the proposed model.
comment: 6 figures, 6 pages
☆ DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos
Being able to simulate the outcomes of actions in varied environments will revolutionize the development of generalist agents at scale. However, modeling these world dynamics, especially for dexterous robotics tasks, poses significant challenges due to limited data coverage and scarce action labels. As an endeavor towards this end, we introduce DreamDojo, a foundation world model that learns diverse interactions and dexterous controls from 44k hours of egocentric human videos. Our data mixture represents the largest video dataset to date for world model pretraining, spanning a wide range of daily scenarios with diverse objects and skills. To address the scarcity of action labels, we introduce continuous latent actions as unified proxy actions, enhancing interaction knowledge transfer from unlabeled videos. After post-training on small-scale target robot data, DreamDojo demonstrates a strong understanding of physics and precise action controllability. We also devise a distillation pipeline that accelerates DreamDojo to a real-time speed of 10.81 FPS and further improves context consistency. Our work enables several important applications based on generative world models, including live teleoperation, policy evaluation, and model-based planning. Systematic evaluation on multiple challenging out-of-distribution (OOD) benchmarks verifies the significance of our method for simulating open-world, contact-rich tasks, paving the way for general-purpose robot world models.
comment: Project page: https://dreamdojo-world.github.io/
☆ Strategizing at Speed: A Learned Model Predictive Game for Multi-Agent Drone Racing
Autonomous drone racing pushes the boundaries of high-speed motion planning and multi-agent strategic decision-making. Success in this domain requires drones not only to navigate at their limits but also to anticipate and counteract competitors' actions. In this paper, we study a fundamental question that arises in this domain: how deeply should an agent strategize before taking an action? To this end, we compare two planning paradigms: the Model Predictive Game (MPG), which finds interaction-aware strategies at the expense of longer computation times, and contouring Model Predictive Control (MPC), which computes strategies rapidly but does not reason about interactions. We perform extensive experiments to study this trade-off, revealing that MPG outperforms MPC at moderate velocities but loses its advantage at higher speeds due to latency. To address this shortcoming, we propose a Learned Model Predictive Game (LMPG) approach that amortizes model predictive gameplay to reduce latency. In both simulation and hardware experiments, we benchmark our approach against MPG and MPC in head-to-head races, finding that LMPG outperforms both baselines.
☆ Consensus-based optimization (CBO): Towards Global Optimality in Robotics
Zero-order optimization has recently received significant attention for designing optimal trajectories and policies for robotic systems. However, most existing methods (e.g., MPPI, CEM, and CMA-ES) are local in nature, as they rely on gradient estimation. In this paper, we introduce consensus-based optimization (CBO) to robotics, which is guaranteed to converge to a global optimum under mild assumptions. We provide theoretical analysis and illustrative examples that give intuition into the fundamental differences between CBO and existing methods. To demonstrate the scalability of CBO for robotics problems, we consider three challenging trajectory optimization scenarios: (1) a long-horizon problem for a simple system, (2) a dynamic balance problem for a highly underactuated system, and (3) a high-dimensional problem with only a terminal cost. Our results show that CBO is able to achieve lower costs with respect to existing methods on all three challenging settings. This opens a new framework to study global trajectory optimization in robotics.
☆ SURE: Safe Uncertainty-Aware Robot-Environment Interaction using Trajectory Optimization
Robotic tasks involving contact interactions pose significant challenges for trajectory optimization due to discontinuous dynamics. Conventional formulations typically assume deterministic contact events, which limit robustness and adaptability in real-world settings. In this work, we propose SURE, a robust trajectory optimization framework that explicitly accounts for contact timing uncertainty. By allowing multiple trajectories to branch from possible pre-impact states and later rejoin a shared trajectory, SURE achieves both robustness and computational efficiency within a unified optimization framework. We evaluate SURE on two representative tasks with unknown impact times. In a cart-pole balancing task involving uncertain wall location, SURE achieves an average improvement of 21.6% in success rate when branch switching is enabled during control. In an egg-catching experiment using a robotic manipulator, SURE improves the success rate by 40%. These results demonstrate that SURE substantially enhances robustness compared to conventional nominal formulations.
☆ Perception-Control Coupled Visual Servoing for Textureless Objects Using Keypoint-Based EKF
Visual servoing is fundamental to robotic applications, enabling precise positioning and control. However, applying it to textureless objects remains a challenge due to the absence of reliable visual features. Moreover, adverse visual conditions, such as occlusions, often corrupt visual feedback, leading to reduced accuracy and instability in visual servoing. In this work, we build upon learning-based keypoint detection for textureless objects and propose a method that enhances robustness by tightly integrating perception and control in a closed loop. Specifically, we employ an Extended Kalman Filter (EKF) that integrates per-frame keypoint measurements to estimate 6D object pose, which drives pose-based visual servoing (PBVS) for control. The resulting camera motion, in turn, enhances the tracking of subsequent keypoints, effectively closing the perception-control loop. Additionally, unlike standard PBVS, we propose a probabilistic control law that computes both camera velocity and its associated uncertainty, enabling uncertainty-aware control for safe and reliable operation. We validate our approach on real-world robotic platforms using quantitative metrics and grasping experiments, demonstrating that our method outperforms traditional visual servoing techniques in both accuracy and practical application.
☆ DynaRetarget: Dynamically-Feasible Retargeting using Sampling-Based Trajectory Optimization
In this paper, we introduce DynaRetarget, a complete pipeline for retargeting human motions to humanoid control policies. The core component of DynaRetarget is a novel Sampling-Based Trajectory Optimization (SBTO) framework that refines imperfect kinematic trajectories into dynamically feasible motions. SBTO incrementally advances the optimization horizon, enabling optimization over the entire trajectory for long-horizon tasks. We validate DynaRetarget by successfully retargeting hundreds of humanoid-object demonstrations and achieving higher success rates than the state of the art. The framework also generalizes across varying object properties, such as mass, size, and geometry, using the same tracking objective. This ability to robustly retarget diverse demonstrations opens the door to generating large-scale synthetic datasets of humanoid loco-manipulation trajectories, addressing a major bottleneck in real-world data collection.
☆ A 26-Gram Butterfly-Inspired Robot Achieving Autonomous Tailless Flight
Flapping-wing micro air vehicles (FWMAVs) have demonstrated remarkable bio-inspired agility, yet tailless two-winged configurations remain largely unexplored due to their complex fluid-structure and wing-body coupling. Here we present \textit{AirPulse}, a 26-gram butterfly-inspired FWMAV that achieves fully onboard, closed-loop, untethered flight without auxiliary control surfaces. The AirPulse robot replicates key biomechanical traits of butterfly flight, including low wing aspect ratio, compliant carbon-fiber-reinforced wings, and low-frequency, high-amplitude flapping that induces cyclic variations in the center of gravity and moment of inertia, producing characteristic body undulation. We establish a quantitative mapping between flapping modulation parameters and force-torque generation, and introduce the Stroke Timing Asymmetry Rhythm (STAR) generator, enabling smooth, stable, and linearly parameterized wingstroke asymmetry for flapping control. Integrating these with an attitude controller, the AirPulse robot maintains pitch and yaw stability despite strong oscillatory dynamics. Free-flight experiments demonstrate stable climbing and turning maneuvers via either angle offset or stroke timing modulation, marking the first onboard controlled flight of the lightest two-winged, tailless butterfly-inspired FWMAV reported in peer-reviewed literature. This work corroborates a foundational platform for lightweight, collision-proof FWMAVs, bridging biological inspiration with practical aerial robotics. Their non-invasive maneuverability is ideally suited for real-world applications, such as confined-space inspection and ecological monitoring, inaccessible to traditional drones, while their biomechanical fidelity provides a physical model to decode the principles underlying the erratic yet efficient flight of real butterflies.
☆ SuReNav: Superpixel Graph-based Constraint Relaxation for Navigation in Over-constrained Environments ICRA 2026
We address the over-constrained planning problem in semi-static environments. The planning objective is to find a best-effort solution that avoids all hard constraint regions while minimally traversing the least risky areas. Conventional methods often rely on pre-defined area costs, limiting generalizations. Further, the spatial continuity of navigation spaces makes it difficult to identify regions that are passable without overestimation. To overcome these challenges, we propose SuReNav, a superpixel graph-based constraint relaxation and navigation method that imitates human-like safe and efficient navigation. Our framework consists of three components: 1) superpixel graph map generation with regional constraints, 2) regional-constraint relaxation using graph neural network trained on human demonstrations for safe and efficient navigation, and 3) interleaving relaxation, planning, and execution for complete navigation. We evaluate our method against state-of-the-art baselines on 2D semantic maps and 3D maps from OpenStreetMap, achieving the highest human-likeness score of complete navigation while maintaining a balanced trade-off between efficiency and safety. We finally demonstrate its scalability and generalization performance in real-world urban navigation with a quadruped robot, Spot.
comment: Accepted by ICRA 2026. Code and videos are available at https://sure-nav.github.io/
☆ Zero-Shot UAV Navigation in Forests via Relightable 3D Gaussian Splatting
UAV navigation in unstructured outdoor environments using passive monocular vision is hindered by the substantial visual domain gap between simulation and reality. While 3D Gaussian Splatting enables photorealistic scene reconstruction from real-world data, existing methods inherently couple static lighting with geometry, severely limiting policy generalization to dynamic real-world illumination. In this paper, we propose a novel end-to-end reinforcement learning framework designed for effective zero-shot transfer to unstructured outdoors. Within a high-fidelity simulation grounded in real-world data, our policy is trained to map raw monocular RGB observations directly to continuous control commands. To overcome photometric limitations, we introduce Relightable 3D Gaussian Splatting, which decomposes scene components to enable explicit, physically grounded editing of environmental lighting within the neural representation. By augmenting training with diverse synthesized lighting conditions ranging from strong directional sunlight to diffuse overcast skies, we compel the policy to learn robust, illumination-invariant visual features. Extensive real-world experiments demonstrate that a lightweight quadrotor achieves robust, collision-free navigation in complex forest environments at speeds up to 10 m/s, exhibiting significant resilience to drastic lighting variations without fine-tuning.
comment: 12 pages, 8 figures
♻ ☆ Information-Theoretic Graph Fusion with Vision-Language-Action Model for Policy Reasoning and Dual Robotic Control
Teaching robots dexterous skills from human videos remains challenging due to the reliance on low-level trajectory imitation, which fails to generalize across object types, spatial layouts, and manipulator configurations. We propose Graph-Fused Vision-Language-Action (GF-VLA), a framework that enables dual-arm robotic systems to perform task-level reasoning and execution directly from RGB and Depth human demonstrations. GF-VLA first extracts Shannon-information-based cues to identify hands and objects with the highest task relevance, then encodes these cues into temporally ordered scene graphs that capture both hand-object and object-object interactions. These graphs are fused with a language-conditioned transformer that generates hierarchical behavior trees and interpretable Cartesian motion commands. To improve execution efficiency in bimanual settings, we further introduce a cross-hand selection policy that infers optimal gripper assignment without explicit geometric reasoning. We evaluate GF-VLA on four structured dual-arm block assembly tasks involving symbolic shape construction and spatial generalization. Experimental results show that the information-theoretic scene representation achieves over 95 percent graph accuracy and 93 percent subtask segmentation, supporting the LLM planner in generating reliable and human-readable task policies. When executed by the dual-arm robot, these policies yield 94 percent grasp success, 89 percent placement accuracy, and 90 percent overall task success across stacking, letter-building, and geometric reconfiguration scenarios, demonstrating strong generalization and robustness across diverse spatial and semantic variations.
comment: Journal accepted by Information Fusion
♻ ☆ Lan-grasp: Using Large Language Models for Semantic Object Grasping and Placement
In this paper, we propose Lan-grasp, a novel approach towards more appropriate semantic grasping and placing. We leverage foundation models to equip the robot with a semantic understanding of object geometry, enabling it to identify the right place to grasp, which parts to avoid, and the natural pose for placement. This is an important contribution to grasping and utilizing objects in a more meaningful and safe manner. We leverage a combination of a Large Language Model, a Vision-Language Model, and a traditional grasp planner to generate grasps that demonstrate a deeper semantic understanding of the objects. Building on foundation models provides us with a zero-shot grasp method that can handle a wide range of objects without requiring further training or fine-tuning. We also propose a method for safely putting down a grasped object. The core idea is to rotate the object upright utilizing a pretrained generative model and the reasoning capabilities of a VLM. We evaluate our method in real-world experiments on a custom object dataset and present the results of a survey that asks participants to choose an object part appropriate for grasping. The results show that the grasps generated by our method are consistently ranked higher by the participants than those generated by a conventional grasping planner and a recent semantic grasping approach. In addition, we propose a Visual Chain-of-Thought feedback loop to assess grasp feasibility in complex scenarios. This mechanism enables dynamic reasoning and generates alternative grasp strategies when needed, ensuring safer and more effective grasping outcomes.
♻ ☆ 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
♻ ☆ Embodying Physical Computing into Soft Robots
Softening and onboarding computers and controllers is one of the final frontiers in soft robotics towards their robustness and intelligence for everyday use. In this regard, embodying soft and physical computing presents exciting potential. Physical computing seeks to encode inputs into a mechanical computing kernel and leverage the internal interactions among this kernel's constituent elements to compute the output. Moreover, such input-to-output evolution can be re-programmable. This perspective paper proposes a framework for embodying physical computing into soft robots and discusses three unique strategies in the literature: analog oscillators, physical reservoir computing, and physical algorithmic computing. These embodied computers enable the soft robot to perform complex behaviors that would otherwise require CMOS-based electronics -- including coordinated locomotion with obstacle avoidance, payload weight and orientation classification, and programmable operation based on logical rules. This paper will detail the working principles of these embodied physical computing methods, survey the current state-of-the-art, and present a perspective for future development.
♻ ☆ Constrained Group Relative Policy Optimization
While Group Relative Policy Optimization (GRPO) has emerged as a scalable framework for critic-free policy learning, extending it to settings with explicit behavioral constraints remains underexplored. We introduce Constrained GRPO, a Lagrangian-based extension of GRPO for constrained policy optimization. Constraints are specified via indicator cost functions, enabling direct optimization of violation rates through a Lagrangian relaxation. We show that a naive multi-component treatment in advantage estimation can break constrained learning: mismatched component-wise standard deviations distort the relative importance of the different objective terms, which in turn corrupts the Lagrangian signal and prevents meaningful constraint enforcement. We formally derive this effect to motivate our scalarized advantage construction that preserves the intended trade-off between reward and constraint terms. Experiments in a toy gridworld confirm the predicted optimization pathology and demonstrate that scalarizing advantages restores stable constraint control. In addition, we evaluate Constrained GRPO on robotics tasks, where it improves constraint satisfaction while increasing task success, establishing a simple and effective recipe for constrained policy optimization in embodied AI domains that increasingly rely on large multimodal foundation models.
comment: 16 pages, 6 figures
♻ ☆ REACT: Real-time Entanglement-Aware Coverage Path Planning for Tethered Underwater Vehicles
Inspection of underwater structures with tethered underwater vehicles is often hindered by the risk of tether entanglement. We propose REACT (real-time entanglement-aware coverage path planning for tethered underwater vehicles), a framework designed to overcome this limitation. REACT comprises a computationally efficient geometry-based tether model using the signed distance field (SDF) map for accurate, real-time simulation of taut tether configurations around arbitrary structures in 3D. This model enables an efficient online replanning strategy by enforcing a maximum tether length constraint, thereby actively preventing entanglement. By integrating REACT into a coverage path planning framework, we achieve safe and entanglement-free inspection paths, previously challenging due to tether constraints. The complete REACT framework's efficacy is validated in a pipe inspection scenario, demonstrating safe navigation and full coverage inspection. Simulation results show that REACT achieves complete coverage while maintaining tether constraints and completing the total mission 20% faster than conventional planners, despite a longer inspection time due to proactive avoidance of entanglement that eliminates extensive post-mission disentanglement. Real-world experiments confirm these benefits, where REACT completes the full mission, while the baseline planner fails due to physical tether entanglement.
comment: Accepted for publication at International Conference on Robotics & Automation 2026
♻ ☆ DarkEQA: Benchmarking Vision-Language Models for Embodied Question Answering in Low-Light Indoor Environments
Vision Language Models (VLMs) are increasingly adopted as central reasoning modules for embodied agents. Existing benchmarks evaluate their capabilities under ideal, well-lit conditions, yet robust 24/7 operation demands performance under a wide range of visual degradations, including low-light conditions at night or in dark environments--a core necessity that has been largely overlooked. To address this underexplored challenge, we present DarkEQA, an open-source benchmark for evaluating EQA-relevant perceptual primitives under multi-level low-light conditions. DarkEQA isolates the perception bottleneck by evaluating question answering from egocentric observations under controlled degradations, enabling attributable robustness analysis. A key design feature of DarkEQA is its physical fidelity: visual degradations are modeled in linear RAW space, simulating physics-based illumination drop and sensor noise followed by an ISP-inspired rendering pipeline. We demonstrate the utility of DarkEQA by evaluating a wide range of state-of-the-art VLMs and Low-Light Image Enhancement (LLIE) models. Our analysis systematically reveals VLMs' limitations when operating under these challenging visual conditions. Project website: https://darkeqa-benchmark.github.io/
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ CRISP -- Compliant ROS2 Controllers for Learning-Based Manipulation Policies and Teleoperation
Learning-based controllers, such as diffusion policies and vision-language action models, often generate low-frequency or discontinuous robot state changes. Achieving smooth reference tracking requires a low-level controller that converts high-level targets commands into joint torques, enabling compliant behavior during contact interactions. We present CRISP, a lightweight C++ implementation of compliant Cartesian and joint-space controllers for the ROS2 control standard, designed for seamless integration with high-level learning-based policies as well as teleoperation. The controllers are compatible with any manipulator that exposes a joint-torque interface. Through our Python and Gymnasium interfaces, CRISP provides a unified pipeline for recording data from hardware and simulation and deploying high-level learning-based policies seamlessly, facilitating rapid experimentation. The system has been validated on hardware with the Franka Robotics FR3 and in simulation with the Kuka IIWA14 and Kinova Gen3. Designed for rapid integration, flexible deployment, and real-time performance, our implementation provides a unified pipeline for data collection and policy execution, lowering the barrier to applying learning-based methods on ROS2-compatible manipulators. Detailed documentation is available at the project website - https://utiasDSL.github.io/crisp_controllers.
comment: 5 pages, 5 figures
Robotics 49
☆ CommCP: Efficient Multi-Agent Coordination via LLM-Based Communication with Conformal Prediction ICRA 2026
To complete assignments provided by humans in natural language, robots must interpret commands, generate and answer relevant questions for scene understanding, and manipulate target objects. Real-world deployments often require multiple heterogeneous robots with different manipulation capabilities to handle different assignments cooperatively. Beyond the need for specialized manipulation skills, effective information gathering is important in completing these assignments. To address this component of the problem, we formalize the information-gathering process in a fully cooperative setting as an underexplored multi-agent multi-task Embodied Question Answering (MM-EQA) problem, which is a novel extension of canonical Embodied Question Answering (EQA), where effective communication is crucial for coordinating efforts without redundancy. To address this problem, we propose CommCP, a novel LLM-based decentralized communication framework designed for MM-EQA. Our framework employs conformal prediction to calibrate the generated messages, thereby minimizing receiver distractions and enhancing communication reliability. To evaluate our framework, we introduce an MM-EQA benchmark featuring diverse, photo-realistic household scenarios with embodied questions. Experimental results demonstrate that CommCP significantly enhances the task success rate and exploration efficiency over baselines. The experiment videos, code, and dataset are available on our project website: https://comm-cp.github.io.
comment: IEEE International Conference on Robotics and Automation (ICRA 2026); Project Website: https://comm-cp.github.io/
☆ InterPrior: Scaling Generative Control for Physics-Based Human-Object Interactions
Humans rarely plan whole-body interactions with objects at the level of explicit whole-body movements. High-level intentions, such as affordance, define the goal, while coordinated balance, contact, and manipulation can emerge naturally from underlying physical and motor priors. Scaling such priors is key to enabling humanoids to compose and generalize loco-manipulation skills across diverse contexts while maintaining physically coherent whole-body coordination. To this end, we introduce InterPrior, a scalable framework that learns a unified generative controller through large-scale imitation pretraining and post-training by reinforcement learning. InterPrior first distills a full-reference imitation expert into a versatile, goal-conditioned variational policy that reconstructs motion from multimodal observations and high-level intent. While the distilled policy reconstructs training behaviors, it does not generalize reliably due to the vast configuration space of large-scale human-object interactions. To address this, we apply data augmentation with physical perturbations, and then perform reinforcement learning finetuning to improve competence on unseen goals and initializations. Together, these steps consolidate the reconstructed latent skills into a valid manifold, yielding a motion prior that generalizes beyond the training data, e.g., it can incorporate new behaviors such as interactions with unseen objects. We further demonstrate its effectiveness for user-interactive control and its potential for real robot deployment.
comment: Webpage: https://sirui-xu.github.io/InterPrior/
☆ Learning Event-Based Shooter Models from Virtual Reality Experiments
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: Preprint under review for conference publication. 9 pages, 4 figures, 4 tables
☆ Visuo-Tactile World Models
We introduce multi-task Visuo-Tactile World Models (VT-WM), which capture the physics of contact through touch reasoning. By complementing vision with tactile sensing, VT-WM better understands robot-object interactions in contact-rich tasks, avoiding common failure modes of vision-only models under occlusion or ambiguous contact states, such as objects disappearing, teleporting, or moving in ways that violate basic physics. Trained across a set of contact-rich manipulation tasks, VT-WM improves physical fidelity in imagination, achieving 33% better performance at maintaining object permanence and 29% better compliance with the laws of motion in autoregressive rollouts. Moreover, experiments show that grounding in contact dynamics also translates to planning. In zero-shot real-robot experiments, VT-WM achieves up to 35% higher success rates, with the largest gains in multi-step, contact-rich tasks. Finally, VT-WM demonstrates significant downstream versatility, effectively adapting its learned contact dynamics to a novel task and achieving reliable planning success with only a limited set of demonstrations.
comment: Preprint
☆ Location-Aware Dispersion on Anonymous Graphs
The well-studied DISPERSION problem is a fundamental coordination problem in distributed robotics, where a set of mobile robots must relocate so that each occupies a distinct node of a network. DISPERSION assumes that a robot can settle at any node as long as no other robot settles on that node. In this work, we introduce LOCATION-AWARE DISPERSION, a novel generalization of DISPERSION that incorporates location awareness: Let $G = (V, E)$ be an anonymous, connected, undirected graph with $n = |V|$ nodes, each labeled with a color $\sf{col}(v) \in C = \{c_1, \dots, c_t\}, t\leq n$. A set $R = \{r_1, \dots, r_k\}$ of $k \leq n$ mobile robots is given, where each robot $r_i$ has an associated color $\mathsf{col}(r_i) \in C$. Initially placed arbitrarily on the graph, the goal is to relocate the robots so that each occupies a distinct node of the same color. When $|C|=1$, LOCATION-AWARE DISPERSION reduces to DISPERSION. There is a solution to DISPERSION in graphs with any $k\leq n$ without knowing $k,n$. Like DISPERSION, the goal is to solve LOCATION-AWARE DISPERSION minimizing both time and memory requirement at each agent. We develop several deterministic algorithms with guaranteed bounds on both time and memory requirement. We also give an impossibility and a lower bound for any deterministic algorithm for LOCATION-AWARE DISPERSION. To the best of our knowledge, the presented results collectively establish the algorithmic feasibility of LOCATION-AWARE DISPERSION in anonymous networks and also highlight the challenges on getting an efficient solution compared to the solutions for DISPERSION.
comment: 3 tables, 2 figures, 6 pseudo-codes
☆ From Bench to Flight: Translating Drone Impact Tests into Operational Safety Limits
Indoor micro-aerial vehicles (MAVs) are increasingly used for tasks that require close proximity to people, yet practitioners lack practical methods to tune motion limits based on measured impact risk. We present an end-to-end, open toolchain that converts benchtop impact tests into deployable safety governors for drones. First, we describe a compact and replicable impact rig and protocol for capturing force-time profiles across drone classes and contact surfaces. Second, we provide data-driven models that map pre-impact speed to impulse and contact duration, enabling direct computation of speed bounds for a target force limit. Third, we release scripts and a ROS2 node that enforce these bounds online and log compliance, with support for facility-specific policies. We validate the workflow on multiple commercial off-the-shelf quadrotors and representative indoor assets, demonstrating that the derived governors preserve task throughput while meeting force constraints specified by safety stakeholders. Our contribution is a practical bridge from measured impacts to runtime limits, with shareable datasets, code, and a repeatable process that teams can adopt to certify indoor MAV operations near humans.
☆ Residual Reinforcement Learning for Waste-Container Lifting Using Large-Scale Cranes with Underactuated Tools
This paper studies the container lifting phase of a waste-container recycling task in urban environments, performed by a hydraulic loader crane equipped with an underactuated discharge unit, and proposes a residual reinforcement learning (RRL) approach that combines a nominal Cartesian controller with a learned residual policy. All experiments are conducted in simulation, where the task is characterized by tight geometric tolerances between the discharge-unit hooks and the container rings relative to the overall crane scale, making precise trajectory tracking and swing suppression essential. The nominal controller uses admittance control for trajectory tracking and pendulum-aware swing damping, followed by damped least-squares inverse kinematics with a nullspace posture term to generate joint velocity commands. A PPO-trained residual policy in Isaac Lab compensates for unmodeled dynamics and parameter variations, improving precision and robustness without requiring end-to-end learning from scratch. We further employ randomized episode initialization and domain randomization over payload properties, actuator gains, and passive joint parameters to enhance generalization. Simulation results demonstrate improved tracking accuracy, reduced oscillations, and higher lifting success rates compared to the nominal controller alone.
comment: 12 pages
☆ Constrained Group Relative Policy Optimization
While Group Relative Policy Optimization (GRPO) has emerged as a scalable framework for critic-free policy learning, extending it to settings with explicit behavioral constraints remains underexplored. We introduce Constrained GRPO, a Lagrangian-based extension of GRPO for constrained policy optimization. Constraints are specified via indicator cost functions, enabling direct optimization of violation rates through a Lagrangian relaxation. We show that a naive multi-component treatment in advantage estimation can break constrained learning: mismatched component-wise standard deviations distort the relative importance of the different objective terms, which in turn corrupts the Lagrangian signal and prevents meaningful constraint enforcement. We formally derive this effect to motivate our scalarized advantage construction that preserves the intended trade-off between reward and constraint terms. Experiments in a toy gridworld confirm the predicted optimization pathology and demonstrate that scalarizing advantages restores stable constraint control. In addition, we evaluate Constrained GRPO on robotics tasks, where it improves constraint satisfaction while increasing task success, establishing a simple and effective recipe for constrained policy optimization in embodied AI domains that increasingly rely on large multimodal foundation models.
comment: 16 pages, 6 figures
☆ A Hybrid Autoencoder for Robust Heightmap Generation from Fused Lidar and Depth Data for Humanoid Robot Locomotion
Reliable terrain perception is a critical prerequisite for the deployment of humanoid robots in unstructured, human-centric environments. While traditional systems often rely on manually engineered, single-sensor pipelines, this paper presents a learning-based framework that uses an intermediate, robot-centric heightmap representation. A hybrid Encoder-Decoder Structure (EDS) is introduced, utilizing a Convolutional Neural Network (CNN) for spatial feature extraction fused with a Gated Recurrent Unit (GRU) core for temporal consistency. The architecture integrates multimodal data from an Intel RealSense depth camera, a LIVOX MID-360 LiDAR processed via efficient spherical projection, and an onboard IMU. Quantitative results demonstrate that multimodal fusion improves reconstruction accuracy by 7.2% over depth-only and 9.9% over LiDAR-only configurations. Furthermore, the integration of a 3.2 s temporal context reduces mapping drift.
☆ Sparse Video Generation Propels Real-World Beyond-the-View Vision-Language Navigation
Why must vision-language navigation be bound to detailed and verbose language instructions? While such details ease decision-making, they fundamentally contradict the goal for navigation in the real-world. Ideally, agents should possess the autonomy to navigate in unknown environments guided solely by simple and high-level intents. Realizing this ambition introduces a formidable challenge: Beyond-the-View Navigation (BVN), where agents must locate distant, unseen targets without dense and step-by-step guidance. Existing large language model (LLM)-based methods, though adept at following dense instructions, often suffer from short-sighted behaviors due to their reliance on short-horimzon supervision. Simply extending the supervision horizon, however, destabilizes LLM training. In this work, we identify that video generation models inherently benefit from long-horizon supervision to align with language instructions, rendering them uniquely suitable for BVN tasks. Capitalizing on this insight, we propose introducing the video generation model into this field for the first time. Yet, the prohibitive latency for generating videos spanning tens of seconds makes real-world deployment impractical. To bridge this gap, we propose SparseVideoNav, achieving sub-second trajectory inference guided by a generated sparse future spanning a 20-second horizon. This yields a remarkable 27x speed-up compared to the unoptimized counterpart. Extensive real-world zero-shot experiments demonstrate that SparseVideoNav achieves 2.5x the success rate of state-of-the-art LLM baselines on BVN tasks and marks the first realization of such capability in challenging night scenes.
☆ Scalable and General Whole-Body Control for Cross-Humanoid Locomotion
Learning-based whole-body controllers have become a key driver for humanoid robots, yet most existing approaches require robot-specific training. In this paper, we study the problem of cross-embodiment humanoid control and show that a single policy can robustly generalize across a wide range of humanoid robot designs with one-time training. We introduce XHugWBC, a novel cross-embodiment training framework that enables generalist humanoid control through: (1) physics-consistent morphological randomization, (2) semantically aligned observation and action spaces across diverse humanoid robots, and (3) effective policy architectures modeling morphological and dynamical properties. XHugWBC is not tied to any specific robot. Instead, it internalizes a broad distribution of morphological and dynamical characteristics during training. By learning motion priors from diverse randomized embodiments, the policy acquires a strong structural bias that supports zero-shot transfer to previously unseen robots. Experiments on twelve simulated humanoids and seven real-world robots demonstrate the strong generalization and robustness of the resulting universal controller.
☆ 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
☆ From Vision to Decision: Neuromorphic Control for Autonomous Navigation and Tracking
Robotic navigation has historically struggled to reconcile reactive, sensor-based control with the decisive capabilities of model-based planners. This duality becomes critical when the absence of a predominant option among goals leads to indecision, challenging reactive systems to break symmetries without computationally-intense planners. We propose a parsimonious neuromorphic control framework that bridges this gap for vision-guided navigation and tracking. Image pixels from an onboard camera are encoded as inputs to dynamic neuronal populations that directly transform visual target excitation into egocentric motion commands. A dynamic bifurcation mechanism resolves indecision by delaying commitment until a critical point induced by the environmental geometry. Inspired by recently proposed mechanistic models of animal cognition and opinion dynamics, the neuromorphic controller provides real-time autonomy with a minimal computational burden, a small number of interpretable parameters, and can be seamlessly integrated with application-specific image processing pipelines. We validate our approach in simulation environments as well as on an experimental quadrotor platform.
☆ HiCrowd: Hierarchical Crowd Flow Alignment for Dense Human Environments ICRA
Navigating through dense human crowds remains a significant challenge for mobile robots. A key issue is the freezing robot problem, where the robot struggles to find safe motions and becomes stuck within the crowd. To address this, we propose HiCrowd, a hierarchical framework that integrates reinforcement learning (RL) with model predictive control (MPC). HiCrowd leverages surrounding pedestrian motion as guidance, enabling the robot to align with compatible crowd flows. A high-level RL policy generates a follow point to align the robot with a suitable pedestrian group, while a low-level MPC safely tracks this guidance with short horizon planning. The method combines long-term crowd aware decision making with safe short-term execution. We evaluate HiCrowd against reactive and learning-based baselines in offline setting (replaying recorded human trajectories) and online setting (human trajectories are updated to react to the robot in simulation). Experiments on a real-world dataset and a synthetic crowd dataset show that our method outperforms in navigation efficiency and safety, while reducing freezing behaviors. Our results suggest that leveraging human motion as guidance, rather than treating humans solely as dynamic obstacles, provides a powerful principle for safe and efficient robot navigation in crowds.
comment: Accepted to the 2026 IEEE International Conference on Robotics and Automation (ICRA)
☆ TOLEBI: Learning Fault-Tolerant Bipedal Locomotion via Online Status Estimation and Fallibility Rewards ICRA
With the growing employment of learning algorithms in robotic applications, research on reinforcement learning for bipedal locomotion has become a central topic for humanoid robotics. While recently published contributions achieve high success rates in locomotion tasks, scarce attention has been devoted to the development of methods that enable to handle hardware faults that may occur during the locomotion process. However, in real-world settings, environmental disturbances or sudden occurrences of hardware faults might yield severe consequences. To address these issues, this paper presents TOLEBI (A faulT-tOlerant Learning framEwork for Bipedal locomotIon) that handles faults on the robot during operation. Specifically, joint locking, power loss and external disturbances are injected in simulation to learn fault-tolerant locomotion strategies. In addition to transferring the learned policy to the real robot via sim-to-real transfer, an online joint status module incorporated. This module enables to classify joint conditions by referring to the actual observations at runtime under real-world conditions. The validation experiments conducted both in real-world and simulation with the humanoid robot TOCABI highlight the applicability of the proposed approach. To our knowledge, this manuscript provides the first learning-based fault-tolerant framework for bipedal locomotion, thereby fostering the development of efficient learning methods in this field.
comment: Accepted for Publication at IEEE International Conference on Robotics and Automation (ICRA) 2026
☆ PIRATR: Parametric Object Inference for Robotic Applications with Transformers in 3D Point Clouds ICRA
We present PIRATR, an end-to-end 3D object detection framework for robotic use cases in point clouds. Extending PI3DETR, our method streamlines parametric 3D object detection by jointly estimating multi-class 6-DoF poses and class-specific parametric attributes directly from occlusion-affected point cloud data. This formulation enables not only geometric localization but also the estimation of task-relevant properties for parametric objects, such as a gripper's opening, where the 3D model is adjusted according to simple, predefined rules. The architecture employs modular, class-specific heads, making it straightforward to extend to novel object types without re-designing the pipeline. We validate PIRATR on an automated forklift platform, focusing on three structurally and functionally diverse categories: crane grippers, loading platforms, and pallets. Trained entirely in a synthetic environment, PIRATR generalizes effectively to real outdoor LiDAR scans, achieving a detection mAP of 0.919 without additional fine-tuning. PIRATR establishes a new paradigm of pose-aware, parameterized perception. This bridges the gap between low-level geometric reasoning and actionable world models, paving the way for scalable, simulation-trained perception systems that can be deployed in dynamic robotic environments. Code available at https://github.com/swingaxe/piratr.
comment: 8 Pages, 11 Figures, Accepted at 2026 IEEE International Conference on Robotics & Automation (ICRA) Vienna
☆ IndustryShapes: An RGB-D Benchmark dataset for 6D object pose estimation of industrial assembly components and tools ICRA 2026
We introduce IndustryShapes, a new RGB-D benchmark dataset of industrial tools and components, designed for both instance-level and novel object 6D pose estimation approaches. The dataset provides a realistic and application-relevant testbed for benchmarking these methods in the context of industrial robotics bridging the gap between lab-based research and deployment in real-world manufacturing scenarios. Unlike many previous datasets that focus on household or consumer products or use synthetic, clean tabletop datasets, or objects captured solely in controlled lab environments, IndustryShapes introduces five new object types with challenging properties, also captured in realistic industrial assembly settings. The dataset has diverse complexity, from simple to more challenging scenes, with single and multiple objects, including scenes with multiple instances of the same object and it is organized in two parts: the classic set and the extended set. The classic set includes a total of 4,6k images and 6k annotated poses. The extended set introduces additional data modalities to support the evaluation of model-free and sequence-based approaches. To the best of our knowledge, IndustryShapes is the first dataset to offer RGB-D static onboarding sequences. We further evaluate the dataset on a representative set of state-of-the art methods for instance-based and novel object 6D pose estimation, including also object detection, segmentation, showing that there is room for improvement in this domain. The dataset page can be found in https://pose-lab.github.io/IndustryShapes.
comment: To appear in ICRA 2026
☆ VLN-Pilot: Large Vision-Language Model as an Autonomous Indoor Drone Operator
This paper introduces VLN-Pilot, a novel framework in which a large Vision-and-Language Model (VLLM) assumes the role of a human pilot for indoor drone navigation. By leveraging the multimodal reasoning abilities of VLLMs, VLN-Pilot interprets free-form natural language instructions and grounds them in visual observations to plan and execute drone trajectories in GPS-denied indoor environments. Unlike traditional rule-based or geometric path-planning approaches, our framework integrates language-driven semantic understanding with visual perception, enabling context-aware, high-level flight behaviors with minimal task-specific engineering. VLN-Pilot supports fully autonomous instruction-following for drones by reasoning about spatial relationships, obstacle avoidance, and dynamic reactivity to unforeseen events. We validate our framework on a custom photorealistic indoor simulation benchmark and demonstrate the ability of the VLLM-driven agent to achieve high success rates on complex instruction-following tasks, including long-horizon navigation with multiple semantic targets. Experimental results highlight the promise of replacing remote drone pilots with a language-guided autonomous agent, opening avenues for scalable, human-friendly control of indoor UAVs in tasks such as inspection, search-and-rescue, and facility monitoring. Our results suggest that VLLM-based pilots may dramatically reduce operator workload while improving safety and mission flexibility in constrained indoor environments.
☆ Virtual-Tube-Based Cooperative Transport Control for Multi-UAV Systems in Constrained Environments
This paper proposes a novel control framework for cooperative transportation of cable-suspended loads by multiple unmanned aerial vehicles (UAVs) operating in constrained environments. Leveraging virtual tube theory and principles from dissipative systems theory, the framework facilitates efficient multi-UAV collaboration for navigating obstacle-rich areas. The proposed framework offers several key advantages. (1) It achieves tension distribution and coordinated transportation within the UAV-cable-load system with low computational overhead, dynamically adapting UAV configurations based on obstacle layouts to facilitate efficient navigation. (2) By integrating dissipative systems theory, the framework ensures high stability and robustness, essential for complex multi-UAV operations. The effectiveness of the proposed approach is validated through extensive simulations, demonstrating its scalability for large-scale multi-UAV systems. Furthermore, the method is experimentally validated in outdoor scenarios, showcasing its practical feasibility and robustness under real-world conditions.
comment: 10 pages, 8 figures
☆ DECO: Decoupled Multimodal Diffusion Transformer for Bimanual Dexterous Manipulation with a Plugin Tactile Adapter
Overview of the Proposed DECO Framework.} DECO is a DiT-based policy that decouples multimodal conditioning. Image and action tokens interact via joint self attention, while proprioceptive states and optional conditions are injected through adaptive layer normalization. Tactile signals are injected via cross attention, while a lightweight LoRA-based adapter is used to efficiently fine-tune the pretrained policy. DECO is also accompanied by DECO-50, a bimanual dexterous manipulation dataset with tactile sensing, consisting of 4 scenarios and 28 sub-tasks, covering more than 50 hours of data, approximately 5 million frames, and 8,000 successful trajectories.
comment: 17 pages, 8 figures
☆ TaSA: Two-Phased Deep Predictive Learning of Tactile Sensory Attenuation for Improving In-Grasp Manipulation ICRA2026
Humans can achieve diverse in-hand manipulations, such as object pinching and tool use, which often involve simultaneous contact between the object and multiple fingers. This is still an open issue for robotic hands because such dexterous manipulation requires distinguishing between tactile sensations generated by their self-contact and those arising from external contact. Otherwise, object/robot breakage happens due to contacts/collisions. Indeed, most approaches ignore self-contact altogether, by constraining motion to avoid/ignore self-tactile information during contact. While this reduces complexity, it also limits generalization to real-world scenarios where self-contact is inevitable. Humans overcome this challenge through self-touch perception, using predictive mechanisms that anticipate the tactile consequences of their own motion, through a principle called sensory attenuation, where the nervous system differentiates predictable self-touch signals, allowing novel object stimuli to stand out as relevant. Deriving from this, we introduce TaSA, a two-phased deep predictive learning framework. In the first phase, TaSA explicitly learns self-touch dynamics, modeling how a robot's own actions generate tactile feedback. In the second phase, this learned model is incorporated into the motion learning phase, to emphasize object contact signals during manipulation. We evaluate TaSA on a set of insertion tasks, which demand fine tactile discrimination: inserting a pencil lead into a mechanical pencil, inserting coins into a slot, and fixing a paper clip onto a sheet of paper, with various orientations, positions, and sizes. Across all tasks, policies trained with TaSA achieve significantly higher success rates than baseline methods, demonstrating that structured tactile perception with self-touch based on sensory attenuation is critical for dexterous robotic manipulation.
comment: 8 pages, 8 figures, 8 tables, ICRA2026 accepted
☆ MerNav: A Highly Generalizable Memory-Execute-Review Framework for Zero-Shot Object Goal Navigation
Visual Language Navigation (VLN) is one of the fundamental capabilities for embodied intelligence and a critical challenge that urgently needs to be addressed. However, existing methods are still unsatisfactory in terms of both success rate (SR) and generalization: Supervised Fine-Tuning (SFT) approaches typically achieve higher SR, while Training-Free (TF) approaches often generalize better, but it is difficult to obtain both simultaneously. To this end, we propose a Memory-Execute-Review framework. It consists of three parts: a hierarchical memory module for providing information support, an execute module for routine decision-making and actions, and a review module for handling abnormal situations and correcting behavior. We validated the effectiveness of this framework on the Object Goal Navigation task. Across 4 datasets, our average SR achieved absolute improvements of 7% and 5% compared to all baseline methods under TF and Zero-Shot (ZS) settings, respectively. On the most commonly used HM3D_v0.1 and the more challenging open vocabulary dataset HM3D_OVON, the SR improved by 8% and 6%, under ZS settings. Furthermore, on the MP3D and HM3D_OVON datasets, our method not only outperformed all TF methods but also surpassed all SFT methods, achieving comprehensive leadership in both SR (5% and 2%) and generalization.
comment: 9 pages, 2 figures, 5 tables, conference
☆ Ontology-Driven Robotic Specification Synthesis
This paper addresses robotic system engineering for safety- and mission-critical applications by bridging the gap between high-level objectives and formal, executable specifications. The proposed method, Robotic System Task to Model Transformation Methodology (RSTM2) is an ontology-driven, hierarchical approach using stochastic timed Petri nets with resources, enabling Monte Carlo simulations at mission, system, and subsystem levels. A hypothetical case study demonstrates how the RSTM2 method supports architectural trades, resource allocation, and performance analysis under uncertainty. Ontological concepts further enable explainable AI-based assistants, facilitating fully autonomous specification synthesis. The methodology offers particular benefits to complex multi-robot systems, such as the NASA CADRE mission, representing decentralized, resource-aware, and adaptive autonomous systems of the future.
comment: 8 pages, 9 figures, 3 tables, journal
☆ Benchmarking Affordance Generalization with BusyBox
Vision-Language-Action (VLA) models have been attracting the attention of researchers and practitioners thanks to their promise of generalization. Although single-task policies still offer competitive performance, VLAs are increasingly able to handle commands and environments unseen in their training set. While generalization in vision and language space is undoubtedly important for robust versatile behaviors, a key meta-skill VLAs need to possess is affordance generalization -- the ability to manipulate new objects with familiar physical features. In this work, we present BusyBox, a physical benchmark for systematic semi-automatic evaluation of VLAs' affordance generalization. BusyBox consists of 6 modules with switches, sliders, wires, buttons, a display, and a dial. The modules can be swapped and rotated to create a multitude of BusyBox variations with different visual appearances but the same set of affordances. We empirically demonstrate that generalization across BusyBox variants is highly challenging even for strong open-weights VLAs such as $π_{0.5}$ and GR00T-N1.6. To encourage the research community to evaluate their own VLAs on BusyBox and to propose new affordance generalization experiments, we have designed BusyBox to be easy to build in most robotics labs. We release the full set of CAD files for 3D-printing its parts as well as a bill of materials for (optionally) assembling its electronics. We also publish a dataset of language-annotated demonstrations that we collected using the common bimanual Mobile Aloha robot on the canonical BusyBox configuration. All of the released materials are available at https://microsoft.github.io/BusyBox.
☆ RoboPaint: From Human Demonstration to Any Robot and Any View
Acquiring large-scale, high-fidelity robot demonstration data remains a critical bottleneck for scaling Vision-Language-Action (VLA) models in dexterous manipulation. We propose a Real-Sim-Real data collection and data editing pipeline that transforms human demonstrations into robot-executable, environment-specific training data without direct robot teleoperation. Standardized data collection rooms are built to capture multimodal human demonstrations (synchronized 3 RGB-D videos, 11 RGB videos, 29-DoF glove joint angles, and 14-channel tactile signals). Based on these human demonstrations, we introduce a tactile-aware retargeting method that maps human hand states to robot dex-hand states via geometry and force-guided optimization. Then the retargeted robot trajectories are rendered in a photorealistic Isaac Sim environment to build robot training data. Real world experiments have demonstrated: (1) The retargeted dex-hand trajectories achieve an 84\% success rate across 10 diverse object manipulation tasks. (2) VLA policies (Pi0.5) trained exclusively on our generated data achieve 80\% average success rate on three representative tasks, i.e., pick-and-place, pushing and pouring. To conclude, robot training data can be efficiently "painted" from human demonstrations using our real-sim-real data pipeline. We offer a scalable, cost-effective alternative to teleoperation with minimal performance loss for complex dexterous manipulation.
comment: 17 pages
☆ A Data Driven Structural Decomposition of Dynamic Games via Best Response Maps
Dynamic games are powerful tools to model multi-agent decision-making, yet computing Nash (generalized Nash) equilibria remains a central challenge in such settings. Complexity arises from tightly coupled optimality conditions, nested optimization structures, and poor numerical conditioning. Existing game-theoretic solvers address these challenges by directly solving the joint game, typically requiring explicit modeling of all agents' objective functions and constraints, while learning-based approaches often decouple interaction through prediction or policy approximation, sacrificing equilibrium consistency. This paper introduces a conceptually novel formulation for dynamic games by restructuring the equilibrium computation. Rather than solving a fully coupled game or decoupling agents through prediction or policy approximation, a data-driven structural reduction of the game is proposed that removes nested optimization layers and derivative coupling by embedding an offline-compiled best-response map as a feasibility constraint. Under standard regularity conditions, when the best-response operator is exact, any converged solution of the reduced problem corresponds to a local open-loop Nash (GNE) equilibrium of the original game; with a learned surrogate, the solution is approximately equilibrium-consistent up to the best-response approximation error. The proposed formulation is supported by mathematical proofs, accompanying a large-scale Monte Carlo study in a two-player open-loop dynamic game motivated by the autonomous racing problem. Comparisons are made against state-of-the-art joint game solvers, and results are reported on solution quality, computational cost, and constraint satisfaction.
comment: 11 pages, 6 figures, 5 tables, Submitted to RSS 2026
☆ Formal Synthesis of Certifiably Robust Neural Lyapunov-Barrier Certificates
Neural Lyapunov and barrier certificates have recently been used as powerful tools for verifying the safety and stability properties of deep reinforcement learning (RL) controllers. However, existing methods offer guarantees only under fixed ideal unperturbed dynamics, limiting their reliability in real-world applications where dynamics may deviate due to uncertainties. In this work, we study the problem of synthesizing \emph{robust neural Lyapunov barrier certificates} that maintain their guarantees under perturbations in system dynamics. We formally define a robust Lyapunov barrier function and specify sufficient conditions based on Lipschitz continuity that ensure robustness against bounded perturbations. We propose practical training objectives that enforce these conditions via adversarial training, Lipschitz neighborhood bound, and global Lipschitz regularization. We validate our approach in two practically relevant environments, Inverted Pendulum and 2D Docking. The former is a widely studied benchmark, while the latter is a safety-critical task in autonomous systems. We show that our methods significantly improve both certified robustness bounds (up to $4.6$ times) and empirical success rates under strong perturbations (up to $2.4$ times) compared to the baseline. Our results demonstrate effectiveness of training robust neural certificates for safe RL under perturbations in dynamics.
☆ Learning Soccer Skills for Humanoid Robots: A Progressive Perception-Action Framework
Soccer presents a significant challenge for humanoid robots, demanding tightly integrated perception-action capabilities for tasks like perception-guided kicking and whole-body balance control. Existing approaches suffer from inter-module instability in modular pipelines or conflicting training objectives in end-to-end frameworks. We propose Perception-Action integrated Decision-making (PAiD), a progressive architecture that decomposes soccer skill acquisition into three stages: motion-skill acquisition via human motion tracking, lightweight perception-action integration for positional generalization, and physics-aware sim-to-real transfer. This staged decomposition establishes stable foundational skills, avoids reward conflicts during perception integration, and minimizes sim-to-real gaps. Experiments on the Unitree G1 demonstrate high-fidelity human-like kicking with robust performance under diverse conditions-including static or rolling balls, various positions, and disturbances-while maintaining consistent execution across indoor and outdoor scenarios. Our divide-and-conquer strategy advances robust humanoid soccer capabilities and offers a scalable framework for complex embodied skill acquisition. The project page is available at https://soccer-humanoid.github.io/.
comment: 13 pages, 9 figures, conference
☆ Affordance-Aware Interactive Decision-Making and Execution for Ambiguous Instructions
Enabling robots to explore and act in unfamiliar environments under ambiguous human instructions by interactively identifying task-relevant objects (e.g., identifying cups or beverages for "I'm thirsty") remains challenging for existing vision-language model (VLM)-based methods. This challenge stems from inefficient reasoning and the lack of environmental interaction, which hinder real-time task planning and execution. To address this, We propose Affordance-Aware Interactive Decision-Making and Execution for Ambiguous Instructions (AIDE), a dual-stream framework that integrates interactive exploration with vision-language reasoning, where Multi-Stage Inference (MSI) serves as the decision-making stream and Accelerated Decision-Making (ADM) as the execution stream, enabling zero-shot affordance analysis and interpretation of ambiguous instructions. Extensive experiments in simulation and real-world environments show that AIDE achieves the task planning success rate of over 80\% and more than 95\% accuracy in closed-loop continuous execution at 10 Hz, outperforming existing VLM-based methods in diverse open-world scenarios.
comment: 14 pages, 10 figures, 8 tables
☆ Low-Cost Underwater In-Pipe Centering and Inspection Using a Minimal-Sensing Robot
Autonomous underwater inspection of submerged pipelines is challenging due to confined geometries, turbidity, and the scarcity of reliable localization cues. This paper presents a minimal-sensing strategy that enables a free-swimming underwater robot to center itself and traverse a flooded pipe of known radius using only an IMU, a pressure sensor, and two sonars: a downward-facing single-beam sonar and a rotating 360 degree sonar. We introduce a computationally efficient method for extracting range estimates from single-beam sonar intensity data, enabling reliable wall detection in noisy and reverberant conditions. A closed-form geometric model leverages the two sonar ranges to estimate the pipe center, and an adaptive, confidence-weighted proportional-derivative (PD) controller maintains alignment during traversal. The system requires no Doppler velocity log, external tracking, or complex multi-sensor arrays. Experiments in a submerged 46 cm-diameter pipe using a Blue Robotics BlueROV2 heavy remotely operated vehicle demonstrate stable centering and successful full-pipe traversal despite ambient flow and structural deformations. These results show that reliable in-pipe navigation and inspection can be achieved with a lightweight, computationally efficient sensing and processing architecture, advancing the practicality of autonomous underwater inspection in confined environments.
☆ RFM-Pose:Reinforcement-Guided Flow Matching for Fast Category-Level 6D Pose Estimation
Object pose estimation is a fundamental problem in computer vision and plays a critical role in virtual reality and embodied intelligence, where agents must understand and interact with objects in 3D space. Recently, score based generative models have to some extent solved the rotational symmetry ambiguity problem in category level pose estimation, but their efficiency remains limited by the high sampling cost of score-based diffusion. In this work, we propose a new framework, RFM-Pose, that accelerates category-level 6D object pose generation while actively evaluating sampled hypotheses. To improve sampling efficiency, we adopt a flow-matching generative model and generate pose candidates along an optimal transport path from a simple prior to the pose distribution. To further refine these candidates, we cast the flow-matching sampling process as a Markov decision process and apply proximal policy optimization to fine-tune the sampling policy. In particular, we interpret the flow field as a learnable policy and map an estimator to a value network, enabling joint optimization of pose generation and hypothesis scoring within a reinforcement learning framework. Experiments on the REAL275 benchmark demonstrate that RFM-Pose achieves favorable performance while significantly reducing computational cost. Moreover, similar to prior work, our approach can be readily adapted to object pose tracking and attains competitive results in this setting.
comment: This work has been submitted to the IEEE for possible publication
☆ MobileManiBench: Simplifying Model Verification for Mobile Manipulation
Vision-language-action models have advanced robotic manipulation but remain constrained by reliance on the large, teleoperation-collected datasets dominated by the static, tabletop scenes. We propose a simulation-first framework to verify VLA architectures before real-world deployment and introduce MobileManiBench, a large-scale benchmark for mobile-based robotic manipulation. Built on NVIDIA Isaac Sim and powered by reinforcement learning, our pipeline autonomously generates diverse manipulation trajectories with rich annotations (language instructions, multi-view RGB-depth-segmentation images, synchronized object/robot states and actions). MobileManiBench features 2 mobile platforms (parallel-gripper and dexterous-hand robots), 2 synchronized cameras (head and right wrist), 630 objects in 20 categories, 5 skills (open, close, pull, push, pick) with over 100 tasks performed in 100 realistic scenes, yielding 300K trajectories. This design enables controlled, scalable studies of robot embodiments, sensing modalities, and policy architectures, accelerating research on data efficiency and generalization. We benchmark representative VLA models and report insights into perception, reasoning, and control in complex simulated environments.
☆ Informative Path Planning with Guaranteed Estimation Uncertainty
Environmental monitoring robots often need to reconstruct spatial fields (e.g., salinity, temperature, bathymetry) under tight distance and energy constraints. Classical boustrophedon lawnmower surveys provide geometric coverage guarantees but can waste effort by oversampling predictable regions. In contrast, informative path planning (IPP) methods leverage spatial correlations to reduce oversampling, yet typically offer no guarantees on reconstruction quality. This paper bridges these approaches by addressing informative path planning with guaranteed estimation uncertainty: computing the shortest path whose measurements ensure that the Gaussian-process (GP) posterior variance -- an intrinsic uncertainty measure that lower-bounds the mean-squared prediction error under the GP model -- falls below a user-specified threshold over the monitoring region. We propose a three-stage approach: (i) learn a GP model from available prior information; (ii) transform the learned GP kernel into binary coverage maps for each candidate sensing location, indicating which locations' uncertainty can be reduced below a specified target; and (iii) plan a near-shortest route whose combined coverage satisfies the global uncertainty constraint. To address heterogeneous phenomena, we incorporate a nonstationary kernel that captures spatially varying correlation structure, and we accommodate non-convex environments with obstacles. Algorithmically, we present methods with provable approximation guarantees for sensing-location selection and for the joint selection-and-routing problem under a travel budget. Experiments on real-world topographic data show that our planners meet the uncertainty target using fewer sensing locations and shorter travel distances than a recent baseline, and field experiments with bathymetry-mapping autonomous surface and underwater vehicles demonstrate real-world feasibility.
comment: 16 pages, 11 figures, preprint
☆ PLATO Hand: Shaping Contact Behavior with Fingernails for Precise Manipulation
We present the PLATO Hand, a dexterous robotic hand with a hybrid fingertip that embeds a rigid fingernail within a compliant pulp. This design shapes contact behavior to enable diverse interaction modes across a range of object geometries. We develop a strain-energy-based bending-indentation model to guide the fingertip design and to explain how guided contact preserves local indentation while suppressing global bending. Experimental results show that the proposed robotic hand design demonstrates improved pinching stability, enhanced force observability, and successful execution of edge-sensitive manipulation tasks, including paper singulation, card picking, and orange peeling. Together, these results show that coupling structured contact geometry with a force-motion transparent mechanism provides a principled, physically embodied approach to precise manipulation.
♻ ☆ Physical Human-Robot Interaction: A Critical Review of Safety Constraints
This paper aims to provide a clear and rigorous understanding of commonly recognized safety constraints in physical human-robot interaction, particularly regarding ISO/TS 15066. We investigate the derivation of these constraints, critically examine the underlying assumptions, and evaluate their practical implications for system-level safety and performance in industrially relevant scenarios. Key design parameters within safety-critical control architectures are identified, and numerical examples are provided to quantify performance degradation arising from typical approximations and design decisions in manufacturing environments. Within this analysis, the fundamental role of energy in safety assessment is emphasized, providing focused insights into energy-based safety methodologies for collaborative industrial robot systems.
♻ ☆ Bench-NPIN: Benchmarking Non-prehensile Interactive Navigation
Mobile robots are increasingly deployed in unstructured environments where obstacles and objects are movable. Navigation in such environments is known as interactive navigation, where task completion requires not only avoiding obstacles but also strategic interactions with movable objects. Non-prehensile interactive navigation focuses on non-grasping interaction strategies, such as pushing, rather than relying on prehensile manipulation. Despite a growing body of research in this field, most solutions are evaluated using case-specific setups, limiting reproducibility and cross-comparison. In this paper, we present Bench-NPIN, the first comprehensive benchmark for non-prehensile interactive navigation. Bench-NPIN includes multiple components: 1) a comprehensive range of simulated environments for non-prehensile interactive navigation tasks, including navigating a maze with movable obstacles, autonomous ship navigation in icy waters, box delivery, and area clearing, each with varying levels of complexity; 2) a set of evaluation metrics that capture unique aspects of interactive navigation, such as efficiency, interaction effort, and partial task completion; and 3) demonstrations using Bench-NPIN to evaluate example implementations of established baselines across environments. Bench-NPIN is an open-source Python library with a modular design. The code, documentation, and trained models can be found at https://github.com/IvanIZ/BenchNPIN.
comment: This paper has been withdrawn by the authors. This paper has been superseded by arXiv:2512.11736
♻ ☆ Enriching physical-virtual interaction in AR gaming by tracking identical objects via an egocentric partial observation frame
Augmented reality (AR) games, particularly those designed for head-mounted displays, have grown increasingly prevalent. However, most existing systems depend on pre-scanned, static environments and rely heavily on continuous tracking or marker-based solutions, which limit adaptability in dynamic physical spaces. This is particularly problematic for AR headsets and glasses, which typically follow the user's head movement and cannot maintain a fixed, stationary view of the scene. Moreover, continuous scene observation is neither power-efficient nor practical for wearable devices, given their limited battery and processing capabilities. A persistent challenge arises when multiple identical objects are present in the environment-standard object tracking pipelines often fail to maintain consistent identities without uninterrupted observation or external sensors. These limitations hinder fluid physical-virtual interactions, especially in dynamic or occluded scenes where continuous tracking is infeasible. To address this, we introduce a novel optimization-based framework for re-identifying identical objects in AR scenes using only one partial egocentric observation frame captured by a headset. We formulate the problem as a label assignment task solved via integer programming, augmented with a Voronoi diagram-based pruning strategy to improve computational efficiency. This method reduces computation time by 50% while preserving 91% accuracy in simulated experiments. Moreover, we evaluated our approach in quantitative synthetic and quantitative real-world experiments. We also conducted three qualitative real-world experiments to demonstrate the practical utility and generalizability for enabling dynamic, markerless object interaction in AR environments. Our video demo is available at https://youtu.be/RwptEfLtW1U.
♻ ☆ 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.
♻ ☆ Improved Bag-of-Words Image Retrieval with Geometric Constraints for Ground Texture Localization ICRA 2025
Ground texture localization using a downward-facing camera offers a low-cost, high-precision localization solution that is robust to dynamic environments and requires no environmental modification. We present a significantly improved bag-of-words (BoW) image retrieval system for ground texture localization, achieving substantially higher accuracy for global localization and higher precision and recall for loop closure detection in SLAM. Our approach leverages an approximate $k$-means (AKM) vocabulary with soft assignment, and exploits the consistent orientation and constant scale constraints inherent to ground texture localization. Identifying the different needs of global localization vs. loop closure detection for SLAM, we present both high-accuracy and high-speed versions of our algorithm. We test the effect of each of our proposed improvements through an ablation study and demonstrate our method's effectiveness for both global localization and loop closure detection. With numerous ground texture localization systems already using BoW, our method can readily replace other generic BoW systems in their pipeline and immediately improve their results.
comment: Accepted to ICRA 2025
♻ ☆ Constraint-Aware Discrete-Time PID Gain Optimization for Robotic Joint Control Under Actuator Saturation
The precise regulation of rotary actuation is fundamental in autonomous robotics, yet practical PID loops deviate from continuous-time theory due to discrete-time execution, actuator saturation, and small delays and measurement imperfections. We present an implementation-aware analysis and tuning workflow for saturated discrete-time joint control. We (i) derive PI stability regions under Euler and exact zero-order-hold (ZOH) discretizations using the Jury criterion, (ii) evaluate a discrete back-calculation anti-windup realization under saturation-dominant regimes, and (iii) propose a hybrid-certified Bayesian optimization workflow that screens analytically unstable candidates and behaviorally unsafe transients while optimizing a robust IAE objective with soft penalties on overshoot and saturation duty. Baseline sweeps ($τ=1.0$~s, $Δt=0.01$~s, $u\in[-10,10]$) quantify rise/settle trends for P/PI/PID. Under a randomized model family emulating uncertainty, delay, noise, quantization, and tighter saturation, robustness-oriented tuning improves median IAE from $0.843$ to $0.430$ while keeping median overshoot below $2\%$. In simulation-only tuning, the certification screen rejects $11.6\%$ of randomly sampled gains within bounds before full robust evaluation, improving sample efficiency.
comment: Pending IEEE Transactions on Robotics Publication
♻ ☆ FilMBot: A High-Speed Soft Parallel Robotic Micromanipulator
Soft robotic manipulators are generally slow despite their great adaptability, resilience, and compliance. This limitation also extends to current soft robotic micromanipulators. Here, we introduce FilMBot, a 3-DOF film-based, electromagnetically actuated, soft kinematic robotic micromanipulator achieving speeds up to 2117 °/s and 2456 °/s in α and \{beta} angular motions, with corresponding linear velocities of 1.61 m/s and 1.92 m/s using a 4-cm needle end-effector, 0.54 m/s along the Z axis, and 1.57 m/s during Z-axis morph switching. The robot can reach ~1.50 m/s in path-following tasks, with an operational bandwidth below ~30 Hz, and remains responsive at 50 Hz. It demonstrates high precision (~6.3 μm, or ~0.05% of its workspace) in path-following tasks, with precision remaining largely stable across frequencies. The novel combination of the low-stiffness soft kinematic film structure and strong electromagnetic actuation in FilMBot opens new avenues for soft robotics. Furthermore, its simple construction and inexpensive, readily accessible components could broaden the application of micromanipulators beyond current academic and professional users.
comment: 13 pages, 16 figures
♻ ☆ Do Robots Really Need Anthropomorphic Hands? -- A Comparison of Human and Robotic Hands
Human manipulation skills represent a pinnacle of their voluntary motor functions, requiring the coordination of many degrees of freedom and processing of high-dimensional sensor input to achieve such a high level of dexterity. Thus, we attempt to answer whether the human hand, with its associated biomechanical properties, sensors, and control mechanisms, is an ideal that we should strive for in robotics-do we really need anthropomorphic robotic hands? This survey can help practitioners to make the trade-off between hand complexity and potential manipulation skills. We provide an overview of the human hand, a comparison of commercially available robotic and prosthetic hands, and a systematic review of hand mechanisms and skills that they are capable of. This leads to follow-up questions. What is the minimum requirement for mechanisms and sensors to implement most skills that a robot needs? What is missing to reach human-level dexterity? Can we improve upon human dexterity? Although complex five-fingered hands are often used as the ultimate goal for robotic manipulators, they are not necessary for all tasks. We found that wrist flexibility and finger abduction/adduction are often more important for manipulation capabilities. Increasing the number of fingers, actuators, or degrees of freedom is not always necessary. Three fingers often are a good compromise between simplicity and dexterity. Non-anthropomorphic hand designs with two opposing pairs of fingers or human hands with six fingers can further increase dexterity, suggesting that the human hand is not the optimum. Consequently, we argue for function-based rather than form-based biomimicry.
♻ ☆ RANGER: A Monocular Zero-Shot Semantic Navigation Framework through Contextual Adaptation ICRA 2026
Efficiently finding targets in complex environments is fundamental to real-world embodied applications. While recent advances in multimodal foundation models have enabled zero-shot object goal navigation, allowing robots to search for arbitrary objects without fine-tuning, existing methods face two key limitations: (1) heavy reliance on precise depth and pose information provided by simulators, which restricts applicability in real-world scenarios; and (2) lack of in-context learning (ICL) capability, making it difficult to quickly adapt to new environments, as in leveraging short videos. To address these challenges, we propose RANGER, a novel zero-shot, open-vocabulary semantic navigation framework that operates using only a monocular camera. Leveraging powerful 3D foundation models, RANGER eliminates the dependency on depth and pose while exhibiting strong ICL capability. By simply observing a short video of a new environment, the system can also significantly improve task efficiency without requiring architectural modifications or fine-tuning. The framework integrates several key components: keyframe-based 3D reconstruction, semantic point cloud generation, vision-language model (VLM)-driven exploration value estimation, high-level adaptive waypoint selection, and low-level action execution. Experiments on the HM3D benchmark and real-world environments demonstrate that RANGER achieves competitive performance in terms of navigation success rate and exploration efficiency, while showing superior ICL adaptability, with no previous 3D mapping of the environment required.
comment: Accepted at ICRA 2026
♻ ☆ TACO: Temporal Consensus Optimization for Continual Neural Mapping
Neural implicit mapping has emerged as a powerful paradigm for robotic navigation and scene understanding. However, real-world robotic deployment requires continual adaptation to changing environments under strict memory and computation constraints, which existing mapping systems fail to support. Most prior methods rely on replaying historical observations to preserve consistency and assume static scenes. As a result, they cannot adapt to continual learning in dynamic robotic settings. To address these challenges, we propose TACO (TemporAl Consensus Optimization), a replay-free framework for continual neural mapping. We reformulate mapping as a temporal consensus optimization problem, where we treat past model snapshots as temporal neighbors. Intuitively, our approach resembles a model consulting its own past knowledge. We update the current map by enforcing weighted consensus with historical representations. Our method allows reliable past geometry to constrain optimization while permitting unreliable or outdated regions to be revised in response to new observations. TACO achieves a balance between memory efficiency and adaptability without storing or replaying previous data. Through extensive simulated and real-world experiments, we show that TACO robustly adapts to scene changes, and consistently outperforms other continual learning baselines.
♻ ☆ RFS: Reinforcement Learning with Residual Flow Steering for Dexterous Manipulation
Imitation learning has emerged as an effective approach for bootstrapping sequential decision-making in robotics, achieving strong performance even in high-dimensional dexterous manipulation tasks. Recent behavior cloning methods further leverage expressive generative models, such as diffusion models and flow matching, to represent multimodal action distributions. However, policies pretrained in this manner often exhibit limited generalization and require additional fine-tuning to achieve robust performance at deployment time. Such adaptation must preserve the global exploration benefits of pretraining while enabling rapid correction of local execution errors. We propose Residual Flow Steering(RFS), a data-efficient reinforcement learning framework for adapting pretrained generative policies. RFS steers a pretrained flow-matching policy by jointly optimizing a residual action and a latent noise distribution, enabling complementary forms of exploration: local refinement through residual corrections and global exploration through latent-space modulation. This design allows efficient adaptation while retaining the expressive structure of the pretrained policy. We demonstrate the effectiveness of RFS on dexterous manipulation tasks, showing efficient fine-tuning in both simulation and real-world settings when adapting pretrained base policies. Project website:https://weirdlabuw.github.io/rfs.
♻ ☆ A Sliced Learning Framework for Online Disturbance Identification in Quadrotor SO(3) Attitude Control
This paper introduces a dimension-decomposed geometric learning framework called Sliced Learning for disturbance identification in quadrotor geometric attitude control. Instead of conventional learning-from-states, this framework adopts a learning-from-error strategy by using the Lie-algebraic error representation as the input feature, enabling axis-wise space decomposition (``slicing") while preserving the SO(3) structure. This is highly consistent with the geometric mechanism of cognitive control observed in neuroscience, where neural systems organize adaptive representations within structured subspaces to enable cognitive flexibility and efficiency. Based on this framework, we develop a lightweight and structurally interpretable Sliced Adaptive-Neuro Mapping (SANM) module. The high-dimensional mapping for online identification is axially ``sliced" into multiple low-dimensional submappings (``slices"), implemented by shallow neural networks and adaptive laws. These neural networks and adaptive laws are updated online via Lyapunov-based adaptation within their respective shared subspaces. To enhance interpretability, we prove exponential convergence despite time-varying disturbances and inertia uncertainties. To our knowledge, Sliced Learning is among the first frameworks to demonstrate lightweight online neural adaptation at 400 Hz on resource-constrained microcontroller units (MCUs), such as STM32, with real-world experimental validation.
comment: v3: Major revision--Revised title; introduced the Sliced Learning framework; added comparative experiments, extended theoretical results, and supplementary materials (such as algorithms and proofs)
♻ ☆ Learning to Plan & Schedule with Reinforcement-Learned Bimanual Robot Skills
Long-horizon contact-rich bimanual manipulation presents a significant challenge, requiring complex coordination involving a mixture of parallel execution and sequential collaboration between arms. In this paper, we introduce a hierarchical framework that frames this challenge as an integrated skill planning & scheduling problem, going beyond purely sequential decision-making to support simultaneous skill invocation. Our approach is built upon a library of single-arm and bimanual primitive skills, each trained using Reinforcement Learning (RL) in GPU-accelerated simulation. We then train a Transformer-based planner on a dataset of skill compositions to act as a high-level scheduler, simultaneously predicting the discrete schedule of skills as well as their continuous parameters. We demonstrate that our method achieves higher success rates on complex, contact-rich tasks than end-to-end RL approaches and produces more efficient, coordinated behaviors than traditional sequential-only planners.
♻ ☆ MindDrive: A Vision-Language-Action Model for Autonomous Driving via Online Reinforcement Learning
Current Vision-Language-Action (VLA) paradigms in autonomous driving primarily rely on Imitation Learning (IL), which introduces inherent challenges such as distribution shift and causal confusion. Online Reinforcement Learning offers a promising pathway to address these issues through trial-and-error learning. However, applying online reinforcement learning to VLA models in autonomous driving is hindered by inefficient exploration in continuous action spaces. To overcome this limitation, we propose MindDrive, a VLA framework comprising a large language model (LLM) with two distinct sets of LoRA parameters. The one LLM serves as a Decision Expert for scenario reasoning and driving decision-making, while the other acts as an Action Expert that dynamically maps linguistic decisions into feasible trajectories. By feeding trajectory-level rewards back into the reasoning space, MindDrive enables trial-and-error learning over a finite set of discrete linguistic driving decisions, instead of operating directly in a continuous action space. This approach effectively balances optimal decision-making in complex scenarios, human-like driving behavior, and efficient exploration in online reinforcement learning. Using the lightweight Qwen-0.5B LLM, MindDrive achieves Driving Score (DS) of 78.04 and Success Rate (SR) of 55.09% on the challenging Bench2Drive benchmark. To the best of our knowledge, this is the first work to demonstrate the effectiveness of online reinforcement learning for the VLA model in autonomous driving.
comment: 16 pages, 12 figures, 6 tables; Project Page: https://xiaomi-mlab.github.io/MindDrive/
♻ ☆ HoRD: Robust Humanoid Control via History-Conditioned Reinforcement Learning and Online Distillation
Humanoid robots can suffer significant performance drops under small changes in dynamics, task specifications, or environment setup. We propose HoRD, a two-stage learning framework for robust humanoid control under domain shift. First, we train a high-performance teacher policy via history-conditioned reinforcement learning, where the policy infers latent dynamics context from recent state--action trajectories to adapt online to diverse randomized dynamics. Second, we perform online distillation to transfer the teacher's robust control capabilities into a transformer-based student policy that operates on sparse root-relative 3D joint keypoint trajectories. By combining history-conditioned adaptation with online distillation, HoRD enables a single policy to adapt zero-shot to unseen domains without per-domain retraining. Extensive experiments show HoRD outperforms strong baselines in robustness and transfer, especially under unseen domains and external perturbations. Code and project page are available at https://tonywang-0517.github.io/hord/.