MyArxiv
Robotics 47
☆ NL2SpaTiaL: Generating Geometric Spatio-Temporal Logic Specifications from Natural Language for Manipulation Tasks
Spatio-Temporal Logic (SpaTiaL) offers a principled formalism for expressing geometric spatial requirements-an essential component of robotic manipulation, where object locations, neighborhood relations, pose constraints, and interactions directly determine task success. Yet prior works have largely relied on standard temporal logic (TL), which models only robot trajectories and overlooks object-level interactions. Existing datasets built from randomly generated TL formulas paired with natural-language descriptions therefore cover temporal operators but fail to represent the layered spatial relations that manipulation tasks depend on. To address this gap, we introduce a dataset generation framework that synthesizes SpaTiaL specifications and converts them into natural-language descriptions through a deterministic, semantics-preserving back-translation procedure. This pipeline produces the NL2SpaTiaL dataset, aligning natural language with multi-level spatial relations and temporal objectives to reflect the compositional structure of manipulation tasks. Building on this foundation, we propose a translation-verification framework equipped with a language-based semantic checker that ensures the generated SpaTiaL formulas faithfully encode the semantics specified by the input description. Experiments across a suite of manipulation tasks show that SpaTiaL-based representations yield more interpretable, verifiable, and compositional grounding for instruction following. Project website: https://sites.google.com/view/nl2spatial
☆ RoboTracer: Mastering Spatial Trace with Reasoning in Vision-Language Models for Robotics
Spatial tracing, as a fundamental embodied interaction ability for robots, is inherently challenging as it requires multi-step metric-grounded reasoning compounded with complex spatial referring and real-world metric measurement. However, existing methods struggle with this compositional task. To this end, we propose RoboTracer, a 3D-aware VLM that first achieves both 3D spatial referring and measuring via a universal spatial encoder and a regression-supervised decoder to enhance scale awareness during supervised fine-tuning (SFT). Moreover, RoboTracer advances multi-step metric-grounded reasoning via reinforcement fine-tuning (RFT) with metric-sensitive process rewards, supervising key intermediate perceptual cues to accurately generate spatial traces. To support SFT and RFT training, we introduce TraceSpatial, a large-scale dataset of 30M QA pairs, spanning outdoor/indoor/tabletop scenes and supporting complex reasoning processes (up to 9 steps). We further present TraceSpatial-Bench, a challenging benchmark filling the gap to evaluate spatial tracing. Experimental results show that RoboTracer surpasses baselines in spatial understanding, measuring, and referring, with an average success rate of 79.1%, and also achieves SOTA performance on TraceSpatial-Bench by a large margin, exceeding Gemini-2.5-Pro by 36% accuracy. Notably, RoboTracer can be integrated with various control policies to execute long-horizon, dynamic tasks across diverse robots (UR5, G1 humanoid) in cluttered real-world scenes.
comment: Project page: https://zhoues.github.io/RoboTracer
World Models Can Leverage Human Videos for Dexterous Manipulation
Dexterous manipulation is challenging because it requires understanding how subtle hand motion influences the environment through contact with objects. We introduce DexWM, a Dexterous Manipulation World Model that predicts the next latent state of the environment conditioned on past states and dexterous actions. To overcome the scarcity of dexterous manipulation datasets, DexWM is trained on over 900 hours of human and non-dexterous robot videos. To enable fine-grained dexterity, we find that predicting visual features alone is insufficient; therefore, we introduce an auxiliary hand consistency loss that enforces accurate hand configurations. DexWM outperforms prior world models conditioned on text, navigation, and full-body actions, achieving more accurate predictions of future states. DexWM also demonstrates strong zero-shot generalization to unseen manipulation skills when deployed on a Franka Panda arm equipped with an Allegro gripper, outperforming Diffusion Policy by over 50% on average in grasping, placing, and reaching tasks.
☆ 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. MindDrive achieves strong closed-loop performance on the challenging Bench2Drive benchmark, with a Driving Score (DS) of 78.04 and a Success Rate (SR) of 55.09%. 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/
☆ Near-Field Perception for Safety Enhancement of Autonomous Mobile Robots in Manufacturing Environments
Near-field perception is essential for the safe operation of autonomous mobile robots (AMRs) in manufacturing environments. Conventional ranging sensors such as light detection and ranging (LiDAR) and ultrasonic devices provide broad situational awareness but often fail to detect small objects near the robot base. To address this limitation, this paper presents a three-tier near-field perception framework. The first approach employs light-discontinuity detection, which projects a laser stripe across the near-field zone and identifies interruptions in the stripe to perform fast, binary cutoff sensing for obstacle presence. The second approach utilizes light-displacement measurement to estimate object height by analyzing the geometric displacement of a projected stripe in the camera image, which provides quantitative obstacle height information with minimal computational overhead. The third approach employs a computer vision-based object detection model on embedded AI hardware to classify objects, enabling semantic perception and context-aware safety decisions. All methods are implemented on a Raspberry Pi 5 system, achieving real-time performance at 25 or 50 frames per second. Experimental evaluation and comparative analysis demonstrate that the proposed hierarchy balances precision, computation, and cost, thereby providing a scalable perception solution for enabling safe operations of AMRs in manufacturing environments.
comment: Submitted to the 54th SME North American Manufacturing Research Conference (NAMRC 54)
☆ Reinforcement Learning based 6-DoF Maneuvers for Microgravity Intravehicular Docking: A Simulation Study with Int-Ball2 in ISS-JEM
Autonomous free-flyers play a critical role in intravehicular tasks aboard the International Space Station (ISS), where their precise docking under sensing noise, small actuation mismatches, and environmental variability remains a nontrivial challenge. This work presents a reinforcement learning (RL) framework for six-degree-of-freedom (6-DoF) docking of JAXA's Int-Ball2 robot inside a high-fidelity Isaac Sim model of the Japanese Experiment Module (JEM). Using Proximal Policy Optimization (PPO), we train and evaluate controllers under domain-randomized dynamics and bounded observation noise, while explicitly modeling propeller drag-torque effects and polarity structure. This enables a controlled study of how Int-Ball2's propulsion physics influence RL-based docking performance in constrained microgravity interiors. The learned policy achieves stable and reliable docking across varied conditions and lays the groundwork for future extensions pertaining to Int-Ball2 in collision-aware navigation, safe RL, propulsion-accurate sim-to-real transfer, and vision-based end-to-end docking.
comment: Presented at AI4OPA Workshop at the International Conference on Space Robotics (iSpaRo) 2025 at Sendai, Japan
☆ Evaluating the Navigation Capabilities of a Modified COAST Guidewire Robot in an Anatomical Phantom Model
To address the issues that arise due to the manual navigation of guidewires in endovascular interventions, research in medical robotics has taken a strong interest in developing robotically steerable guidewires, which offer the possibility of enhanced maneuverability and navigation, as the tip of the guidewire can be actively steered. The COaxially Aligned STeerable (COAST) guidewire robot has the ability to generate a wide variety of motions including bending motion with different bending lengths, follow-the-leader motion, and feedforward motion. In our past studies, we have explored different designs of the COAST guidewire robot and developed modeling, control, and sensing strategies for the COAST guidewire robot. In this study, the performance of a modified COAST guidewire robot is evaluated by conducting navigation experiments in an anatomical phantom model with pulsatile flow. The modified COAST guidewire robot is a simplified version of the COAST guidewire robot and consists of two tubes as opposed to three tubes. Through this study, we demonstrate the effectiveness of the modified COAST guidewire robot in navigating the tortuous phantom vasculature.
comment: Presented at the 14th Conference on New Technologies for Computer and Robot Assisted Surgery (CRAS 2025)
☆ Universal Dexterous Functional Grasping via Demonstration-Editing Reinforcement Learning
Reinforcement learning (RL) has achieved great success in dexterous grasping, significantly improving grasp performance and generalization from simulation to the real world. However, fine-grained functional grasping, which is essential for downstream manipulation tasks, remains underexplored and faces several challenges: the complexity of specifying goals and reward functions for functional grasps across diverse objects, the difficulty of multi-task RL exploration, and the challenge of sim-to-real transfer. In this work, we propose DemoFunGrasp for universal dexterous functional grasping. We factorize functional grasping conditions into two complementary components - grasping style and affordance - and integrate them into an RL framework that can learn to grasp any object with any functional grasping condition. To address the multi-task optimization challenge, we leverage a single grasping demonstration and reformulate the RL problem as one-step demonstration editing, substantially enhancing sample efficiency and performance. Experimental results in both simulation and the real world show that DemoFunGrasp generalizes to unseen combinations of objects, affordances, and grasping styles, outperforming baselines in both success rate and functional grasping accuracy. In addition to strong sim-to-real capability, by incorporating a vision-language model (VLM) for planning, our system achieves autonomous instruction-following grasp execution.
comment: 19 pages
☆ Fast Policy Learning for 6-DOF Position Control of Underwater Vehicles
Autonomous Underwater Vehicles (AUVs) require reliable six-degree-of-freedom (6-DOF) position control to operate effectively in complex and dynamic marine environments. Traditional controllers are effective under nominal conditions but exhibit degraded performance when faced with unmodeled dynamics or environmental disturbances. Reinforcement learning (RL) provides a powerful alternative but training is typically slow and sim-to-real transfer remains challenging. This work introduces a GPU-accelerated RL training pipeline built in JAX and MuJoCo-XLA (MJX). By jointly JIT-compiling large-scale parallel physics simulation and learning updates, we achieve training times of under two minutes.Through systematic evaluation of multiple RL algorithms, we show robust 6-DOF trajectory tracking and effective disturbance rejection in real underwater experiments, with policies transferred zero-shot from simulation. Our results provide the first explicit real-world demonstration of RL-based AUV position control across all six degrees of freedom.
☆ Control of a Twin Rotor using Twin Delayed Deep Deterministic Policy Gradient (TD3)
This paper proposes a reinforcement learning (RL) framework for controlling and stabilizing the Twin Rotor Aerodynamic System (TRAS) at specific pitch and azimuth angles and tracking a given trajectory. The complex dynamics and non-linear characteristics of the TRAS make it challenging to control using traditional control algorithms. However, recent developments in RL have attracted interest due to their potential applications in the control of multirotors. The Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm was used in this paper to train the RL agent. This algorithm is used for environments with continuous state and action spaces, similar to the TRAS, as it does not require a model of the system. The simulation results illustrated the effectiveness of the RL control method. Next, external disturbances in the form of wind disturbances were used to test the controller's effectiveness compared to conventional PID controllers. Lastly, experiments on a laboratory setup were carried out to confirm the controller's effectiveness in real-world applications.
comment: This is the Author Accepted Manuscript version of a paper accepted for publication. The final published version is available via IEEE Xplore
☆ Humanoid Robot Running Through Random Stepping Stones and Jumping Over Obstacles: Step Adaptation Using Spring-Mass Trajectories
This study proposes a step adaptation framework for running through spring-mass trajectories and deadbeat control gain libraries. It includes four main parts: (1) Automatic spring-mass trajectory library generation; (2) Deadbeat control gain library generation through an actively controlled template model that resembles the whole-body dynamics well; (3) Trajectory selection policy development for step adaptation; (4) Mapping spring-mass trajectories to a humanoid model through a whole-body control (WBC) framework also accounting for closed-kinematic chain systems, self collisions, and reactive limb swinging. We show the inclusiveness and the robustness of the proposed framework through various challenging and agile behaviors such as running through randomly generated stepping stones, jumping over random obstacles, performing slalom motions, changing the running direction suddenly with a random leg, and rejecting significant disturbances and uncertainties through the MuJoCo physics simulator. We also perform additional simulations under a comprehensive set of uncertainties and noise to better justify the proposed method's robustness to real-world challenges, including signal noise, imprecision, modeling errors, and delays. All the aforementioned behaviors are performed with a single library and the same set of WBC control parameters without additional tuning. The spring-mass and the deadbeat control gain library are automatically computed in 4.5 seconds in total for 315 different trajectories.
comment: Accepted for publication in Biomimetic Intelligence and Robotics. Supplemental video: https://youtu.be/HlAg2nbNct4
☆ Intrinsic-Motivation Multi-Robot Social Formation Navigation with Coordinated Exploration
This paper investigates the application of reinforcement learning (RL) to multi-robot social formation navigation, a critical capability for enabling seamless human-robot coexistence. While RL offers a promising paradigm, the inherent unpredictability and often uncooperative dynamics of pedestrian behavior pose substantial challenges, particularly concerning the efficiency of coordinated exploration among robots. To address this, we propose a novel coordinated-exploration multi-robot RL algorithm introducing an intrinsic motivation exploration. Its core component is a self-learning intrinsic reward mechanism designed to collectively alleviate policy conservatism. Moreover, this algorithm incorporates a dual-sampling mode within the centralized training and decentralized execution framework to enhance the representation of both the navigation policy and the intrinsic reward, leveraging a two-time-scale update rule to decouple parameter updates. Empirical results on social formation navigation benchmarks demonstrate the proposed algorithm's superior performance over existing state-of-the-art methods across crucial metrics. Our code and video demos are available at: https://github.com/czxhunzi/CEMRRL.
☆ SAMAY: System for Acoustic Measurement and Analysis
This paper describes an automatic bird call recording system called SAMAY, which is developed to study bird species by creating a database of large amounts of bird acoustic data. By analysing the recorded bird call data, the system can also be used for automatic classification of bird species, monitoring bird populations and analysing the impact of environmental changes. The system is driven through a powerful STM32F407 series microcontroller, supports 4 microphones, is equipped with 128 GB of storage capacity, and is powered by a 10400 mAh battery pack interfaced with a solar charger. In addition, the device is user-configurable over USB and Wi-Fi during runtime, ensuring user-friendly operation during field deployment.
☆ Lightweight Dynamic Modeling of Cable-Driven Continuum Robots Based on Actuation-Space Energy Formulation
Cable-driven continuum robots (CDCRs) require accurate, real-time dynamic models for high-speed dynamics prediction or model-based control, making such capability an urgent need. In this paper, we propose the Lightweight Actuation-Space Energy Modeling (LASEM) framework for CDCRs, which formulates actuation potential energy directly in actuation space to enable lightweight yet accurate dynamic modeling. Through a unified variational derivation, the governing dynamics reduce to a single partial differential equation (PDE), requiring only the Euler moment balance while implicitly incorporating the Newton force balance. By also avoiding explicit computation of cable-backbone contact forces, the formulation simplifies the model structure and improves computational efficiency while preserving geometric accuracy and physical consistency. Importantly, the proposed framework for dynamic modeling natively supports both force-input and displacement-input actuation modes, a capability seldom achieved in existing dynamic formulations. Leveraging this lightweight structure, a Galerkin space-time modal discretization with analytical time-domain derivatives of the reduced state further enables an average 62.3% computational speedup over state-of-the-art real-time dynamic modeling approaches.
☆ Post-Training and Test-Time Scaling of Generative Agent Behavior Models for Interactive Autonomous Driving
Learning interactive motion behaviors among multiple agents is a core challenge in autonomous driving. While imitation learning models generate realistic trajectories, they often inherit biases from datasets dominated by safe demonstrations, limiting robustness in safety-critical cases. Moreover, most studies rely on open-loop evaluation, overlooking compounding errors in closed-loop execution. We address these limitations with two complementary strategies. First, we propose Group Relative Behavior Optimization (GRBO), a reinforcement learning post-training method that fine-tunes pretrained behavior models via group relative advantage maximization with human regularization. Using only 10% of the training dataset, GRBO improves safety performance by over 40% while preserving behavioral realism. Second, we introduce Warm-K, a warm-started Top-K sampling strategy that balances consistency and diversity in motion selection. Our Warm-K method-based test-time scaling enhances behavioral consistency and reactivity at test time without retraining, mitigating covariate shift and reducing performance discrepancies. Demo videos are available in the supplementary material.
comment: 11 pages, 5 figures
☆ A Unified Framework for Automated Assembly Sequence and Production Line Planning using Graph-based Optimization
This paper presents PyCAALP (Python-based Computer-Aided Assembly Line Planning), a framework for automated Assembly Sequence Planning (ASP) and Production Line Planning (PLP), employing a graph-based approach to model components and joints within production modules. The framework integrates kinematic boundary conditions, such as potential part collisions, to guarantee the feasibility of automated assembly planning. The developed algorithm computes all feasible production sequences, integrating modules for detecting spatial relationships and formulating geometric constraints. The algorithm incorporates additional attributes, including handling feasibility, tolerance matching, and joint compatibility, to manage the high combinatorial complexity inherent in assembly sequence generation. Heuristics, such as Single-Piece Flow assembly and geometrical constraint enforcement, are utilized to further refine the solution space, facilitating more efficient planning for complex assemblies. The PLP stage is formulated as a Mixed-Integer Program (MIP), balancing the total times of a fixed number of manufacturing stations. While some complexity reduction techniques may sacrifice optimality, they significantly reduce the MIPs computational time. Furthermore, the framework enables customization of engineering constraints and supports a flexible trade-off between ASP and PLP. The open-source nature of the framework, available at https://github.com/TUM-utg/PyCAALP, promotes further collaboration and adoption in both industrial and production research applications.
comment: Code available at https://github.com/TUM-utg/PyCAALP (repository will be made public prior to publication)
☆ Multi-directional Safe Rectangle Corridor-Based MPC for Nonholonomic Robots Navigation in Cluttered Environment
Autonomous Mobile Robots (AMRs) have become indispensable in industrial applications due to their operational flexibility and efficiency. Navigation serves as a crucial technical foundation for accomplishing complex tasks. However, navigating AMRs in dense, cluttered, and semi-structured environments remains challenging, primarily due to nonholonomic vehicle dynamics, interactions with mixed static/dynamic obstacles, and the non-convex constrained nature of such operational spaces. To solve these problems, this paper proposes an Improved Sequential Model Predictive Control (ISMPC) navigation framework that systematically reformulates navigation tasks as sequential switched optimal control problems. The framework addresses the aforementioned challenges through two key innovations: 1) Implementation of a Multi-Directional Safety Rectangular Corridor (MDSRC) algorithm, which encodes the free space through rectangular convex regions to avoid collision with static obstacles, eliminating redundant computational burdens and accelerating solver convergence; 2) A sequential MPC navigation framework that integrates corridor constraints with barrier function constraints is proposed to achieve static and dynamic obstacle avoidance. The ISMPC navigation framework enables direct velocity generation for AMRs, simplifying traditional navigation algorithm architectures. Comparative experiments demonstrate the framework's superiority in free-space utilization ( an increase of 41.05$\%$ in the average corridor area) while maintaining real-time computational performance (average corridors generation latency of 3 ms).
comment: 9 pages, 11 figures, conference paper for the 2025 International Conference on Advanced Robotics and Mechatronics (ICARM), accepted
☆ Differentiable Material Point Method for the Control of Deformable Objects
Controlling the deformation of flexible objects is challenging due to their non-linear dynamics and high-dimensional configuration space. This work presents a differentiable Material Point Method (MPM) simulator targeted at control applications. We exploit the differentiability of the simulator to optimize a control trajectory in an active damping problem for a hyperelastic rope. The simulator effectively minimizes the kinetic energy of the rope around 2$\times$ faster than a baseline MPPI method and to a 20% lower energy level, while using about 3% of the computation time.
comment: 7 Pages, 4 Figures, 1 Table
☆ ALBATROSS: A robotised system for high-throughput electrolyte screening via automated electrolyte formulation, coin-cell fabrication, and electrochemical evaluation
As battery technologies advance toward higher stability and energy density, the need for extensive cell-level testing across various component configurations becomes critical. To evaluate performance and understand the operating principles of batteries in laboratory scale, fabrication and evaluation of coin cells are essential processes. However, the conventional coin-cell assembly and testing processes require significant time and labor from researchers, posing challenges to high-throughput screening research. In this study, we introduce an Automated Li-ion BAttery Testing RObot SyStem (ALBATROSS), an automated system capable of electrolyte formulation, coin-cell assembly, and electrochemical evaluation. The system, integrated within a argon-filled glovebox, enables fully automated assembly and testing of up to 48 cells without researcher intervention. By incorporating custom-designed robot gripper and 3D-printed structures optimized for precise cell handling, ALBATROSS achieved high assembly reliability, yielding a relative standard deviation (RSD) of less than 1.2% in discharge capacity and a standard deviation of less than 3 Ω in EIS measurements for NCM811||Li half cells. Owing to its high reliability and automation capability, ALBATROSS allows for the acquisition of high-quality coin-cell datasets, which are expected to accelerate the development of next-generation electrolytes.
☆ Efficient Generation of Smooth Paths with Curvature Guarantees by Mollification
Most path following and trajectory tracking algorithms in mobile robotics require the desired path or trajectory to be defined by at least twice continuously differentiable functions to guarantee key properties such as global convergence, especially for nonholonomic robots like unicycles with speed constraints. Consequently, these algorithms typically exclude continuous but non-differentiable paths, such as piecewise functions. Despite this exclusion, such paths provide convenient high-level inputs for describing robot missions or behavior. While techniques such as spline interpolation or optimization-based methods are commonly used to smooth non-differentiable paths or create feasible ones from sequences of waypoints, they either can produce unnecessarily complex trajectories or are computationally expensive. In this work, we present a method to regularize non-differentiable functions and generate feasible paths through mollification. Specifically, we approximate an arbitrary path with a differentiable function that can converge to it with arbitrary precision. Additionally, we provide a systematic method for bounding the curvature of generated paths, which we demonstrate by applying it to paths resulting from linking a sequence of waypoints with segments. The proposed approach is computationally efficient, enabling real-time implementation on microcontrollers and compatibility with standard trajectory tracking and path following algorithms.
☆ MMDrive: Interactive Scene Understanding Beyond Vision with Multi-representational Fusion
Vision-language models enable the understanding and reasoning of complex traffic scenarios through multi-source information fusion, establishing it as a core technology for autonomous driving. However, existing vision-language models are constrained by the image understanding paradigm in 2D plane, which restricts their capability to perceive 3D spatial information and perform deep semantic fusion, resulting in suboptimal performance in complex autonomous driving environments. This study proposes MMDrive, an multimodal vision-language model framework that extends traditional image understanding to a generalized 3D scene understanding framework. MMDrive incorporates three complementary modalities, including occupancy maps, LiDAR point clouds, and textual scene descriptions. To this end, it introduces two novel components for adaptive cross-modal fusion and key information extraction. Specifically, the Text-oriented Multimodal Modulator dynamically weights the contributions of each modality based on the semantic cues in the question, guiding context-aware feature integration. The Cross-Modal Abstractor employs learnable abstract tokens to generate compact, cross-modal summaries that highlight key regions and essential semantics. Comprehensive evaluations on the DriveLM and NuScenes-QA benchmarks demonstrate that MMDrive achieves significant performance gains over existing vision-language models for autonomous driving, with a BLEU-4 score of 54.56 and METEOR of 41.78 on DriveLM, and an accuracy score of 62.7% on NuScenes-QA. MMDrive effectively breaks the traditional image-only understanding barrier, enabling robust multimodal reasoning in complex driving environments and providing a new foundation for interpretable autonomous driving scene understanding.
☆ Iterative Tuning of Nonlinear Model Predictive Control for Robotic Manufacturing Tasks
Manufacturing processes are often perturbed by drifts in the environment and wear in the system, requiring control re-tuning even in the presence of repetitive operations. This paper presents an iterative learning framework for automatic tuning of Nonlinear Model Predictive Control (NMPC) weighting matrices based on task-level performance feedback. Inspired by norm-optimal Iterative Learning Control (ILC), the proposed method adaptively adjusts NMPC weights Q and R across task repetitions to minimize key performance indicators (KPIs) related to tracking accuracy, control effort, and saturation. Unlike gradient-based approaches that require differentiating through the NMPC solver, we construct an empirical sensitivity matrix, enabling structured weight updates without analytic derivatives. The framework is validated through simulation on a UR10e robot performing carbon fiber winding on a tetrahedral core. Results demonstrate that the proposed approach converges to near-optimal tracking performance (RMSE within 0.3% of offline Bayesian Optimization (BO)) in just 4 online repetitions, compared to 100 offline evaluations required by BO algorithm. The method offers a practical solution for adaptive NMPC tuning in repetitive robotic tasks, combining the precision of carefully optimized controllers with the flexibility of online adaptation.
☆ START: Traversing Sparse Footholds with Terrain Reconstruction
Traversing terrains with sparse footholds like legged animals presents a promising yet challenging task for quadruped robots, as it requires precise environmental perception and agile control to secure safe foot placement while maintaining dynamic stability. Model-based hierarchical controllers excel in laboratory settings, but suffer from limited generalization and overly conservative behaviors. End-to-end learning-based approaches unlock greater flexibility and adaptability, but existing state-of-the-art methods either rely on heightmaps that introduce noise and complex, costly pipelines, or implicitly infer terrain features from egocentric depth images, often missing accurate critical geometric cues and leading to inefficient learning and rigid gaits. To overcome these limitations, we propose START, a single-stage learning framework that enables agile, stable locomotion on highly sparse and randomized footholds. START leverages only low-cost onboard vision and proprioception to accurately reconstruct local terrain heightmap, providing an explicit intermediate representation to convey essential features relevant to sparse foothold regions. This supports comprehensive environmental understanding and precise terrain assessment, reducing exploration cost and accelerating skill acquisition. Experimental results demonstrate that START achieves zero-shot transfer across diverse real-world scenarios, showcasing superior adaptability, precise foothold placement, and robust locomotion.
☆ OXE-AugE: A Large-Scale Robot Augmentation of OXE for Scaling Cross-Embodiment Policy Learning
Large and diverse datasets are needed for training generalist robot policies that have potential to control a variety of robot embodiments -- robot arm and gripper combinations -- across diverse tasks and environments. As re-collecting demonstrations and retraining for each new hardware platform are prohibitively costly, we show that existing robot data can be augmented for transfer and generalization. The Open X-Embodiment (OXE) dataset, which aggregates demonstrations from over 60 robot datasets, has been widely used as the foundation for training generalist policies. However, it is highly imbalanced: the top four robot types account for over 85\% of its real data, which risks overfitting to robot-scene combinations. We present AugE-Toolkit, a scalable robot augmentation pipeline, and OXE-AugE, a high-quality open-source dataset that augments OXE with 9 different robot embodiments. OXE-AugE provides over 4.4 million trajectories, more than triple the size of the original OXE. We conduct a systematic study of how scaling robot augmentation impacts cross-embodiment learning. Results suggest that augmenting datasets with diverse arms and grippers improves policy performance not only on the augmented robots, but also on unseen robots and even the original robots under distribution shifts. In physical experiments, we demonstrate that state-of-the-art generalist policies such as OpenVLA and $π_0$ benefit from fine-tuning on OXE-AugE, improving success rates by 24-45% on previously unseen robot-gripper combinations across four real-world manipulation tasks. Project website: https://OXE-AugE.github.io/.
☆ Sequence of Expert: Boosting Imitation Planners for Autonomous Driving through Temporal Alternation
Imitation learning (IL) has emerged as a central paradigm in autonomous driving. While IL excels in matching expert behavior in open-loop settings by minimizing per-step prediction errors, its performance degrades unexpectedly in closed-loop due to the gradual accumulation of small, often imperceptible errors over time.Over successive planning cycles, these errors compound, potentially resulting in severe failures.Current research efforts predominantly rely on increasingly sophisticated network architectures or high-fidelity training datasets to enhance the robustness of IL planners against error accumulation, focusing on the state-level robustness at a single time point. However, autonomous driving is inherently a continuous-time process, and leveraging the temporal scale to enhance robustness may provide a new perspective for addressing this issue.To this end, we propose a method termed Sequence of Experts (SoE), a temporal alternation policy that enhances closed-loop performance without increasing model size or data requirements. Our experiments on large-scale autonomous driving benchmarks nuPlan demonstrate that SoE method consistently and significantly improves the performance of all the evaluated models, and achieves state-of-the-art performance.This module may provide a key and widely applicable support for improving the training efficiency of autonomous driving models.
☆ PvP: Data-Efficient Humanoid Robot Learning with Proprioceptive-Privileged Contrastive Representations
Achieving efficient and robust whole-body control (WBC) is essential for enabling humanoid robots to perform complex tasks in dynamic environments. Despite the success of reinforcement learning (RL) in this domain, its sample inefficiency remains a significant challenge due to the intricate dynamics and partial observability of humanoid robots. To address this limitation, we propose PvP, a Proprioceptive-Privileged contrastive learning framework that leverages the intrinsic complementarity between proprioceptive and privileged states. PvP learns compact and task-relevant latent representations without requiring hand-crafted data augmentations, enabling faster and more stable policy learning. To support systematic evaluation, we develop SRL4Humanoid, the first unified and modular framework that provides high-quality implementations of representative state representation learning (SRL) methods for humanoid robot learning. Extensive experiments on the LimX Oli robot across velocity tracking and motion imitation tasks demonstrate that PvP significantly improves sample efficiency and final performance compared to baseline SRL methods. Our study further provides practical insights into integrating SRL with RL for humanoid WBC, offering valuable guidance for data-efficient humanoid robot learning.
comment: 13 pages, 12 figures
☆ Multi-Robot Motion Planning from Vision and Language using Heat-Inspired Diffusion
Diffusion models have recently emerged as powerful tools for robot motion planning by capturing the multi-modal distribution of feasible trajectories. However, their extension to multi-robot settings with flexible, language-conditioned task specifications remains limited. Furthermore, current diffusion-based approaches incur high computational cost during inference and struggle with generalization because they require explicit construction of environment representations and lack mechanisms for reasoning about geometric reachability. To address these limitations, we present Language-Conditioned Heat-Inspired Diffusion (LCHD), an end-to-end vision-based framework that generates language-conditioned, collision-free trajectories. LCHD integrates CLIP-based semantic priors with a collision-avoiding diffusion kernel serving as a physical inductive bias that enables the planner to interpret language commands strictly within the reachable workspace. This naturally handles out-of-distribution scenarios -- in terms of reachability -- by guiding robots toward accessible alternatives that match the semantic intent, while eliminating the need for explicit obstacle information at inference time. Extensive evaluations on diverse real-world-inspired maps, along with real-robot experiments, show that LCHD consistently outperforms prior diffusion-based planners in success rate, while reducing planning latency.
☆ Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos
Vision-Language-Action (VLA) models provide a promising paradigm for robot learning by integrating visual perception with language-guided policy learning. However, most existing approaches rely on 2D visual inputs to perform actions in 3D physical environments, creating a significant gap between perception and action grounding. To bridge this gap, we propose a Spatial-Aware VLA Pretraining paradigm that performs explicit alignment between visual space and physical space during pretraining, enabling models to acquire 3D spatial understanding before robot policy learning. Starting from pretrained vision-language models, we leverage large-scale human demonstration videos to extract 3D visual and 3D action annotations, forming a new source of supervision that aligns 2D visual observations with 3D spatial reasoning. We instantiate this paradigm with VIPA-VLA, a dual-encoder architecture that incorporates a 3D visual encoder to augment semantic visual representations with 3D-aware features. When adapted to downstream robot tasks, VIPA-VLA achieves significantly improved grounding between 2D vision and 3D action, resulting in more robust and generalizable robotic policies.
☆ Motus: A Unified Latent Action World Model
While a general embodied agent must function as a unified system, current methods are built on isolated models for understanding, world modeling, and control. This fragmentation prevents unifying multimodal generative capabilities and hinders learning from large-scale, heterogeneous data. In this paper, we propose Motus, a unified latent action world model that leverages existing general pretrained models and rich, sharable motion information. Motus introduces a Mixture-of-Transformer (MoT) architecture to integrate three experts (i.e., understanding, video generation, and action) and adopts a UniDiffuser-style scheduler to enable flexible switching between different modeling modes (i.e., world models, vision-language-action models, inverse dynamics models, video generation models, and video-action joint prediction models). Motus further leverages the optical flow to learn latent actions and adopts a recipe with three-phase training pipeline and six-layer data pyramid, thereby extracting pixel-level "delta action" and enabling large-scale action pretraining. Experiments show that Motus achieves superior performance against state-of-the-art methods in both simulation (a +15% improvement over X-VLA and a +45% improvement over Pi0.5) and real-world scenarios(improved by +11~48%), demonstrating unified modeling of all functionalities and priors significantly benefits downstream robotic tasks.
☆ K-VARK: Kernelized Variance-Aware Residual Kalman Filter for Sensorless Force Estimation in Collaborative Robots
Reliable estimation of contact forces is crucial for ensuring safe and precise interaction of robots with unstructured environments. However, accurate sensorless force estimation remains challenging due to inherent modeling errors and complex residual dynamics and friction. To address this challenge, in this paper, we propose K-VARK (Kernelized Variance-Aware Residual Kalman filter), a novel approach that integrates a kernelized, probabilistic model of joint residual torques into an adaptive Kalman filter framework. Through Kernelized Movement Primitives trained on optimized excitation trajectories, K-VARK captures both the predictive mean and input-dependent heteroscedastic variance of residual torques, reflecting data variability and distance-to-training effects. These statistics inform a variance-aware virtual measurement update by augmenting the measurement noise covariance, while the process noise covariance adapts online via variational Bayesian optimization to handle dynamic disturbances. Experimental validation on a 6-DoF collaborative manipulator demonstrates that K-VARK achieves over 20% reduction in RMSE compared to state-of-the-art sensorless force estimation methods, yielding robust and accurate external force/torque estimation suitable for advanced tasks such as polishing and assembly.
☆ Learning Terrain Aware Bipedal Locomotion via Reduced Dimensional Perceptual Representations
This work introduces a hierarchical strategy for terrain-aware bipedal locomotion that integrates reduced-dimensional perceptual representations to enhance reinforcement learning (RL)-based high-level (HL) policies for real-time gait generation. Unlike end-to-end approaches, our framework leverages latent terrain encodings via a Convolutional Variational Autoencoder (CNN-VAE) alongside reduced-order robot dynamics, optimizing the locomotion decision process with a compact state. We systematically analyze the impact of latent space dimensionality on learning efficiency and policy robustness. Additionally, we extend our method to be history-aware, incorporating sequences of recent terrain observations into the latent representation to improve robustness. To address real-world feasibility, we introduce a distillation method to learn the latent representation directly from depth camera images and provide preliminary hardware validation by comparing simulated and real sensor data. We further validate our framework using the high-fidelity Agility Robotics (AR) simulator, incorporating realistic sensor noise, state estimation, and actuator dynamics. The results confirm the robustness and adaptability of our method, underscoring its potential for hardware deployment.
☆ Tackling Snow-Induced Challenges: Safe Autonomous Lane-Keeping with Robust Reinforcement Learning
This paper proposes two new algorithms for the lane keeping system (LKS) in autonomous vehicles (AVs) operating under snowy road conditions. These algorithms use deep reinforcement learning (DRL) to handle uncertainties and slippage. They include Action-Robust Recurrent Deep Deterministic Policy Gradient (AR-RDPG) and end-to-end Action-Robust convolutional neural network Attention Deterministic Policy Gradient (AR-CADPG), two action-robust approaches for decision-making. In the AR-RDPG method, within the perception layer, camera images are first denoised using multi-scale neural networks. Then, the centerline coefficients are extracted by a pre-trained deep convolutional neural network (DCNN). These coefficients, concatenated with the driving characteristics, are used as input to the control layer. The AR-CADPG method presents an end-to-end approach in which a convolutional neural network (CNN) and an attention mechanism are integrated within a DRL framework. Both methods are first trained in the CARLA simulator and validated under various snowy scenarios. Real-world experiments on a Jetson Nano-based autonomous vehicle confirm the feasibility and stability of the learned policies. Among the two models, the AR-CADPG approach demonstrates superior path-tracking accuracy and robustness, highlighting the effectiveness of combining temporal memory, adversarial resilience, and attention mechanisms in AVs.
☆ SLIM-VDB: A Real-Time 3D Probabilistic Semantic Mapping Framework
This paper introduces SLIM-VDB, a new lightweight semantic mapping system with probabilistic semantic fusion for closed-set or open-set dictionaries. Advances in data structures from the computer graphics community, such as OpenVDB, have demonstrated significantly improved computational and memory efficiency in volumetric scene representation. Although OpenVDB has been used for geometric mapping in robotics applications, semantic mapping for scene understanding with OpenVDB remains unexplored. In addition, existing semantic mapping systems lack support for integrating both fixed-category and open-language label predictions within a single framework. In this paper, we propose a novel 3D semantic mapping system that leverages the OpenVDB data structure and integrates a unified Bayesian update framework for both closed- and open-set semantic fusion. Our proposed framework, SLIM-VDB, achieves significant reduction in both memory and integration times compared to current state-of-the-art semantic mapping approaches, while maintaining comparable mapping accuracy. An open-source C++ codebase with a Python interface is available at https://github.com/umfieldrobotics/slim-vdb.
comment: Accepted into R-AL
♻ ☆ Active 6D Pose Estimation for Textureless Objects using Multi-View RGB Frames
Estimating the 6D pose of textureless objects from RGB images is an important problem in robotics. Due to appearance ambiguities, rotational symmetries, and severe occlusions, single-view based 6D pose estimators are still unable to handle a wide range of objects, motivating research towards multi-view pose estimation and next-best-view prediction that addresses these limitations. In this work, we propose a comprehensive active perception framework for estimating the 6D poses of textureless objects using only RGB images. Our approach is built upon a key idea: decoupling the 6D pose estimation into a two-step sequential process can greatly improve both accuracy and efficiency. First, we estimate the 3D translation of each object, resolving scale and depth ambiguities inherent to RGB images. These estimates are then used to simplify the subsequent task of determining the 3D orientation, which we achieve through canonical scale template matching. Building on this formulation, we then introduce an active perception strategy that predicts the next best camera viewpoint to capture an RGB image, effectively reducing object pose uncertainty and enhancing pose accuracy. We evaluate our method on the public ROBI and TOD datasets, as well as on our reconstructed transparent object dataset, T-ROBI. Under the same camera viewpoints, our multi-view pose estimation significantly outperforms state-of-the-art approaches. Furthermore, by leveraging our next-best-view strategy, our approach achieves high pose accuracy with fewer viewpoints than heuristic-based policies across all evaluated datasets. The accompanying video and T-ROBI dataset will be released on our project page: https://trailab.github.io/ActiveODPE.
♻ ☆ Accelerated Rotation-Invariant Convolution for UAV Image Segmentation
Rotation invariance is essential for precise, object-level segmentation in UAV aerial imagery, where targets can have arbitrary orientations and exhibit fine-scale details. Conventional segmentation architectures like U-Net rely on convolution operators that are not rotation-invariant, leading to degraded segmentation accuracy across varying viewpoints. Rotation invariance can be achieved by expanding the filter bank across multiple orientations; however, this will significantly increase computational cost and memory traffic. In this paper, we introduce a GPU-optimized rotation-invariant convolution framework that eliminates the traditional data-lowering (im2col) step required for matrix-multiplication-based convolution. By exploiting structured data sharing among symmetrically rotated filters, our method achieves multi-orientation convolution with greatly reduced memory traffic and computational redundancy. We further generalize the approach to accelerate convolution with arbitrary (non-symmetric) rotation angles. Across extensive benchmarks, the proposed convolution achieves 20--55% faster training and 15--45% lower energy consumption than CUDNN, while maintaining accuracy comparable to state-of-the-art rotation-invariant methods. In the eight-orientation setting, our approach achieves up to 45% speedup and 41% energy savings on 256\(\times\)256 inputs, and 32% speedup and 23% lower energy usage on 1024\(\times\)1024 inputs. Integrated into a U-Net segmentation model, the framework yields up to 6% improvement in accuracy over the non-rotation-aware baseline. These results demonstrate that the proposed method provides an effective and highly efficient alternative to existing rotation-invariant CNN frameworks.
♻ ☆ RoboCOIN: An Open-Sourced Bimanual Robotic Data COllection for INtegrated Manipulation
Bimanual manipulation is essential for achieving human-like dexterity in robots, but the large-scale and diverse bimanual robot datasets remain scarce due to hardware heterogeneity across robotic platforms. To address the challenge, we present RoboCOIN, a comprehensive multi-embodiment bimanual manipulation dataset with over 180,000 demonstrations collected from 15 distinct robotic platforms. The dataset covers 16 scenarios, including residential, commercial, and working environments, with 421 tasks systematically organized by bimanual coordination patterns and object properties. Our key innovation is a hierarchical capability pyramid that provides multi-level annotations, spanning trajectory-level concepts, segment-level subtasks, and frame-level kinematics. We further develop CoRobot, a comprehensive processing framework featuring Robot Trajectory Markup Language (RTML) for quality assessment, automated annotation generation, and unified multi-embodiment management. Extensive experiments demonstrate the reliability and effectiveness of RoboCOIN in multi-embodiment bimanual learning, with significant performance improvements across various model architectures and robotic platforms. The complete dataset and framework are open-sourced and publicly available for further research purposes. Project website: https://FlagOpen.github.io/RoboCOIN/.
comment: Fixed typos
♻ ☆ CT-UIO: Continuous-Time UWB-Inertial-Odometer Localization Using Non-Uniform B-spline with Fewer Anchors
Ultra-wideband (UWB) based positioning with fewer anchors has attracted significant research interest in recent years, especially under energy-constrained conditions. However, most existing methods rely on discrete-time representations and smoothness priors to infer a robot's motion states, which often struggle with ensuring multi-sensor data synchronization. In this article, we present a continuous-time UWB-Inertial-Odometer localization system (CT-UIO), utilizing a non-uniform B-spline framework with fewer anchors. Unlike traditional uniform B-spline-based continuous-time methods, we introduce an adaptive knot-span adjustment strategy for non-uniform continuous-time trajectory representation. This is accomplished by adjusting control points dynamically based on movement speed. To enable efficient fusion of {inertial measurement unit (IMU) and odometer data, we propose an improved extended Kalman filter (EKF) with innovation-based adaptive estimation to provide short-term accurate motion prior. Furthermore, to address the challenge of achieving a fully observable UWB localization system under few-anchor conditions, the virtual anchor (VA) generation method based on multiple hypotheses is proposed. At the backend, we propose an adaptive sliding window strategy for global trajectory estimation. Comprehensive experiments are conducted on three self-collected datasets with different UWB anchor numbers and motion modes. The result shows that the proposed CT-UIO achieves 0.403m, 0.150m, and 0.189m localization accuracy in corridor, exhibition hall, and office environments, yielding 17.2%, 26.1%, and 15.2% improvements compared with competing state-of-the-art UIO systems, respectively. The codebase and datasets of this work will be open-sourced at https://github.com/JasonSun623/CT-UIO.
comment: Accepted to IEEE Transactions on Mobile Computing (TMC). The codebase and datasets will be open-sourced at https://github.com/JasonSun623/CT-UIO
♻ ☆ Enhancing Interpretability and Interactivity in Robot Manipulation: A Neurosymbolic Approach
In this paper we present a neurosymbolic architecture for coupling language-guided visual reasoning with robot manipulation. A non-expert human user can prompt the robot using unconstrained natural language, providing a referring expression (REF), a question (VQA), or a grasp action instruction. The system tackles all cases in a task-agnostic fashion through the utilization of a shared library of primitive skills. Each primitive handles an independent sub-task, such as reasoning about visual attributes, spatial relation comprehension, logic and enumeration, as well as arm control. A language parser maps the input query to an executable program composed of such primitives, depending on the context. While some primitives are purely symbolic operations (e.g. counting), others are trainable neural functions (e.g. visual grounding), therefore marrying the interpretability and systematic generalization benefits of discrete symbolic approaches with the scalability and representational power of deep networks. We generate a 3D vision-and-language synthetic dataset of tabletop scenes in a simulation environment to train our approach and perform extensive evaluations in both synthetic and real-world scenes. Results showcase the benefits of our approach in terms of accuracy, sample-efficiency, and robustness to the user's vocabulary, while being transferable to real-world scenes with few-shot visual fine-tuning. Finally, we integrate our method with a robot framework and demonstrate how it can serve as an interpretable solution for an interactive object-picking task, achieving an average success rate of 80.2\%, both in simulation and with a real robot. We make supplementary material available in https://gtziafas.github.io/neurosymbolic-manipulation.
comment: Published in International Journal of Robotics Research (IJRR) (2025)
♻ ☆ 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
♻ ☆ WholeBodyVLA: Towards Unified Latent VLA for Whole-Body Loco-Manipulation Control
Humanoid robots require precise locomotion and dexterous manipulation to perform challenging loco-manipulation tasks. Yet existing approaches, modular or end-to-end, are deficient in manipulation-aware locomotion. This confines the robot to a limited workspace, preventing it from performing large-space loco-manipulation. We attribute this to: (1) the challenge of acquiring loco-manipulation knowledge due to the scarcity of humanoid teleoperation data, and (2) the difficulty of faithfully and reliably executing locomotion commands, stemming from the limited precision and stability of existing RL controllers. To acquire richer loco-manipulation knowledge, we propose a unified latent learning framework that enables Vision-Language-Action (VLA) system to learn from low-cost action-free egocentric videos. Moreover, an efficient human data collection pipeline is devised to augment the dataset and scale the benefits. To execute the desired locomotion commands more precisely, we present a loco-manipulation-oriented (LMO) RL policy specifically tailored for accurate and stable core loco-manipulation movements, such as advancing, turning, and squatting. Building on these components, we introduce WholeBodyVLA, a unified framework for humanoid loco-manipulation. To the best of our knowledge, WholeBodyVLA is one of its kind enabling large-space humanoid loco-manipulation. It is verified via comprehensive experiments on the AgiBot X2 humanoid, outperforming prior baseline by 21.3%. It also demonstrates strong generalization and high extensibility across a broad range of tasks.
♻ ☆ HACTS: a Human-As-Copilot Teleoperation System for Robot Learning
Teleoperation is essential for autonomous robot learning, especially in manipulation tasks that require human demonstrations or corrections. However, most existing systems only offer unilateral robot control and lack the ability to synchronize the robot's status with the teleoperation hardware, preventing real-time, flexible intervention. In this work, we introduce HACTS (Human-As-Copilot Teleoperation System), a novel system that establishes bilateral, real-time joint synchronization between a robot arm and teleoperation hardware. This simple yet effective feedback mechanism, akin to a steering wheel in autonomous vehicles, enables the human copilot to intervene seamlessly while collecting action-correction data for future learning. Implemented using 3D-printed components and low-cost, off-the-shelf motors, HACTS is both accessible and scalable. Our experiments show that HACTS significantly enhances performance in imitation learning (IL) and reinforcement learning (RL) tasks, boosting IL recovery capabilities and data efficiency, and facilitating human-in-the-loop RL. HACTS paves the way for more effective and interactive human-robot collaboration and data-collection, advancing the capabilities of robot manipulation.
♻ ☆ DRO-EDL-MPC: Evidential Deep Learning-Based Distributionally Robust Model Predictive Control for Safe Autonomous Driving
Safety is a critical concern in motion planning for autonomous vehicles. Modern autonomous vehicles rely on neural network-based perception, but making control decisions based on these inference results poses significant safety risks due to inherent uncertainties. To address this challenge, we present a distributionally robust optimization (DRO) framework that accounts for both aleatoric and epistemic perception uncertainties using evidential deep learning (EDL). Our approach introduces a novel ambiguity set formulation based on evidential distributions that dynamically adjusts the conservativeness according to perception confidence levels. We integrate this uncertainty-aware constraint into model predictive control (MPC), proposing the DRO-EDL-MPC algorithm with computational tractability for autonomous driving applications. Validation in the CARLA simulator demonstrates that our approach maintains efficiency under high perception confidence while enforcing conservative constraints under low confidence.
comment: Accepted to IEEE Robotics and Automation Letters (RA-L)
♻ ☆ DualMap: Online Open-Vocabulary Semantic Mapping for Natural Language Navigation in Dynamic Changing Scenes
We introduce DualMap, an online open-vocabulary mapping system that enables robots to understand and navigate dynamically changing environments through natural language queries. Designed for efficient semantic mapping and adaptability to changing environments, DualMap meets the essential requirements for real-world robot navigation applications. Our proposed hybrid segmentation frontend and object-level status check eliminate the costly 3D object merging required by prior methods, enabling efficient online scene mapping. The dual-map representation combines a global abstract map for high-level candidate selection with a local concrete map for precise goal-reaching, effectively managing and updating dynamic changes in the environment. Through extensive experiments in both simulation and real-world scenarios, we demonstrate state-of-the-art performance in 3D open-vocabulary segmentation, efficient scene mapping, and online language-guided navigation. Project page: https://eku127.github.io/DualMap/
comment: 14 pages, 14 figures. Published in IEEE Robotics and Automation Letters (RA-L), 2025. Code: https://github.com/Eku127/DualMap Project page: https://eku127.github.io/DualMap/
♻ ☆ MDE-AgriVLN: Agricultural Vision-and-Language Navigation with Monocular Depth Estimation
Agricultural robots are serving as powerful assistants across a wide range of agricultural tasks, nevertheless, still heavily relying on manual operations or railway systems for movement. The AgriVLN method and the A2A benchmark pioneeringly extended Vision-and-Language Navigation (VLN) to the agricultural domain, enabling a robot to navigate to a target position following a natural language instruction. Unlike human binocular vision, most agricultural robots are only given a single camera for monocular vision, which results in limited spatial perception. To bridge this gap, we present the method of Agricultural Vision-and-Language Navigation with Monocular Depth Estimation (MDE-AgriVLN), in which we propose the MDE module generating depth features from RGB images, to assist the decision-maker on multimodal reasoning. When evaluated on the A2A benchmark, our MDE-AgriVLN method successfully increases Success Rate from 0.23 to 0.32 and decreases Navigation Error from 4.43m to 4.08m, demonstrating the state-of-the-art performance in the agricultural VLN domain. Code: https://github.com/AlexTraveling/MDE-AgriVLN.
♻ ☆ SoMi-ToM: Evaluating Multi-Perspective Theory of Mind in Embodied Social Interactions
Humans continuously infer the states, goals, and behaviors of others by perceiving their surroundings in dynamic, real-world social interactions. However, most Theory of Mind (ToM) benchmarks only evaluate static, text-based scenarios, which have a significant gap compared to real interactions. We propose the SoMi-ToM benchmark, designed to evaluate multi-perspective ToM in embodied multi-agent complex social interactions. This benchmark is based on rich multimodal interaction data generated by the interaction environment SoMi, covering diverse crafting goals and social relationships. Our framework supports multi-level evaluation: (1) first-person evaluation provides multimodal (visual, dialogue, action, etc.) input from a first-person perspective during a task for real-time state inference, (2) third-person evaluation provides complete third-person perspective video and text records after a task for goal and behavior inference. This evaluation method allows for a more comprehensive examination of a model's ToM capabilities from both the subjective immediate experience and the objective global observation. We constructed a challenging dataset containing 35 third-person perspective videos, 363 first-person perspective images, and 1225 expert-annotated multiple-choice questions (three options). On this dataset, we systematically evaluated the performance of human subjects and several state-of-the-art large vision-language models (LVLMs). The results show that LVLMs perform significantly worse than humans on SoMi-ToM: the average accuracy gap between humans and models is 40.1% in first-person evaluation and 26.4% in third-person evaluation. This indicates that future LVLMs need to further improve their ToM capabilities in embodied, complex social interactions.
comment: 24 pages, 6 figures
♻ ☆ MSG-Loc: Multi-Label Likelihood-based Semantic Graph Matching for Object-Level Global Localization
Robots are often required to localize in environments with unknown object classes and semantic ambiguity. However, when performing global localization using semantic objects, high semantic ambiguity intensifies object misclassification and increases the likelihood of incorrect associations, which in turn can cause significant errors in the estimated pose. Thus, in this letter, we propose a multi-label likelihood-based semantic graph matching framework for object-level global localization. The key idea is to exploit multi-label graph representations, rather than single-label alternatives, to capture and leverage the inherent semantic context of object observations. Based on these representations, our approach enhances semantic correspondence across graphs by combining the likelihood of each node with the maximum likelihood of its neighbors via context-aware likelihood propagation. For rigorous validation, data association and pose estimation performance are evaluated under both closed-set and open-set detection configurations. In addition, we demonstrate the scalability of our approach to large-vocabulary object categories in both real-world indoor scenes and synthetic environments.
comment: Accepted in IEEE Robotics and Automation Letters (2025)
♻ ☆ ESPADA: Execution Speedup via Semantics Aware Demonstration Data Downsampling for Imitation Learning
Behavior-cloning based visuomotor policies enable precise manipulation but often inherit the slow, cautious tempo of human demonstrations, limiting practical deployment. However, prior studies on acceleration methods mainly rely on statistical or heuristic cues that ignore task semantics and can fail across diverse manipulation settings. We present ESPADA, a semantic and spatially aware framework that segments demonstrations using a VLM-LLM pipeline with 3D gripper-object relations, enabling aggressive downsampling only in non-critical segments while preserving precision-critical phases, without requiring extra data or architectural modifications, or any form of retraining. To scale from a single annotated episode to the full dataset, ESPADA propagates segment labels via Dynamic Time Warping (DTW) on dynamics-only features. Across both simulation and real-world experiments with ACT and DP baselines, ESPADA achieves approximately a 2x speed-up while maintaining success rates, narrowing the gap between human demonstrations and efficient robot control.
comment: project page: https://project-espada.github.io/espada/
Robotics 18
☆ SignRAG: A Retrieval-Augmented System for Scalable Zero-Shot Road Sign Recognition
Automated road sign recognition is a critical task for intelligent transportation systems, but traditional deep learning methods struggle with the sheer number of sign classes and the impracticality of creating exhaustive labeled datasets. This paper introduces a novel zero-shot recognition framework that adapts the Retrieval-Augmented Generation (RAG) paradigm to address this challenge. Our method first uses a Vision Language Model (VLM) to generate a textual description of a sign from an input image. This description is used to retrieve a small set of the most relevant sign candidates from a vector database of reference designs. Subsequently, a Large Language Model (LLM) reasons over the retrieved candidates to make a final, fine-grained recognition. We validate this approach on a comprehensive set of 303 regulatory signs from the Ohio MUTCD. Experimental results demonstrate the framework's effectiveness, achieving 95.58% accuracy on ideal reference images and 82.45% on challenging real-world road data. This work demonstrates the viability of RAG-based architectures for creating scalable and accurate systems for road sign recognition without task-specific training.
comment: Submitted to IV 2026
☆ Cross-Level Sensor Fusion with Object Lists via Transformer for 3D Object Detection
In automotive sensor fusion systems, smart sensors and Vehicle-to-Everything (V2X) modules are commonly utilized. Sensor data from these systems are typically available only as processed object lists rather than raw sensor data from traditional sensors. Instead of processing other raw data separately and then fusing them at the object level, we propose an end-to-end cross-level fusion concept with Transformer, which integrates highly abstract object list information with raw camera images for 3D object detection. Object lists are fed into a Transformer as denoising queries and propagated together with learnable queries through the latter feature aggregation process. Additionally, a deformable Gaussian mask, derived from the positional and size dimensional priors from the object lists, is explicitly integrated into the Transformer decoder. This directs attention toward the target area of interest and accelerates model training convergence. Furthermore, as there is no public dataset containing object lists as a standalone modality, we propose an approach to generate pseudo object lists from ground-truth bounding boxes by simulating state noise and false positives and negatives. As the first work to conduct cross-level fusion, our approach shows substantial performance improvements over the vision-based baseline on the nuScenes dataset. It demonstrates its generalization capability over diverse noise levels of simulated object lists and real detectors.
comment: 6 pages, 3 figures, accepted at IV2025
☆ MPC-Guided Safe Reinforcement Learning and Lipschitz-Based Filtering for Structured Nonlinear Systems
Modern engineering systems, such as autonomous vehicles, flexible robotics, and intelligent aerospace platforms, require controllers that are robust to uncertainties, adaptive to environmental changes, and safety-aware under real-time constraints. RL offers powerful data-driven adaptability for systems with nonlinear dynamics that interact with uncertain environments. RL, however, lacks built-in mechanisms for dynamic constraint satisfaction during exploration. MPC offers structured constraint handling and robustness, but its reliance on accurate models and computationally demanding online optimization may pose significant challenges. This paper proposes an integrated MPC-RL framework that combines stability and safety guarantees of MPC with the adaptability of RL. During training, MPC defines safe control bounds that guide the RL component and that enable constraint-aware policy learning. At deployment, the learned policy operates in real time with a lightweight safety filter based on Lipschitz continuity to ensure constraint satisfaction without heavy online optimizations. The approach, which is validated on a nonlinear aeroelastic wing system, demonstrates improved disturbance rejection, reduced actuator effort, and robust performance under turbulence. The architecture generalizes to other domains with structured nonlinearities and bounded disturbances, offering a scalable solution for safe artificial-intelligence-driven control in engineering applications.
☆ SAGA: Open-World Mobile Manipulation via Structured Affordance Grounding
We present SAGA, a versatile and adaptive framework for visuomotor control that can generalize across various environments, task objectives, and user specifications. To efficiently learn such capability, our key idea is to disentangle high-level semantic intent from low-level visuomotor control by explicitly grounding task objectives in the observed environment. Using an affordance-based task representation, we express diverse and complex behaviors in a unified, structured form. By leveraging multimodal foundation models, SAGA grounds the proposed task representation to the robot's visual observation as 3D affordance heatmaps, highlighting task-relevant entities while abstracting away spurious appearance variations that would hinder generalization. These grounded affordances enable us to effectively train a conditional policy on multi-task demonstration data for whole-body control. In a unified framework, SAGA can solve tasks specified in different forms, including language instructions, selected points, and example demonstrations, enabling both zero-shot execution and few-shot adaptation. We instantiate SAGA on a quadrupedal manipulator and conduct extensive experiments across eleven real-world tasks. SAGA consistently outperforms end-to-end and modular baselines by substantial margins. Together, these results demonstrate that structured affordance grounding offers a scalable and effective pathway toward generalist mobile manipulation.
comment: 9 pages, 7 figures
☆ VLG-Loc: Vision-Language Global Localization from Labeled Footprint Maps
This paper presents Vision-Language Global Localization (VLG-Loc), a novel global localization method that uses human-readable labeled footprint maps containing only names and areas of distinctive visual landmarks in an environment. While humans naturally localize themselves using such maps, translating this capability to robotic systems remains highly challenging due to the difficulty of establishing correspondences between observed landmarks and those in the map without geometric and appearance details. To address this challenge, VLG-Loc leverages a vision-language model (VLM) to search the robot's multi-directional image observations for the landmarks noted in the map. The method then identifies robot poses within a Monte Carlo localization framework, where the found landmarks are used to evaluate the likelihood of each pose hypothesis. Experimental validation in simulated and real-world retail environments demonstrates superior robustness compared to existing scan-based methods, particularly under environmental changes. Further improvements are achieved through the probabilistic fusion of visual and scan-based localization.
☆ High Order Control Lyapunov Function - Control Barrier Function - Quadratic Programming Based Autonomous Driving Controller for Bicyclist Safety
Ensuring the safety of Vulnerable Road Users (VRUs) is a critical challenge in the development of advanced autonomous driving systems in smart cities. Among vulnerable road users, bicyclists present unique characteristics that make their safety both critical and also manageable. Vehicles often travel at significantly higher relative speeds when interacting with bicyclists as compared to their interactions with pedestrians which makes collision avoidance system design for bicyclist safety more challenging. Yet, bicyclist movements are generally more predictable and governed by clear traffic rules as compared to the sudden and sometimes erratic pedestrian motion, offering opportunities for model-based control strategies. To address bicyclist safety in complex traffic environments, this study proposes and develops a High Order Control Lyapunov Function High Order Control Barrier Function Quadratic Programming (HOCLF HOCBF QP) control framework. Through this framework, CLFs constraints guarantee system stability so that the vehicle can track its reference trajectory, whereas CBFs constraints ensure system safety by letting vehicle avoiding potential collisions region with surrounding obstacles. Then by solving a QP problem, an optimal control command that simultaneously satisfies stability and safety requirements can be calculated. Three key bicyclist crash scenarios recorded in the Fatality Analysis Reporting System (FARS) are recreated and used to comprehensively evaluate the proposed autonomous driving bicyclist safety control strategy in a simulation study. Simulation results demonstrate that the HOCLF HOCBF QP controller can help the vehicle perform robust, and collision-free maneuvers, highlighting its potential for improving bicyclist safety in complex traffic environments.
☆ Making Robots Play by the Rules: The ROS 2 CLIPS-Executive
CLIPS is a rule-based programming language for building knowledge-driven applications, well suited for the complex task of coordinating autonomous robots. Inspired by the CLIPS-Executive originally developed for the lesser known Fawkes robotics framework, we present an Integration of CLIPS into the ROS ecosystem. Additionally, we show the flexibility of CLIPS by describing a PDDL-based planning framework integration.
☆ HMPCC: Human-Aware Model Predictive Coverage Control
We address the problem of coordinating a team of robots to cover an unknown environment while ensuring safe operation and avoiding collisions with non-cooperative agents. Traditional coverage strategies often rely on simplified assumptions, such as known or convex environments and static density functions, and struggle to adapt to real-world scenarios, especially when humans are involved. In this work, we propose a human-aware coverage framework based on Model Predictive Control (MPC), namely HMPCC, where human motion predictions are integrated into the planning process. By anticipating human trajectories within the MPC horizon, robots can proactively coordinate their actions %avoid redundant exploration, and adapt to dynamic conditions. The environment is modeled as a Gaussian Mixture Model (GMM), representing regions of interest. Team members operate in a fully decentralized manner, without relying on explicit communication, an essential feature in hostile or communication-limited scenarios. Our results show that human trajectory forecasting enables more efficient and adaptive coverage, improving coordination between human and robotic agents.
☆ Bayesian Optimization Parameter Tuning Framework for a Lyapunov Based Path Following Controller
Parameter tuning in real-world experiments is constrained by the limited evaluation budget available on hardware. The path-following controller studied in this paper reflects a typical situation in nonlinear geometric controller, where multiple gains influence the dynamics through coupled nonlinear terms. Such interdependence makes manual tuning inefficient and unlikely to yield satisfactory performance within a practical number of trials. To address this challenge, we propose a Bayesian optimization (BO) framework that treats the closed-loop system as a black box and selects controller gains using a Gaussian-process surrogate. BO offers model-free exploration, quantified uncertainty, and data-efficient search, making it well suited for tuning tasks where each evaluation is costly. The framework is implemented on Honda's AI-Formula three-wheeled robot and assessed through repeated full-lap experiments on a fixed test track. The results show that BO improves controller performance within 32 trials, including 15 warm-start initial evaluations, indicating that it can efficiently locate high-performing regions of the parameter space under real-world conditions. These findings demonstrate that BO provides a practical, reliable, and data-efficient tuning approach for nonlinear path-following controllers on real robotic platforms.
☆ Optimized Conflict Management for Urban Air Mobility Using Swarm UAV Networks
Urban Air Mobility (UAM) poses unprecedented traffic coordination challenges, especially with increasing UAV densities in dense urban corridors. This paper introduces a mathematical model using a control algorithm to optimize an Edge AI-driven decentralized swarm architecture for intelligent conflict resolution, enabling real-time decision-making with low latency. Using lightweight neural networks, the system leverages edge nodes to perform distributed conflict detection and resolution. A simulation platform was developed to evaluate the scheme under various UAV densities. Results indicate that the conflict resolution time is dramatically minimized up to 3.8 times faster, and accuracy is enhanced compared to traditional centralized control models. The proposed architecture is highly promising for scalable, efficient, and safe aerial traffic management in future UAM systems.
comment: Preprint. Under review for conference submission
☆ D3D-VLP: Dynamic 3D Vision-Language-Planning Model for Embodied Grounding and Navigation
Embodied agents face a critical dilemma that end-to-end models lack interpretability and explicit 3D reasoning, while modular systems ignore cross-component interdependencies and synergies. To bridge this gap, we propose the Dynamic 3D Vision-Language-Planning Model (D3D-VLP). Our model introduces two key innovations: 1) A Dynamic 3D Chain-of-Thought (3D CoT) that unifies planning, grounding, navigation, and question answering within a single 3D-VLM and CoT pipeline; 2) A Synergistic Learning from Fragmented Supervision (SLFS) strategy, which uses a masked autoregressive loss to learn from massive and partially-annotated hybrid data. This allows different CoT components to mutually reinforce and implicitly supervise each other. To this end, we construct a large-scale dataset with 10M hybrid samples from 5K real scans and 20K synthetic scenes that are compatible with online learning methods such as RL and DAgger. Our D3D-VLP achieves state-of-the-art results on multiple benchmarks, including Vision-and-Language Navigation (R2R-CE, REVERIE-CE, NavRAG-CE), Object-goal Navigation (HM3D-OVON), and Task-oriented Sequential Grounding and Navigation (SG3D). Real-world mobile manipulation experiments further validate the effectiveness.
♻ ☆ LocoMamba: Vision-Driven Locomotion via End-to-End Deep Reinforcement Learning with Mamba
We introduce LocoMamba, a vision-driven cross-modal DRL framework built on selective state-space models, specifically leveraging Mamba, that achieves near-linear-time sequence modeling, effectively captures long-range dependencies, and enables efficient training with longer sequences. First, we embed proprioceptive states with a multilayer perceptron and patchify depth images with a lightweight convolutional neural network, producing compact tokens that improve state representation. Second, stacked Mamba layers fuse these tokens via near-linear-time selective scanning, reducing latency and memory footprint, remaining robust to token length and image resolution, and providing an inductive bias that mitigates overfitting. Third, we train the policy end-to-end with Proximal Policy Optimization under terrain and appearance randomization and an obstacle-density curriculum, using a compact state-centric reward that balances progress, smoothness, and safety. We evaluate our method in challenging simulated environments with static and moving obstacles as well as uneven terrain. Compared with state-of-the-art baselines, our method achieves higher returns and success rates with fewer collisions, exhibits stronger generalization to unseen terrains and obstacle densities, and improves training efficiency by converging in fewer updates under the same compute budget.
comment: 14 pages. This paper has been published in Advanced Engineering Informatics. Please cite the journal version: DOI: 10.1016/j.aei.2025.104230
♻ ☆ Learning Obstacle Avoidance using Double DQN for Quadcopter Navigation
One of the challenges faced by Autonomous Aerial Vehicles is reliable navigation through urban environments. Factors like reduction in precision of Global Positioning System (GPS), narrow spaces and dynamically moving obstacles make the path planning of an aerial robot a complicated task. One of the skills required for the agent to effectively navigate through such an environment is to develop an ability to avoid collisions using information from onboard depth sensors. In this paper, we propose Reinforcement Learning of a virtual quadcopter robot agent equipped with a Depth Camera to navigate through a simulated urban environment.
comment: Fixed typo in second author's name throughout the paper
♻ ☆ Human-Like Robot Impedance Regulation Skill Learning from Human-Human Demonstrations
Humans are experts in physical collaboration by leveraging cognitive abilities such as perception, reasoning, and decision-making to regulate compliance behaviors based on their partners' states and task requirements. Equipping robots with similar cognitive-inspired collaboration skills can significantly enhance the efficiency and adaptability of human-robot collaboration (HRC). This paper introduces an innovative HumanInspired Impedance Regulation Skill Learning framework (HIImpRSL) for robotic systems to achieve leader-follower and mutual adaptation in multiple physical collaborative tasks. The proposed framework enables the robot to adapt its compliance based on human states and reference trajectories derived from human-human demonstrations. By integrating electromyography (EMG) signals and motion data, we extract endpoint impedance profiles and reference trajectories to construct a joint representation via imitation learning. An LSTM-based module then learns task-oriented impedance regulation policies, which are implemented through a whole-body impedance controller for online impedance adaptation. Experimental validation was conducted through collaborative transportation, two interactive Tai Chi pushing hands, and collaborative sawing tasks with multiple human subjects, demonstrating the ability of our framework to achieve human-like collaboration skills and the superior performance from the perspective of interactive forces compared to four other related methods.
comment: 13 pages, 9 figures
♻ ☆ EfficientFlow: Efficient Equivariant Flow Policy Learning for Embodied AI AAAI 2026
Generative modeling has recently shown remarkable promise for visuomotor policy learning, enabling flexible and expressive control across diverse embodied AI tasks. However, existing generative policies often struggle with data inefficiency, requiring large-scale demonstrations, and sampling inefficiency, incurring slow action generation during inference. We introduce EfficientFlow, a unified framework for efficient embodied AI with flow-based policy learning. To enhance data efficiency, we bring equivariance into flow matching. We theoretically prove that when using an isotropic Gaussian prior and an equivariant velocity prediction network, the resulting action distribution remains equivariant, leading to improved generalization and substantially reduced data demands. To accelerate sampling, we propose a novel acceleration regularization strategy. As direct computation of acceleration is intractable for marginal flow trajectories, we derive a novel surrogate loss that enables stable and scalable training using only conditional trajectories. Across a wide range of robotic manipulation benchmarks, the proposed algorithm achieves competitive or superior performance under limited data while offering dramatically faster inference. These results highlight EfficientFlow as a powerful and efficient paradigm for high-performance embodied AI.
comment: Accepted by AAAI 2026. Project Page: https://efficientflow.github.io/
♻ ☆ Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics
Learning robust and generalizable world models is crucial for enabling efficient and scalable robotic control in real-world environments. In this work, we introduce a novel framework for learning world models that accurately capture complex, partially observable, and stochastic dynamics. The proposed method employs a dual-autoregressive mechanism and self-supervised training to achieve reliable long-horizon predictions without relying on domain-specific inductive biases, ensuring adaptability across diverse robotic tasks. We further propose a policy optimization framework that leverages world models for efficient training in imagined environments and seamless deployment in real-world systems. This work advances model-based reinforcement learning by addressing the challenges of long-horizon prediction, error accumulation, and sim-to-real transfer. By providing a scalable and robust framework, the introduced methods pave the way for adaptive and efficient robotic systems in real-world applications.
♻ ☆ Enhancing Sample Efficiency and Exploration in Reinforcement Learning through the Integration of Diffusion Models and Proximal Policy Optimization
Proximal Policy Optimization (PPO) is widely used in continuous control due to its robustness and stable training, yet it remains sample-inefficient in tasks with expensive interactions and high-dimensional action spaces. This paper proposes PPO-DAP (PPO with Diffusion Action Prior), a strictly on-policy framework that improves exploration quality and learning efficiency without modifying the PPO objective. PPO-DAP follows a two-stage protocol. Offline, we pretrain a conditional diffusion action prior on logged trajectories to cover the action distribution supported by the behavior policy. Online, PPO updates the actor-critic only using newly collected on-policy rollouts, while the diffusion prior is adapted around the on-policy state distribution via parameter-efficient tuning (Adapter/LoRA) over a small parameter subset. For each on-policy state, the prior generates multiple action proposals and concentrates them toward high-value regions using critic-based energy reweighting and in-denoising gradient guidance. These proposals affect the actor only through a low-weight imitation loss and an optional soft KL regularizer to the prior; importantly, PPO gradients are never backpropagated through offline logs or purely synthetic trajectories. We further analyze the method from a dual-proximal perspective and derive a one-step performance lower bound. Across eight MuJoCo continuous-control tasks under a unified online budget of 1.0M environment steps, PPO-DAP consistently improves early learning efficiency (area under the learning curve over the first 40 epochs, ALC@40) and matches or exceeds the strongest on-policy baselines in final return on 6/8 tasks, with modest overhead (1.18+/-0.04x wall-clock time and 1.05+/-0.02x peak GPU memory relative to PPO).
♻ ☆ Symphony: A Heuristic Normalized Calibrated Advantage Actor and Critic Algorithm in application for Humanoid Robots
In our work we not explicitly hint that it is a misconception to think that humans learn fast. Learning process takes time. Babies start learning to move in the restricted liquid area called placenta. Children often are 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 dozen 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 no secret that continuous increase in Gaussian noise without appropriate smoothing is harmful for motors and gearboxes. Compared to Stochastic algorithms, we set a limited parametric noise and promote a reduced strength of actions, safely increasing entropy, since the actions are kind of immersed 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 Actor and Critic in one pass, as well as combine Actor and Critic in one Object and implement their Losses in one line.
comment: https://github.com/SuspensionRailway/symphony
Robotics 19
☆ Autonomously Unweaving Multiple Cables Using Visual Feedback
Many cable management tasks involve separating out the different cables and removing tangles. Automating this task is challenging because cables are deformable and can have combinations of knots and multiple interwoven segments. Prior works have focused on untying knots in one cable, which is one subtask of cable management. However, in this paper, we focus on a different subtask called multi-cable unweaving, which refers to removing the intersections among multiple interwoven cables to separate them and facilitate further manipulation. We propose a method that utilizes visual feedback to unweave a bundle of loosely entangled cables. We formulate cable unweaving as a pick-and-place problem, where the grasp position is selected from discrete nodes in a graph-based cable state representation. Our cable state representation encodes both topological and geometric information about the cables from the visual image. To predict future cable states and identify valid actions, we present a novel state transition model that takes into account the straightening and bending of cables during manipulation. Using this state transition model, we select between two high-level action primitives and calculate predicted immediate costs to optimize the lower-level actions. We experimentally demonstrate that iterating the above perception-planning-action process enables unweaving electric cables and shoelaces with an 84% success rate on average.
comment: 6 pages, 5 figures
☆ Sim2Real Reinforcement Learning for Soccer skills
This thesis work presents a more efficient and effective approach to training control-related tasks for humanoid robots using Reinforcement Learning (RL). The traditional RL methods are limited in adapting to real-world environments, complexity, and natural motions, but the proposed approach overcomes these limitations by using curriculum training and Adversarial Motion Priors (AMP) technique. The results show that the developed RL policies for kicking, walking, and jumping are more dynamic, and adaptive, and outperformed previous methods. However, the transfer of the learned policy from simulation to the real world was unsuccessful, highlighting the limitations of current RL methods in fully adapting to real-world scenarios.
comment: Undergrad Thesis
☆ Unifying Quadrotor Motion Planning and Control by Chaining Different Fidelity Models
Many aerial tasks involving quadrotors demand both instant reactivity and long-horizon planning. High-fidelity models enable accurate control but are too slow for long horizons; low-fidelity planners scale but degrade closed-loop performance. We present Unique, a unified MPC that cascades models of different fidelity within a single optimization: a short-horizon, high-fidelity model for accurate control, and a long-horizon, low-fidelity model for planning. We align costs across horizons, derive feasibility-preserving thrust and body-rate constraints for the point-mass model, and introduce transition constraints that match the different states, thrust-induced acceleration, and jerk-body-rate relations. To prevent local minima emerging from nonsmooth clutter, we propose a 3D progressive smoothing schedule that morphs norm-based obstacles along the horizon. In addition, we deploy parallel randomly initialized MPC solvers to discover lower-cost local minima on the long, low-fidelity horizon. In simulation and real flights, under equal computational budgets, Unique improves closed-loop position or velocity tracking by up to 75% compared with standard MPC and hierarchical planner-tracker baselines. Ablations and Pareto analyses confirm robust gains across horizon variations, constraint approximations, and smoothing schedules.
☆ INDOOR-LiDAR: Bridging Simulation and Reality for Robot-Centric 360 degree Indoor LiDAR Perception -- A Robot-Centric Hybrid Dataset
We present INDOOR-LIDAR, a comprehensive hybrid dataset of indoor 3D LiDAR point clouds designed to advance research in robot perception. Existing indoor LiDAR datasets often suffer from limited scale, inconsistent annotation formats, and human-induced variability during data collection. INDOOR-LIDAR addresses these limitations by integrating simulated environments with real-world scans acquired using autonomous ground robots, providing consistent coverage and realistic sensor behavior under controlled variations. Each sample consists of dense point cloud data enriched with intensity measurements and KITTI-style annotations. The annotation schema encompasses common indoor object categories within various scenes. The simulated subset enables flexible configuration of layouts, point densities, and occlusions, while the real-world subset captures authentic sensor noise, clutter, and domain-specific artifacts characteristic of real indoor settings. INDOOR-LIDAR supports a wide range of applications including 3D object detection, bird's-eye-view (BEV) perception, SLAM, semantic scene understanding, and domain adaptation between simulated and real indoor domains. By bridging the gap between synthetic and real-world data, INDOOR-LIDAR establishes a scalable, realistic, and reproducible benchmark for advancing robotic perception in complex indoor environments.
☆ Programmable Deformation Design of Porous Soft Actuator through Volumetric-Pattern-Induced Anisotropy
Conventional soft pneumatic actuators, typically based on hollow elastomeric chambers, often suffer from small structural support and require costly geometry-specific redesigns for multimodal functionality. Porous materials such as foam, filled into chambers, can provide structural stability for the actuators. However, methods to achieve programmable deformation by tailoring the porous body itself remain underexplored. In this paper, a novel design method is presented to realize soft porous actuators with programmable deformation by incising specific patterns into the porous foam body. This approach introduces localized structural anisotropy of the foam guiding the material's deformation under a global vacuum input. Furthermore, three fundamental patterns on a cylindrical foam substrate are discussed: transverse for bending, longitudinal for tilting, and diagonal for twisting. A computational model is built with Finite Element Analysis (FEA), to investigate the mechanism of the incision-patterning method. Experiments demonstrate that with a potential optimal design of the pattern array number N, actuators can achieve bending up to $80^{\circ}$ (N=2), tilting of $18^{\circ}$ (N=1), and twisting of $115^{\circ}$ (N=8). The versatility of our approach is demonstrated via pattern transferability, scalability, and mold-less rapid prototyping of complex designs. As a comprehensive application, we translate the human hand crease map into a functional incision pattern, creating a bio-inspired soft robot hand capable of human-like adaptive grasping. Our work provides a new, efficient, and scalable paradigm for the design of multi-functional soft porous robots.
☆ From Human Intention to Action Prediction: A Comprehensive Benchmark for Intention-driven End-to-End Autonomous Driving
Current end-to-end autonomous driving systems operate at a level of intelligence akin to following simple steering commands. However, achieving genuinely intelligent autonomy requires a paradigm shift: moving from merely executing low-level instructions to understanding and fulfilling high-level, abstract human intentions. This leap from a command-follower to an intention-fulfiller, as illustrated in our conceptual framework, is hindered by a fundamental challenge: the absence of a standardized benchmark to measure and drive progress on this complex task. To address this critical gap, we introduce Intention-Drive, the first comprehensive benchmark designed to evaluate the ability to translate high-level human intent into safe and precise driving actions. Intention-Drive features two core contributions: (1) a new dataset of complex scenarios paired with corresponding natural language intentions, and (2) a novel evaluation protocol centered on the Intent Success Rate (ISR), which assesses the semantic fulfillment of the human's goal beyond simple geometric accuracy. Through an extensive evaluation of a spectrum of baseline models on Intention-Drive, we reveal a significant performance deficit, showing that the baseline model struggle to achieve the comprehensive scene and intention understanding required for this advanced task.
☆ CAR-CHASE: Car-Like Robot Conflict-Aware Heuristic Adaptive Search Enhancement
Multi-Agent Path Finding (MAPF) for car-like robots, addressed by algorithms such as Conflict-Based Search with Continuous Time (CL-CBS), faces significant computational challenges due to expensive kinematic heuristic calculations. Traditional heuristic caching assumes that the heuristic function depends only on the state, which is incorrect in CBS where constraints from conflict resolution make the search space context-dependent. We propose \textbf{CAR-CHASE} (Car-Like Robot Conflict-Aware Heuristic Adaptive Search Enhancement), a novel approach that combines \textbf{conflict-aware heuristic caching} -- which caches heuristic values based on both state and relevant constraint context -- with an \textbf{adaptive hybrid heuristic} that intelligently switches between fast approximate and exact computations. Our key innovations are (1) a compact \emph{conflict fingerprint} that efficiently encodes which constraints affect a state's heuristic, (2) a relevance filter using spatial, temporal, and geometric criteria, and (3) an adaptive switching strategy with theoretical quality bounds. Experimental evaluation on 480 benchmark instances with varying agent counts (10 to 30) and obstacle densities (0\% and 50\%) demonstrates a geometric mean speedup of 2.46$\times$ over the baseline CL-CBS implementation while maintaining solution optimality. The optimizations improve success rate from 77.9\% to 84.8\% (+6.9 percentage points), reduce total runtime by 70.1\%, and enable solving 33 additional instances that previously timed out. Performance gains scale with problem complexity, reaching up to 4.06$\times$ speedup for challenging 30-agent obstacle scenarios. Our techniques are general and applicable to other CBS variants.
☆ Robust Underwater Localization of Buoyancy Driven microFloats Using Acoustic Time-of-Flight Measurements
Accurate underwater localization remains a challenge for inexpensive autonomous platforms that require highfrequency position updates. In this paper, we present a robust, low-cost localization pipeline for buoyancy-driven microFloats operating in coastal waters. We build upon previous work by introducing a bidirectional acoustic Time-of-Flight (ToF) localization framework, which incorporates both float-to-buoy and buoy-to-float transmissions, thereby increasing the number of usable measurements. The method integrates nonlinear trilateration with a filtering of computed position estimates based on geometric cost and Cramer-Rao Lower Bounds (CRLB). This approach removes outliers caused by multipath effects and other acoustic errors from the ToF estimation and improves localization robustness without relying on heavy smoothing. We validate the framework in two field deployments in Puget Sound, Washington, USA. The localization pipeline achieves median positioning errors below 4 m relative to GPS positions. The filtering technique shows a reduction in mean error from 139.29 m to 12.07 m, and improved alignment of trajectories with GPS paths. Additionally, we demonstrate a Time-Difference-of-Arrival (TDoA) localization for unrecovered floats that were transmitting during the experiment. Range-based acoustic localization techniques are widely used and generally agnostic to hardware-this work aims to maximize their utility by improving positioning frequency and robustness through careful algorithmic design.
comment: 9 pages
☆ Learning to Get Up Across Morphologies: Zero-Shot Recovery with a Unified Humanoid Policy
Fall recovery is a critical skill for humanoid robots in dynamic environments such as RoboCup, where prolonged downtime often decides the match. Recent techniques using deep reinforcement learning (DRL) have produced robust get-up behaviors, yet existing methods require training of separate policies for each robot morphology. This paper presents a single DRL policy capable of recovering from falls across seven humanoid robots with diverse heights (0.48 - 0.81 m), weights (2.8 - 7.9 kg), and dynamics. Trained with CrossQ, the unified policy transfers zero-shot up to 86 +/- 7% (95% CI [81, 89]) on unseen morphologies, eliminating the need for robot-specific training. Comprehensive leave-one-out experiments, morph scaling analysis, and diversity ablations show that targeted morphological coverage improves zero-shot generalization. In some cases, the shared policy even surpasses the specialist baselines. These findings illustrate the practicality of morphology-agnostic control for fall recovery, laying the foundation for generalist humanoid control. The software is open-source and available at: https://github.com/utra-robosoccer/unified-humanoid-getup
comment: Accepted at 28th RoboCup International Symposium
☆ Semantic Zone based 3D Map Management for Mobile Robot
Mobile robots in large-scale indoor environments, such as hospitals and logistics centers, require accurate 3D spatial representations. However, 3D maps consume substantial memory, making it difficult to maintain complete map data within limited computational resources. Existing SLAM frameworks typically rely on geometric distance or temporal metrics for memory management, often resulting in inefficient data retrieval in spatially compartmentalized environments. To address this, we propose a semantic zone-based 3D map management method that shifts the paradigm from geometry-centric to semantics-centric control. Our approach partitions the environment into meaningful spatial units (e.g., lobbies, hallways) and designates these zones as the primary unit for memory management. By dynamically loading only task-relevant zones into Working Memory (WM) and offloading inactive zones to Long-Term Memory (LTM), the system strictly enforces user-defined memory thresholds. Implemented within the RTAB-Map framework, our method demonstrates substantial reductions in unnecessary signature load/unload cycles and cumulative memory utilization compared to standard approaches. The results confirm that semantic zone-based management ensures stable, predictable memory usage while preserving map availability for navigation. Code is available at: https://github.com/huichangs/rtabmap/tree/segment
comment: 12 pages, 11 figures
☆ Measuring What Matters: Scenario-Driven Evaluation for Trajectory Predictors in Autonomous Driving
Being able to anticipate the motion of surrounding agents is essential for the safe operation of autonomous driving systems in dynamic situations. While various methods have been proposed for trajectory prediction, the current evaluation practices still rely on error-based metrics (e.g., ADE, FDE), which reveal the accuracy from a post-hoc view but ignore the actual effect the predictor brings to the self-driving vehicles (SDVs), especially in complex interactive scenarios: a high-quality predictor not only chases accuracy, but should also captures all possible directions a neighbor agent might move, to support the SDVs' cautious decision-making. Given that the existing metrics hardly account for this standard, in our work, we propose a comprehensive pipeline that adaptively evaluates the predictor's performance by two dimensions: accuracy and diversity. Based on the criticality of the driving scenario, these two dimensions are dynamically combined and result in a final score for the predictor's performance. Extensive experiments on a closed-loop benchmark using real-world datasets show that our pipeline yields a more reasonable evaluation than traditional metrics by better reflecting the correlation of the predictors' evaluation with the autonomous vehicles' driving performance. This evaluation pipeline shows a robust way to select a predictor that potentially contributes most to the SDV's driving performance.
comment: 9 Pages, 8 Figures
☆ A Hybrid Deep Learning Framework for Emotion Recognition in Children with Autism During NAO Robot-Mediated Interaction
Understanding emotional responses in children with Autism Spectrum Disorder (ASD) during social interaction remains a critical challenge in both developmental psychology and human-robot interaction. This study presents a novel deep learning pipeline for emotion recognition in autistic children in response to a name-calling event by a humanoid robot (NAO), under controlled experimental settings. The dataset comprises of around 50,000 facial frames extracted from video recordings of 15 children with ASD. A hybrid model combining a fine-tuned ResNet-50-based Convolutional Neural Network (CNN) and a three-layer Graph Convolutional Network (GCN) trained on both visual and geometric features extracted from MediaPipe FaceMesh landmarks. Emotions were probabilistically labeled using a weighted ensemble of two models: DeepFace's and FER, each contributing to soft-label generation across seven emotion classes. Final classification leveraged a fused embedding optimized via Kullback-Leibler divergence. The proposed method demonstrates robust performance in modeling subtle affective responses and offers significant promise for affective profiling of ASD children in clinical and therapeutic human-robot interaction contexts, as the pipeline effectively captures micro emotional cues in neurodivergent children, addressing a major gap in autism-specific HRI research. This work represents the first such large-scale, real-world dataset and pipeline from India on autism-focused emotion analysis using social robotics, contributing an essential foundation for future personalized assistive technologies.
comment: 12 pages, journal paper
☆ Navigation Around Unknown Space Objects Using Visible-Thermal Image Fusion SC
As the popularity of on-orbit operations grows, so does the need for precise navigation around unknown resident space objects (RSOs) such as other spacecraft, orbital debris, and asteroids. The use of Simultaneous Localization and Mapping (SLAM) algorithms is often studied as a method to map out the surface of an RSO and find the inspector's relative pose using a lidar or conventional camera. However, conventional cameras struggle during eclipse or shadowed periods, and lidar, though robust to lighting conditions, tends to be heavier, bulkier, and more power-intensive. Thermal-infrared cameras can track the target RSO throughout difficult illumination conditions without these limitations. While useful, thermal-infrared imagery lacks the resolution and feature-richness of visible cameras. In this work, images of a target satellite in low Earth orbit are photo-realistically simulated in both visible and thermal-infrared bands. Pixel-level fusion methods are used to create visible/thermal-infrared composites that leverage the best aspects of each camera. Navigation errors from a monocular SLAM algorithm are compared between visible, thermal-infrared, and fused imagery in various lighting and trajectories. Fused imagery yields substantially improved navigation performance over visible-only and thermal-only methods.
comment: 18 pages, 11 figures. To be published in proceedings of AIAA SCITECH 2026 Forum
☆ B-ActiveSEAL: Scalable Uncertainty-Aware Active Exploration with Tightly Coupled Localization-Mapping
Active robot exploration requires decision-making processes that integrate localization and mapping under tightly coupled uncertainty. However, managing these interdependent uncertainties over long-term operations in large-scale environments rapidly becomes computationally intractable. To address this challenge, we propose B-ActiveSEAL, a scalable information-theoretic active exploration framework that explicitly accounts for coupled uncertainties-from perception through mapping-into the decision-making process. Our framework (i) adaptively balances map uncertainty (exploration) and localization uncertainty (exploitation), (ii) accommodates a broad class of generalized entropy measures, enabling flexible and uncertainty-aware active exploration, and (iii) establishes Behavioral entropy (BE) as an effective information measure for active exploration by enabling intuitive and adaptive decision-making under coupled uncertainties. We establish a theoretical foundation for propagating coupled uncertainties and integrating them into general entropy formulations, enabling uncertainty-aware active exploration under tightly coupled localization-mapping. The effectiveness of the proposed approach is validated through rigorous theoretical analysis and extensive experiments on open-source maps and ROS-Unity simulations across diverse and complex environments. The results demonstrate that B-ActiveSEAL achieves a well-balanced exploration-exploitation trade-off and produces diverse, adaptive exploration behaviors across environments, highlighting clear advantages over representative baselines.
comment: 18 pages, 17 figures
♻ ☆ Ctrl-Crash: Controllable Diffusion for Realistic Car Crashes
Video diffusion techniques have advanced significantly in recent years; however, they struggle to generate realistic imagery of car crashes due to the scarcity of accident events in most driving datasets. Improving traffic safety requires realistic and controllable accident simulations. To tackle the problem, we propose Ctrl-Crash, a controllable car crash video generation model that conditions on signals such as bounding boxes, crash types, and an initial image frame. Our approach enables counterfactual scenario generation where minor variations in input can lead to dramatically different crash outcomes. To support fine-grained control at inference time, we leverage classifier-free guidance with independently tunable scales for each conditioning signal. Ctrl-Crash achieves state-of-the-art performance across quantitative video quality metrics (e.g., FVD and JEDi) and qualitative measurements based on a human-evaluation of physical realism and video quality compared to prior diffusion-based methods.
comment: Under review at Pattern Recognition Letters
♻ ☆ End-to-End Dexterous Arm-Hand VLA Policies via Shared Autonomy: VR Teleoperation Augmented by Autonomous Hand VLA Policy for Efficient Data Collection
Achieving human-like dexterous manipulation remains a major challenge for general-purpose robots. While Vision-Language-Action (VLA) models show potential in learning skills from demonstrations, their scalability is limited by scarce high-quality training data. Existing data collection methods face inherent constraints: manual teleoperation overloads human operators, while automated planning often produces unnatural motions. We propose a Shared Autonomy framework that divides control between macro and micro motions. A human operator guides the robot's arm pose through intuitive VR teleoperation, while an autonomous DexGrasp-VLA policy handles fine-grained hand control using real-time tactile and visual feedback. This division significantly reduces cognitive load and enables efficient collection of high-quality coordinated arm-hand demonstrations. Using this data, we train an end-to-end VLA policy enhanced with our novel Arm-Hand Feature Enhancement module, which captures both distinct and shared representations of macro and micro movements for more natural coordination. Our Corrective Teleoperation system enables continuous policy improvement through human-in-the-loop failure recovery. Experiments demonstrate that our framework generates high-quality data with minimal manpower and achieves a 90% success rate across diverse objects, including unseen instances. Comprehensive evaluations validate the system's effectiveness in developing dexterous manipulation capabilities.
♻ ☆ Entropy-Controlled Intrinsic Motivation Reinforcement Learning for Quadruped Robot Locomotion in Complex Terrains
Learning is the basis of both biological and artificial systems when it comes to mimicking intelligent behaviors. From the classical PPO (Proximal Policy Optimization), there is a series of deep reinforcement learning algorithms which are widely used in training locomotion policies for quadrupedal robots because of their stability and sample efficiency. However, among all these variants, experiments and simulations often converge prematurely, leading to suboptimal locomotion and reduced task performance. Therefore, in this paper, we introduce Entropy-Controlled Intrinsic Motivation (ECIM), an entropy-based reinforcement learning algorithm in contrast with the PPO series, that can reduce premature convergence by combining intrinsic motivation with adaptive exploration. For experiments, in order to parallel with other baselines, we chose to apply it in Isaac Gym across six terrain categories: upward slopes, downward slopes, uneven rough terrain, ascending stairs, descending stairs, and flat ground as widely used. For comparison, our experiments consistently achieve better performance: task rewards increase by 4--12%, peak body pitch oscillation is reduced by 23--29%, joint acceleration decreases by 20--32%, and joint torque consumption declines by 11--20%. Overall, our model ECIM, by combining entropy control and intrinsic motivation control, achieves better results in stability across different terrains for quadrupedal locomotion, and at the same time reduces energetic cost and makes it a practical choice for complex robotic control tasks.
♻ ☆ BEASST: Behavioral Entropic Gradient based Adaptive Source Seeking for Mobile Robots
This paper presents BEASST (Behavioral Entropic Gradient-based Adaptive Source Seeking for Mobile Robots), a novel framework for robotic source seeking in complex, unknown environments. Our approach enables mobile robots to efficiently balance exploration and exploitation by modeling normalized signal strength as a surrogate probability of source location. Building on Behavioral Entropy(BE) with Prelec's probability weighting function, we define an objective function that adapts robot behavior from risk-averse to risk-seeking based on signal reliability and mission urgency. The framework provides theoretical convergence guarantees under unimodal signal assumptions and practical stability under bounded disturbances. Experimental validation across DARPA SubT and multi-room scenarios demonstrates that BEASST consistently outperforms state-of-the-art methods, achieving 15% reduction in path length and 20% faster source localization through intelligent uncertainty-driven navigation that dynamically transitions between aggressive pursuit and cautious exploration.
♻ ☆ Camera Calibration via Circular Patterns: A Comprehensive Framework with Detection Uncertainty and Unbiased Projection Model
Camera calibration using planar targets has been widely favored, and two types of control points have been mainly considered as measurements: the corners of the checkerboard and the centroid of circles. Since a centroid is derived from numerous pixels, the circular pattern provides more precise measurements than the checkerboard. However, the existing projection model of circle centroids is biased under lens distortion, resulting in low performance. To surmount this limitation, we propose an unbiased projection model of the circular pattern and demonstrate its superior accuracy compared to the checkerboard. Complementing this, we introduce uncertainty into circular patterns to enhance calibration robustness and completeness. Defining centroid uncertainty improves the performance of calibration components, including pattern detection, optimization, and evaluation metrics. We also provide guidelines for performing good camera calibration based on the evaluation metric. The core concept of this approach is to model the boundary points of a two-dimensional shape as a Markov random field, considering its connectivity. The shape distribution is propagated to the centroid uncertainty through an appropriate shape representation based on the Green theorem. Consequently, the resulting framework achieves marked gains in calibration accuracy and robustness. The complete source code and demonstration video are available at https://github.com/chaehyeonsong/discocal.
Robotics 48
☆ AnchorDream: Repurposing Video Diffusion for Embodiment-Aware Robot Data Synthesis
The collection of large-scale and diverse robot demonstrations remains a major bottleneck for imitation learning, as real-world data acquisition is costly and simulators offer limited diversity and fidelity with pronounced sim-to-real gaps. While generative models present an attractive solution, existing methods often alter only visual appearances without creating new behaviors, or suffer from embodiment inconsistencies that yield implausible motions. To address these limitations, we introduce AnchorDream, an embodiment-aware world model that repurposes pretrained video diffusion models for robot data synthesis. AnchorDream conditions the diffusion process on robot motion renderings, anchoring the embodiment to prevent hallucination while synthesizing objects and environments consistent with the robot's kinematics. Starting from only a handful of human teleoperation demonstrations, our method scales them into large, diverse, high-quality datasets without requiring explicit environment modeling. Experiments show that the generated data leads to consistent improvements in downstream policy learning, with relative gains of 36.4% in simulator benchmarks and nearly double performance in real-world studies. These results suggest that grounding generative world models in robot motion provides a practical path toward scaling imitation learning.
comment: Project page: https://jay-ye.github.io/AnchorDream/
☆ Toward a Decision Support System for Energy-Efficient Ferry Operation on Lake Constance based on Optimal Control
The maritime sector is undergoing a disruptive technological change driven by three main factors: autonomy, decarbonization, and digital transformation. Addressing these factors necessitates a reassessment of inland vessel operations. This paper presents the design and development of a decision support system for ferry operations based on a shrinking-horizon optimal control framework. The problem formulation incorporates a mathematical model of the ferry's dynamics and environmental disturbances, specifically water currents and wind, which can significantly influence the dynamics. Real-world data and illustrative scenarios demonstrate the potential of the proposed system to effectively support ferry crews by providing real-time guidance. This enables enhanced operational efficiency while maintaining predefined maneuver durations. The findings suggest that optimal control applications hold substantial promise for advancing future ferry operations on inland waters. A video of the real-world ferry MS Insel Mainau operating on Lake Constance is available at: https://youtu.be/i1MjCdbEQyE
comment: 6 pages, 8 figures
☆ Agile Flight Emerges from Multi-Agent Competitive Racing
Through multi-agent competition and the sparse high-level objective of winning a race, we find that both agile flight (e.g., high-speed motion pushing the platform to its physical limits) and strategy (e.g., overtaking or blocking) emerge from agents trained with reinforcement learning. We provide evidence in both simulation and the real world that this approach outperforms the common paradigm of training agents in isolation with rewards that prescribe behavior, e.g., progress on the raceline, in particular when the complexity of the environment increases, e.g., in the presence of obstacles. Moreover, we find that multi-agent competition yields policies that transfer more reliably to the real world than policies trained with a single-agent progress-based reward, despite the two methods using the same simulation environment, randomization strategy, and hardware. In addition to improved sim-to-real transfer, the multi-agent policies also exhibit some degree of generalization to opponents unseen at training time. Overall, our work, following in the tradition of multi-agent competitive game-play in digital domains, shows that sparse task-level rewards are sufficient for training agents capable of advanced low-level control in the physical world. Code: https://github.com/Jirl-upenn/AgileFlight_MultiAgent
☆ ProbeMDE: Uncertainty-Guided Active Proprioception for Monocular Depth Estimation in Surgical Robotics
Monocular depth estimation (MDE) provides a useful tool for robotic perception, but its predictions are often uncertain and inaccurate in challenging environments such as surgical scenes where textureless surfaces, specular reflections, and occlusions are common. To address this, we propose ProbeMDE, a cost-aware active sensing framework that combines RGB images with sparse proprioceptive measurements for MDE. Our approach utilizes an ensemble of MDE models to predict dense depth maps conditioned on both RGB images and on a sparse set of known depth measurements obtained via proprioception, where the robot has touched the environment in a known configuration. We quantify predictive uncertainty via the ensemble's variance and measure the gradient of the uncertainty with respect to candidate measurement locations. To prevent mode collapse while selecting maximally informative locations to propriocept (touch), we leverage Stein Variational Gradient Descent (SVGD) over this gradient map. We validate our method in both simulated and physical experiments on central airway obstruction surgical phantoms. Our results demonstrate that our approach outperforms baseline methods across standard depth estimation metrics, achieving higher accuracy while minimizing the number of required proprioceptive measurements.
comment: 9 pages, 5 figures
☆ BLURR: A Boosted Low-Resource Inference for Vision-Language-Action Models
Vision-language-action (VLA) models enable impressive zero shot manipulation, but their inference stacks are often too heavy for responsive web demos or high frequency robot control on commodity GPUs. We present BLURR, a lightweight inference wrapper that can be plugged into existing VLA controllers without retraining or changing model checkpoints. Instantiated on the pi-zero VLA controller, BLURR keeps the original observation interfaces and accelerates control by combining an instruction prefix key value cache, mixed precision execution, and a single step rollout schedule that reduces per step computation. In our SimplerEnv based evaluation, BLURR maintains task success rates comparable to the original controller while significantly lowering effective FLOPs and wall clock latency. We also build an interactive web demo that allows users to switch between controllers and toggle inference options in real time while watching manipulation episodes. This highlights BLURR as a practical approach for deploying modern VLA policies under tight compute budgets.
comment: 10 pages, 3 figures. Code and integration scripts will be released at this http URL: https://github.com/JijiKing-Sam/BLURR-A-Boosted-Low-Resource-Inference-for-Vision-Language-Action-Model
☆ The Influence of Human-like Appearance on Expected Robot Explanations
A robot's appearance is a known factor influencing user's mental model and human-robot interaction, that has not been studied in the context of its influence in expected robot explanations. In this study, we investigate whether and to what extent the human-like appearance of robots elicits anthropomorphism, which is conceptualised as an attribution of mental capacities, and how the level of anthropomorphism is revealed in explanations that people expect to receive. We designed a between-subject study comprising conditions with visual stimuli of three domestic service robots with varying human-like appearance, and we prompted respondents to provide explanations they would expect to receive from the robot for the same robot actions. We found that most explanations were anthropomorphic across all conditions. However, there is a positive correlation between the anthropomorphic explanations and human-like appearance. We also report on more nuanced trends observed in non-anthropomorphic explanations and trends in robot descriptions.
☆ Bench-Push: Benchmarking Pushing-based Navigation and Manipulation Tasks for Mobile Robots ICRA 2026
Mobile robots are increasingly deployed in cluttered environments with movable objects, posing challenges for traditional methods that prohibit interaction. In such settings, the mobile robot must go beyond traditional obstacle avoidance, leveraging pushing or nudging strategies to accomplish its goals. While research in pushing-based robotics is growing, evaluations rely on ad hoc setups, limiting reproducibility and cross-comparison. To address this, we present Bench-Push, the first unified benchmark for pushing-based mobile robot navigation and manipulation tasks. Bench-Push includes multiple components: 1) a comprehensive range of simulated environments that capture the fundamental challenges in pushing-based tasks, including navigating a maze with movable obstacles, autonomous ship navigation in ice-covered waters, box delivery, and area clearing, each with varying levels of complexity; 2) novel evaluation metrics to capture efficiency, interaction effort, and partial task completion; and 3) demonstrations using Bench-Push to evaluate example implementations of established baselines across environments. Bench-Push is open-sourced as a Python library with a modular design. The code, documentation, and trained models can be found at https://github.com/IvanIZ/BenchNPIN.
comment: Under review for ICRA 2026
☆ Two-dimensional Decompositions of High-dimensional Configurations for Efficient Multi-vehicle Coordination at Intelligent Intersections
For multi-vehicle complex traffic scenarios in shared spaces such as intelligent intersections, safe coordination and trajectory planning is challenging due to computational complexity. To meet this challenge, we introduce a computationally efficient method for generating collision-free trajectories along predefined vehicle paths. We reformulate a constrained minimum-time trajectory planning problem as a problem in a high-dimensional configuration space, where conflict zones are modeled by high-dimensional polyhedra constructed from two-dimensional rectangles. Still, in such a formulation, as the number of vehicles involved increases, the computational complexity increases significantly. To address this, we propose two algorithms for near-optimal local optimization that significantly reduce the computational complexity by decomposing the high-dimensional problem into a sequence of 2D graph search problems. The resulting trajectories are then incorporated into a Nonlinear Model Predictive Control (NMPC) framework to ensure safe and smooth vehicle motion. We furthermore show in numerical evaluation that this approach significantly outperforms existing MILP-based time-scheduling; both in terms of objective-value and computational time.
☆ Architecting Large Action Models for Human-in-the-Loop Intelligent Robots
The realization of intelligent robots, operating autonomously and interacting with other intelligent agents, human or artificial, requires the integration of environment perception, reasoning, and action. Classic Artificial Intelligence techniques for this purpose, focusing on symbolic approaches, have long-ago hit the scalability wall on compute and memory costs. Advances in Large Language Models in the past decade (neural approaches) have resulted in unprecedented displays of capability, at the cost of control, explainability, and interpretability. Large Action Models aim at extending Large Language Models to encompass the full perception, reasoning, and action cycle; however, they typically require substantially more comprehensive training and suffer from the same deficiencies in reliability. Here, we show it is possible to build competent Large Action Models by composing off-the-shelf foundation models, and that their control, interpretability, and explainability can be effected by incorporating symbolic wrappers and associated verification on their outputs, achieving verifiable neuro-symbolic solutions for intelligent robots. Our experiments on a multi-modal robot demonstrate that Large Action Model intelligence does not require massive end-to-end training, but can be achieved by integrating efficient perception models with a logic-driven core. We find that driving action execution through the generation of Planning Domain Definition Language (PDDL) code enables a human-in-the-loop verification stage that effectively mitigates action hallucinations. These results can support practitioners in the design and development of robotic Large Action Models across novel industries, and shed light on the ongoing challenges that must be addressed to ensure safety in the field.
☆ UniBYD: A Unified Framework for Learning Robotic Manipulation Across Embodiments Beyond Imitation of Human Demonstrations
In embodied intelligence, the embodiment gap between robotic and human hands brings significant challenges for learning from human demonstrations. Although some studies have attempted to bridge this gap using reinforcement learning, they remain confined to merely reproducing human manipulation, resulting in limited task performance. In this paper, we propose UniBYD, a unified framework that uses a dynamic reinforcement learning algorithm to discover manipulation policies aligned with the robot's physical characteristics. To enable consistent modeling across diverse robotic hand morphologies, UniBYD incorporates a unified morphological representation (UMR). Building on UMR, we design a dynamic PPO with an annealed reward schedule, enabling reinforcement learning to transition from imitation of human demonstrations to explore policies adapted to diverse robotic morphologies better, thereby going beyond mere imitation of human hands. To address the frequent failures of learning human priors in the early training stage, we design a hybrid Markov-based shadow engine that enables reinforcement learning to imitate human manipulations in a fine-grained manner. To evaluate UniBYD comprehensively, we propose UniManip, the first benchmark encompassing robotic manipulation tasks spanning multiple hand morphologies. Experiments demonstrate a 67.90% improvement in success rate over the current state-of-the-art. Upon acceptance of the paper, we will release our code and benchmark at https://github.com/zhanheng-creator/UniBYD.
☆ Atomic Action Slicing: Planner-Aligned Options for Generalist VLA Agents
Current vision-language-action (VLA) models generalize poorly, particularly when tasks require new compositions of skills or objects. We introduce Atomic Action Slicing (AAS), a planner-aligned approach that decomposes long-horizon demonstrations into short, typed atomic actions that are easier for planners to use and policies to learn. Using LIBERO demonstrations, AAS produces a validated dataset of 2,124 atomic segments labeled with action type, temporal span, and confidence. A stronger segmenter (Gemini 2.5 Pro) closely matches planner-defined plans and remains robust under keyframe jitter, while smaller models perform worse on multi-object tasks. Fine-tuning CLIP-RT+ on our atomic dataset improves task success from 94.2% to 95.3% on LIBERO-Goal and 83.8% to 88.8% on LIBERO-Long. We publicly release the GATE-VLAP dataset on HuggingFace(https://huggingface.co/datasets/gate-institute/GATE-VLAP-datasets)
comment: The 41st ACM/SIGAPP Symposium On Applied Computing
☆ Cross-Entropy Optimization of Physically Grounded Task and Motion Plans
Autonomously performing tasks often requires robots to plan high-level discrete actions and continuous low-level motions to realize them. Previous TAMP algorithms have focused mainly on computational performance, completeness, or optimality by making the problem tractable through simplifications and abstractions. However, this comes at the cost of the resulting plans potentially failing to account for the dynamics or complex contacts necessary to reliably perform the task when object manipulation is required. Additionally, approaches that ignore effects of the low-level controllers may not obtain optimal or feasible plan realizations for the real system. We investigate the use of a GPU-parallelized physics simulator to compute realizations of plans with motion controllers, explicitly accounting for dynamics, and considering contacts with the environment. Using cross-entropy optimization, we sample the parameters of the controllers, or actions, to obtain low-cost solutions. Since our approach uses the same controllers as the real system, the robot can directly execute the computed plans. We demonstrate our approach for a set of tasks where the robot is able to exploit the environment's geometry to move an object. Website and code: https://andreumatoses.github.io/research/parallel-realization
comment: Preprint
☆ CarlaNCAP: A Framework for Quantifying the Safety of Vulnerable Road Users in Infrastructure-Assisted Collective Perception Using EuroNCAP Scenarios
The growing number of road users has significantly increased the risk of accidents in recent years. Vulnerable Road Users (VRUs) are particularly at risk, especially in urban environments where they are often occluded by parked vehicles or buildings. Autonomous Driving (AD) and Collective Perception (CP) are promising solutions to mitigate these risks. In particular, infrastructure-assisted CP, where sensor units are mounted on infrastructure elements such as traffic lights or lamp posts, can help overcome perceptual limitations by providing enhanced points of view, which significantly reduces occlusions. To encourage decision makers to adopt this technology, comprehensive studies and datasets demonstrating safety improvements for VRUs are essential. In this paper, we propose a framework for evaluating the safety improvement by infrastructure-based CP specifically targeted at VRUs including a dataset with safety-critical EuroNCAP scenarios (CarlaNCAP) with 11k frames. Using this dataset, we conduct an in-depth simulation study and demonstrate that infrastructure-assisted CP can significantly reduce accident rates in safety-critical scenarios, achieving up to 100% accident avoidance compared to a vehicle equipped with sensors with only 33%. Code is available at https://github.com/ekut-es/carla_ncap
☆ Mirror Skin: In Situ Visualization of Robot Touch Intent on Robotic Skin
Effective communication of robotic touch intent is a key factor in promoting safe and predictable physical human-robot interaction (pHRI). While intent communication has been widely studied, existing approaches lack the spatial specificity and semantic depth necessary to convey robot touch actions. We present Mirror Skin, a cephalopod-inspired concept that utilizes high-resolution, mirror-like visual feedback on robotic skin. By mapping in-situ visual representations of a human's body parts onto the corresponding robot's touch region, Mirror Skin communicates who shall initiate touch, where it will occur, and when it is imminent. To inform the design of Mirror Skin, we conducted a structured design exploration with experts in virtual reality (VR), iteratively refining six key dimensions. A subsequent controlled user study demonstrated that Mirror Skin significantly enhances accuracy and reduces response times for interpreting touch intent. These findings highlight the potential of visual feedback on robotic skin to communicate human-robot touch interactions.
☆ An Anatomy of Vision-Language-Action Models: From Modules to Milestones and Challenges
Vision-Language-Action (VLA) models are driving a revolution in robotics, enabling machines to understand instructions and interact with the physical world. This field is exploding with new models and datasets, making it both exciting and challenging to keep pace with. This survey offers a clear and structured guide to the VLA landscape. We design it to follow the natural learning path of a researcher: we start with the basic Modules of any VLA model, trace the history through key Milestones, and then dive deep into the core Challenges that define recent research frontier. Our main contribution is a detailed breakdown of the five biggest challenges in: (1) Representation, (2) Execution, (3) Generalization, (4) Safety, and (5) Dataset and Evaluation. This structure mirrors the developmental roadmap of a generalist agent: establishing the fundamental perception-action loop, scaling capabilities across diverse embodiments and environments, and finally ensuring trustworthy deployment-all supported by the essential data infrastructure. For each of them, we review existing approaches and highlight future opportunities. We position this paper as both a foundational guide for newcomers and a strategic roadmap for experienced researchers, with the dual aim of accelerating learning and inspiring new ideas in embodied intelligence. A live version of this survey, with continuous updates, is maintained on our \href{https://suyuz1.github.io/Survery/}{project page}.
☆ Incremental Validation of Automated Driving Functions using Generic Volumes in Micro- Operational Design Domains
The validation of highly automated, perception-based driving systems must ensure that they function correctly under the full range of real-world conditions. Scenario-based testing is a prominent approach to addressing this challenge, as it involves the systematic simulation of objects and environments. Operational Design Domains (ODDs) are usually described using a taxonomy of qualitative designations for individual objects. However, the process of transitioning from taxonomy to concrete test cases remains unstructured, and completeness is theoretical. This paper introduces a structured method of subdividing the ODD into manageable sections, termed micro-ODDs (mODDs), and deriving test cases with abstract object representations. This concept is demonstrated using a one-dimensional, laterally guided manoeuvre involving a shunting locomotive within a constrained ODD. In this example, mODDs are defined and refined into narrow taxonomies that enable test case generation. Obstacles are represented as generic cubes of varying sizes, providing a simplified yet robust means of evaluating perception performance. A series of tests were conducted in a closed-loop, co-simulated virtual environment featuring photorealistic rendering and simulated LiDAR, GNSS and camera sensors. The results demonstrate how edge cases in obstacle detection can be systematically explored and how perception quality can be evaluated based on observed vehicle behaviour, using crash versus safe stop as the outcome metrics. These findings support the development of a standardised framework for safety argumentation and offer a practical step towards the validation and authorisation of automated driving functions.
☆ Symmetry-Aware Steering of Equivariant Diffusion Policies: Benefits and Limits
Equivariant diffusion policies (EDPs) combine the generative expressivity of diffusion models with the strong generalization and sample efficiency afforded by geometric symmetries. While steering these policies with reinforcement learning (RL) offers a promising mechanism for fine-tuning beyond demonstration data, directly applying standard (non-equivariant) RL can be sample-inefficient and unstable, as it ignores the symmetries that EDPs are designed to exploit. In this paper, we theoretically establish that the diffusion process of an EDP is equivariant, which in turn induces a group-invariant latent-noise MDP that is well-suited for equivariant diffusion steering. Building on this theory, we introduce a principled symmetry-aware steering framework and compare standard, equivariant, and approximately equivariant RL strategies through comprehensive experiments across tasks with varying degrees of symmetry. While we identify the practical boundaries of strict equivariance under symmetry breaking, we show that exploiting symmetry during the steering process yields substantial benefits-enhancing sample efficiency, preventing value divergence, and achieving strong policy improvements even when EDPs are trained from extremely limited demonstrations.
☆ Towards Logic-Aware Manipulation: A Knowledge Primitive for VLM-Based Assistants in Smart Manufacturing
Existing pipelines for vision-language models (VLMs) in robotic manipulation prioritize broad semantic generalization from images and language, but typically omit execution-critical parameters required for contact-rich actions in manufacturing cells. We formalize an object-centric manipulation-logic schema, serialized as an eight-field tuple τ, which exposes object, interface, trajectory, tolerance, and force/impedance information as a first-class knowledge signal between human operators, VLM-based assistants, and robot controllers. We instantiate τ and a small knowledge base (KB) on a 3D-printer spool-removal task in a collaborative cell, and analyze τ-conditioned VLM planning using plan-quality metrics adapted from recent VLM/LLM planning benchmarks, while demonstrating how the same schema supports taxonomy-tagged data augmentation at training time and logic-aware retrieval-augmented prompting at test time as a building block for assistant systems in smart manufacturing enterprises.
comment: 8 pages, 2 figures, submitted to the 2026 IFAC World Congress
☆ Optimal Control and Structurally-Informed Gradient Optimization of a Custom 4-DOF Rigid-Body Manipulator
This work develops a control-centric framework for a custom 4-DOF rigid-body manipulator by coupling a reduced-order Pontryagin's Maximum Principle (PMP) controller with a physics-informed Gradient Descent stage. The reduced PMP model provides a closed-form optimal control law for the joint accelerations, while the Gradient Descent module determines the corresponding time horizons by minimizing a cost functional built directly from the full Rigid-Body Dynamics. Structural-mechanics reaction analysis is used only to initialize feasible joint velocities-most critically the azimuthal component-ensuring that the optimizer begins in a physically admissible region. The resulting kinematic trajectories and dynamically consistent time horizons are then supplied to the symbolic Euler-Lagrange model to yield closed-form inverse-dynamics inputs. This pipeline preserves a strict control-theoretic structure while embedding the physical constraints and loading behavior of the manipulator in a computationally efficient way.
comment: 6 pages + 18 page appendix
☆ Elevation Aware 2D/3D Co-simulation Framework for Large-scale Traffic Flow and High-fidelity Vehicle Dynamics
Reliable testing of autonomous driving systems requires simulation environments that combine large-scale traffic modeling with realistic 3D perception and terrain. Existing tools rarely capture real-world elevation, limiting their usefulness in cities with complex topography. This paper presents an automated, elevation-aware co-simulation framework that integrates SUMO with CARLA using a pipeline that fuses OpenStreetMap road networks and USGS elevation data into physically consistent 3D environments. The system generates smooth elevation profiles, validates geometric accuracy, and enables synchronized 2D-3D simulation across platforms. Demonstrations on multiple regions of San Francisco show the framework's scalability and ability to reproduce steep and irregular terrain. The result is a practical foundation for high-fidelity autonomous vehicle testing in realistic, elevation-rich urban settings.
☆ Seeing to Act, Prompting to Specify: A Bayesian Factorization of Vision Language Action Policy
The pursuit of out-of-distribution generalization in Vision-Language-Action (VLA) models is often hindered by catastrophic forgetting of the Vision-Language Model (VLM) backbone during fine-tuning. While co-training with external reasoning data helps, it requires experienced tuning and data-related overhead. Beyond such external dependencies, we identify an intrinsic cause within VLA datasets: modality imbalance, where language diversity is much lower than visual and action diversity. This imbalance biases the model toward visual shortcuts and language forgetting. To address this, we introduce BayesVLA, a Bayesian factorization that decomposes the policy into a visual-action prior, supporting seeing-to-act, and a language-conditioned likelihood, enabling prompt-to-specify. This inherently preserves generalization and promotes instruction following. We further incorporate pre- and post-contact phases to better leverage pre-trained foundation models. Information-theoretic analysis formally validates our effectiveness in mitigating shortcut learning. Extensive experiments show superior generalization to unseen instructions, objects, and environments compared to existing methods. Project page is available at: https://xukechun.github.io/papers/BayesVLA.
☆ A Stochastic Approach to Terrain Maps for Safe Lunar Landing
Safely landing on the lunar surface is a challenging task, especially in the heavily-shadowed South Pole region where traditional vision-based hazard detection methods are not reliable. The potential existence of valuable resources at the lunar South Pole has made landing in that region a high priority for many space agencies and commercial companies. However, relying on a LiDAR for hazard detection during descent is risky, as this technology is fairly untested in the lunar environment. There exists a rich log of lunar surface data from the Lunar Reconnaissance Orbiter (LRO), which could be used to create informative prior maps of the surface before descent. In this work, we propose a method for generating stochastic elevation maps from LRO data using Gaussian processes (GPs), which are a powerful Bayesian framework for non-parametric modeling that produce accompanying uncertainty estimates. In high-risk environments such as autonomous spaceflight, interpretable estimates of terrain uncertainty are critical. However, no previous approaches to stochastic elevation mapping have taken LRO Digital Elevation Model (DEM) confidence maps into account, despite this data containing key information about the quality of the DEM in different areas. To address this gap, we introduce a two-stage GP model in which a secondary GP learns spatially varying noise characteristics from DEM confidence data. This heteroscedastic information is then used to inform the noise parameters for the primary GP, which models the lunar terrain. Additionally, we use stochastic variational GPs to enable scalable training. By leveraging GPs, we are able to more accurately model the impact of heteroscedastic sensor noise on the resulting elevation map. As a result, our method produces more informative terrain uncertainty, which can be used for downstream tasks such as hazard detection and safe landing site selection.
comment: Accepted to IEEE Aerospace 2026
☆ Goal Reaching with Eikonal-Constrained Hierarchical Quasimetric Reinforcement Learning
Goal-Conditioned Reinforcement Learning (GCRL) mitigates the difficulty of reward design by framing tasks as goal reaching rather than maximizing hand-crafted reward signals. In this setting, the optimal goal-conditioned value function naturally forms a quasimetric, motivating Quasimetric RL (QRL), which constrains value learning to quasimetric mappings and enforces local consistency through discrete, trajectory-based constraints. We propose Eikonal-Constrained Quasimetric RL (Eik-QRL), a continuous-time reformulation of QRL based on the Eikonal Partial Differential Equation (PDE). This PDE-based structure makes Eik-QRL trajectory-free, requiring only sampled states and goals, while improving out-of-distribution generalization. We provide theoretical guarantees for Eik-QRL and identify limitations that arise under complex dynamics. To address these challenges, we introduce Eik-Hierarchical QRL (Eik-HiQRL), which integrates Eik-QRL into a hierarchical decomposition. Empirically, Eik-HiQRL achieves state-of-the-art performance in offline goal-conditioned navigation and yields consistent gains over QRL in manipulation tasks, matching temporal-difference methods.
☆ Modified Hybrid A* Collision-Free Path-Planning for Automated Reverse Parking
Parking a vehicle in tight spaces is a challenging task to perform due to the scarcity of feasible paths that are also collision-free. This paper presents a strategy to tackle this kind of maneuver with a modified Hybrid-A* path-planning algorithm that combines the feasibility guarantee inherent in the standard Hybrid A* algorithm with the addition of static obstacle collision avoidance. A kinematic single-track model is derived to describe the low-speed motion of the vehicle, which is subsequently used as the motion model in the Hybrid A* path-planning algorithm to generate feasible motion primitive branches. The model states are also used to reconstruct the vehicle centerline, which, in conjunction with an inflated binary occupancy map, facilitates static obstacle collision avoidance functions. Simulation study and animation are set up to test the efficacy of the approach, and the proposed algorithm proves to consistently provide kinematically feasible trajectories that are also collision-free.
☆ Semantic-Drive: Democratizing Long-Tail Data Curation via Open-Vocabulary Grounding and Neuro-Symbolic VLM Consensus
The development of robust Autonomous Vehicles (AVs) is bottlenecked by the scarcity of "Long-Tail" training data. While fleets collect petabytes of video logs, identifying rare safety-critical events (e.g., erratic jaywalking, construction diversions) remains a manual, cost-prohibitive process. Existing solutions rely on coarse metadata search, which lacks precision, or cloud-based VLMs, which are privacy-invasive and expensive. We introduce Semantic-Drive, a local-first, neuro-symbolic framework for semantic data mining. Our approach decouples perception into two stages: (1) Symbolic Grounding via a real-time open-vocabulary detector (YOLOE) to anchor attention, and (2) Cognitive Analysis via a Reasoning VLM that performs forensic scene analysis. To mitigate hallucination, we implement a "System 2" inference-time alignment strategy, utilizing a multi-model "Judge-Scout" consensus mechanism. Benchmarked on the nuScenes dataset against the Waymo Open Dataset (WOD-E2E) taxonomy, Semantic-Drive achieves a Recall of 0.966 (vs. 0.475 for CLIP) and reduces Risk Assessment Error by 40\% compared to single models. The system runs entirely on consumer hardware (NVIDIA RTX 3090), offering a privacy-preserving alternative to the cloud.
☆ Taylor-Lagrange Control for Safety-Critical Systems
This paper proposes a novel Taylor-Lagrange Control (TLC) method for nonlinear control systems to ensure the safety and stability through Taylor's theorem with Lagrange remainder. To achieve this, we expand a safety or stability function with respect to time along the system dynamics using the Lie derivative and Taylor's theorem. This expansion enables the control input to appear in the Taylor series at an order equivalent to the relative degree of the function. We show that the proposed TLC provides necessary and sufficient conditions for system safety and is applicable to systems and constraints of arbitrary relative degree. The TLC exhibits connections with existing Control Barrier Function (CBF) and Control Lyapunov Function (CLF) methods, and it further extends the CBF and CLF methods to the complex domain, especially for higher order cases. Compared to High-Order CBFs (HOCBFs), TLC is less restrictive as it does not require forward invariance of the intersection of a set of safe sets while HOCBFs do. We employ TLC to reformulate a constrained optimal control problem as a sequence of quadratic programs with a zero-order hold implementation method, and demonstrate the safety of zero-order hold TLC using an event-triggered control method to address inter-sampling effects. Finally, we illustrate the effectiveness of the proposed TLC method through an adaptive cruise control system and a robot control problem, and compare it with existing CBF methods.
comment: 13 pages
☆ A Review of Learning-Based Motion Planning: Toward a Data-Driven Optimal Control Approach
Motion planning for high-level autonomous driving is constrained by a fundamental trade-off between the transparent, yet brittle, nature of pipeline methods and the adaptive, yet opaque, "black-box" characteristics of modern learning-based systems. This paper critically synthesizes the evolution of the field -- from pipeline methods through imitation learning, reinforcement learning, and generative AI -- to demonstrate how this persistent dilemma has hindered the development of truly trustworthy systems. To resolve this impasse, we conduct a comprehensive review of learning-based motion planning methods. Based on this review, we outline a data-driven optimal control paradigm as a unifying framework that synergistically integrates the verifiable structure of classical control with the adaptive capacity of machine learning, leveraging real-world data to continuously refine key components such as system dynamics, cost functions, and safety constraints. We explore this framework's potential to enable three critical next-generation capabilities: "Human-Centric" customization, "Platform-Adaptive" dynamics adaptation, and "System Self-Optimization" via self-tuning. We conclude by proposing future research directions based on this paradigm, aimed at developing intelligent transportation systems that are simultaneously safe, interpretable, and capable of human-like autonomy.
comment: 34 pages, 11 figures
♻ ☆ Openpi Comet: Competition Solution For 2025 BEHAVIOR Challenge
The 2025 BEHAVIOR Challenge is designed to rigorously track progress toward solving long-horizon tasks by physical agents in simulated environments. BEHAVIOR-1K focuses on everyday household tasks that people most want robots to assist with and these tasks introduce long-horizon mobile manipulation challenges in realistic settings, bridging the gap between current research and real-world, human-centric applications. This report presents our solution to the 2025 BEHAVIOR Challenge in a very close 2nd place and substantially outperforms the rest of the submissions. Building on $π_{0.5}$, we focus on systematically building our solution by studying the effects of training techniques and data. Through careful ablations, we show the scaling power in pre-training and post-training phases for competitive performance. We summarize our practical lessons and design recommendations that we hope will provide actionable insights for the broader embodied AI community when adapting powerful foundation models to complex embodied scenarios.
comment: preprint
♻ ☆ Decoupled Q-Chunking
Temporal-difference (TD) methods learn state and action values efficiently by bootstrapping from their own future value predictions, but such a self-bootstrapping mechanism is prone to bootstrapping bias, where the errors in the value targets accumulate across steps and result in biased value estimates. Recent work has proposed to use chunked critics, which estimate the value of short action sequences ("chunks") rather than individual actions, speeding up value backup. However, extracting policies from chunked critics is challenging: policies must output the entire action chunk open-loop, which can be sub-optimal for environments that require policy reactivity and also challenging to model especially when the chunk length grows. Our key insight is to decouple the chunk length of the critic from that of the policy, allowing the policy to operate over shorter action chunks. We propose a novel algorithm that achieves this by optimizing the policy against a distilled critic for partial action chunks, constructed by optimistically backing up from the original chunked critic to approximate the maximum value achievable when a partial action chunk is extended to a complete one. This design retains the benefits of multi-step value propagation while sidestepping both the open-loop sub-optimality and the difficulty of learning action chunking policies for long action chunks. We evaluate our method on challenging, long-horizon offline goal-conditioned tasks and show that it reliably outperforms prior methods. Code: github.com/ColinQiyangLi/dqc.
comment: 76 pages, 14 figures
♻ ☆ From "Thumbs Up" to "10 out of 10": Reconsidering Scalar Feedback in Interactive Reinforcement Learning
Learning from human feedback is an effective way to improve robotic learning in exploration-heavy tasks. Compared to the wide application of binary human feedback, scalar human feedback has been used less because it is believed to be noisy and unstable. In this paper, we compare scalar and binary feedback, and demonstrate that scalar feedback benefits learning when properly handled. We collected binary or scalar feedback respectively from two groups of crowdworkers on a robot task. We found that when considering how consistently a participant labeled the same data, scalar feedback led to less consistency than binary feedback; however, the difference vanishes if small mismatches are allowed. Additionally, scalar and binary feedback show no significant differences in their correlations with key Reinforcement Learning targets. We then introduce Stabilizing TEacher Assessment DYnamics (STEADY) to improve learning from scalar feedback. Based on the idea that scalar feedback is muti-distributional, STEADY re-constructs underlying positive and negative feedback distributions and re-scales scalar feedback based on feedback statistics. We show that models trained with \textit{scalar feedback + STEADY } outperform baselines, including binary feedback and raw scalar feedback, in a robot reaching task with non-expert human feedback. Our results show that both binary feedback and scalar feedback are dynamic, and scalar feedback is a promising signal for use in interactive Reinforcement Learning.
♻ ☆ Iterative Compositional Data Generation for Robot Control
Collecting robotic manipulation data is expensive, making it impractical to acquire demonstrations for the combinatorially large space of tasks that arise in multi-object, multi-robot, and multi-environment settings. While recent generative models can synthesize useful data for individual tasks, they do not exploit the compositional structure of robotic domains and struggle to generalize to unseen task combinations. We propose a semantic compositional diffusion transformer that factorizes transitions into robot-, object-, obstacle-, and objective-specific components and learns their interactions through attention. Once trained on a limited subset of tasks, we show that our model can zero-shot generate high-quality transitions from which we can learn control policies for unseen task combinations. Then, we introduce an iterative self-improvement procedure in which synthetic data is validated via offline reinforcement learning and incorporated into subsequent training rounds. Our approach substantially improves zero-shot performance over monolithic and hard-coded compositional baselines, ultimately solving nearly all held-out tasks and demonstrating the emergence of meaningful compositional structure in the learned representations.
comment: Corrected reference chronological order and added acknowledgements; results unchanged
♻ ☆ An effective control of large systems of active particles: An application to evacuation problem
Manipulation of large systems of active particles is a serious challenge across diverse domains, including crowd management, control of robotic swarms, and coordinated material transport. The development of advanced control strategies for complex scenarios is hindered, however, by the lack of scalability and robustness of the existing methods, in particular, due to the need of an individual control for each agent. One possible solution involves controlling a system through a leader or a group of leaders, which other agents tend to follow. Using such an approach we develop an effective control strategy for a leader, combining reinforcement learning (RL) with artificial forces acting on the system. To describe the guidance of active particles by a leader we introduce the generalized Vicsek model. This novel method is then applied to the problem of the effective evacuation by a robot-rescuer (leader) of large groups of people from hazardous places. We demonstrate, that while a straightforward application of RL yields suboptimal results, even for advanced architectures, our approach provides a robust and efficient evacuation strategy. The source code supporting this study is publicly available at: https://github.com/cinemere/evacuation.
♻ ☆ Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning AAAI 2026
Language-conditioned manipulation facilitates human-robot interaction via behavioral cloning (BC), which learns control policies from human demonstrations and serves as a cornerstone of embodied AI. Overcoming compounding errors in sequential action decisions remains a central challenge to improving BC performance. Existing approaches mitigate compounding errors through data augmentation, expressive representation, or temporal abstraction. However, they suffer from physical discontinuities and semantic-physical misalignment, leading to inaccurate action cloning and intermittent execution. In this paper, we present Continuous vision-language-action Co-Learning with Semantic-Physical Alignment (CCoL), a novel BC framework that ensures temporally consistent execution and fine-grained semantic grounding. It generates robust and smooth action execution trajectories through continuous co-learning across vision, language, and proprioceptive inputs (e.g., robot internal states). Meanwhile, we anchor language semantics to visuomotor representations by a bidirectional cross-attention to learn contextual information for action generation, successfully overcoming the problem of semantic-physical misalignment. Extensive experiments show that CCoL achieves an average 8.0% relative improvement across three simulation suites, with up to 19.2% relative gain in human-demonstrated bimanual insertion tasks. Real-world tests on a 7-DoF robot further confirm CCoL's generalization under unseen and noisy object states.
comment: Accepted at AAAI 2026, the Project website is available at https://qhemu.github.io/CCoL/
♻ ☆ Fast Multi-Party Open-Ended Conversation with a Social Robot
Multi-party open-ended conversation remains a major challenge in human-robot interaction, particularly when robots must recognise speakers, allocate turns, and respond coherently under overlapping or rapidly shifting dialogue. This paper presents a multi-party conversational system that combines multimodal perception (voice direction of arrival, speaker diarisation, face recognition) with a large language model for response generation. Implemented on the Furhat robot, the system was evaluated with 30 participants across two scenarios: (i) parallel, separate conversations and (ii) shared group discussion. Results show that the system maintains coherent and engaging conversations, achieving high addressee accuracy in parallel settings (92.6%) and strong face recognition reliability (80-94%). Participants reported clear social presence and positive engagement, although technical barriers such as audio-based speaker recognition errors and response latency affected the fluidity of group interactions. The results highlight both the promise and limitations of LLM-based multi-party interaction and outline concrete directions for improving multimodal cue integration and responsiveness in future social robots.
comment: 15 pages, 5 figures, 4 tables; 2 appendices
♻ ☆ Feedback-MPPI: Fast Sampling-Based MPC via Rollout Differentiation -- Adios low-level controllers
Model Predictive Path Integral control is a powerful sampling-based approach suitable for complex robotic tasks due to its flexibility in handling nonlinear dynamics and non-convex costs. However, its applicability in real-time, highfrequency robotic control scenarios is limited by computational demands. This paper introduces Feedback-MPPI (F-MPPI), a novel framework that augments standard MPPI by computing local linear feedback gains derived from sensitivity analysis inspired by Riccati-based feedback used in gradient-based MPC. These gains allow for rapid closed-loop corrections around the current state without requiring full re-optimization at each timestep. We demonstrate the effectiveness of F-MPPI through simulations and real-world experiments on two robotic platforms: a quadrupedal robot performing dynamic locomotion on uneven terrain and a quadrotor executing aggressive maneuvers with onboard computation. Results illustrate that incorporating local feedback significantly improves control performance and stability, enabling robust, high-frequency operation suitable for complex robotic systems.
♻ ☆ Real-Time QP Solvers: A Concise Review and Practical Guide Towards Legged Robots
Quadratic programming (QP) underpins real-time robotics by enabling efficient, constrained optimization in state estimation, motion planning, and control. In legged locomotion and manipulation, essential modules like inverse dynamics, Model Predictive Control (MPC), and Whole-Body Control (WBC) are inherently QP-based, demanding reliable solutions amid tight timing, energy, and computational resources on embedded platforms. This paper presents a comprehensive analysis and benchmarking study of QP solvers for legged robotics. We begin by formulating the standard convex QP and classify solvers into principal algorithmic approaches: interior-point methods, active-set strategies, operator-splitting schemes, and augmented Lagrangian/proximal approaches, while also discussing solver code generation for fixed-structure QPs. Each solver is examined in terms of algorithmic structure, computational characteristics, and its ability to exploit problem structure and warm-starting. Performance is reviewed using publicly available benchmarks, with a focus on metrics such as computation time, constraint satisfaction, and robustness under perturbations. Unified comparison tables yield practical guidance for solver selection, underscoring trade-offs in speed, accuracy, and energy efficiency. Our findings emphasize the synergy between solvers, tasks, and hardware -- e.g., sparse structured IPMs for long-horizon MPC and dense active-set for high-frequency WBC to advance agile, autonomous legged systems, with emerging trends toward ill-conditioned, conic, and code-generated deployments.
comment: 12 pages, 1 figure, 2 tables
♻ ☆ Driving Through Uncertainty: Risk-Averse Control with LLM Commonsense for Autonomous Driving under Perception Deficits
Partial perception deficits can compromise autonomous vehicle safety by disrupting environmental understanding. Existing protocols typically default to entirely risk-avoidant actions such as immediate stops, which are detrimental to navigation goals and lack flexibility for rare driving scenarios. Yet, in cases of minor risk, halting the vehicle may be unnecessary, and more adaptive responses are preferable. In this paper, we propose LLM-RCO, a risk-averse framework leveraging large language models (LLMs) to integrate human-like driving commonsense into autonomous systems facing perception deficits. LLM-RCO features four key modules interacting with the dynamic driving environment: hazard inference, short-term motion planner, action condition verifier, and safety constraint generator, enabling proactive and context-aware actions in such challenging conditions. To enhance the driving decision-making of LLMs, we construct DriveLM-Deficit, a dataset of 53,895 video clips featuring deficits of safety-critical objects, annotated for LLM fine-tuning in hazard detection and motion planning. Extensive experiments in adverse driving conditions with the CARLA simulator demonstrate that LLM-RCO promotes proactive maneuvers over purely risk-averse actions in perception deficit scenarios, underscoring its value for boosting autonomous driving resilience against perception loss challenges.
♻ ☆ Social Mediation through Robots -- A Scoping Review on Improving Group Interactions through Directed Robot Action using an Extended Group Process Model
Group processes refer to the dynamics that occur within a group and are critical for understanding how groups function. With robots being increasingly placed within small groups, improving these processes has emerged as an important application of social robotics. Social Mediation Robots elicit behavioral change within groups by deliberately influencing the processes of groups. While research in this field has demonstrated that robots can effectively affect interpersonal dynamics, there is a notable gap in integrating these insights to develop coherent understanding and theory. We present a scoping review of literature targeting changes in social interactions between multiple humans through intentional action from robotic agents. To guide our review, we adapt the classical Input-Process-Output (I-P-O) models that we call "Mediation I-P-O model". We evaluated 1633 publications, which yielded 89 distinct social mediation concepts. We construct 11 mediation approaches robots can use to shape processes in small groups and teams. This work strives to produce generalizable insights and evaluate the extent to which the potential of social mediation through robots has been realized thus far. We hope that the proposed framework encourages a holistic approach to the study of social mediation and provides a foundation to standardize future reporting in the domain.
comment: Early version of the published journal paper: Weisswange, T. H., Javed, H., Dietrich, M., Jung, M. F., & Jamali, N. (2025). Design Implications for Robots that Facilitate Groups-A Scoping Review on Improving Group Interactions through Directed Robot Action. ACM Transactions on Human-Robot Interaction
♻ ☆ WARPD: World model Assisted Reactive Policy Diffusion NeurIPS 2025
With the increasing availability of open-source robotic data, imitation learning has become a promising approach for both manipulation and locomotion. Diffusion models are now widely used to train large, generalized policies that predict controls or trajectories, leveraging their ability to model multimodal action distributions. However, this generality comes at the cost of larger model sizes and slower inference, an acute limitation for robotic tasks requiring high control frequencies. Moreover, Diffusion Policy (DP), a popular trajectory-generation approach, suffers from a trade-off between performance and action horizon: fewer diffusion queries lead to larger trajectory chunks, which in turn accumulate tracking errors. To overcome these challenges, we introduce WARPD (World model Assisted Reactive Policy Diffusion), a method that generates closed-loop policies (weights for neural policies) directly, instead of open-loop trajectories. By learning behavioral distributions in parameter space rather than trajectory space, WARPD offers two major advantages: (1) extended action horizons with robustness to perturbations, while maintaining high task performance, and (2) significantly reduced inference costs. Empirically, WARPD outperforms DP in long-horizon and perturbed environments, and achieves multitask performance on par with DP while requiring only ~ 1/45th of the inference-time FLOPs per step.
comment: Outstanding Paper Award at the Embodied World Models for Decision Making Workshop at NeurIPS 2025
♻ ☆ Communication-Efficient Module-Wise Federated Learning for Grasp Pose Detection in Cluttered Environments
Grasp pose detection (GPD) is a fundamental capability for robotic autonomy, but its reliance on large, diverse datasets creates significant data privacy and centralization challenges. Federated Learning (FL) offers a privacy-preserving solution, but its application to GPD is hindered by the substantial communication overhead of large models, a key issue for resource-constrained robots. To address this, we propose a novel module-wise FL framework that begins by analyzing the learning dynamics of the GPD model's functional components. This analysis identifies slower-converging modules, to which our framework then allocates additional communication effort. This is realized through a two-phase process: a standard full-model training phase is followed by a communication-efficient phase where only the identified subset of slower-converging modules is trained and their partial updates are aggregated. Extensive experiments on the GraspNet-1B dataset demonstrate that our method outperforms standard FedAvg and other baselines, achieving higher accuracy for a given communication budget. Furthermore, real-world experiments on a physical robot validate our approach, showing a superior grasp success rate compared to baseline methods in cluttered scenes. Our work presents a communication-efficient framework for training robust, generalized GPD models in a decentralized manner, effectively improving the trade-off between communication cost and model performance.
comment: Accepted to IEEE Robotics and Automation Letters (RA-L). 8 pages, 5 figures
♻ ☆ MOFU: Development of a MOrphing Fluffy Unit with Expansion and Contraction Capabilities and Evaluation of the Animacy of Its Movements
Robots designed for therapy and social interaction aim to evoke a sense of animacy in humans. While many studies have focused on life like appearance or joint based movements, the effect of whole body volume changing movements commonly observed in living organisms has received little attention. In this study, we developed MOFU MOrphing Fluffy Unit, a mobile robot capable of whole body expansion and contraction using a single motor enclosed in a fluffy exterior. MOFU employs a Jitterbug geometric transformation mechanism that enables smooth diameter changes from approximately 210 mm to 280 mm with a single actuator, and is equipped with a differential two wheel drive mechanism for locomotion. We conducted an online survey using videos of MOFU behaviors and evaluated perceived animacy using the Godspeed Questionnaire Series. First, we compared stationary conditions with and without expansion contraction and with and without rotational motion. Both expansion contraction and rotation independently increased perceived animacy. Second, we examined whether presenting two MOFUs simultaneously would further enhance animacy perception, but no significant difference was observed. Exploratory analyses were also conducted across four dual robot motion conditions. Third, when expansion contraction was combined with locomotion, animacy ratings were higher than for locomotion alone. These results suggest that whole body volume changing movements enhance perceived animacy in robots, indicating that physical volume change is an important design element for future social and therapeutic robots.
♻ ☆ Bio-inspired reconfigurable stereo vision for robotics using omnidirectional cameras ICRA 2025
This work introduces a novel bio-inspired reconfigurable stereo vision system for robotics, leveraging omnidirectional cameras and a novel algorithm to achieve flexible visual capabilities. Inspired by the adaptive vision of various species, our visual system addresses traditional stereo vision limitations, i.e., immutable camera alignment with narrow fields of view, by introducing a reconfigurable stereo vision system to robotics. Our key innovations include the reconfigurable stereo vision strategy that allows dynamic camera alignment, a robust depth measurement system utilizing a nonrectified geometrical method combined with a deep neural network for feature matching, and a geometrical compensation technique to enhance visual accuracy. Implemented on a metamorphic robot, this vision system demonstrates its great adaptability to various scenarios by switching its configurations of 316° monocular with 79° binocular field for fast target seeking and 242° monocular with 150° binocular field for detailed close inspection.
comment: 7 pages, 8 figures, submitted to IEEE ICRA 2025
♻ ☆ Adaptive Compressive Tactile Subsampling: Enabling High Spatiotemporal Resolution in Scalable Robotic Skin
Robots require full-body, high-resolution tactile sensing to operate safely in unstructured environments, enabling reflexive responses and closed-loop control. However, the pixel counts needed for dense, large-area coverage limit readout rates of most tactile arrays to <100 Hz, hindering their use in high-speed tasks. We present Adaptive Compressive Tactile Subsampling (ACTS), a scalable and data-driven method that greatly enhances traditional tactile matrices by leveraging adaptive sensor sampling and sparse recovery. By adaptively allocating measurements to informative regions, ACTS is especially effective for spatially sparse signals common in real-world interactions. Tested on a 1024-pixel tactile sensor array (32x32), ACTS achieved frame rates up to 1,000 Hz, an 18X improvement over conventional raster scanning, with minimal reconstruction error. For the first time, ACTS enables wearable, large-area, high-density tactile sensing systems that can deliver high-speed results. We demonstrate rapid object classification within 20 ms of contact, high-speed projectile detection, ricochet angle estimation, and soft deformation tracking, in tactile and robotics applications, all using flexible, high-density tactile arrays. These include high-resolution tactile gloves, pressure insoles, and full-body configurations covering robotic arms and human-sized mannequins. We further showcase tactile-based closed-loop control by guiding a metallic ball to trace letters using tactile feedback and by executing tactile-only whole-hand reflexes on a fully sensorized LEAP hand to stabilize grasps, prevent slip, and avoid sharp objects, validating ACTS for real-time interaction and motion control. ACTS transforms standard, low-cost, and robust tactile sensors into high-speed systems enabling scalable, responsive, and adaptive tactile perception for robots and wearables operating in dynamic environments.
comment: 51 pages, 11 main figures, 16 supplemental figures, Videos can be accessed at https://tinyurl.com/ACTS-videos
♻ ☆ In-situ Value-aligned Human-Robot Interactions with Physical Constraints IROS 2025
Equipped with Large Language Models (LLMs), human-centered robots are now capable of performing a wide range of tasks that were previously deemed challenging or unattainable. However, merely completing tasks is insufficient for cognitive robots, who should learn and apply human preferences to future scenarios. In this work, we propose a framework that combines human preferences with physical constraints, requiring robots to complete tasks while considering both. Firstly, we developed a benchmark of everyday household activities, which are often evaluated based on specific preferences. We then introduced In-Context Learning from Human Feedback (ICLHF), where human feedback comes from direct instructions and adjustments made intentionally or unintentionally in daily life. Extensive sets of experiments, testing the ICLHF to generate task plans and balance physical constraints with preferences, have demonstrated the efficiency of our approach. Project page: https://iclhf.github.io .
comment: 8 pages, 7 figures. Accepted by IROS 2025
♻ ☆ UMArm: Untethered, Modular, Portable, Soft Pneumatic Arm
Robotic arms are essential to modern industries, however, their adaptability to unstructured environments remains limited. Soft robotic arms, particularly those actuated pneumatically, offer greater adaptability in unstructured environments and enhanced safety for human-robot interaction. However, current pneumatic soft arms are constrained by limited degrees of freedom, precision, payload capacity, and reliance on bulky external pressure regulators. In this work, a novel pneumatically driven rigid-soft hybrid arm, ``UMArm'', is presented. The shortcomings of pneumatically actuated soft arms are addressed by densely integrating high-force-to-weight-ratio, self-regulated McKibben actuators onto a lightweight rigid spine structure. The modified McKibben actuators incorporate valves and controllers directly inside, eliminating the need for individual pressure lines and external regulators, significantly reducing system weight and complexity. Full untethered operation, high payload capacity, precision, and directionally tunable compliance are achieved by the UMArm. Portability is demonstrated through a wearable assistive arm experiment, and versatility is showcased by reconfiguring the system into an inchworm robot. The results of this work show that the high-degree-of-freedom, external-regulator-free pneumatically driven arm systems like the UMArm possess great potential for real-world unstructured environments.
♻ ☆ Gaze on the Prize: Shaping Visual Attention with Return-Guided Contrastive Learning
Visual Reinforcement Learning (RL) agents must learn to act based on high-dimensional image data where only a small fraction of the pixels is task-relevant. This forces agents to waste exploration and computational resources on irrelevant features, leading to sample-inefficient and unstable learning. To address this, inspired by human visual foveation, we introduce Gaze on the Prize. This framework augments visual RL with a learnable foveal attention mechanism (Gaze), guided by a self-supervised signal derived from the agent's experience pursuing higher returns (the Prize). Our key insight is that return differences reveal what matters most: If two similar representations produce different outcomes, their distinguishing features are likely task-relevant, and the gaze should focus on them accordingly. This is realized through return-guided contrastive learning that trains the attention to distinguish between the features relevant to success and failure. We group similar visual representations into positives and negatives based on their return differences and use the resulting labels to construct contrastive triplets. These triplets provide the training signal that teaches the attention mechanism to produce distinguishable representations for states associated with different outcomes. Our method achieves up to 2.52x improvement in sample efficiency and can solve challenging tasks from the ManiSkill3 benchmark that the baseline fails to learn, without modifying the underlying algorithm or hyperparameters.
comment: Project page: https://andrewcwlee.github.io/gaze-on-the-prize
♻ ☆ STITCHER: Constrained Trajectory Planning in Complex Environments with Real-Time Motion Primitive Search
Autonomous high-speed navigation through large, complex environments requires real-time generation of agile trajectories that are dynamically feasible, collision-free, and satisfy state or actuator constraints. Modern trajectory planning techniques primarily use numerical optimization, as they enable the systematic computation of high-quality, expressive trajectories that satisfy various constraints. However, stringent requirements on computation time and the risk of numerical instability can limit the use of optimization-based planners in safety-critical scenarios. This work presents an optimization-free planning framework called STITCHER that stitches short trajectory segments together with graph search to compute long-range, expressive, and near-optimal trajectories in real-time. STITCHER outperforms modern optimization-based planners through our innovative planning architecture and several algorithmic developments that make real-time planning possible. Extensive simulation testing is performed to analyze the algorithmic components that make up STITCHER, along with a thorough comparison with two state-of-the-art optimization planners. Simulation tests show that safe trajectories can be created within a few milliseconds for paths that span the entirety of two 50 m x 50 m environments. Hardware tests with a custom quadrotor verify that STITCHER can produce trackable paths in real-time while respecting nonconvex constraints, such as limits on tilt angle and motor forces, which are otherwise hard to include in optimization-based planners.
♻ ☆ STITCHER: Real-Time Trajectory Planning with Motion Primitive Search
Autonomous high-speed navigation through large, complex environments requires real-time generation of agile trajectories that are dynamically feasible, collision-free, and satisfy constraints. Most modern trajectory planning techniques rely on numerical optimization because high-quality, expressive trajectories that satisfy constraints can be systematically computed. However, strict requirements on computation time and the risk of numerical instability can limit the use of optimization-based planners in safety-critical situations. This work presents an optimization-free planning framework called STITCHER that leverages graph search to generate long-range trajectories by stitching short trajectory segments together in real time. STITCHER is shown to outperform modern optimization-based planners through its innovative planning architecture and several algorithmic developments that make real-time planning possible. Simulation results show safe trajectories through complex environments can be generated in milliseconds that cover tens of meters.
Robotics 58
☆ ImplicitRDP: An End-to-End Visual-Force Diffusion Policy with Structural Slow-Fast Learning
Human-level contact-rich manipulation relies on the distinct roles of two key modalities: vision provides spatially rich but temporally slow global context, while force sensing captures rapid, high-frequency local contact dynamics. Integrating these signals is challenging due to their fundamental frequency and informational disparities. In this work, we propose ImplicitRDP, a unified end-to-end visual-force diffusion policy that integrates visual planning and reactive force control within a single network. We introduce Structural Slow-Fast Learning, a mechanism utilizing causal attention to simultaneously process asynchronous visual and force tokens, allowing the policy to perform closed-loop adjustments at the force frequency while maintaining the temporal coherence of action chunks. Furthermore, to mitigate modality collapse where end-to-end models fail to adjust the weights across different modalities, we propose Virtual-target-based Representation Regularization. This auxiliary objective maps force feedback into the same space as the action, providing a stronger, physics-grounded learning signal than raw force prediction. Extensive experiments on contact-rich tasks demonstrate that ImplicitRDP significantly outperforms both vision-only and hierarchical baselines, achieving superior reactivity and success rates with a streamlined training pipeline. Code and videos will be publicly available at https://implicit-rdp.github.io.
comment: Project page: https://implicit-rdp.github.io
☆ Any4D: Unified Feed-Forward Metric 4D Reconstruction
We present Any4D, a scalable multi-view transformer for metric-scale, dense feed-forward 4D reconstruction. Any4D directly generates per-pixel motion and geometry predictions for N frames, in contrast to prior work that typically focuses on either 2-view dense scene flow or sparse 3D point tracking. Moreover, unlike other recent methods for 4D reconstruction from monocular RGB videos, Any4D can process additional modalities and sensors such as RGB-D frames, IMU-based egomotion, and Radar Doppler measurements, when available. One of the key innovations that allows for such a flexible framework is a modular representation of a 4D scene; specifically, per-view 4D predictions are encoded using a variety of egocentric factors (depthmaps and camera intrinsics) represented in local camera coordinates, and allocentric factors (camera extrinsics and scene flow) represented in global world coordinates. We achieve superior performance across diverse setups - both in terms of accuracy (2-3X lower error) and compute efficiency (15X faster), opening avenues for multiple downstream applications.
comment: Project Website: https://any-4d.github.io/
☆ Curriculum-Based Reinforcement Learning for Autonomous UAV Navigation in Unknown Curved Tubular Conduit
Autonomous drone navigation in confined tubular environments remains a major challenge due to the constraining geometry of the conduits, the proximity of the walls, and the perceptual limitations inherent to such scenarios. We propose a reinforcement learning approach enabling a drone to navigate unknown three-dimensional tubes without any prior knowledge of their geometry, relying solely on local observations from LiDAR and a conditional visual detection of the tube center. In contrast, the Pure Pursuit algorithm, used as a deterministic baseline, benefits from explicit access to the centerline, creating an information asymmetry designed to assess the ability of RL to compensate for the absence of a geometric model. The agent is trained through a progressive Curriculum Learning strategy that gradually exposes it to increasingly curved geometries, where the tube center frequently disappears from the visual field. A turning-negotiation mechanism, based on the combination of direct visibility, directional memory, and LiDAR symmetry cues, proves essential for ensuring stable navigation under such partial observability conditions. Experiments show that the PPO policy acquires robust and generalizable behavior, consistently outperforming the deterministic controller despite its limited access to geometric information. Validation in a high-fidelity 3D environment further confirms the transferability of the learned behavior to a continuous physical dynamics. The proposed approach thus provides a complete framework for autonomous navigation in unknown tubular environments and opens perspectives for industrial, underground, or medical applications where progressing through narrow and weakly perceptive conduits represents a central challenge.
☆ Decoupled Q-Chunking
Temporal-difference (TD) methods learn state and action values efficiently by bootstrapping from their own future value predictions, but such a self-bootstrapping mechanism is prone to bootstrapping bias, where the errors in the value targets accumulate across steps and result in biased value estimates. Recent work has proposed to use chunked critics, which estimate the value of short action sequences ("chunks") rather than individual actions, speeding up value backup. However, extracting policies from chunked critics is challenging: policies must output the entire action chunk open-loop, which can be sub-optimal for environments that require policy reactivity and also challenging to model especially when the chunk length grows. Our key insight is to decouple the chunk length of the critic from that of the policy, allowing the policy to operate over shorter action chunks. We propose a novel algorithm that achieves this by optimizing the policy against a distilled critic for partial action chunks, constructed by optimistically backing up from the original chunked critic to approximate the maximum value achievable when a partial action chunk is extended to a complete one. This design retains the benefits of multi-step value propagation while sidestepping both the open-loop sub-optimality and the difficulty of learning action chunking policies for long action chunks. We evaluate our method on challenging, long-horizon offline goal-conditioned tasks and show that it reliably outperforms prior methods. Code: github.com/ColinQiyangLi/dqc.
comment: 76 pages, 14 figures
☆ Digital Twin Supervised Reinforcement Learning Framework for Autonomous Underwater Navigation
Autonomous navigation in underwater environments remains a major challenge due to the absence of GPS, degraded visibility, and the presence of submerged obstacles. This article investigates these issues through the case of the BlueROV2, an open platform widely used for scientific experimentation. We propose a deep reinforcement learning approach based on the Proximal Policy Optimization (PPO) algorithm, using an observation space that combines target-oriented navigation information, a virtual occupancy grid, and ray-casting along the boundaries of the operational area. The learned policy is compared against a reference deterministic kinematic planner, the Dynamic Window Approach (DWA), commonly employed as a robust baseline for obstacle avoidance. The evaluation is conducted in a realistic simulation environment and complemented by validation on a physical BlueROV2 supervised by a 3D digital twin of the test site, helping to reduce risks associated with real-world experimentation. The results show that the PPO policy consistently outperforms DWA in highly cluttered environments, notably thanks to better local adaptation and reduced collisions. Finally, the experiments demonstrate the transferability of the learned behavior from simulation to the real world, confirming the relevance of deep RL for autonomous navigation in underwater robotics.
☆ Iterative Compositional Data Generation for Robot Control
Collecting robotic manipulation data is expensive, making it impractical to acquire demonstrations for the combinatorially large space of tasks that arise in multi-object, multi-robot, and multi-environment settings. While recent generative models can synthesize useful data for individual tasks, they do not exploit the compositional structure of robotic domains and struggle to generalize to unseen task combinations. We propose a semantic compositional diffusion transformer that factorizes transitions into robot-, object-, obstacle-, and objective-specific components and learns their interactions through attention. Once trained on a limited subset of tasks, we show that our model can zero-shot generate high-quality transitions from which we can learn control policies for unseen task combinations. Then, we introduce an iterative self-improvement procedure in which synthetic data is validated via offline reinforcement learning and incorporated into subsequent training rounds. Our approach substantially improves zero-shot performance over monolithic and hard-coded compositional baselines, ultimately solving nearly all held-out tasks and demonstrating the emergence of meaningful compositional structure in the learned representations.
☆ V-OCBF: Learning Safety Filters from Offline Data via Value-Guided Offline Control Barrier Functions
Ensuring safety in autonomous systems requires controllers that satisfy hard, state-wise constraints without relying on online interaction. While existing Safe Offline RL methods typically enforce soft expected-cost constraints, they do not guarantee forward invariance. Conversely, Control Barrier Functions (CBFs) provide rigorous safety guarantees but usually depend on expert-designed barrier functions or full knowledge of the system dynamics. We introduce Value-Guided Offline Control Barrier Functions (V-OCBF), a framework that learns a neural CBF entirely from offline demonstrations. Unlike prior approaches, V-OCBF does not assume access to the dynamics model; instead, it derives a recursive finite-difference barrier update, enabling model-free learning of a barrier that propagates safety information over time. Moreover, V-OCBF incorporates an expectile-based objective that avoids querying the barrier on out-of-distribution actions and restricts updates to the dataset-supported action set. The learned barrier is then used with a Quadratic Program (QP) formulation to synthesize real-time safe control. Across multiple case studies, V-OCBF yields substantially fewer safety violations than baseline methods while maintaining strong task performance, highlighting its scalability for offline synthesis of safety-critical controllers without online interaction or hand-engineered barriers.
comment: 23 pages, 8 figure, 7 tables
☆ AERMANI-Diffusion: Regime-Conditioned Diffusion for Dynamics Learning in Aerial Manipulators
Aerial manipulators undergo rapid, configuration-dependent changes in inertial coupling forces and aerodynamic forces, making accurate dynamics modeling a core challenge for reliable control. Analytical models lose fidelity under these nonlinear and nonstationary effects, while standard data-driven methods such as deep neural networks and Gaussian processes cannot represent the diverse residual behaviors that arise across different operating conditions. We propose a regime-conditioned diffusion framework that models the full distribution of residual forces using a conditional diffusion process and a lightweight temporal encoder. The encoder extracts a compact summary of recent motion and configuration, enabling consistent residual predictions even through abrupt transitions or unseen payloads. When combined with an adaptive controller, the framework enables dynamics uncertainty compensation and yields markedly improved tracking accuracy in real-world tests.
☆ Distribution-Free Stochastic MPC for Joint-in-Time Chance-Constrained Linear Systems
This work presents a stochastic model predictive control (MPC) framework for linear systems subject to joint-in-time chance constraints under unknown disturbance distributions. Unlike existing stochastic MPC formulations that rely on parametric or Gaussian assumptions or require expensive offline computations, the proposed method leverages conformal prediction (CP) as a streamlined tool to construct finite-sample confidence regions for the system's stochastic error trajectories with minimal computational effort. These regions enable the relaxation of probabilistic constraints while providing formal guarantees. By employing an indirect feedback mechanism and a probabilistic set-based formulation, we prove recursive feasibility of the relaxed optimization problem and establish chance constraint satisfaction in closed-loop. Furthermore, we extend the approach to the more general output feedback setting with unknown measurement noise distributions. Given available noise samples, we establish satisfaction of the joint chance constraints and recursive feasibility via output measurements alone. Numerical examples demonstrate the effectiveness and advantages of the proposed method compared to existing approaches.
☆ On the Stabilization of Rigid Formations on Regular Curves
This work deals with the problem of stabilizing a multi-agent rigid formation on a general class of planar curves. Namely, we seek to stabilize an equilateral polygonal formation on closed planar differentiable curves after a path sweep. The task of finding an inscribed regular polygon centered at the point of interest is solved via a randomized multi-start Newton-Like algorithm for which one is able to ascertain the existence of a minimizer. Then we design a continuous feedback law that guarantees convergence to, and sufficient sweeping of the curve, followed by convergence to the desired formation vertices while ensuring inter-agent avoidance. The proposed approach is validated through numerical simulations for different classes of curves and different rigid formations. Code: https://github.com/mebbaid/paper-elobaid-ifacwc-2026
☆ How to Brake? Ethical Emergency Braking with Deep Reinforcement Learning
Connected and automated vehicles (CAVs) have the potential to enhance driving safety, for example by enabling safe vehicle following and more efficient traffic scheduling. For such future deployments, safety requirements should be addressed, where the primary such are avoidance of vehicle collisions and substantial mitigating of harm when collisions are unavoidable. However, conservative worst-case-based control strategies come at the price of reduced flexibility and may compromise overall performance. In light of this, we investigate how Deep Reinforcement Learning (DRL) can be leveraged to improve safety in multi-vehicle-following scenarios involving emergency braking. Specifically, we investigate how DRL with vehicle-to-vehicle communication can be used to ethically select an emergency breaking profile in scenarios where overall, or collective, three-vehicle harm reduction or collision avoidance shall be obtained instead of single-vehicle such. As an algorithm, we provide a hybrid approach that combines DRL with a previously published method based on analytical expressions for selecting optimal constant deceleration. By combining DRL with the previous method, the proposed hybrid approach increases the reliability compared to standalone DRL, while achieving superior performance in terms of overall harm reduction and collision avoidance.
☆ Evaluating Gemini Robotics Policies in a Veo World Simulator
Generative world models hold significant potential for simulating interactions with visuomotor policies in varied environments. Frontier video models can enable generation of realistic observations and environment interactions in a scalable and general manner. However, the use of video models in robotics has been limited primarily to in-distribution evaluations, i.e., scenarios that are similar to ones used to train the policy or fine-tune the base video model. In this report, we demonstrate that video models can be used for the entire spectrum of policy evaluation use cases in robotics: from assessing nominal performance to out-of-distribution (OOD) generalization, and probing physical and semantic safety. We introduce a generative evaluation system built upon a frontier video foundation model (Veo). The system is optimized to support robot action conditioning and multi-view consistency, while integrating generative image-editing and multi-view completion to synthesize realistic variations of real-world scenes along multiple axes of generalization. We demonstrate that the system preserves the base capabilities of the video model to enable accurate simulation of scenes that have been edited to include novel interaction objects, novel visual backgrounds, and novel distractor objects. This fidelity enables accurately predicting the relative performance of different policies in both nominal and OOD conditions, determining the relative impact of different axes of generalization on policy performance, and performing red teaming of policies to expose behaviors that violate physical or semantic safety constraints. We validate these capabilities through 1600+ real-world evaluations of eight Gemini Robotics policy checkpoints and five tasks for a bimanual manipulator.
☆ LEO-RobotAgent: A General-purpose Robotic Agent for Language-driven Embodied Operator
We propose LEO-RobotAgent, a general-purpose language-driven intelligent agent framework for robots. Under this framework, LLMs can operate different types of robots to complete unpredictable complex tasks across various scenarios. This framework features strong generalization, robustness, and efficiency. The application-level system built around it can fully enhance bidirectional human-robot intent understanding and lower the threshold for human-robot interaction. Regarding robot task planning, the vast majority of existing studies focus on the application of large models in single-task scenarios and for single robot types. These algorithms often have complex structures and lack generalizability. Thus, the proposed LEO-RobotAgent framework is designed with a streamlined structure as much as possible, enabling large models to independently think, plan, and act within this clear framework. We provide a modular and easily registrable toolset, allowing large models to flexibly call various tools to meet different requirements. Meanwhile, the framework incorporates a human-robot interaction mechanism, enabling the algorithm to collaborate with humans like a partner. Experiments have verified that this framework can be easily adapted to mainstream robot platforms including unmanned aerial vehicles (UAVs), robotic arms, and wheeled robot, and efficiently execute a variety of carefully designed tasks with different complexity levels. Our code is available at https://github.com/LegendLeoChen/LEO-RobotAgent.
☆ Motion Planning for Safe Landing of a Human-Piloted Parafoil
Most skydiving accidents occur during the parafoil-piloting and landing stages and result from human lapses in judgment while piloting the parafoil. Training of novice pilots is protracted due to the lack of functional and easily accessible training simulators. Moreover, work on parafoil trajectory planning suitable for aiding human training remains limited. To bridge this gap, we study the problem of computing safe trajectories for human-piloted parafoil flight and examine how such trajectories fare against human-generated solutions. For the algorithmic part, we adapt the sampling-based motion planner Stable Sparse RRT (SST) by Li et al., to cope with the problem constraints while minimizing the bank angle (control effort) as a proxy for safety. We then compare the computer-generated solutions with data from human-generated parafoil flight, where the algorithm offers a relative cost improvement of 20\%-80\% over the performance of the human pilot. We observe that human pilots tend to, first, close the horizontal distance to the landing area, and then address the vertical gap by spiraling down to the suitable altitude for starting a landing maneuver. The algorithm considered here makes smoother and more gradual descents, arriving at the landing area at the precise altitude necessary for the final approach while maintaining safety constraints. Overall, the study demonstrates the potential of computer-generated guidelines, rather than traditional rules of thumb, which can be integrated into future simulators to train pilots for safer and more cost-effective flights.
☆ Mr. Virgil: Learning Multi-robot Visual-range Relative Localization IROS
Ultra-wideband (UWB)-vision fusion localization has achieved extensive applications in the domain of multi-agent relative localization. The challenging matching problem between robots and visual detection renders existing methods highly dependent on identity-encoded hardware or delicate tuning algorithms. Overconfident yet erroneous matches may bring about irreversible damage to the localization system. To address this issue, we introduce Mr. Virgil, an end-to-end learning multi-robot visual-range relative localization framework, consisting of a graph neural network for data association between UWB rangings and visual detections, and a differentiable pose graph optimization (PGO) back-end. The graph-based front-end supplies robust matching results, accurate initial position predictions, and credible uncertainty estimates, which are subsequently integrated into the PGO back-end to elevate the accuracy of the final pose estimation. Additionally, a decentralized system is implemented for real-world applications. Experiments spanning varying robot numbers, simulation and real-world, occlusion and non-occlusion conditions showcase the stability and exactitude under various scenes compared to conventional methods. Our code is available at: https://github.com/HiOnes/Mr-Virgil.
comment: Accepted by 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
☆ Neural Ranging Inertial Odometry ICRA
Ultra-wideband (UWB) has shown promising potential in GPS-denied localization thanks to its lightweight and drift-free characteristics, while the accuracy is limited in real scenarios due to its sensitivity to sensor arrangement and non-Gaussian pattern induced by multi-path or multi-signal interference, which commonly occurs in many typical applications like long tunnels. We introduce a novel neural fusion framework for ranging inertial odometry which involves a graph attention UWB network and a recurrent neural inertial network. Our graph net learns scene-relevant ranging patterns and adapts to any number of anchors or tags, realizing accurate positioning without calibration. Additionally, the integration of least squares and the incorporation of nominal frame enhance overall performance and scalability. The effectiveness and robustness of our methods are validated through extensive experiments on both public and self-collected datasets, spanning indoor, outdoor, and tunnel environments. The results demonstrate the superiority of our proposed IR-ULSG in handling challenging conditions, including scenarios outside the convex envelope and cases where only a single anchor is available.
comment: Accepted by 2025 IEEE International Conference on Robotics and Automation (ICRA)
☆ Contact SLAM: An Active Tactile Exploration Policy Based on Physical Reasoning Utilized in Robotic Fine Blind Manipulation Tasks
Contact-rich manipulation is difficult for robots to execute and requires accurate perception of the environment. In some scenarios, vision is occluded. The robot can then no longer obtain real-time scene state information through visual feedback. This is called ``blind manipulation". In this manuscript, a novel physically-driven contact cognition method, called ``Contact SLAM", is proposed. It estimates the state of the environment and achieves manipulation using only tactile sensing and prior knowledge of the scene. To maximize exploration efficiency, this manuscript also designs an active exploration policy. The policy gradually reduces uncertainties in the manipulation scene. The experimental results demonstrated the effectiveness and accuracy of the proposed method in several contact-rich tasks, including the difficult and delicate socket assembly task and block-pushing task.
comment: 8 pages, 8 figures
☆ Seamless Outdoor-Indoor Pedestrian Positioning System with GNSS/UWB/IMU Fusion: A Comparison of EKF, FGO, and PF
Accurate and continuous pedestrian positioning across outdoor-indoor environments remains challenging because GNSS, UWB, and inertial PDR are complementary yet individually fragile under signal blockage, multipath, and drift. This paper presents a unified GNSS/UWB/IMU fusion framework for seamless pedestrian localization and provides a controlled comparison of three probabilistic back-ends: an error-state extended Kalman filter, sliding-window factor graph optimization, and a particle filter. The system uses chest-mounted IMU-based PDR as the motion backbone and integrates absolute updates from GNSS outdoors and UWB indoors. To enhance transition robustness and mitigate urban GNSS degradation, we introduce a lightweight map-based feasibility constraint derived from OpenStreetMap building footprints, treating most building interiors as non-navigable while allowing motion inside a designated UWB-instrumented building. The framework is implemented in ROS 2 and runs in real time on a wearable platform, with visualization in Foxglove. We evaluate three scenarios: indoor (UWB+PDR), outdoor (GNSS+PDR), and seamless outdoor-indoor (GNSS+UWB+PDR). Results show that the ESKF provides the most consistent overall performance in our implementation.
comment: 8 pages, 4 figures, submitted to The 17th International Conference on Ambient Systems, Networks and Technologies
☆ Symphony: A Heuristic Normalized Calibrated Advantage Actor and Critic Algorithm in application for Humanoid Robots
In our work we not explicitly hint that it is a misconception to think that humans learn fast. Learning process takes time. Babies start learning to move in the restricted liquid area called placenta. Children often are 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 dozen 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 no secret that continuous increase in Gaussian noise without appropriate smoothing is harmful for motors and gearboxes. Compared to Stochastic algorithms, we set a limited parametric noise and promote a reduced strength of actions, safely increasing entropy, since the actions are kind of immersed 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 Actor and Critic in one pass, as well as combine Actor and Critic in one Object and implement their Losses in one line.
☆ Design and Implementation of a High-Precision Wind-Estimation UAV with Onboard Sensors
Accurate real-time wind vector estimation is essential for enhancing the safety, navigation accuracy, and energy efficiency of unmanned aerial vehicles (UAVs). Traditional approaches rely on external sensors or simplify vehicle dynamics, which limits their applicability during agile flight or in resource-constrained platforms. This paper proposes a real-time wind estimation method based solely on onboard sensors. The approach first estimates external aerodynamic forces using a disturbance observer (DOB), and then maps these forces to wind vectors using a thin-plate spline (TPS) model. A custom-designed wind barrel mounted on the UAV enhances aerodynamic sensitivity, further improving estimation accuracy. The system is validated through comprehensive experiments in wind tunnels, indoor and outdoor flights. Experimental results demonstrate that the proposed method achieves consistently high-accuracy wind estimation across controlled and real-world conditions, with speed RMSEs as low as \SI{0.06}{m/s} in wind tunnel tests, \SI{0.22}{m/s} during outdoor hover, and below \SI{0.38}{m/s} in indoor and outdoor dynamic flights, and direction RMSEs under \ang{7.3} across all scenarios, outperforming existing baselines. Moreover, the method provides vertical wind estimates -- unavailable in baselines -- with RMSEs below \SI{0.17}{m/s} even during fast indoor translations.
comment: https://www.sciencedirect.com/science/article/abs/pii/S0263224125032415?via%3Dihub
☆ RoboNeuron: A Modular Framework Linking Foundation Models and ROS for Embodied AI
Current embodied AI systems face severe engineering impediments, primarily characterized by poor cross-scenario adaptability, rigid inter-module coupling, and fragmented inference acceleration. To overcome these limitations, we propose RoboNeuron, a universal deployment framework for embodied intelligence. RoboNeuron is the first framework to deeply integrate the cognitive capabilities of Large Language Models (LLMs) and Vision-Language-Action (VLA) models with the real-time execution backbone of the Robot Operating System (ROS). We utilize the Model Context Protocol (MCP) as a semantic bridge, enabling the LLM to dynamically orchestrate underlying robotic tools. The framework establishes a highly modular architecture that strictly decouples sensing, reasoning, and control by leveraging ROS's unified communication interfaces. Crucially, we introduce an automated tool to translate ROS messages into callable MCP functions, significantly streamlining development. RoboNeuron significantly enhances cross-scenario adaptability and component flexibility, while establishing a systematic platform for horizontal performance benchmarking, laying a robust foundation for scalable real-world embodied applications.
☆ CLASH: Collaborative Large-Small Hierarchical Framework for Continuous Vision-and-Language Navigation
Vision-and-Language Navigation (VLN) requires robots to follow natural language instructions and navigate complex environments without prior maps. While recent vision-language large models demonstrate strong reasoning abilities, they often underperform task-specific panoramic small models in VLN tasks. To address this, we propose CLASH (Collaborative Large-Small Hierarchy), a VLN-CE framework that integrates a reactive small-model planner (RSMP) with a reflective large-model reasoner (RLMR). RSMP adopts a causal-learning-based dual-branch architecture to enhance generalization, while RLMR leverages panoramic visual prompting with chain-of-thought reasoning to support interpretable spatial understanding and navigation. We further introduce an uncertainty-aware collaboration mechanism (UCM) that adaptively fuses decisions from both models. For obstacle avoidance, in simulation, we replace the rule-based controller with a fully learnable point-goal policy, and in real-world deployment, we design a LiDAR-based clustering module for generating navigable waypoints and pair it with an online SLAM-based local controller. CLASH achieves state-of-the-art (SoTA) results (ranking 1-st) on the VLN-CE leaderboard, significantly improving SR and SPL on the test-unseen set over the previous SoTA methods. Real-world experiments demonstrate CLASH's strong robustness, validating its effectiveness in both simulation and deployment scenarios.
☆ Design and Validation of an Under-actuated Robotic Finger with Synchronous Tendon Routing
Tendon-driven under-actuated robotic fingers provide advantages for dexterous manipulation through reduced actuator requirements and simplified mechanical design. However, achieving both high load capacity and adaptive compliance in a compact form remains challenging. This paper presents an under-actuated tendon-driven robotic finger (UTRF) featuring a synchronous tendon routing that mechanically couples all joints with fixed angular velocity ratios, enabling the entire finger to be actuated by a single actuator. This approach significantly reduces the number of actuators required in multi-finger hands, resulting in a lighter and more compact structure without sacrificing stiffness or compliance. The kinematic and static models of the finger are derived, incorporating tendon elasticity to predict structural stiffness. A single-finger prototype was fabricated and tested under static loading, showing an average deflection prediction error of 1.0 mm (0.322% of total finger length) and a measured stiffness of 1.2x10^3 N/m under a 3 kg tip load. Integration into a five-finger robotic hand (UTRF-RoboHand) demonstrates effective object manipulation across diverse scenarios, confirming that the proposed routing achieves predictable stiffness and reliable grasping performance with a minimal actuator count.
comment: 7 pages and 11 figures
☆ Design of a six wheel suspension and a three-axis linear actuation mechanism for a laser weeding robot
Mobile robots are increasingly utilized in agriculture to automate labor-intensive tasks such as weeding, sowing, harvesting and soil analysis. Recently, agricultural robots have been developed to detect and remove weeds using mechanical tools or precise herbicide sprays. Mechanical weeding is inefficient over large fields, and herbicides harm the soil ecosystem. Laser weeding with mobile robots has emerged as a sustainable alternative in precision farming. In this paper, we present an autonomous weeding robot that uses controlled exposure to a low energy laser beam for weed removal. The proposed robot is six-wheeled with a novel double four-bar suspension for higher stability. The laser is guided towards the detected weeds by a three-dimensional linear actuation mechanism. Field tests have demonstrated the robot's capability to navigate agricultural terrains effectively by overcoming obstacles up to 15 cm in height. At an optimal speed of 42.5 cm/s, the robot achieves a weed detection rate of 86.2\% and operating time of 87 seconds per meter. The laser actuation mechanism maintains a minimal mean positional error of 1.54 mm, combined with a high hit rate of 97\%, ensuring effective and accurate weed removal. This combination of speed, accuracy, and efficiency highlights the robot's potential for significantly enhancing precision farming practices.
comment: 15 Pages, 10 figures
☆ Lies We Can Trust: Quantifying Action Uncertainty with Inaccurate Stochastic Dynamics through Conformalized Nonholonomic Lie Groups
We propose Conformal Lie-group Action Prediction Sets (CLAPS), a symmetry-aware conformal prediction-based algorithm that constructs, for a given action, a set guaranteed to contain the resulting system configuration at a user-defined probability. Our assurance holds under both aleatoric and epistemic uncertainty, non-asymptotically, and does not require strong assumptions about the true system dynamics, the uncertainty sources, or the quality of the approximate dynamics model. Typically, uncertainty quantification is tackled by making strong assumptions about the error distribution or magnitude, or by relying on uncalibrated uncertainty estimates - i.e., with no link to frequentist probabilities - which are insufficient for safe control. Recently, conformal prediction has emerged as a statistical framework capable of providing distribution-free probabilistic guarantees on test-time prediction accuracy. While current conformal methods treat robots as Euclidean points, many systems have non-Euclidean configurations, e.g., some mobile robots have SE(2). In this work, we rigorously analyze configuration errors using Lie groups, extending previous Euclidean Space theoretical guarantees to SE(2). Our experiments on a simulated JetBot, and on a real MBot, suggest that by considering the configuration space's structure, our symmetry-informed nonconformity score leads to more volume-efficient prediction regions which represent the underlying uncertainty better than existing approaches.
comment: 13 pages, 7 figures. Under review
☆ Task-Oriented Grasping Using Reinforcement Learning with a Contextual Reward Machine
This paper presents a reinforcement learning framework that incorporates a Contextual Reward Machine for task-oriented grasping. The Contextual Reward Machine reduces task complexity by decomposing grasping tasks into manageable sub-tasks. Each sub-task is associated with a stage-specific context, including a reward function, an action space, and a state abstraction function. This contextual information enables efficient intra-stage guidance and improves learning efficiency by reducing the state-action space and guiding exploration within clearly defined boundaries. In addition, transition rewards are introduced to encourage or penalize transitions between stages which guides the model toward desirable stage sequences and further accelerates convergence. When integrated with the Proximal Policy Optimization algorithm, the proposed method achieved a 95% success rate across 1,000 simulated grasping tasks encompassing diverse objects, affordances, and grasp topologies. It outperformed the state-of-the-art methods in both learning speed and success rate. The approach was transferred to a real robot, where it achieved a success rate of 83.3% in 60 grasping tasks over six affordances. These experimental results demonstrate superior accuracy, data efficiency, and learning efficiency. They underscore the model's potential to advance task-oriented grasping in both simulated and real-world settings.
☆ Latent Chain-of-Thought World Modeling for End-to-End Driving
Recent Vision-Language-Action (VLA) models for autonomous driving explore inference-time reasoning as a way to improve driving performance and safety in challenging scenarios. Most prior work uses natural language to express chain-of-thought (CoT) reasoning before producing driving actions. However, text may not be the most efficient representation for reasoning. In this work, we present Latent-CoT-Drive (LCDrive): a model that expresses CoT in a latent language that captures possible outcomes of the driving actions being considered. Our approach unifies CoT reasoning and decision making by representing both in an action-aligned latent space. Instead of natural language, the model reasons by interleaving (1) action-proposal tokens, which use the same vocabulary as the model's output actions; and (2) world model tokens, which are grounded in a learned latent world model and express future outcomes of these actions. We cold start latent CoT by supervising the model's action proposals and world model tokens based on ground-truth future rollouts of the scene. We then post-train with closed-loop reinforcement learning to strengthen reasoning capabilities. On a large-scale end-to-end driving benchmark, LCDrive achieves faster inference, better trajectory quality, and larger improvements from interactive reinforcement learning compared to both non-reasoning and text-reasoning baselines.
comment: Technical Report
☆ Active Optics for Hyperspectral Imaging of Reflective Agricultural Leaf Sensors
Monitoring plant health increasingly relies on leaf-mounted sensors that provide real-time physiological data, yet efficiently locating and sampling these sensors in complex agricultural environments remains a major challenge. We present an integrated, adaptive, and scalable system that autonomously detects and interrogates plant sensors using a coordinated suite of low-cost optical components including a LiDAR, liquid lens, monochrome camera, filter wheel, and Fast Steering Mirror (FSM). The system first uses LiDAR to identify the distinct reflective signatures of sensors within the field, then dynamically redirects the camera s field of view via the FSM to target each sensor for hyperspectral imaging. The liquid lens continuously adjusts focus to maintain image sharpness across varying depths, enabling precise spectral measurements. We validated the system in controlled indoor experiments, demonstrating accurate detection and tracking of reflective plant sensors and successful acquisition of their spectral data. To our knowledge, no other system currently integrates these sensing and optical modalities for agricultural monitoring. This work establishes a foundation for adaptive, low-cost, and automated plant sensor interrogation, representing a significant step toward scalable, real-time plant health monitoring in precision agriculture.
☆ 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/
☆ Fast-FoundationStereo: Real-Time Zero-Shot Stereo Matching
Stereo foundation models achieve strong zero-shot generalization but remain computationally prohibitive for real-time applications. Efficient stereo architectures, on the other hand, sacrifice robustness for speed and require costly per-domain fine-tuning. To bridge this gap, we present Fast-FoundationStereo, a family of architectures that achieve, for the first time, strong zero-shot generalization at real-time frame rate. We employ a divide-and-conquer acceleration strategy with three components: (1) knowledge distillation to compress the hybrid backbone into a single efficient student; (2) blockwise neural architecture search for automatically discovering optimal cost filtering designs under latency budgets, reducing search complexity exponentially; and (3) structured pruning for eliminating redundancy in the iterative refinement module. Furthermore, we introduce an automatic pseudo-labeling pipeline used to curate 1.4M in-the-wild stereo pairs to supplement synthetic training data and facilitate knowledge distillation. The resulting model can run over 10x faster than FoundationStereo while closely matching its zero-shot accuracy, thus establishing a new state-of-the-art among real-time methods. Project page: https://nvlabs.github.io/Fast-FoundationStereo/
☆ Design and Experimental Validation of Closed-Form CBF-Based Safe Control for Stewart Platform Under Multiple Constraints
This letter presents a closed-form solution of Control Barrier Function (CBF) framework for enforcing safety constraints on a Stewart robotic platform. The proposed method simultaneously handles multiple position and velocity constraints through an explicit closed-form control law, eliminating the need to solve a Quadratic Program (QP) at every control step and enabling efficient real-time implementation. This letter derives necessary and sufficient conditions under which the closed-form expression remains non-singular, thereby ensuring well-posedness of the CBF solution to multi-constraint problem. The controller is validated in both simulation and hardware experiments on a custom-built Stewart platform prototype, demonstrating safetyguaranteed performance that is comparable to the QP-based formulation, while reducing computation time by more than an order of magnitude. The results confirm that the proposed approach provides a reliable and computationally lightweight framework for real-time safe control of parallel robotic systems. The experimental videos are available on the project website. (https://nail-uh.github.io/StewartPlatformSafeControl.github.io/)
comment: 9 pages
☆ Vision-Language Models for Infrared Industrial Sensing in Additive Manufacturing Scene Description
Many manufacturing environments operate in low-light conditions or within enclosed machines where conventional vision systems struggle. Infrared cameras provide complementary advantages in such environments. Simultaneously, supervised AI systems require large labeled datasets, which makes zero-shot learning frameworks more practical for applications including infrared cameras. Recent advances in vision-language foundation models (VLMs) offer a new path in zero-shot predictions from paired image-text representations. However, current VLMs cannot understand infrared camera data since they are trained on RGB data. This work introduces VLM-IRIS (Vision-Language Models for InfraRed Industrial Sensing), a zero-shot framework that adapts VLMs to infrared data by preprocessing infrared images captured by a FLIR Boson sensor into RGB-compatible inputs suitable for CLIP-based encoders. We demonstrate zero-shot workpiece presence detection on a 3D printer bed where temperature differences between the build plate and workpieces make the task well-suited for thermal imaging. VLM-IRIS converts the infrared images to magma representation and applies centroid prompt ensembling with a CLIP ViT-B/32 encoder to achieve high accuracy on infrared images without any model retraining. These findings demonstrate that the proposed improvements to VLMs can be effectively extended to thermal applications for label-free monitoring.
☆ Taxonomy and Modular Tool System for Versatile and Effective Non-Prehensile Manipulations
General-purpose robotic end-effectors of limited complexity, like the parallel-jaw gripper, are appealing for their balance of simplicity and effectiveness in a wide range of manipulation tasks. However, while many such manipulators offer versatility in grasp-like interactions, they are not optimized for non-prehensile actions like pressing, rubbing, or scraping -- manipulations needed for many common tasks. To perform such tasks, humans use a range of different body parts or tools with different rigidity, friction, etc., according to the properties most effective for a given task. Here, we discuss a taxonomy for the key properties of a non-actuated end-effector, laying the groundwork for a systematic understanding of the affordances of non-prehensile manipulators. We then present a modular tool system, based on the taxonomy, that can be used by a standard two-fingered gripper to extend its versatility and effectiveness in performing such actions. We demonstrate the application of the tool system in aerospace and household scenarios that require a range of non-prehensile and prehensile manipulations.
comment: 34 pages, 10 figures, 2 tables, supplementary videos: https://youtu.be/Hcefy53PY0M, https://youtu.be/nFF9k91hsfU, https://youtu.be/EulPLskNIZQ
☆ WholeBodyVLA: Towards Unified Latent VLA for Whole-Body Loco-Manipulation Control
Humanoid robots require precise locomotion and dexterous manipulation to perform challenging loco-manipulation tasks. Yet existing approaches, modular or end-to-end, are deficient in manipulation-aware locomotion. This confines the robot to a limited workspace, preventing it from performing large-space loco-manipulation. We attribute this to: (1) the challenge of acquiring loco-manipulation knowledge due to the scarcity of humanoid teleoperation data, and (2) the difficulty of faithfully and reliably executing locomotion commands, stemming from the limited precision and stability of existing RL controllers. To acquire richer loco-manipulation knowledge, we propose a unified latent learning framework that enables Vision-Language-Action (VLA) system to learn from low-cost action-free egocentric videos. Moreover, an efficient human data collection pipeline is devised to augment the dataset and scale the benefits. To more precisely execute the desired locomotion commands, we present a loco-manipulation-oriented (LMO) RL policy specifically tailored for accurate and stable core loco-manipulation movements, such as advancing, turning, and squatting. Building on these components, we introduce WholeBodyVLA, a unified framework for humanoid loco-manipulation. To the best of our knowledge, WholeBodyVLA is one of its kind enabling large-space humanoid loco-manipulation. It is verified via comprehensive experiments on the AgiBot X2 humanoid, outperforming prior baseline by 21.3%. It also demonstrates strong generalization and high extensibility across a broad range of tasks.
☆ Towards Accessible Physical AI: LoRA-Based Fine-Tuning of VLA Models for Real-World Robot Control
Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in robotic manipulation,enabling robots to execute natural language commands through end-to-end learning from visual observations.However, deploying large-scale VLA models on affordable robotic platforms remains challenging due to computational constraints and the need for efficient adaptation to new robot embodiments. This paper presents an efficient fine-tuning methodology and real-world deployment analysis for adapting VLA models to low-cost robotic manipulation systems.We propose a resource-efficient fine-tuning strategy using Low-Rank Adaptation (LoRA) and quantization techniques that enable multi-billion parameter VLA models ( 3.1B parameters) to run on consumer-grade GPUs with 8GB VRAM. Our methodology addresses the critical challenge of adapting pre-trained VLA models to new robot embodiments with limited demonstration data, focusing on the trade-offs between frozen and unfrozen vision encoders. Through real-world deployment on the SO101 robotic arm for a button-pressing manipulation task, we demonstrate that our approach achieves effective manipulation performance while maintaining computational efficiency. We provide detailed analysis of deployment challenges, failure modes, and the relationship between training data quantity and real-world performance,trained on 200 demonstration episodes. Our results show that with proper fine-tuning methodology, VLA models can be successfully deployed on affordable robotic platforms,making advanced manipulation capabilities accessible beyond expensive research robots.
From Generated Human Videos to Physically Plausible Robot Trajectories
Video generation models are rapidly improving in their ability to synthesize human actions in novel contexts, holding the potential to serve as high-level planners for contextual robot control. To realize this potential, a key research question remains open: how can a humanoid execute the human actions from generated videos in a zero-shot manner? This challenge arises because generated videos are often noisy and exhibit morphological distortions that make direct imitation difficult compared to real video. To address this, we introduce a two-stage pipeline. First, we lift video pixels into a 4D human representation and then retarget to the humanoid morphology. Second, we propose GenMimic-a physics-aware reinforcement learning policy conditioned on 3D keypoints, and trained with symmetry regularization and keypoint-weighted tracking rewards. As a result, GenMimic can mimic human actions from noisy, generated videos. We curate GenMimicBench, a synthetic human-motion dataset generated using two video generation models across a spectrum of actions and contexts, establishing a benchmark for assessing zero-shot generalization and policy robustness. Extensive experiments demonstrate improvements over strong baselines in simulation and confirm coherent, physically stable motion tracking on a Unitree G1 humanoid robot without fine-tuning. This work offers a promising path to realizing the potential of video generation models as high-level policies for robot control.
comment: For project website, see https://genmimic.github.io
♻ ☆ MaskedManipulator: Versatile Whole-Body Manipulation SIGGRAPH
We tackle the challenges of synthesizing versatile, physically simulated human motions for full-body object manipulation. Unlike prior methods that are focused on detailed motion tracking, trajectory following, or teleoperation, our framework enables users to specify versatile high-level objectives such as target object poses or body poses. To achieve this, we introduce MaskedManipulator, a generative control policy distilled from a tracking controller trained on large-scale human motion capture data. This two-stage learning process allows the system to perform complex interaction behaviors, while providing intuitive user control over both character and object motions. MaskedManipulator produces goal-directed manipulation behaviors that expand the scope of interactive animation systems beyond task-specific solutions.
comment: SIGGRAPH Asia 2025 (Project page: https://research.nvidia.com/labs/par/maskedmanipulator/ )
♻ ☆ Reframing Human-Robot Interaction Through Extended Reality: Unlocking Safer, Smarter, and More Empathic Interactions with Virtual Robots and Foundation Models
This perspective reframes human-robot interaction (HRI) through extended reality (XR), arguing that virtual robots powered by large foundation models (FMs) can serve as cognitively grounded, empathic agents. Unlike physical robots, XR-native agents are unbound by hardware constraints and can be instantiated, adapted, and scaled on demand, while still affording embodiment and co-presence. We synthesize work across XR, HRI, and cognitive AI to show how such agents can support safety-critical scenarios, socially and cognitively empathic interaction across domains, and outreaching physical capabilities with XR and AI integration. We then discuss how multimodal large FMs (e.g., large language model, large vision model, and vision-language model) enable context-aware reasoning, affect-sensitive situations, and long-term adaptation, positioning virtual robots as cognitive and empathic mediators rather than mere simulation assets. At the same time, we highlight challenges and potential risks, including overtrust, cultural and representational bias, privacy concerns around biometric sensing, and data governance and transparency. The paper concludes by outlining a research agenda for human-centered, ethically grounded XR agents - emphasizing multi-layered evaluation frameworks, multi-user ecosystems, mixed virtual-physical embodiment, and societal and ethical design practices to envision XR-based virtual agents powered by FMs as reshaping future HRI into a more efficient and adaptive paradigm.
comment: This paper is under review
♻ ☆ WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving
We introduce WAM-Flow, a vision-language-action (VLA) model that casts ego-trajectory planning as discrete flow matching over a structured token space. In contrast to autoregressive decoders, WAM-Flow performs fully parallel, bidirectional denoising, enabling coarse-to-fine refinement with a tunable compute-accuracy trade-off. Specifically, the approach combines a metric-aligned numerical tokenizer that preserves scalar geometry via triplet-margin learning, a geometry-aware flow objective and a simulator-guided GRPO alignment that integrates safety, ego progress, and comfort rewards while retaining parallel generation. A multi-stage adaptation converts a pre-trained auto-regressive backbone (Janus-1.5B) from causal decoding to non-causal flow model and strengthens road-scene competence through continued multimodal pretraining. Thanks to the inherent nature of consistency model training and parallel decoding inference, WAM-Flow achieves superior closed-loop performance against autoregressive and diffusion-based VLA baselines, with 1-step inference attaining 89.1 PDMS and 5-step inference reaching 90.3 PDMS on NAVSIM v1 benchmark. These results establish discrete flow matching as a new promising paradigm for end-to-end autonomous driving. The code will be publicly available soon.
comment: 18 pages, 11 figures. Code & Model: https://github.com/fudan-generative-vision/WAM-Flow
♻ ☆ Panoramic Out-of-Distribution Segmentation
Panoramic imaging enables capturing 360° images with an ultra-wide Field-of-View (FoV) for dense omnidirectional perception, which is critical to applications, such as autonomous driving and augmented reality, etc. However, current panoramic semantic segmentation methods fail to identify outliers, and pinhole Out-of-distribution Segmentation (OoS) models perform unsatisfactorily in the panoramic domain due to pixel distortions and background clutter. To address these issues, we introduce a new task, Panoramic Out-of-distribution Segmentation (PanOoS), with the aim of achieving comprehensive and safe scene understanding. Furthermore, we propose the first solution, POS, which adapts to the characteristics of panoramic images through text-guided prompt distribution learning. Specifically, POS integrates a disentanglement strategy designed to materialize the cross-domain generalization capability of CLIP. The proposed Prompt-based Restoration Attention (PRA) optimizes semantic decoding by prompt guidance and self-adaptive correction, while Bilevel Prompt Distribution Learning (BPDL) refines the manifold of per-pixel mask embeddings via semantic prototype supervision. Besides, to compensate for the scarcity of PanOoS datasets, we establish two benchmarks: DenseOoS, which features diverse outliers in complex environments, and QuadOoS, captured by a quadruped robot with a panoramic annular lens system. Extensive experiments demonstrate superior performance of POS, with AuPRC improving by 34.25% and FPR95 decreasing by 21.42% on DenseOoS, outperforming state-of-the-art pinhole-OoS methods. Moreover, POS achieves leading closed-set segmentation capabilities and advances the development of panoramic understanding. Code and datasets will be available at https://github.com/MengfeiD/PanOoS.
comment: Code and datasets will be available at https://github.com/MengfeiD/PanOoS
♻ ☆ Towards Open-World Human Action Segmentation Using Graph Convolutional Networks IROS25
Human-object interaction segmentation is a fundamental task of daily activity understanding, which plays a crucial role in applications such as assistive robotics, healthcare, and autonomous systems. Most existing learning-based methods excel in closed-world action segmentation, they struggle to generalize to open-world scenarios where novel actions emerge. Collecting exhaustive action categories for training is impractical due to the dynamic diversity of human activities, necessitating models that detect and segment out-of-distribution actions without manual annotation. To address this issue, we formally define the open-world action segmentation problem and propose a structured framework for detecting and segmenting unseen actions. Our framework introduces three key innovations: 1) an Enhanced Pyramid Graph Convolutional Network (EPGCN) with a novel decoder module for robust spatiotemporal feature upsampling. 2) Mixup-based training to synthesize out-of-distribution data, eliminating reliance on manual annotations. 3) A novel Temporal Clustering loss that groups in-distribution actions while distancing out-of-distribution samples. We evaluate our framework on two challenging human-object interaction recognition datasets: Bimanual Actions and 2 Hands and Object (H2O) datasets. Experimental results demonstrate significant improvements over state-of-the-art action segmentation models across multiple open-set evaluation metrics, achieving 16.9% and 34.6% relative gains in open-set segmentation (F1@50) and out-of-distribution detection performances (AUROC), respectively. Additionally, we conduct an in-depth ablation study to assess the impact of each proposed component, identifying the optimal framework configuration for open-world action segmentation.
comment: 8 pages, 3 figures, accepted in IROS25, Hangzhou, China
♻ ☆ Multi-Modal Graph Convolutional Network with Sinusoidal Encoding for Robust Human Action Segmentation IROS25
Accurate temporal segmentation of human actions is critical for intelligent robots in collaborative settings, where a precise understanding of sub-activity labels and their temporal structure is essential. However, the inherent noise in both human pose estimation and object detection often leads to over-segmentation errors, disrupting the coherence of action sequences. To address this, we propose a Multi-Modal Graph Convolutional Network (MMGCN) that integrates low-frame-rate (e.g., 1 fps) visual data with high-frame-rate (e.g., 30 fps) motion data (skeleton and object detections) to mitigate fragmentation. Our framework introduces three key contributions. First, a sinusoidal encoding strategy that maps 3D skeleton coordinates into a continuous sin-cos space to enhance spatial representation robustness. Second, a temporal graph fusion module that aligns multi-modal inputs with differing resolutions via hierarchical feature aggregation, Third, inspired by the smooth transitions inherent to human actions, we design SmoothLabelMix, a data augmentation technique that mixes input sequences and labels to generate synthetic training examples with gradual action transitions, enhancing temporal consistency in predictions and reducing over-segmentation artifacts. Extensive experiments on the Bimanual Actions Dataset, a public benchmark for human-object interaction understanding, demonstrate that our approach outperforms state-of-the-art methods, especially in action segmentation accuracy, achieving F1@10: 94.5% and F1@25: 92.8%.
comment: 8 pages, 5 figures, accepted in IROS25, Hangzhou, China
♻ ☆ Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning AAAI 2026
Language-conditioned manipulation facilitates human-robot interaction via behavioral cloning (BC), which learns control policies from human demonstrations and serves as a cornerstone of embodied AI. Overcoming compounding errors in sequential action decisions remains a central challenge to improving BC performance. Existing approaches mitigate compounding errors through data augmentation, expressive representation, or temporal abstraction. However, they suffer from physical discontinuities and semantic-physical misalignment, leading to inaccurate action cloning and intermittent execution. In this paper, we present Continuous vision-language-action Co-Learning with Semantic-Physical Alignment (CCoL), a novel BC framework that ensures temporally consistent execution and fine-grained semantic grounding. It generates robust and smooth action execution trajectories through continuous co-learning across vision, language, and proprioceptive inputs (e.g., robot internal states). Meanwhile, we anchor language semantics to visuomotor representations by a bidirectional cross-attention to learn contextual information for action generation, successfully overcoming the problem of semantic-physical misalignment. Extensive experiments show that CCoL achieves an average 8.0% relative improvement across three simulation suites, with up to 19.2% relative gain in human-demonstrated bimanual insertion tasks. Real-world tests on a 7-DoF robot further confirm CCoL's generalization under unseen and noisy object states.
comment: Accepted at AAAI 2026, the Project website is available at https://qhemu.github.io/CCoL/
♻ ☆ Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Corrections
We address key challenges in Dataset Aggregation (DAgger) for real-world contact-rich manipulation: how to collect informative human correction data and how to effectively update policies with this new data. We introduce Compliant Residual DAgger (CR-DAgger), which contains two novel components: 1) a Compliant Intervention Interface that leverages compliance control, allowing humans to provide gentle, accurate delta action corrections without interrupting the ongoing robot policy execution; and 2) a Compliant Residual Policy formulation that learns from human corrections while incorporating force feedback and force control. Our system significantly enhances performance on precise contact-rich manipulation tasks using minimal correction data, improving base policy success rates by over 50\% on two challenging tasks (book flipping and belt assembly) while outperforming both retraining-from-scratch and finetuning approaches. Through extensive real-world experiments, we provide practical guidance for implementing effective DAgger in real-world robot learning tasks. Result videos are available at: https://compliant-residual-dagger.github.io/
♻ ☆ Enhancing Hand Palm Motion Gesture Recognition by Eliminating Reference Frame Bias via Frame-Invariant Similarity Measures
The ability of robots to recognize human gestures facilitates a natural and accessible human-robot collaboration. However, most work in gesture recognition remains rooted in reference frame-dependent representations. This poses a challenge when reference frames vary due to different work cell layouts, imprecise frame calibrations, or other environmental changes. This paper investigated the use of invariant trajectory descriptors for robust hand palm motion gesture recognition under reference frame changes. First, a novel dataset of recorded Hand Palm Motion (HPM) gestures is introduced. The motion gestures in this dataset were specifically designed to be distinguishable without dependence on specific reference frames or directional cues. Afterwards, multiple invariant trajectory descriptor approaches were benchmarked to assess how their performances generalize to this novel HPM dataset. After this offline benchmarking, the best scoring approach is validated for online recognition by developing a real-time Proof of Concept (PoC). In this PoC, hand palm motion gestures were used to control the real-time movement of a manipulator arm. The PoC demonstrated a high recognition reliability in real-time operation, achieving an $F_1$-score of 92.3%. This work demonstrates the effectiveness of the invariant descriptor approach as a standalone solution. Moreover, we believe that the invariant descriptor approach can also be utilized within other state-of-the-art pattern recognition and learning systems to improve their robustness against reference frame variations.
comment: This is the preprint version of a paper accepted for publication at the 2025 IEEE International Conference on Automation Science and Engineering (CASE). The final published version is available at DOI: 10.1109/CASE58245.2025.11163910
♻ ☆ Never too Cocky to Cooperate: An FIM and RL-based USV-AUV Collaborative System for Underwater Tasks in Extreme Sea Conditions
This paper develops a novel unmanned surface vehicle (USV)-autonomous underwater vehicle (AUV) collaborative system designed to enhance underwater task performance in extreme sea conditions. The system integrates a dual strategy: (1) high-precision multi-AUV localization enabled by Fisher information matrix-optimized USV path planning, and (2) reinforcement learning-based cooperative planning and control method for multi-AUV task execution. Extensive experimental evaluations in the underwater data collection task demonstrate the system's operational feasibility, with quantitative results showing significant performance improvements over baseline methods. The proposed system exhibits robust coordination capabilities between USV and AUVs while maintaining stability in extreme sea conditions. To facilitate reproducibility and community advancement, we provide an open-source simulation toolkit available at: https://github.com/360ZMEM/USV-AUV-colab .
comment: This paper has been submitted to IEEE Transactions on Mobile Computing, and is currently under minor revision
♻ ☆ ShapeForce: Low-Cost Soft Robotic Wrist for Contact-Rich Manipulation
Contact feedback is essential for contact-rich robotic manipulation, as it allows the robot to detect subtle interaction changes and adjust its actions accordingly. Six-axis force-torque sensors are commonly used to obtain contact feedback, but their high cost and fragility have discouraged many researchers from adopting them in contact-rich tasks. To offer a more cost-efficient and easy-accessible source of contact feedback, we present ShapeForce, a low-cost, plug-and-play soft wrist that provides force-like signals for contact-rich robotic manipulation. Inspired by how humans rely on relative force changes in contact rather than precise force magnitudes, ShapeForce converts external force and torque into measurable deformations of its compliant core, which are then estimated via marker-based pose tracking and converted into force-like signals. Our design eliminates the need for calibration or specialized electronics to obtain exact values, and instead focuses on capturing force and torque changes sufficient for enabling contact-rich manipulation. Extensive experiments across diverse contact-rich tasks and manipulation policies demonstrate that ShapeForce delivers performance comparable to six-axis force-torque sensors at an extremely low cost.
♻ ☆ Transformer Driven Visual Servoing and Dual Arm Impedance Control for Fabric Texture Matching
In this paper, we propose a method to align and place a fabric piece on top of another using a dual-arm manipulator and a grayscale camera, so that their surface textures are accurately matched. We propose a novel control scheme that combines Transformer-driven visual servoing with dualarm impedance control. This approach enables the system to simultaneously control the pose of the fabric piece and place it onto the underlying one while applying tension to keep the fabric piece flat. Our transformer-based network incorporates pretrained backbones and a newly introduced Difference Extraction Attention Module (DEAM), which significantly enhances pose difference prediction accuracy. Trained entirely on synthetic images generated using rendering software, the network enables zero-shot deployment in real-world scenarios without requiring prior training on specific fabric textures. Real-world experiments demonstrate that the proposed system accurately aligns fabric pieces with different textures.
comment: 8 pages, 11 figures. Accepted to IEEE Robotics and Automation Letters (RA-L)
♻ ☆ SEA: Semantic Map Prediction for Active Exploration of Uncertain Areas
In this paper, we propose SEA, a novel approach for active robot exploration through semantic map prediction and a reinforcement learning-based hierarchical exploration policy. Unlike existing learning-based methods that rely on one-step waypoint prediction, our approach enhances the agent's long-term environmental understanding to facilitate more efficient exploration. We propose an iterative prediction-exploration framework that explicitly predicts the missing areas of the map based on current observations. The difference between the actual accumulated map and the predicted global map is then used to guide exploration. Additionally, we design a novel reward mechanism that leverages reinforcement learning to update the long-term exploration strategies, enabling us to construct an accurate semantic map within limited steps. Experimental results demonstrate that our method significantly outperforms state-of-the-art exploration strategies, achieving superior coverage ares of the global map within the same time constraints.
comment: Project page: https://robo-lavira.github.io/sea-active-exp
♻ ☆ Semantic Trajectory Generation for Goal-Oriented Spacecraft Rendezvous SC
Reliable real-time trajectory generation is essential for future autonomous spacecraft. While recent progress in nonconvex guidance and control is paving the way for onboard autonomous trajectory optimization, these methods still rely on extensive expert input (e.g., waypoints, constraints, mission timelines, etc.), which limits the operational scalability in real rendezvous missions. This paper introduces SAGES (Semantic Autonomous Guidance Engine for Space), a trajectory-generation framework that translates natural-language commands into spacecraft trajectories that reflect high-level intent while respecting nonconvex constraints. Experiments in two settings -- fault-tolerant proximity operations with continuous-time constraint enforcement and a free-flying robotic platform -- demonstrate that SAGES reliably produces trajectories aligned with human commands, achieving over 90% semantic-behavioral consistency across diverse behavior modes. Ultimately, this work marks an initial step toward language-conditioned, constraint-aware spacecraft trajectory generation, enabling operators to interactively guide both safety and behavior through intuitive natural-language commands with reduced expert burden.
comment: 28 pages, 12 figures. Submitted to AIAA SCITECH 2026
♻ ☆ Development and Testing for Perception Based Autonomous Landing of a Long-Range QuadPlane
QuadPlanes combine the range efficiency of fixed-wing aircraft with the maneuverability of multi-rotor platforms for long-range autonomous missions. In GPS-denied or cluttered urban environments, perception-based landing is vital for reliable operation. Unlike structured landing zones, real-world sites are unstructured and highly variable, requiring strong generalization capabilities from the perception system. Deep neural networks (DNNs) provide a scalable solution for learning landing site features across diverse visual and environmental conditions. While perception-driven landing has been shown in simulation, real-world deployment introduces significant challenges. Payload and volume constraints limit high-performance edge AI devices like the NVIDIA Jetson Orin Nano, which are crucial for real-time detection and control. Accurate pose estimation during descent is necessary, especially in the absence of GPS, and relies on dependable visual-inertial odometry. Achieving this with limited edge AI resources requires careful optimization of the entire deployment framework. The flight characteristics of large QuadPlanes further complicate the problem. These aircraft exhibit high inertia, reduced thrust vectoring, and slow response times further complicate stable landing maneuvers. This work presents a lightweight QuadPlane system for efficient vision-based autonomous landing and visual-inertial odometry, specifically developed for long-range QuadPlane operations such as aerial monitoring. It describes the hardware platform, sensor configuration, and embedded computing architecture designed to meet demanding real-time, physical constraints. This establishes a foundation for deploying autonomous landing in dynamic, unstructured, GPS-denied environments.
♻ ☆ Mastering Diverse, Unknown, and Cluttered Tracks for Robust Vision-Based Drone Racing
Most reinforcement learning(RL)-based methods for drone racing target fixed, obstacle-free tracks, leaving the generalization to unknown, cluttered environments largely unaddressed. This challenge stems from the need to balance racing speed and collision avoidance, limited feasible space causing policy exploration trapped in local optima during training, and perceptual ambiguity between gates and obstacles in depth maps-especially when gate positions are only coarsely specified. To overcome these issues, we propose a two-phase learning framework: an initial soft-collision training phase that preserves policy exploration for high-speed flight, followed by a hard-collision refinement phase that enforces robust obstacle avoidance. An adaptive, noise-augmented curriculum with an asymmetric actor-critic architecture gradually shifts the policy's reliance from privileged gate-state information to depth-based visual input. We further impose Lipschitz constraints and integrate a track-primitive generator to enhance motion stability and cross-environment generalization. We evaluate our framework through extensive simulation and ablation studies, and validate it in real-world experiments on a computationally constrained quadrotor. The system achieves agile flight while remaining robust to gate-position errors, developing a generalizable drone racing framework with the capability to operate in diverse, partially unknown and cluttered environments. https://yufengsjtu.github.io/MasterRacing.github.io/
comment: 8 pages, 9 figures, accepted to Robotics and Automation Letters
♻ ☆ Multi-Robot Path Planning Combining Heuristics and Multi-Agent Reinforcement Learning
Multi-robot path finding in dynamic environments is a highly challenging classic problem. In the movement process, robots need to avoid collisions with other moving robots while minimizing their travel distance. Previous methods for this problem either continuously replan paths using heuristic search methods to avoid conflicts or choose appropriate collision avoidance strategies based on learning approaches. The former may result in long travel distances due to frequent replanning, while the latter may have low learning efficiency due to low sample exploration and utilization, and causing high training costs for the model. To address these issues, we propose a path planning method, MAPPOHR, which combines heuristic search, empirical rules, and multi-agent reinforcement learning. The method consists of two layers: a real-time planner based on the multi-agent reinforcement learning algorithm, MAPPO, which embeds empirical rules in the action output layer and reward functions, and a heuristic search planner used to create a global guiding path. During movement, the heuristic search planner replans new paths based on the instructions of the real-time planner. We tested our method in 10 different conflict scenarios. The experiments show that the planning performance of MAPPOHR is better than that of existing learning and heuristic methods. Due to the utilization of empirical knowledge and heuristic search, the learning efficiency of MAPPOHR is higher than that of existing learning methods.
♻ ☆ An Introduction to Deep Reinforcement and Imitation Learning
Embodied agents, such as robots and virtual characters, must continuously select actions to execute tasks effectively, solving complex sequential decision-making problems. Given the difficulty of designing such controllers manually, learning-based approaches have emerged as promising alternatives, most notably Deep Reinforcement Learning (DRL) and Deep Imitation Learning (DIL). DRL leverages reward signals to optimize behavior, while DIL uses expert demonstrations to guide learning. This document introduces DRL and DIL in the context of embodied agents, adopting a concise, depth-first approach to the literature. It is self-contained, presenting all necessary mathematical and machine learning concepts as they are needed. It is not intended as a survey of the field; rather, it focuses on a small set of foundational algorithms and techniques, prioritizing in-depth understanding over broad coverage. The material ranges from Markov Decision Processes to REINFORCE and Proximal Policy Optimization (PPO) for DRL, and from Behavioral Cloning to Dataset Aggregation (DAgger) and Generative Adversarial Imitation Learning (GAIL) for DIL.
♻ ☆ Aligning Humans and Robots via Reinforcement Learning from Implicit Human Feedback
Conventional reinforcement learning (RL) ap proaches often struggle to learn effective policies under sparse reward conditions, necessitating the manual design of complex, task-specific reward functions. To address this limitation, rein forcement learning from human feedback (RLHF) has emerged as a promising strategy that complements hand-crafted rewards with human-derived evaluation signals. However, most existing RLHF methods depend on explicit feedback mechanisms such as button presses or preference labels, which disrupt the natural interaction process and impose a substantial cognitive load on the user. We propose a novel reinforcement learning from implicit human feedback (RLIHF) framework that utilizes non-invasive electroencephalography (EEG) signals, specifically error-related potentials (ErrPs), to provide continuous, implicit feedback without requiring explicit user intervention. The proposed method adopts a pre-trained decoder to transform raw EEG signals into probabilistic reward components, en abling effective policy learning even in the presence of sparse external rewards. We evaluate our approach in a simulation environment built on the MuJoCo physics engine, using a Kinova Gen2 robotic arm to perform a complex pick-and-place task that requires avoiding obstacles while manipulating target objects. The results show that agents trained with decoded EEG feedback achieve performance comparable to those trained with dense, manually designed rewards. These findings validate the potential of using implicit neural feedback for scalable and human-aligned reinforcement learning in interactive robotics.
comment: Accepted to IEEE Int. Conf. Syst., Man, Cybern. (SMC) 2025
♻ ☆ 3DSGrasp: 3D Shape-Completion for Robotic Grasp
Real-world robotic grasping can be done robustly if a complete 3D Point Cloud Data (PCD) of an object is available. However, in practice, PCDs are often incomplete when objects are viewed from few and sparse viewpoints before the grasping action, leading to the generation of wrong or inaccurate grasp poses. We propose a novel grasping strategy, named 3DSGrasp, that predicts the missing geometry from the partial PCD to produce reliable grasp poses. Our proposed PCD completion network is a Transformer-based encoder-decoder network with an Offset-Attention layer. Our network is inherently invariant to the object pose and point's permutation, which generates PCDs that are geometrically consistent and completed properly. Experiments on a wide range of partial PCD show that 3DSGrasp outperforms the best state-of-the-art method on PCD completion tasks and largely improves the grasping success rate in real-world scenarios. The code and dataset will be made available upon acceptance.
♻ ☆ Model-Based Lookahead Reinforcement Learning for in-hand manipulation
In-Hand Manipulation, as many other dexterous tasks, remains a difficult challenge in robotics by combining complex dynamic systems with the capability to control and manoeuvre various objects using its actuators. This work presents the application of a previously developed hybrid Reinforcement Learning (RL) Framework to In-Hand Manipulation task, verifying that it is capable of improving the performance of the task. The model combines concepts of both Model-Free and Model-Based Reinforcement Learning, by guiding a trained policy with the help of a dynamic model and value-function through trajectory evaluation, as done in Model Predictive Control. This work evaluates the performance of the model by comparing it with the policy that will be guided. To fully explore this, various tests are performed using both fully-actuated and under-actuated simulated robotic hands to manipulate different objects for a given task. The performance of the model will also be tested for generalization tests, by changing the properties of the objects in which both the policy and dynamic model were trained, such as density and size, and additionally by guiding a trained policy in a certain object to perform the same task in a different one. The results of this work show that, given a policy with high average reward and an accurate dynamic model, the hybrid framework improves the performance of in-hand manipulation tasks for most test cases, even when the object properties are changed. However, this improvement comes at the expense of increasing the computational cost, due to the complexity of trajectory evaluation.
♻ ☆ See What I Mean? Expressiveness and Clarity in Robot Display Design
Nonverbal visual symbols and displays play an important role in communication when humans and robots work collaboratively. However, few studies have investigated how different types of non-verbal cues affect objective task performance, especially in a dynamic environment that requires real time decision-making. In this work, we designed a collaborative navigation task where the user and the robot only had partial information about the map on each end and thus the users were forced to communicate with a robot to complete the task. We conducted our study in a public space and recruited 37 participants who randomly passed by our setup. Each participant collaborated with a robot utilizing either animated anthropomorphic eyes and animated icons, or static anthropomorphic eyes and static icons. We found that participants that interacted with a robot with animated displays reported the greatest level of trust and satisfaction; that participants interpreted static icons the best; and that participants with a robot with static eyes had the highest completion success. These results suggest that while animation can foster trust with robots, human-robot communication can be optimized by the addition of familiar static icons that may be easier for users to interpret. We published our code, designed symbols, and collected results online at: https://github.com/mattufts/huamn_Cozmo_interaction.
Robotics 52
Closing the Train-Test Gap in World Models for Gradient-Based Planning
World models paired with model predictive control (MPC) can be trained offline on large-scale datasets of expert trajectories and enable generalization to a wide range of planning tasks at inference time. Compared to traditional MPC procedures, which rely on slow search algorithms or on iteratively solving optimization problems exactly, gradient-based planning offers a computationally efficient alternative. However, the performance of gradient-based planning has thus far lagged behind that of other approaches. In this paper, we propose improved methods for training world models that enable efficient gradient-based planning. We begin with the observation that although a world model is trained on a next-state prediction objective, it is used at test-time to instead estimate a sequence of actions. The goal of our work is to close this train-test gap. To that end, we propose train-time data synthesis techniques that enable significantly improved gradient-based planning with existing world models. At test time, our approach outperforms or matches the classical gradient-free cross-entropy method (CEM) across a variety of object manipulation and navigation tasks in 10% of the time budget.
☆ HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Models
Vision-Language-Action (VLA) models have recently enabled robotic manipulation by grounding visual and linguistic cues into actions. However, most VLAs assume the Markov property, relying only on the current observation and thus suffering from temporal myopia that degrades long-horizon coherence. In this work, we view motion as a more compact and informative representation of temporal context and world dynamics, capturing inter-state changes while filtering static pixel-level noise. Building on this idea, we propose HiF-VLA (Hindsight, Insight, and Foresight for VLAs), a unified framework that leverages motion for bidirectional temporal reasoning. HiF-VLA encodes past dynamics through hindsight priors, anticipates future motion via foresight reasoning, and integrates both through a hindsight-modulated joint expert to enable a ''think-while-acting'' paradigm for long-horizon manipulation. As a result, HiF-VLA surpasses strong baselines on LIBERO-Long and CALVIN ABC-D benchmarks, while incurring negligible additional inference latency. Furthermore, HiF-VLA achieves substantial improvements in real-world long-horizon manipulation tasks, demonstrating its broad effectiveness in practical robotic settings.
comment: Project page: https://hifvla.github.io Github: https://github.com/OpenHelix-Team/HiF-VLA
☆ Token Expand-Merge: Training-Free Token Compression for Vision-Language-Action Models
Vision-Language-Action (VLA) models pretrained on large-scale multimodal datasets have emerged as powerful foundations for robotic perception and control. However, their massive scale, often billions of parameters, poses significant challenges for real-time deployment, as inference becomes computationally expensive and latency-sensitive in dynamic environments. To address this, we propose Token Expand-and-Merge-VLA (TEAM-VLA), a training-free token compression framework that accelerates VLA inference while preserving task performance. TEAM-VLA introduces a dynamic token expansion mechanism that identifies and samples additional informative tokens in the spatial vicinity of attention-highlighted regions, enhancing contextual completeness. These expanded tokens are then selectively merged in deeper layers under action-aware guidance, effectively reducing redundancy while maintaining semantic coherence. By coupling expansion and merging within a single feed-forward pass, TEAM-VLA achieves a balanced trade-off between efficiency and effectiveness, without any retraining or parameter updates. Extensive experiments on LIBERO benchmark demonstrate that TEAM-VLA consistently improves inference speed while maintaining or even surpassing the task success rate of full VLA models. The code is public available on \href{https://github.com/Jasper-aaa/TEAM-VLA}{https://github.com/Jasper-aaa/TEAM-VLA}
comment: 8 pages, 5 figures
☆ LISN: Language-Instructed Social Navigation with VLM-based Controller Modulating
Towards human-robot coexistence, socially aware navigation is significant for mobile robots. Yet existing studies on this area focus mainly on path efficiency and pedestrian collision avoidance, which are essential but represent only a fraction of social navigation. Beyond these basics, robots must also comply with user instructions, aligning their actions to task goals and social norms expressed by humans. In this work, we present LISN-Bench, the first simulation-based benchmark for language-instructed social navigation. Built on Rosnav-Arena 3.0, it is the first standardized social navigation benchmark to incorporate instruction following and scene understanding across diverse contexts. To address this task, we further propose Social-Nav-Modulator, a fast-slow hierarchical system where a VLM agent modulates costmaps and controller parameters. Decoupling low-level action generation from the slower VLM loop reduces reliance on high-frequency VLM inference while improving dynamic avoidance and perception adaptability. Our method achieves an average success rate of 91.3%, which is greater than 63% than the most competitive baseline, with most of the improvements observed in challenging tasks such as following a person in a crowd and navigating while strictly avoiding instruction-forbidden regions. The project website is at: https://social-nav.github.io/LISN-project/
comment: 8 pages
☆ Py-DiSMech: A Scalable and Efficient Framework for Discrete Differential Geometry-Based Modeling and Control of Soft Robots
High-fidelity simulation has become essential to the design and control of soft robots, where large geometric deformations and complex contact interactions challenge conventional modeling tools. Recent advances in the field demand simulation frameworks that combine physical accuracy, computational scalability, and seamless integration with modern control and optimization pipelines. In this work, we present Py-DiSMech, a Python-based, open-source simulation framework for modeling and control of soft robotic structures grounded in the principles of Discrete Differential Geometry (DDG). By discretizing geometric quantities such as curvature and strain directly on meshes, Py-DiSMech captures the nonlinear deformation of rods, shells, and hybrid structures with high fidelity and reduced computational cost. The framework introduces (i) a fully vectorized NumPy implementation achieving order-of-magnitude speed-ups over existing geometry-based simulators; (ii) a penalty-energy-based fully implicit contact model that supports rod-rod, rod-shell, and shell-shell interactions; (iii) a natural-strain-based feedback-control module featuring a proportional-integral (PI) controller for shape regulation and trajectory tracking; and (iv) a modular, object-oriented software design enabling user-defined elastic energies, actuation schemes, and integration with machine-learning libraries. Benchmark comparisons demonstrate that Py-DiSMech substantially outperforms the state-of-the-art simulator Elastica in computational efficiency while maintaining physical accuracy. Together, these features establish Py-DiSMech as a scalable, extensible platform for simulation-driven design, control validation, and sim-to-real research in soft robotics.
comment: https://github.com/structuresComp/dismech-python
☆ YOPO-Nav: Visual Navigation using 3DGS Graphs from One-Pass Videos
Visual navigation has emerged as a practical alternative to traditional robotic navigation pipelines that rely on detailed mapping and path planning. However, constructing and maintaining 3D maps is often computationally expensive and memory-intensive. We address the problem of visual navigation when exploration videos of a large environment are available. The videos serve as a visual reference, allowing a robot to retrace the explored trajectories without relying on metric maps. Our proposed method, YOPO-Nav (You Only Pass Once), encodes an environment into a compact spatial representation composed of interconnected local 3D Gaussian Splatting (3DGS) models. During navigation, the framework aligns the robot's current visual observation with this representation and predicts actions that guide it back toward the demonstrated trajectory. YOPO-Nav employs a hierarchical design: a visual place recognition (VPR) module provides coarse localization, while the local 3DGS models refine the goal and intermediate poses to generate control actions. To evaluate our approach, we introduce the YOPO-Campus dataset, comprising 4 hours of egocentric video and robot controller inputs from over 6 km of human-teleoperated robot trajectories. We benchmark recent visual navigation methods on trajectories from YOPO-Campus using a Clearpath Jackal robot. Experimental results show YOPO-Nav provides excellent performance in image-goal navigation for real-world scenes on a physical robot. The dataset and code will be made publicly available for visual navigation and scene representation research.
☆ Visual Heading Prediction for Autonomous Aerial Vehicles
The integration of Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs) is increasingly central to the development of intelligent autonomous systems for applications such as search and rescue, environmental monitoring, and logistics. However, precise coordination between these platforms in real-time scenarios presents major challenges, particularly when external localization infrastructure such as GPS or GNSS is unavailable or degraded [1]. This paper proposes a vision-based, data-driven framework for real-time UAV-UGV integration, with a focus on robust UGV detection and heading angle prediction for navigation and coordination. The system employs a fine-tuned YOLOv5 model to detect UGVs and extract bounding box features, which are then used by a lightweight artificial neural network (ANN) to estimate the UAV's required heading angle. A VICON motion capture system was used to generate ground-truth data during training, resulting in a dataset of over 13,000 annotated images collected in a controlled lab environment. The trained ANN achieves a mean absolute error of 0.1506° and a root mean squared error of 0.1957°, offering accurate heading angle predictions using only monocular camera inputs. Experimental evaluations achieve 95% accuracy in UGV detection. This work contributes a vision-based, infrastructure- independent solution that demonstrates strong potential for deployment in GPS/GNSS-denied environments, supporting reliable multi-agent coordination under realistic dynamic conditions. A demonstration video showcasing the system's real-time performance, including UGV detection, heading angle prediction, and UAV alignment under dynamic conditions, is available at: https://github.com/Kooroshraf/UAV-UGV-Integration
☆ 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 alternating tactile-visual (66.3%) and vision-only (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.
☆ Bridging the Basilisk Astrodynamics Framework with ROS 2 for Modular Spacecraft Simulation and Hardware Integration
Integrating high-fidelity spacecraft simulators with modular robotics frameworks remains a challenge for autonomy development. This paper presents a lightweight, open-source communication bridge between the Basilisk astrodynamics simulator and the Robot Operating System 2 (ROS 2), enabling real-time, bidirectional data exchange for spacecraft control. The bridge requires no changes to Basilisk's core and integrates seamlessly with ROS 2 nodes. We demonstrate its use in a leader-follower formation flying scenario using nonlinear model predictive control, deployed identically in both simulation and on the ATMOS planar microgravity testbed. This setup supports rapid development, hardware-in-the-loop testing, and seamless transition from simulation to hardware. The bridge offers a flexible and scalable platform for modular spacecraft autonomy and reproducible research workflows.
comment: Presented at the International Conference on Space Robotics (iSpaRo) 2025. To appear in IEEE Xplore
☆ High-Resolution Water Sampling via a Solar-Powered Autonomous Surface Vehicle
Accurate water quality assessment requires spatially resolved sampling, yet most unmanned surface vehicles (USVs) can collect only a limited number of samples or rely on single-point sensors with poor representativeness. This work presents a solar-powered, fully autonomous USV featuring a novel syringe-based sampling architecture capable of acquiring 72 discrete, contamination-minimized water samples per mission. The vehicle incorporates a ROS 2 autonomy stack with GPS-RTK navigation, LiDAR and stereo-vision obstacle detection, Nav2-based mission planning, and long-range LoRa supervision, enabling dependable execution of sampling routes in unstructured environments. The platform integrates a behavior-tree autonomy architecture adapted from Nav2, enabling mission-level reasoning and perception-aware navigation. A modular 6x12 sampling system, controlled by distributed micro-ROS nodes, provides deterministic actuation, fault isolation, and rapid module replacement, achieving spatial coverage beyond previously reported USV-based samplers. Field trials in Achocalla Lagoon (La Paz, Bolivia) demonstrated 87% waypoint accuracy, stable autonomous navigation, and accurate physicochemical measurements (temperature, pH, conductivity, total dissolved solids) comparable to manually collected references. These results demonstrate that the platform enables reliable high-resolution sampling and autonomous mission execution, providing a scalable solution for aquatic monitoring in remote environments.
☆ ReMoSPLAT: Reactive Mobile Manipulation Control on a Gaussian Splat
Reactive control can gracefully coordinate the motion of the base and the arm of a mobile manipulator. However, incorporating an accurate representation of the environment to avoid obstacles without involving costly planning remains a challenge. In this work, we present ReMoSPLAT, a reactive controller based on a quadratic program formulation for mobile manipulation that leverages a Gaussian Splat representation for collision avoidance. By integrating additional constraints and costs into the optimisation formulation, a mobile manipulator platform can reach its intended end effector pose while avoiding obstacles, even in cluttered scenes. We investigate the trade-offs of two methods for efficiently calculating robot-obstacle distances, comparing a purely geometric approach with a rasterisation-based approach. Our experiments in simulation on both synthetic and real-world scans demonstrate the feasibility of our method, showing that the proposed approach achieves performance comparable to controllers that rely on perfect ground-truth information.
comment: 9 pages, 5 figures
☆ GLaD: Geometric Latent Distillation for Vision-Language-Action Models
Most existing Vision-Language-Action (VLA) models rely primarily on RGB information, while ignoring geometric cues crucial for spatial reasoning and manipulation. In this work, we introduce GLaD, a geometry-aware VLA framework that incorporates 3D geometric priors during pretraining through knowledge distillation. Rather than distilling geometric features solely into the vision encoder, we align the LLM's hidden states corresponding to visual tokens with features from a frozen geometry-aware vision transformer (VGGT), ensuring that geometric understanding is deeply integrated into the multimodal representations that drive action prediction. Pretrained on the Bridge dataset with this geometry distillation mechanism, GLaD achieves 94.1% average success rate across four LIBERO task suites, outperforming UniVLA (92.5%) which uses identical pretraining data. These results validate that geometry-aware pretraining enhances spatial reasoning and policy generalization without requiring explicit depth sensors or 3D annotations.
☆ Super4DR: 4D Radar-centric Self-supervised Odometry and Gaussian-based Map Optimization
Conventional SLAM systems using visual or LiDAR data often struggle in poor lighting and severe weather. Although 4D radar is suited for such environments, its sparse and noisy point clouds hinder accurate odometry estimation, while the radar maps suffer from obscure and incomplete structures. Thus, we propose Super4DR, a 4D radar-centric framework for learning-based odometry estimation and gaussian-based map optimization. First, we design a cluster-aware odometry network that incorporates object-level cues from the clustered radar points for inter-frame matching, alongside a hierarchical self-supervision mechanism to overcome outliers through spatio-temporal consistency, knowledge transfer, and feature contrast. Second, we propose using 3D gaussians as an intermediate representation, coupled with a radar-specific growth strategy, selective separation, and multi-view regularization, to recover blurry map areas and those undetected based on image texture. Experiments show that Super4DR achieves a 67% performance gain over prior self-supervised methods, nearly matches supervised odometry, and narrows the map quality disparity with LiDAR while enabling multi-modal image rendering.
comment: 17 pages, 20 figures
☆ UrbanNav: Learning Language-Guided Urban Navigation from Web-Scale Human Trajectories AAAI 2026
Navigating complex urban environments using natural language instructions poses significant challenges for embodied agents, including noisy language instructions, ambiguous spatial references, diverse landmarks, and dynamic street scenes. Current visual navigation methods are typically limited to simulated or off-street environments, and often rely on precise goal formats, such as specific coordinates or images. This limits their effectiveness for autonomous agents like last-mile delivery robots navigating unfamiliar cities. To address these limitations, we introduce UrbanNav, a scalable framework that trains embodied agents to follow free-form language instructions in diverse urban settings. Leveraging web-scale city walking videos, we develop an scalable annotation pipeline that aligns human navigation trajectories with language instructions grounded in real-world landmarks. UrbanNav encompasses over 1,500 hours of navigation data and 3 million instruction-trajectory-landmark triplets, capturing a wide range of urban scenarios. Our model learns robust navigation policies to tackle complex urban scenarios, demonstrating superior spatial reasoning, robustness to noisy instructions, and generalization to unseen urban settings. Experimental results show that UrbanNav significantly outperforms existing methods, highlighting the potential of large-scale web video data to enable language-guided, real-world urban navigation for embodied agents.
comment: 9 pages, 5 figures, accepted to AAAI 2026
☆ CS3D: An Efficient Facial Expression Recognition via Event Vision
Responsive and accurate facial expression recognition is crucial to human-robot interaction for daily service robots. Nowadays, event cameras are becoming more widely adopted as they surpass RGB cameras in capturing facial expression changes due to their high temporal resolution, low latency, computational efficiency, and robustness in low-light conditions. Despite these advantages, event-based approaches still encounter practical challenges, particularly in adopting mainstream deep learning models. Traditional deep learning methods for facial expression analysis are energy-intensive, making them difficult to deploy on edge computing devices and thereby increasing costs, especially for high-frequency, dynamic, event vision-based approaches. To address this challenging issue, we proposed the CS3D framework by decomposing the Convolutional 3D method to reduce the computational complexity and energy consumption. Additionally, by utilizing soft spiking neurons and a spatial-temporal attention mechanism, the ability to retain information is enhanced, thus improving the accuracy of facial expression detection. Experimental results indicate that our proposed CS3D method attains higher accuracy on multiple datasets compared to architectures such as the RNN, Transformer, and C3D, while the energy consumption of the CS3D method is just 21.97\% of the original C3D required on the same device.
☆ Mastering Diverse, Unknown, and Cluttered Tracks for Robust Vision-Based Drone Racing
Most reinforcement learning(RL)-based methods for drone racing target fixed, obstacle-free tracks, leaving the generalization to unknown, cluttered environments largely unaddressed. This challenge stems from the need to balance racing speed and collision avoidance, limited feasible space causing policy exploration trapped in local optima during training, and perceptual ambiguity between gates and obstacles in depth maps-especially when gate positions are only coarsely specified. To overcome these issues, we propose a two-phase learning framework: an initial soft-collision training phase that preserves policy exploration for high-speed flight, followed by a hard-collision refinement phase that enforces robust obstacle avoidance. An adaptive, noise-augmented curriculum with an asymmetric actor-critic architecture gradually shifts the policy's reliance from privileged gate-state information to depth-based visual input. We further impose Lipschitz constraints and integrate a track-primitive generator to enhance motion stability and cross-environment generalization. We evaluate our framework through extensive simulation and ablation studies, and validate it in real-world experiments on a computationally constrained quadrotor. The system achieves agile flight while remaining robust to gate-position errors, developing a generalizable drone racing framework with the capability to operate in diverse, partially unknown and cluttered environments. https://yufengsjtu.github.io/MasterRacing.github.io/
comment: 8 pages, 9 figures, Robotics and Automation Letters accept
☆ REASAN: Learning Reactive Safe Navigation for Legged Robots
We present a novel modularized end-to-end framework for legged reactive navigation in complex dynamic environments using a single light detection and ranging (LiDAR) sensor. The system comprises four simulation-trained modules: three reinforcement-learning (RL) policies for locomotion, safety shielding, and navigation, and a transformer-based exteroceptive estimator that processes raw point-cloud inputs. This modular decomposition of complex legged motor-control tasks enables lightweight neural networks with simple architectures, trained using standard RL practices with targeted reward shaping and curriculum design, without reliance on heuristics or sophisticated policy-switching mechanisms. We conduct comprehensive ablations to validate our design choices and demonstrate improved robustness compared to existing approaches in challenging navigation tasks. The resulting reactive safe navigation (REASAN) system achieves fully onboard and real-time reactive navigation across both single- and multi-robot settings in complex environments. We release our training and deployment code at https://github.com/ASIG-X/REASAN.
comment: 8 pages
☆ ViTA-Seg: Vision Transformer for Amodal Segmentation in Robotics
Occlusions in robotic bin picking compromise accurate and reliable grasp planning. We present ViTA-Seg, a class-agnostic Vision Transformer framework for real-time amodal segmentation that leverages global attention to recover complete object masks, including hidden regions. We proposte two architectures: a) Single-Head for amodal mask prediction; b) Dual-Head for amodal and occluded mask prediction. We also introduce ViTA-SimData, a photo-realistic synthetic dataset tailored to industrial bin-picking scenario. Extensive experiments on two amodal benchmarks, COOCA and KINS, demonstrate that ViTA-Seg Dual Head achieves strong amodal and occlusion segmentation accuracy with computational efficiency, enabling robust, real-time robotic manipulation.
☆ On Mobile Ad Hoc Networks for Coverage of Partially Observable Worlds
This paper addresses the movement and placement of mobile agents to establish a communication network in initially unknown environments. We cast the problem in a computational-geometric framework by relating the coverage problem and line-of-sight constraints to the Cooperative Guard Art Gallery Problem, and introduce its partially observable variant, the Partially Observable Cooperative Guard Art Gallery Problem (POCGAGP). We then present two algorithms that solve POCGAGP: CADENCE, a centralized planner that incrementally selects 270 degree corners at which to deploy agents, and DADENCE, a decentralized scheme that coordinates agents using local information and lightweight messaging. Both approaches operate under partial observability and target simultaneous coverage and connectivity. We evaluate the methods in simulation across 1,500 test cases of varied size and structure, demonstrating consistent success in forming connected networks while covering and exploring unknown space. These results highlight the value of geometric abstractions for communication-driven exploration and show that decentralized policies are competitive with centralized performance while retaining scalability.
☆ Development of a Compliant Gripper for Safe Robot-Assisted Trouser Dressing-Undressing
In recent years, many countries, including Japan, have rapidly aging populations, making the preservation of seniors' quality of life a significant concern. For elderly people with impaired physical abilities, support for toileting is one of the most important issues. This paper details the design, development, experimental assessment, and potential application of the gripper system, with a focus on the unique requirements and obstacles involved in aiding elderly or hemiplegic individuals in dressing and undressing trousers. The gripper we propose seeks to find the right balance between compliance and grasping forces, ensuring precise manipulation while maintaining a safe and compliant interaction with the users. The gripper's integration into a custom--built robotic manipulator system provides a comprehensive solution for assisting hemiplegic individuals in their dressing and undressing tasks. Experimental evaluations and comparisons with existing studies demonstrate the gripper's ability to successfully assist in both dressing and dressing of trousers in confined spaces with a high success rate. This research contributes to the advancement of assistive robotics, empowering elderly, and physically impaired individuals to maintain their independence and improve their quality of life.
☆ Sequential Testing for Descriptor-Agnostic LiDAR Loop Closure in Repetitive Environments
We propose a descriptor-agnostic, multi-frame loop closure verification method that formulates LiDAR loop closure as a truncated Sequential Probability Ratio Test (SPRT). Instead of deciding from a single descriptor comparison or using fixed thresholds with late-stage Iterative Closest Point (ICP) vetting, the verifier accumulates a short temporal stream of descriptor similarities between a query and each candidate. It then issues an accept/reject decision adaptively once sufficient multi-frame evidence has been observed, according to user-specified Type-I/II error design targets. This precision-first policy is designed to suppress false positives in structurally repetitive indoor environments. We evaluate the verifier on a five-sequence library dataset, using a fixed retrieval front-end with several representative LiDAR global descriptors. Performance is assessed via segment-level K-hit precision-recall and absolute trajectory error (ATE) and relative pose error (RPE) after pose graph optimization. Across descriptors, the sequential verifier consistently improves precision and reduces the impact of aliased loops compared with single-frame and heuristic multi-frame baselines. Our implementation and dataset will be released at: https://github.com/wanderingcar/snu_library_dataset.
comment: 8 pages, 4 figures
☆ A Hierarchical, Model-Based System for High-Performance Humanoid Soccer
The development of athletic humanoid robots has gained significant attention as advances in actuation, sensing, and control enable increasingly dynamic, real-world capabilities. RoboCup, an international competition of fully autonomous humanoid robots, provides a uniquely challenging benchmark for such systems, culminating in the long-term goal of competing against human soccer players by 2050. This paper presents the hardware and software innovations underlying our team's victory in the RoboCup 2024 Adult-Sized Humanoid Soccer Competition. On the hardware side, we introduce an adult-sized humanoid platform built with lightweight structural components, high-torque quasi-direct-drive actuators, and a specialized foot design that enables powerful in-gait kicks while preserving locomotion robustness. On the software side, we develop an integrated perception and localization framework that combines stereo vision, object detection, and landmark-based fusion to provide reliable estimates of the ball, goals, teammates, and opponents. A mid-level navigation stack then generates collision-aware, dynamically feasible trajectories, while a centralized behavior manager coordinates high-level decision making, role selection, and kick execution based on the evolving game state. The seamless integration of these subsystems results in fast, precise, and tactically effective gameplay, enabling robust performance under the dynamic and adversarial conditions of real matches. This paper presents the design principles, system architecture, and experimental results that contributed to ARTEMIS's success as the 2024 Adult-Sized Humanoid Soccer champion.
☆ D$^2$GSLAM: 4D Dynamic Gaussian Splatting SLAM
Recent advances in Dense Simultaneous Localization and Mapping (SLAM) have demonstrated remarkable performance in static environments. However, dense SLAM in dynamic environments remains challenging. Most methods directly remove dynamic objects and focus solely on static scene reconstruction, which ignores the motion information contained in these dynamic objects. In this paper, we present D$^2$GSLAM, a novel dynamic SLAM system utilizing Gaussian representation, which simultaneously performs accurate dynamic reconstruction and robust tracking within dynamic environments. Our system is composed of four key components: (i) We propose a geometric-prompt dynamic separation method to distinguish between static and dynamic elements of the scene. This approach leverages the geometric consistency of Gaussian representation and scene geometry to obtain coarse dynamic regions. The regions then serve as prompts to guide the refinement of the coarse mask for achieving accurate motion mask. (ii) To facilitate accurate and efficient mapping of the dynamic scene, we introduce dynamic-static composite representation that integrates static 3D Gaussians with dynamic 4D Gaussians. This representation allows for modeling the transitions between static and dynamic states of objects in the scene for composite mapping and optimization. (iii) We employ a progressive pose refinement strategy that leverages both the multi-view consistency of static scene geometry and motion information from dynamic objects to achieve accurate camera tracking. (iv) We introduce a motion consistency loss, which leverages the temporal continuity in object motions for accurate dynamic modeling. Our D$^2$GSLAM demonstrates superior performance on dynamic scenes in terms of mapping and tracking accuracy, while also showing capability in accurate dynamic modeling.
☆ Generalizable Collaborative Search-and-Capture in Cluttered Environments via Path-Guided MAPPO and Directional Frontier Allocation
Collaborative pursuit-evasion in cluttered environments presents significant challenges due to sparse rewards and constrained Fields of View (FOV). Standard Multi-Agent Reinforcement Learning (MARL) often suffers from inefficient exploration and fails to scale to large scenarios. We propose PGF-MAPPO (Path-Guided Frontier MAPPO), a hierarchical framework bridging topological planning with reactive control. To resolve local minima and sparse rewards, we integrate an A*-based potential field for dense reward shaping. Furthermore, we introduce Directional Frontier Allocation, combining Farthest Point Sampling (FPS) with geometric angle suppression to enforce spatial dispersion and accelerate coverage. The architecture employs a parameter-shared decentralized critic, maintaining O(1) model complexity suitable for robotic swarms. Experiments demonstrate that PGF-MAPPO achieves superior capture efficiency against faster evaders. Policies trained on 10x10 maps exhibit robust zero-shot generalization to unseen 20x20 environments, significantly outperforming rule-based and learning-based baselines.
comment: 7 pages, 7 figures
☆ H2R-Grounder: A Paired-Data-Free Paradigm for Translating Human Interaction Videos into Physically Grounded Robot Videos
Robots that learn manipulation skills from everyday human videos could acquire broad capabilities without tedious robot data collection. We propose a video-to-video translation framework that converts ordinary human-object interaction videos into motion-consistent robot manipulation videos with realistic, physically grounded interactions. Our approach does not require any paired human-robot videos for training only a set of unpaired robot videos, making the system easy to scale. We introduce a transferable representation that bridges the embodiment gap: by inpainting the robot arm in training videos to obtain a clean background and overlaying a simple visual cue (a marker and arrow indicating the gripper's position and orientation), we can condition a generative model to insert the robot arm back into the scene. At test time, we apply the same process to human videos (inpainting the person and overlaying human pose cues) and generate high-quality robot videos that mimic the human's actions. We fine-tune a SOTA video diffusion model (Wan 2.2) in an in-context learning manner to ensure temporal coherence and leveraging of its rich prior knowledge. Empirical results demonstrate that our approach achieves significantly more realistic and grounded robot motions compared to baselines, pointing to a promising direction for scaling up robot learning from unlabeled human videos. Project page: https://showlab.github.io/H2R-Grounder/
comment: 13 pages, 6 figures
☆ Observability Analysis and Composite Disturbance Filtering for a Bar Tethered to Dual UAVs Subject to Multi-source Disturbances
Cooperative suspended aerial transportation is highly susceptible to multi-source disturbances such as aerodynamic effects and thrust uncertainties. To achieve precise load manipulation, existing methods often rely on extra sensors to measure cable directions or the payload's pose, which increases the system cost and complexity. A fundamental question remains: is the payload's pose observable under multi-source disturbances using only the drones' odometry information? To answer this question, this work focuses on the two-drone-bar system and proves that the whole system is observable when only two or fewer types of lumped disturbances exist by using the observability rank criterion. To the best of our knowledge, we are the first to present such a conclusion and this result paves the way for more cost-effective and robust systems by minimizing their sensor suites. Next, to validate this analysis, we consider the situation where the disturbances are only exerted on the drones, and develop a composite disturbance filtering scheme. A disturbance observer-based error-state extended Kalman filter is designed for both state and disturbance estimation, which renders improved estimation performance for the whole system evolving on the manifold $(\mathbb{R}^3)^2\times(TS^2)^3$. Our simulation and experimental tests have validated that it is possible to fully estimate the state and disturbance of the system with only odometry information of the drones.
☆ COVLM-RL: Critical Object-Oriented Reasoning for Autonomous Driving Using VLM-Guided Reinforcement Learning
End-to-end autonomous driving frameworks face persistent challenges in generalization, training efficiency, and interpretability. While recent methods leverage Vision-Language Models (VLMs) through supervised learning on large-scale datasets to improve reasoning, they often lack robustness in novel scenarios. Conversely, reinforcement learning (RL)-based approaches enhance adaptability but remain data-inefficient and lack transparent decision-making. % contribution To address these limitations, we propose COVLM-RL, a novel end-to-end driving framework that integrates Critical Object-oriented (CO) reasoning with VLM-guided RL. Specifically, we design a Chain-of-Thought (CoT) prompting strategy that enables the VLM to reason over critical traffic elements and generate high-level semantic decisions, effectively transforming multi-view visual inputs into structured semantic decision priors. These priors reduce the input dimensionality and inject task-relevant knowledge into the RL loop, accelerating training and improving policy interpretability. However, bridging high-level semantic guidance with continuous low-level control remains non-trivial. To this end, we introduce a consistency loss that encourages alignment between the VLM's semantic plans and the RL agent's control outputs, enhancing interpretability and training stability. Experiments conducted in the CARLA simulator demonstrate that COVLM-RL significantly improves the success rate by 30\% in trained driving environments and by 50\% in previously unseen environments, highlighting its strong generalization capability.
comment: 8 pages, 5 figures
☆ Development and Testing for Perception Based Autonomous Landing of a Long-Range QuadPlane
QuadPlanes combine the range efficiency of fixed-wing aircraft with the maneuverability of multi-rotor platforms for long-range autonomous missions. In GPS-denied or cluttered urban environments, perception-based landing is vital for reliable operation. Unlike structured landing zones, real-world sites are unstructured and highly variable, requiring strong generalization capabilities from the perception system. Deep neural networks (DNNs) provide a scalable solution for learning landing site features across diverse visual and environmental conditions. While perception-driven landing has been shown in simulation, real-world deployment introduces significant challenges. Payload and volume constraints limit high-performance edge AI devices like the NVIDIA Jetson Orin Nano, which are crucial for real-time detection and control. Accurate pose estimation during descent is necessary, especially in the absence of GPS, and relies on dependable visual-inertial odometry. Achieving this with limited edge AI resources requires careful optimization of the entire deployment framework. The flight characteristics of large QuadPlanes further complicate the problem. These aircraft exhibit high inertia, reduced thrust vectoring, and slow response times further complicate stable landing maneuvers. This work presents a lightweight QuadPlane system for efficient vision-based autonomous landing and visual-inertial odometry, specifically developed for long-range QuadPlane operations such as aerial monitoring. It describes the hardware platform, sensor configuration, and embedded computing architecture designed to meet demanding real-time, physical constraints. This establishes a foundation for deploying autonomous landing in dynamic, unstructured, GPS-denied environments.
☆ Scene-agnostic Hierarchical Bimanual Task Planning via Visual Affordance Reasoning
Embodied agents operating in open environments must translate high-level instructions into grounded, executable behaviors, often requiring coordinated use of both hands. While recent foundation models offer strong semantic reasoning, existing robotic task planners remain predominantly unimanual and fail to address the spatial, geometric, and coordination challenges inherent to bimanual manipulation in scene-agnostic settings. We present a unified framework for scene-agnostic bimanual task planning that bridges high-level reasoning with 3D-grounded two-handed execution. Our approach integrates three key modules. Visual Point Grounding (VPG) analyzes a single scene image to detect relevant objects and generate world-aligned interaction points. Bimanual Subgoal Planner (BSP) reasons over spatial adjacency and cross-object accessibility to produce compact, motion-neutralized subgoals that exploit opportunities for coordinated two-handed actions. Interaction-Point-Driven Bimanual Prompting (IPBP) binds these subgoals to a structured skill library, instantiating synchronized unimanual or bimanual action sequences that satisfy hand-state and affordance constraints. Together, these modules enable agents to plan semantically meaningful, physically feasible, and parallelizable two-handed behaviors in cluttered, previously unseen scenes. Experiments show that it produces coherent, feasible, and compact two-handed plans, and generalizes to cluttered scenes without retraining, demonstrating robust scene-agnostic affordance reasoning for bimanual tasks.
comment: 8 pages, 4 figures
☆ One-Shot Real-World Demonstration Synthesis for Scalable Bimanual Manipulation
Learning dexterous bimanual manipulation policies critically depends on large-scale, high-quality demonstrations, yet current paradigms face inherent trade-offs: teleoperation provides physically grounded data but is prohibitively labor-intensive, while simulation-based synthesis scales efficiently but suffers from sim-to-real gaps. We present BiDemoSyn, a framework that synthesizes contact-rich, physically feasible bimanual demonstrations from a single real-world example. The key idea is to decompose tasks into invariant coordination blocks and variable, object-dependent adjustments, then adapt them through vision-guided alignment and lightweight trajectory optimization. This enables the generation of thousands of diverse and feasible demonstrations within several hour, without repeated teleoperation or reliance on imperfect simulation. Across six dual-arm tasks, we show that policies trained on BiDemoSyn data generalize robustly to novel object poses and shapes, significantly outperforming recent baselines. By bridging the gap between efficiency and real-world fidelity, BiDemoSyn provides a scalable path toward practical imitation learning for complex bimanual manipulation without compromising physical grounding.
comment: under review
☆ UPETrack: Unidirectional Position Estimation for Tracking Occluded Deformable Linear Objects
Real-time state tracking of Deformable Linear Objects (DLOs) is critical for enabling robotic manipulation of DLOs in industrial assembly, medical procedures, and daily-life applications. However, the high-dimensional configuration space, nonlinear dynamics, and frequent partial occlusions present fundamental barriers to robust real-time DLO tracking. To address these limitations, this study introduces UPETrack, a geometry-driven framework based on Unidirectional Position Estimation (UPE), which facilitates tracking without the requirement for physical modeling, virtual simulation, or visual markers. The framework operates in two phases: (1) visible segment tracking is based on a Gaussian Mixture Model (GMM) fitted via the Expectation Maximization (EM) algorithm, and (2) occlusion region prediction employing UPE algorithm we proposed. UPE leverages the geometric continuity inherent in DLO shapes and their temporal evolution patterns to derive a closed-form positional estimator through three principal mechanisms: (i) local linear combination displacement term, (ii) proximal linear constraint term, and (iii) historical curvature term. This analytical formulation allows efficient and stable estimation of occluded nodes through explicit linear combinations of geometric components, eliminating the need for additional iterative optimization. Experimental results demonstrate that UPETrack surpasses two state-of-the-art tracking algorithms, including TrackDLO and CDCPD2, in both positioning accuracy and computational efficiency.
☆ MPC for momentum counter-balanced and zero-impulse contact with a free-spinning satellite
In on-orbit robotics, a servicer satellite's ability to make contact with a free-spinning target satellite is essential to completing most on-orbit servicing (OOS) tasks. This manuscript develops a nonlinear model predictive control (MPC) framework that generates feasible controls for a servicer satellite to achieve zero-impulse contact with a free-spinning target satellite. The overall maneuver requires coordination between two separately actuated modules of the servicer satellite: (1) a moment generation module and (2) a manipulation module. We apply MPC to control both modules by explicitly modeling the cross-coupling dynamics between them. We demonstrate that the MPC controller can enforce actuation and state constraints that prior control approaches could not account for. We evaluate the performance of the MPC controller by simulating zero-impulse contact scenarios with a free-spinning target satellite via numerical Monte Carlo (MC) trials and comparing the simulation results with prior control approaches. Our simulation results validate the effectiveness of the MPC controller in maintaining spin synchronization and zero-impulse contact under operation constraints, moving contact location, and observation and actuation noise.
comment: 21 pages, 4 figures, 5 tables, submission for AIAA SciTech 2026 conference
☆ Inertial Magnetic SLAM Systems Using Low-Cost Sensors
Spatially inhomogeneous magnetic fields offer a valuable, non-visual information source for positioning. Among systems leveraging this, magnetic field-based simultaneous localization and mapping (SLAM) systems are particularly attractive because they can provide positioning information and build a magnetic field map on the fly. Moreover, they have bounded error within mapped regions. However, state-of-the-art methods typically require low-drift odometry data provided by visual odometry or a wheel encoder, etc. This is because these systems need to minimize/reduce positioning errors while exploring, which happens when they are in unmapped regions. To address these limitations, this work proposes a loosely coupled and a tightly coupled inertial magnetic SLAM (IM-SLAM) system. The proposed systems use commonly available low-cost sensors: an inertial measurement unit (IMU), a magnetometer array, and a barometer. The use of non-visual data provides a significant advantage over visual-based systems, making it robust to low-visibility conditions. Both systems employ state-space representations, and magnetic field models on different scales. The difference lies in how they use a local and global magnetic field model. The loosely coupled system uses these models separately in two state-space models, while the tightly coupled system integrates them into one state-space model. Experiment results show that the tightly coupled IM-SLAM system achieves lower positioning errors than the loosely coupled system in most scenarios, with typical errors on the order of meters per 100 meters traveled. These results demonstrate the feasiblity of developing a full 3D IM-SLAM systems using low-cost sensors and the potential of applying these systems in emergency response scenarios such as mine/fire rescue.
☆ CHyLL: Learning Continuous Neural Representations of Hybrid Systems
Learning the flows of hybrid systems that have both continuous and discrete time dynamics is challenging. The existing method learns the dynamics in each discrete mode, which suffers from the combination of mode switching and discontinuities in the flows. In this work, we propose CHyLL (Continuous Hybrid System Learning in Latent Space), which learns a continuous neural representation of a hybrid system without trajectory segmentation, event functions, or mode switching. The key insight of CHyLL is that the reset map glues the state space at the guard surface, reformulating the state space as a piecewise smooth quotient manifold where the flow becomes spatially continuous. Building upon these insights and the embedding theorems grounded in differential topology, CHyLL concurrently learns a singularity-free neural embedding in a higher-dimensional space and the continuous flow in it. We showcase that CHyLL can accurately predict the flow of hybrid systems with superior accuracy and identify the topological invariants of the hybrid systems. Finally, we apply CHyLL to the stochastic optimal control problem.
☆ Fast Functionally Redundant Inverse Kinematics for Robotic Toolpath Optimisation in Manufacturing Tasks
Industrial automation with six-axis robotic arms is critical for many manufacturing tasks, including welding and additive manufacturing applications; however, many of these operations are functionally redundant due to the symmetrical tool axis, which effectively makes the operation a five-axis task. Exploiting this redundancy is crucial for achieving the desired workspace and dexterity required for the feasibility and optimisation of toolpath planning. Inverse kinematics algorithms can solve this in a fast, reactive framework, but these techniques are underutilised over the more computationally expensive offline planning methods. We propose a novel algorithm to solve functionally redundant inverse kinematics for robotic manipulation utilising a task space decomposition approach, the damped least-squares method and Halley's method to achieve fast and robust solutions with reduced joint motion. We evaluate our methodology in the case of toolpath optimisation in a cold spray coating application on a non-planar surface. The functionally redundant inverse kinematics algorithm can quickly solve motion plans that minimise joint motion, expanding the feasible operating space of the complex toolpath. We validate our approach on an industrial ABB manipulator and cold-spray gun executing the computed toolpath.
comment: Published at the Australasian Conference on Robotics and Automation (ACRA 2025) https://ssl.linklings.net/conferences/acra/acra2025_proceedings/views/includes/files/pap149s2.pdf
☆ Push Smarter, Not Harder: Hierarchical RL-Diffusion Policy for Efficient Nonprehensile Manipulation
Nonprehensile manipulation, such as pushing objects across cluttered environments, presents a challenging control problem due to complex contact dynamics and long-horizon planning requirements. In this work, we propose HeRD, a hierarchical reinforcement learning-diffusion policy that decomposes pushing tasks into two levels: high-level goal selection and low-level trajectory generation. We employ a high-level reinforcement learning (RL) agent to select intermediate spatial goals, and a low-level goal-conditioned diffusion model to generate feasible, efficient trajectories to reach them. This architecture combines the long-term reward maximizing behaviour of RL with the generative capabilities of diffusion models. We evaluate our method in a 2D simulation environment and show that it outperforms the state-of-the-art baseline in success rate, path efficiency, and generalization across multiple environment configurations. Our results suggest that hierarchical control with generative low-level planning is a promising direction for scalable, goal-directed nonprehensile manipulation. Code, documentation, and trained models are available: https://github.com/carosteven/HeRD.
comment: 8 pages, 8 figures
☆ Openpi Comet: Competition Solution For 2025 BEHAVIOR Challenge
The 2025 BEHAVIOR Challenge is designed to rigorously track progress toward solving long-horizon tasks by physical agents in simulated environments. BEHAVIOR-1K focuses on everyday household tasks that people most want robots to assist with and these tasks introduce long-horizon mobile manipulation challenges in realistic settings, bridging the gap between current research and real-world, human-centric applications. This report presents our solution to the 2025 BEHAVIOR Challenge in a very close 2nd place and substantially outperforms the rest of the submissions. Building on $π_{0.5}$, we focus on systematically building our solution by studying the effects of training techniques and data. Through careful ablations, we show the scaling power in pre-training and post-training phases for competitive performance. We summarize our practical lessons and design recommendations that we hope will provide actionable insights for the broader embodied AI community when adapting powerful foundation models to complex embodied scenarios.
comment: preprint
♻ ☆ C*: A Coverage Path Planning Algorithm for Unknown Environments using Rapidly Covering Graphs
The paper presents a novel sample-based algorithm, called C*, for real-time coverage path planning (CPP) of unknown environments. C* is built upon the concept of a Rapidly Covering Graph (RCG), which is incrementally constructed during robot navigation via progressive sampling of the search space. By using efficient sampling and pruning techniques, the RCG is constructed to be a minimum-sufficient graph, where its nodes and edges form the potential waypoints and segments of the coverage trajectory, respectively. The RCG tracks the coverage progress, generates the coverage trajectory and helps the robot to escape from the dead-end situations. To minimize coverage time, C* produces the desired back-and-forth coverage pattern, while adapting to the TSP-based optimal coverage of local isolated regions, called coverage holes, which are surrounded by obstacles and covered regions. It is analytically proven that C* provides complete coverage of unknown environments. The algorithmic simplicity and low computational complexity of C* make it easy to implement and suitable for real-time on-board applications. The performance of C* is validated by 1) extensive high-fidelity simulations and 2) laboratory experiments using an autonomous robot. C* yields near optimal trajectories, and a comparative evaluation with seven existing CPP methods demonstrates significant improvements in performance in terms of coverage time, number of turns, trajectory length, and overlap ratio, while preventing the formation of coverage holes. Finally, C* is comparatively evaluated on two different CPP applications using 1) energy-constrained robots and 2) multi-robot teams.
♻ ☆ Toward Efficient and Robust Behavior Models for Multi-Agent Driving Simulation
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: This work has been submitted to the IEEE for possible publication
♻ ☆ Model-free Adaptive Output Feedback Vibration Suppression in a Cantilever Beam
This paper presents a model-free adaptive control approach to suppress vibrations in a cantilevered beam excited by an unknown disturbance. The cantilevered beam under harmonic excitation is modeled using a lumped parameter approach. Based on retrospective cost optimization, a sampled-data adaptive controller is developed to suppress vibrations caused by external disturbances. Both displacement and acceleration measurements are considered for feedback. Since acceleration measurements are more sensitive to spillover, which excites higher frequency modes, a filter is developed to extract key displacement information from the acceleration data and enhance suppression performance. The vibration suppression performance is compared using both displacement and acceleration measurements.
comment: 16 pages, 14 figures, to be presented at Scitech 2026
♻ ☆ PUL-SLAM: Path-Uncertainty Co-Optimization with Lightweight Stagnation Detection for Efficient Robotic Exploration
Existing Active SLAM methodologies face issues such as slow exploration speed and suboptimal paths. To address these limitations, we propose a hybrid framework combining a Path-Uncertainty Co-Optimization Deep Reinforcement Learning framework and a Lightweight Stagnation Detection mechanism. The Path-Uncertainty Co-Optimization framework jointly optimizes travel distance and map uncertainty through a dual-objective reward function, balancing exploration and exploitation. The Lightweight Stagnation Detection reduces redundant exploration through Lidar Static Anomaly Detection and Map Update Stagnation Detection, terminating episodes on low expansion rates. Experimental results show that compared with the frontier-based method and RRT method, our approach shortens exploration time by up to 65% and reduces path distance by up to 42%, significantly improving exploration efficiency in complex environments while maintaining reliable map completeness. Ablation studies confirm that the collaborative mechanism accelerates training convergence. Empirical validation on a physical robotic platform demonstrates the algorithm's practical applicability and its successful transferability from simulation to real-world environments.
♻ ☆ Breaking the Circle: An Autonomous Control-Switching Strategy for Stable Orographic Soaring in MAVs
Orographic soaring can significantly extend the endurance of micro aerial vehicles (MAVs), but circling behavior, arising from control conflicts between the longitudinal and vertical axes, increases energy consumption and the risk of divergence. We propose a control switching method, named SAOS: Switched Control for Autonomous Orographic Soaring, which mitigates circling behavior by selectively controlling either the horizontal or vertical axis, effectively transforming the system from underactuated to fully actuated during soaring. Additionally, the angle of attack is incorporated into the INDI controller to improve force estimation. Simulations with randomized initial positions and wind tunnel experiments on two MAVs demonstrate that the SAOS improves position convergence, reduces throttle usage, and mitigates roll oscillations caused by pitch-roll coupling. These improvements enhance energy efficiency and flight stability in constrained soaring environments.
comment: 13 pages, 15 figures
♻ ☆ Hybrid A* Path Planning with Multi-Modal Motion Extension for Four-Wheel Steering Mobile Robots
Four-wheel independent steering (4WIS) systems provide mobile robots with a rich set of motion modes, such as Ackermann steering, lateral steering, and parallel movement, offering superior maneuverability in constrained environments. However, existing path planning methods generally assume a single kinematic model and thus fail to fully exploit the multi-modal capabilities of 4WIS platforms. To address this limitation, we propose an extended Hybrid A* framework that operates in a four-dimensional state space incorporating both spatial states and motion modes. Within this framework, we design multi-modal Reeds-Shepp curves tailored to the distinct kinematic constraints of each motion mode, develop an enhanced heuristic function that accounts for mode-switching costs, and introduce a terminal connection strategy with intelligent mode selection to ensure smooth transitions between different steering patterns. The proposed planner enables seamless integration of multiple motion modalities within a single path, significantly improving flexibility and adaptability in complex environments. Results demonstrate significantly improved planning performance for 4WIS robots in complex environments.
comment: Corrected some formatting errors
♻ ☆ Mind to Hand: Purposeful Robotic Control via Embodied Reasoning
Humans act with context and intention, with reasoning playing a central role. While internet-scale data has enabled broad reasoning capabilities in AI systems, grounding these abilities in physical action remains a major challenge. We introduce Lumo-1, a generalist vision-language-action (VLA) model that unifies robot reasoning ("mind") with robot action ("hand"). Our approach builds upon the general multi-modal reasoning capabilities of pre-trained vision-language models (VLMs), progressively extending them to embodied reasoning and action prediction, and ultimately towards structured reasoning and reasoning-action alignment. This results in a three-stage pre-training pipeline: (1) Continued VLM pre-training on curated vision-language data to enhance embodied reasoning skills such as planning, spatial understanding, and trajectory prediction; (2) Co-training on cross-embodiment robot data alongside vision-language data; and (3) Action training with reasoning process on trajectories collected on Astribot S1, a bimanual mobile manipulator with human-like dexterity and agility. Finally, we integrate reinforcement learning to further refine reasoning-action consistency and close the loop between semantic inference and motor control. Extensive experiments demonstrate that Lumo-1 achieves significant performance improvements in embodied vision-language reasoning, a critical component for generalist robotic control. Real-world evaluations further show that Lumo-1 surpasses strong baselines across a wide range of challenging robotic tasks, with strong generalization to novel objects and environments, excelling particularly in long-horizon tasks and responding to human-natural instructions that require reasoning over strategy, concepts and space.
comment: 49 pages, 25 figures
♻ ☆ SensHRPS: Sensing Comfortable Human-Robot Proxemics and Personal Space With Eye-Tracking
Social robots must adjust to human proxemic norms to ensure user comfort and engagement. While prior research demonstrates that eye-tracking features reliably estimate comfort in human-human interactions, their applicability to interactions with humanoid robots remains unexplored. In this study, we investigate user comfort with the robot "Ameca" across four experimentally controlled distances (0.5 m to 2.0 m) using mobile eye-tracking and subjective reporting (N=19). We evaluate multiple machine learning and deep learning models to estimate comfort based on gaze features. Contrary to previous human-human studies where Transformer models excelled, a Decision Tree classifier achieved the highest performance (F1-score = 0.73), with minimum pupil diameter identified as the most critical predictor. These findings suggest that physiological comfort thresholds in human-robot interaction differ from human-human dynamics and can be effectively modeled using interpretable logic.
♻ ☆ Flow-Aided Flight Through Dynamic Clutters From Point To Motion
Challenges in traversing dynamic clutters lie mainly in the efficient perception of the environmental dynamics and the generation of evasive behaviors considering obstacle movement. Previous solutions have made progress in explicitly modeling the dynamic obstacle motion for avoidance, but this key dependency of decision-making is time-consuming and unreliable in highly dynamic scenarios with occlusions. On the contrary, without introducing object detection, tracking, and prediction, we empower the reinforcement learning (RL) with single LiDAR sensing to realize an autonomous flight system directly from point to motion. For exteroception, a depth sensing distance map achieving fixed-shape, low-resolution, and detail-safe is encoded from raw point clouds, and an environment change sensing point flow is adopted as motion features extracted from multi-frame observations. These two are integrated into a lightweight and easy-to-learn representation of complex dynamic environments. For action generation, the behavior of avoiding dynamic threats in advance is implicitly driven by the proposed change-aware sensing representation, where the policy optimization is indicated by the relative motion modulated distance field. With the deployment-friendly sensing simulation and dynamics model-free acceleration control, the proposed system shows a superior success rate and adaptability to alternatives, and the policy derived from the simulator can drive a real-world quadrotor with safe maneuvers.
comment: Accepted to IEEE Robotics and Automation Letters (RA-L), November, 2025
♻ ☆ More than Segmentation: Benchmarking SAM 3 for Segmentation, 3D Perception, and Reconstruction in Robotic Surgery
The recent SAM 3 and SAM 3D have introduced significant advancements over the predecessor, SAM 2, particularly with the integration of language-based segmentation and enhanced 3D perception capabilities. SAM 3 supports zero-shot segmentation across a wide range of prompts, including point, bounding box, and language-based prompts, allowing for more flexible and intuitive interactions with the model. In this empirical evaluation, we assess the performance of SAM 3 in robot-assisted surgery, benchmarking its zero-shot segmentation with point and bounding box prompts and exploring its effectiveness in dynamic video tracking, alongside its newly introduced language prompt segmentation. While language prompts show potential, their performance in the surgical domain is currently suboptimal, highlighting the need for further domain-specific training. Additionally, we investigate SAM 3D's depth reconstruction abilities, demonstrating its capacity to process surgical scene data and reconstruct 3D anatomical structures from 2D images. Through comprehensive testing on the MICCAI EndoVis 2017 and EndoVis 2018 benchmarks, SAM 3 shows clear improvements over SAM and SAM 2 in both image and video segmentation under spatial prompts, while the zero-shot evaluations of SAM 3D on SCARED, StereoMIS, and EndoNeRF indicate strong monocular depth estimation and realistic 3D instrument reconstruction, yet also reveal remaining limitations in complex, highly dynamic surgical scenes.
comment: Technical Report
♻ ☆ Proportional integral derivative booster for neural networks-based time-series prediction: Case of water demand prediction
Multi-step time-series prediction is an essential supportive step for decision-makers in several industrial areas. Artificial intelligence techniques, which use a neural network component in various forms, have recently frequently been used to accomplish this step. However, the complexity of the neural network structure still stands up as a critical problem against prediction accuracy. In this paper, a method inspired by the proportional-integral-derivative (PID) control approach is investigated to enhance the performance of neural network models used for multi-step ahead prediction of periodic time-series information while maintaining a negligible impact on the complexity of the system. The PID-based method is applied to the predicted value at each time step to bring that value closer to the real value. The water demand forecasting problem is considered as a case study, where two deep neural network models from the literature are used to prove the effectiveness of the proposed boosting method. Furthermore, to prove the applicability of this PID-based booster to other types of periodic time-series prediction problems, it is applied to enhance the accuracy of a neural network model used for multi-step forecasting of hourly energy consumption. The comparison between the results of the original prediction models and the results after using the proposed technique demonstrates the superiority of the proposed method in terms of prediction accuracy and system complexity.
comment: Engineering Applications of Artificial Intelligence 2022
♻ ☆ From the Laboratory to Real-World Application: Evaluating Zero-Shot Scene Interpretation on Edge Devices for Mobile Robotics
Video Understanding, Scene Interpretation and Commonsense Reasoning are highly challenging tasks enabling the interpretation of visual information, allowing agents to perceive, interact with and make rational decisions in its environment. Large Language Models (LLMs) and Visual Language Models (VLMs) have shown remarkable advancements in these areas in recent years, enabling domain-specific applications as well as zero-shot open vocabulary tasks, combining multiple domains. However, the required computational complexity poses challenges for their application on edge devices and in the context of Mobile Robotics, especially considering the trade-off between accuracy and inference time. In this paper, we investigate the capabilities of state-of-the-art VLMs for the task of Scene Interpretation and Action Recognition, with special regard to small VLMs capable of being deployed to edge devices in the context of Mobile Robotics. The proposed pipeline is evaluated on a diverse dataset consisting of various real-world cityscape, on-campus and indoor scenarios. The experimental evaluation discusses the potential of these small models on edge devices, with particular emphasis on challenges, weaknesses, inherent model biases and the application of the gained information. Supplementary material is provided via the following repository: https://datahub.rz.rptu.de/hstr-csrl-public/publications/scene-interpretation-on-edge-devices/
comment: 15 pages, 6 figures, 1 table; accepted for AI-2025 Forty-fifth SGAI International Conference on Artificial Intelligence CAMBRIDGE, ENGLAND 16-18 DECEMBER 2025
♻ ☆ FICO: Finite-Horizon Closed-Loop Factorization for Unified Multi-Agent Path Finding
Multi-Agent Path Finding is a fundamental problem in robotics and AI, yet most existing formulations treat planning and execution separately and address variants of the problem in an ad hoc manner. This paper presents a system-level framework for MAPF that integrates planning and execution, generalizes across variants, and explicitly models uncertainties. At its core is the MAPF system, a formal model that casts MAPF as a control design problem encompassing classical and uncertainty-aware formulations. To solve it, we introduce Finite-Horizon Closed-Loop Factorization (FICO), a factorization-based algorithm inspired by receding-horizon control that exploits compositional structure for efficient closed-loop operation. FICO enables real-time responses -- commencing execution within milliseconds -- while scaling to thousands of agents and adapting seamlessly to execution-time uncertainties. Extensive case studies demonstrate that it reduces computation time by up to two orders of magnitude compared with open-loop baselines, while delivering significantly higher throughput under stochastic delays and agent arrivals. These results establish a principled foundation for analyzing and advancing MAPF through system-level modeling, factorization, and closed-loop design.
♻ ☆ Risk-Bounded Multi-Agent Visual Navigation via Iterative Risk Allocation
Safe navigation is essential for autonomous systems operating in hazardous environments, especially when multiple agents must coordinate using only high-dimensional visual observations. While recent approaches successfully combine Goal-Conditioned RL (GCRL) for graph construction with Conflict-Based Search (CBS) for planning, they typically rely on static edge pruning to enforce safety. This binary strategy is overly conservative, precluding feasible missions that require traversing high-risk regions, even when the aggregate risk is acceptable. To address this, we introduce a framework for Risk-Bounded Multi-Agent Path Finding (\problem{}), where agents share a user-specified global risk budget ($Δ$). Rather than permanently discarding edges, our framework dynamically distributes per-agent risk budgets ($δ_i$) during search via an Iterative Risk Allocation (IRA) layer that integrates with a standard CBS planner. We investigate two distribution strategies: a greedy surplus-deficit scheme for rapid feasibility repair, and a market-inspired mechanism that treats risk as a priced resource to guide improved allocation. This yields a tunable trade-off wherein agents exploit available risk to secure shorter, more efficient paths, but revert to longer, safer detours under tighter budgets. Experiments in complex visual environments show that, our dynamic allocation framework achieves higher success rates than baselines and effectively leverages the available safety budget to reduce travel time.
♻ ☆ Learning Agile Striker Skills for Humanoid Soccer Robots from Noisy Sensory Input
Learning fast and robust ball-kicking skills is a critical capability for humanoid soccer robots, yet it remains a challenging problem due to the need for rapid leg swings, postural stability on a single support foot, and robustness under noisy sensory input and external perturbations (e.g., opponents). This paper presents a reinforcement learning (RL)-based system that enables humanoid robots to execute robust continual ball-kicking with adaptability to different ball-goal configurations. The system extends a typical teacher-student training framework -- in which a "teacher" policy is trained with ground truth state information and the "student" learns to mimic it with noisy, imperfect sensing -- by including four training stages: (1) long-distance ball chasing (teacher); (2) directional kicking (teacher); (3) teacher policy distillation (student); and (4) student adaptation and refinement (student). Key design elements -- including tailored reward functions, realistic noise modeling, and online constrained RL for adaptation and refinement -- are critical for closing the sim-to-real gap and sustaining performance under perceptual uncertainty. Extensive evaluations in both simulation and on a real robot demonstrate strong kicking accuracy and goal-scoring success across diverse ball-goal configurations. Ablation studies further highlight the necessity of the constrained RL, noise modeling, and the adaptation stage. This work presents a system for learning robust continual humanoid ball-kicking under imperfect perception, establishing a benchmark task for visuomotor skill learning in humanoid whole-body control.
Robotics 64
☆ OSMO: Open-Source Tactile Glove for Human-to-Robot Skill Transfer
Human video demonstrations provide abundant training data for learning robot policies, but video alone cannot capture the rich contact signals critical for mastering manipulation. We introduce OSMO, an open-source wearable tactile glove designed for human-to-robot skill transfer. The glove features 12 three-axis tactile sensors across the fingertips and palm and is designed to be compatible with state-of-the-art hand-tracking methods for in-the-wild data collection. We demonstrate that a robot policy trained exclusively on human demonstrations collected with OSMO, without any real robot data, is capable of executing a challenging contact-rich manipulation task. By equipping both the human and the robot with the same glove, OSMO minimizes the visual and tactile embodiment gap, enabling the transfer of continuous shear and normal force feedback while avoiding the need for image inpainting or other vision-based force inference. On a real-world wiping task requiring sustained contact pressure, our tactile-aware policy achieves a 72% success rate, outperforming vision-only baselines by eliminating contact-related failure modes. We release complete hardware designs, firmware, and assembly instructions to support community adoption.
comment: Project website: https://jessicayin.github.io/osmo_tactile_glove/
☆ LiDAS: Lighting-driven Dynamic Active Sensing for Nighttime Perception
Nighttime environments pose significant challenges for camera-based perception, as existing methods passively rely on the scene lighting. We introduce Lighting-driven Dynamic Active Sensing (LiDAS), a closed-loop active illumination system that combines off-the-shelf visual perception models with high-definition headlights. Rather than uniformly brightening the scene, LiDAS dynamically predicts an optimal illumination field that maximizes downstream perception performance, i.e., decreasing light on empty areas to reallocate it on object regions. LiDAS enables zero-shot nighttime generalization of daytime-trained models through adaptive illumination control. Trained on synthetic data and deployed zero-shot in real-world closed-loop driving scenarios, LiDAS enables +18.7% mAP50 and +5.0% mIoU over standard low-beam at equal power. It maintains performances while reducing energy use by 40%. LiDAS complements domain-generalization methods, further strengthening robustness without retraining. By turning readily available headlights into active vision actuators, LiDAS offers a cost-effective solution to robust nighttime perception.
comment: Preprint. 12 pages, 9 figures. Project page: https://simondemoreau.github.io/LiDAS/
☆ Accelerated Rotation-Invariant Convolution for UAV Image Segmentation
Rotation invariance is essential for precise, object-level segmentation in UAV aerial imagery, where targets can have arbitrary orientations and exhibit fine-scale details. Conventional segmentation architectures like U-Net rely on convolution operators that are not rotation-invariant, leading to degraded segmentation accuracy across varying viewpoints. Rotation invariance can be achieved by expanding the filter bank across multiple orientations; however, this will significantly increase computational cost and memory traffic. In this paper, we introduce a GPU-optimized rotation-invariant convolution framework that eliminates the traditional data-lowering (im2col) step required for matrix-multiplication-based convolution. By exploiting structured data sharing among symmetrically rotated filters, our method achieves multi-orientation convolution with greatly reduced memory traffic and computational redundancy. We further generalize the approach to accelerate convolution with arbitrary (non-symmetric) rotation angles. Across extensive benchmarks, the proposed convolution achieves 20--55% faster training and 15--45% lower energy consumption than CUDNN, while maintaining accuracy comparable to state-of-the-art rotation-invariant methods. In the eight-orientation setting, our approach achieves up to 45% speedup and 41% energy savings on 256\(\times\)256 inputs, and 32% speedup and 23% lower energy usage on 1024\(\times\)1024 inputs. Integrated into a U-Net segmentation model, the framework yields up to 6% improvement in accuracy over the non-rotation-aware baseline. These results demonstrate that the proposed method provides an effective and highly efficient alternative to existing rotation-invariant CNN frameworks.
☆ IPPO Learns the Game, Not the Team: A Study on Generalization in Heterogeneous Agent Teams
Multi-Agent Reinforcement Learning (MARL) is commonly deployed in settings where agents are trained via self-play with homogeneous teammates, often using parameter sharing and a single policy architecture. This opens the question: to what extent do self-play PPO agents learn general coordination strategies grounded in the underlying game, compared to overfitting to their training partners' behaviors? This paper investigates the question using the Heterogeneous Multi-Agent Challenge (HeMAC) environment, which features distinct Observer and Drone agents with complementary capabilities. We introduce Rotating Policy Training (RPT), an approach that rotates heterogeneous teammate policies of different learning algorithms during training, to expose the agent to a broader range of partner strategies. When playing alongside a withheld teammate policy (DDQN), we find that RPT achieves similar performance to a standard self-play baseline, IPPO, where all agents were trained sharing a single PPO policy. This result indicates that in this heterogeneous multi-agent setting, the IPPO baseline generalizes to novel teammate algorithms despite not experiencing teammate diversity during training. This shows that a simple IPPO baseline may possess the level of generalization to novel teammates that a diverse training regimen was designed to achieve.
comment: 4 pages, 3 figures, appendix
☆ Heterogeneity in Multi-Robot Environmental Monitoring for Resolving Time-Conflicting Tasks
Multi-robot systems performing continuous tasks face a performance trade-off when interrupted by urgent, time-critical sub-tasks. We investigate this trade-off in a scenario where a team must balance area patrolling with locating an anomalous radio signal. To address this trade-off, we evaluate both behavioral heterogeneity through agent role specialization ("patrollers" and "searchers") and sensing heterogeneity (i.e., only the searchers can sense the radio signal). Through simulation, we identify the Pareto-optimal trade-offs under varying team compositions, with behaviorally heterogeneous teams demonstrating the most balanced trade-offs in the majority of cases. When sensing capability is restricted, heterogeneous teams with half of the sensing-capable agents perform comparably to homogeneous teams, providing cost-saving rationale for restricting sensor payload deployment. Our findings demonstrate that pre-deployment role and sensing specialization are powerful design considerations for multi-robot systems facing time-conflicting tasks, where varying the degree of behavioral heterogeneity can tune system performance toward either task.
comment: Accepted to SAC '26. To appear, DOI: https://doi.org/10.1145/3748522.3779970
☆ Data-Driven Dynamic Parameter Learning of manipulator robots
Bridging the sim-to-real gap remains a fundamental challenge in robotics, as accurate dynamic parameter estimation is essential for reliable model-based control, realistic simulation, and safe deployment of manipulators. Traditional analytical approaches often fall short when faced with complex robot structures and interactions. Data-driven methods offer a promising alternative, yet conventional neural networks such as recurrent models struggle to capture long-range dependencies critical for accurate estimation. In this study, we propose a Transformer-based approach for dynamic parameter estimation, supported by an automated pipeline that generates diverse robot models and enriched trajectory data using Jacobian-derived features. The dataset consists of 8,192 robots with varied inertial and frictional properties. Leveraging attention mechanisms, our model effectively captures both temporal and spatial dependencies. Experimental results highlight the influence of sequence length, sampling rate, and architecture, with the best configuration (sequence length 64, 64 Hz, four layers, 32 heads) achieving a validation R2 of 0.8633. Mass and inertia are estimated with near-perfect accuracy, Coulomb friction with moderate-to-high accuracy, while viscous friction and distal link center-of-mass remain more challenging. These results demonstrate that combining Transformers with automated dataset generation and kinematic enrichment enables scalable, accurate dynamic parameter estimation, contributing to improved sim-to-real transfer in robotic systems
comment: Accepted for publication at SII 2026. 6 pages, 7 figures. Code is available at: https://github.com/MohamedAlsiagy/dynamic_parameter_est
☆ A Multi-Robot Platform for Robotic Triage Combining Onboard Sensing and Foundation Models
This report presents a heterogeneous robotic system designed for remote primary triage in mass-casualty incidents (MCIs). The system employs a coordinated air-ground team of unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) to locate victims, assess their injuries, and prioritize medical assistance without risking the lives of first responders. The UAV identify and provide overhead views of casualties, while UGVs equipped with specialized sensors measure vital signs and detect and localize physical injuries. Unlike previous work that focused on exploration or limited medical evaluation, this system addresses the complete triage process: victim localization, vital sign measurement, injury severity classification, mental status assessment, and data consolidation for first responders. Developed as part of the DARPA Triage Challenge, this approach demonstrates how multi-robot systems can augment human capabilities in disaster response scenarios to maximize lives saved.
comment: Technical Report for the DARPA Triage Challenge PRONTO team
☆ Non Normalized Shared-Constraint Dynamic Games for Human-Robot Collaboration with Asymmetric Responsibility
This paper proposes a dynamic game formulation for cooperative human-robot navigation in shared workspaces with obstacles, where the human and robot jointly satisfy shared safety constraints while pursuing a common task. A key contribution is the introduction of a non-normalized equilibrium structure for the shared constraints. This structure allows the two agents to contribute different levels of effort towards enforcing safety requirements such as collision avoidance and inter-players spacing. We embed this non-normalized equilibrium into a receding-horizon optimal control scheme.
☆ Ergodic Trajectory Planning with Dynamic Sensor Footprints
This paper addresses the problem of trajectory planning for information gathering with a dynamic and resolution-varying sensor footprint. Ergodic planning offers a principled framework that balances exploration (visiting all areas) and exploitation (focusing on high-information regions) by planning trajectories such that the time spent in a region is proportional to the amount of information in that region. Existing ergodic planning often oversimplifies the sensing model by assuming a point sensor or a footprint with constant shape and resolution. In practice, the sensor footprint can drastically change over time as the robot moves, such as aerial robots equipped with downward-facing cameras, whose field of view depends on the orientation and altitude. To overcome this limitation, we propose a new metric that accounts for dynamic sensor footprints, analyze the theoretic local optimality conditions, and propose numerical trajectory optimization algorithms. Experimental results show that the proposed approach can simultaneously optimize both the trajectories and sensor footprints, with up to an order of magnitude better ergodicity than conventional methods. We also deploy our approach in a multi-drone system to ergodically cover an object in 3D space.
comment: 12 figures
☆ Sim2Swim: Zero-Shot Velocity Control for Agile AUV Maneuvering in 3 Minutes
Holonomic autonomous underwater vehicles (AUVs) have the hardware ability for agile maneuvering in both translational and rotational degrees of freedom (DOFs). However, due to challenges inherent to underwater vehicles, such as complex hydrostatics and hydrodynamics, parametric uncertainties, and frequent changes in dynamics due to payload changes, control is challenging. Performance typically relies on carefully tuned controllers targeting unique platform configurations, and a need for re-tuning for deployment under varying payloads and hydrodynamic conditions. As a consequence, agile maneuvering with simultaneous tracking of time-varying references in both translational and rotational DOFs is rarely utilized in practice. To the best of our knowledge, this paper presents the first general zero-shot sim2real deep reinforcement learning-based (DRL) velocity controller enabling path following and agile 6DOF maneuvering with a training duration of just 3 minutes. Sim2Swim, the proposed approach, inspired by state-of-the-art DRL-based position control, leverages domain randomization and massively parallelized training to converge to field-deployable control policies for AUVs of variable characteristics without post-processing or tuning. Sim2Swim is extensively validated in pool trials for a variety of configurations, showcasing robust control for highly agile motions.
comment: 7 pages, 4 figures
☆ A Sensor-Aware Phenomenological Framework for Lidar Degradation Simulation and SLAM Robustness Evaluation
Lidar-based SLAM systems are highly sensitive to adverse conditions such as occlusion, noise, and field-of-view (FoV) degradation, yet existing robustness evaluation methods either lack physical grounding or do not capture sensor-specific behavior. This paper presents a sensor-aware, phenomenological framework for simulating interpretable lidar degradations directly on real point clouds, enabling controlled and reproducible SLAM stress testing. Unlike image-derived corruption benchmarks (e.g., SemanticKITTI-C) or simulation-only approaches (e.g., lidarsim), the proposed system preserves per-point geometry, intensity, and temporal structure while applying structured dropout, FoV reduction, Gaussian noise, occlusion masking, sparsification, and motion distortion. The framework features autonomous topic and sensor detection, modular configuration with four severity tiers (light--extreme), and real-time performance (less than 20 ms per frame) compatible with ROS workflows. Experimental validation across three lidar architectures and five state-of-the-art SLAM systems reveals distinct robustness patterns shaped by sensor design and environmental context. The open-source implementation provides a practical foundation for benchmarking lidar-based SLAM under physically meaningful degradation scenarios.
☆ Multi-Task Bayesian Optimization for Tuning Decentralized Trajectory Generation in Multi-UAV Systems
This paper investigates the use of Multi-Task Bayesian Optimization for tuning decentralized trajectory generation algorithms in multi-drone systems. We treat each task as a trajectory generation scenario defined by a specific number of drone-to-drone interactions. To model relationships across scenarios, we employ Multi-Task Gaussian Processes, which capture shared structure across tasks and enable efficient information transfer during optimization. We compare two strategies: optimizing the average mission time across all tasks and optimizing each task individually. Through a comprehensive simulation campaign, we show that single-task optimization leads to progressively shorter mission times as swarm size grows, but requires significantly more optimization time than the average-task approach.
☆ Decoupled Design of Time-Varying Control Barrier Functions via Equivariances
This article presents a systematic method for designing time-varying Control Barrier Functions (CBF) composed of a time-invariant component and multiple time-dependent components, leveraging structural properties of the system dynamics. The method involves the construction of a specific class of time-invariant CBFs that encode the system's dynamic capabilities with respect to a given constraint, and augments them subsequently with appropriately designed time-dependent transformations. While transformations uniformly varying the time-invariant CBF can be applied to arbitrary systems, transformations exploiting structural properties in the dynamics - equivariances in particular - enable the handling of a broader and more expressive class of time-varying constraints. The article shows how to leverage such properties in the design of time-varying CBFs. The proposed method decouples the design of time variations from the computationally expensive construction of the underlying CBFs, thereby providing a computationally attractive method to the design of time-varying CBFs. The method accounts for input constraints and under-actuation, and requires only qualitative knowledge on the time-variation of the constraints making it suitable to the application in uncertain environments.
comment: 7 pages, 3 figures
☆ Mind to Hand: Purposeful Robotic Control via Embodied Reasoning
Humans act with context and intention, with reasoning playing a central role. While internet-scale data has enabled broad reasoning capabilities in AI systems, grounding these abilities in physical action remains a major challenge. We introduce Lumo-1, a generalist vision-language-action (VLA) model that unifies robot reasoning ("mind") with robot action ("hand"). Our approach builds upon the general multi-modal reasoning capabilities of pre-trained vision-language models (VLMs), progressively extending them to embodied reasoning and action prediction, and ultimately towards structured reasoning and reasoning-action alignment. This results in a three-stage pre-training pipeline: (1) Continued VLM pre-training on curated vision-language data to enhance embodied reasoning skills such as planning, spatial understanding, and trajectory prediction; (2) Co-training on cross-embodiment robot data alongside vision-language data; and (3) Action training with reasoning process on trajectories collected on Astribot S1, a bimanual mobile manipulator with human-like dexterity and agility. Finally, we integrate reinforcement learning to further refine reasoning-action consistency and close the loop between semantic inference and motor control. Extensive experiments demonstrate that Lumo-1 achieves significant performance improvements in embodied vision-language reasoning, a critical component for generalist robotic control. Real-world evaluations further show that Lumo-1 surpasses strong baselines across a wide range of challenging robotic tasks, with strong generalization to novel objects and environments, excelling particularly in long-horizon tasks and responding to human-natural instructions that require reasoning over strategy, concepts and space.
comment: 49 pages, 25 figures
☆ Disturbance-Free Surgical Video Generation from Multi-Camera Shadowless Lamps for Open Surgery
Video recordings of open surgeries are greatly required for education and research purposes. However, capturing unobstructed videos is challenging since surgeons frequently block the camera field of view. To avoid occlusion, the positions and angles of the camera must be frequently adjusted, which is highly labor-intensive. Prior work has addressed this issue by installing multiple cameras on a shadowless lamp and arranging them to fully surround the surgical area. This setup increases the chances of some cameras capturing an unobstructed view. However, manual image alignment is needed in post-processing since camera configurations change every time surgeons move the lamp for optimal lighting. This paper aims to fully automate this alignment task. The proposed method identifies frames in which the lighting system moves, realigns them, and selects the camera with the least occlusion to generate a video that consistently presents the surgical field from a fixed perspective. A user study involving surgeons demonstrated that videos generated by our method were superior to those produced by conventional methods in terms of the ease of confirming the surgical area and the comfort during video viewing. Additionally, our approach showed improvements in video quality over existing techniques. Furthermore, we implemented several synthesis options for the proposed view-synthesis method and conducted a user study to assess surgeons' preferences for each option.
☆ RVC-NMPC: Nonlinear Model Predictive Control with Reciprocal Velocity Constraints for Mutual Collision Avoidance in Agile UAV Flight
This paper presents an approach to mutual collision avoidance based on Nonlinear Model Predictive Control (NMPC) with time-dependent Reciprocal Velocity Constraints (RVCs). Unlike most existing methods, the proposed approach relies solely on observable information about other robots, eliminating the necessity of excessive communication use. The computationally efficient algorithm for computing RVCs, together with the direct integration of these constraints into NMPC problem formulation on a controller level, allows the whole pipeline to run at 100 Hz. This high processing rate, combined with modeled nonlinear dynamics of the controlled Uncrewed Aerial Vehicles (UAVs), is a key feature that facilitates the use of the proposed approach for an agile UAV flight. The proposed approach was evaluated through extensive simulations emulating real-world conditions in scenarios involving up to 10 UAVs and velocities of up to 25 m/s, and in real-world experiments with accelerations up to 30 m/s$^2$. Comparison with state of the art shows 31% improvement in terms of flight time reduction in challenging scenarios, while maintaining a collision-free navigation in all trials.
comment: 8 pages, 8 figures
☆ Bridging Scale Discrepancies in Robotic Control via Language-Based Action Representations
Recent end-to-end robotic manipulation research increasingly adopts architectures inspired by large language models to enable robust manipulation. However, a critical challenge arises from severe distribution shifts between robotic action data, primarily due to substantial numerical variations in action commands across diverse robotic platforms and tasks, hindering the effective transfer of pretrained knowledge. To address this limitation, we propose a semantically grounded linguistic representation to normalize actions for efficient pretraining. Unlike conventional discretized action representations that are sensitive to numerical scales, the motion representation specifically disregards numeric scale effects, emphasizing directionality instead. This abstraction mitigates distribution shifts, yielding a more generalizable pretraining representation. Moreover, using the motion representation narrows the feature distance between action tokens and standard vocabulary tokens, mitigating modality gaps. Multi-task experiments on two benchmarks demonstrate that the proposed method significantly improves generalization performance and transferability in robotic manipulation tasks.
☆ vEDGAR -- Can CARLA Do HiL?
Simulation offers advantages throughout the development process of automated driving functions, both in research and product development. Common open-source simulators like CARLA are extensively used in training, evaluation, and software-in-the-loop testing of new automated driving algorithms. However, the CARLA simulator lacks an evaluation where research and automated driving vehicles are simulated with their entire sensor and actuation stack in real time. The goal of this work is therefore to create a simulation framework for testing the automation software on its dedicated hardware and identifying its limits. Achieving this goal would greatly benefit the open-source development workflow of automated driving functions, designating CARLA as a consistent evaluation tool along the entire development process. To achieve this goal, in a first step, requirements are derived, and a simulation architecture is specified and implemented. Based on the formulated requirements, the proposed vEDGAR software is evaluated, resulting in a final conclusion on the applicability of CARLA for HiL testing of automated vehicles. The tool is available open source: Modified CARLA fork: https://github.com/TUMFTM/carla, vEDGAR Framework: https://github.com/TUMFTM/vEDGAR
☆ SensHRPS: Sensing Comfortable Human-Robot Proxemics and Personal Space With Eye-Tracking
Social robots must adjust to human proxemic norms to ensure user comfort and engagement. While prior research demonstrates that eye-tracking features reliably estimate comfort in human-human interactions, their applicability to interactions with humanoid robots remains unexplored. In this study, we investigate user comfort with the robot "Ameca" across four experimentally controlled distances (0.5 m to 2.0 m) using mobile eye-tracking and subjective reporting (N=19). We evaluate multiple machine learning and deep learning models to estimate comfort based on gaze features. Contrary to previous human-human studies where Transformer models excelled, a Decision Tree classifier achieved the highest performance (F1-score = 0.73), with minimum pupil diameter identified as the most critical predictor. These findings suggest that physiological comfort thresholds in human-robot interaction differ from human-human dynamics and can be effectively modeled using interpretable logic.
☆ Prospect Theory in Physical Human-Robot Interaction: A Pilot Study of Probability Perception
Understanding how humans respond to uncertainty is critical for designing safe and effective physical human-robot interaction (pHRI), as physically working with robots introduces multiple sources of uncertainty, including trust, comfort, and perceived safety. Conventional pHRI control frameworks typically build on optimal control theory, which assumes that human actions minimize a cost function; however, human behavior under uncertainty often departs from such optimal patterns. To address this gap, additional understanding of human behavior under uncertainty is needed. This pilot study implemented a physically coupled target-reaching task in which the robot delivered assistance or disturbances with systematically varied probabilities (10\% to 90\%). Analysis of participants' force inputs and decision-making strategies revealed two distinct behavioral clusters: a "trade-off" group that modulated their physical responses according to disturbance likelihood, and an "always-compensate" group characterized by strong risk aversion irrespective of probability. These findings provide empirical evidence that human decision-making in pHRI is highly individualized and that the perception of probability can differ to its true value. Accordingly, the study highlights the need for more interpretable behavioral models, such as cumulative prospect theory (CPT), to more accurately capture these behaviors and inform the design of future adaptive robot controllers.
comment: 9 pages, 6 figures
☆ A Multi-Agent LLM Framework for Design Space Exploration in Autonomous Driving Systems
Designing autonomous driving systems requires efficient exploration of large hardware/software configuration spaces under diverse environmental conditions, e.g., with varying traffic, weather, and road layouts. Traditional design space exploration (DSE) approaches struggle with multi-modal execution outputs and complex performance trade-offs, and often require human involvement to assess correctness based on execution outputs. This paper presents a multi-agent, large language model (LLM)-based DSE framework, which integrates multi-modal reasoning with 3D simulation and profiling tools to automate the interpretation of execution outputs and guide the exploration of system designs. Specialized LLM agents are leveraged to handle user input interpretation, design point generation, execution orchestration, and analysis of both visual and textual execution outputs, which enables identification of potential bottlenecks without human intervention. A prototype implementation is developed and evaluated on a robotaxi case study (an SAE Level 4 autonomous driving application). Compared with a genetic algorithm baseline, the proposed framework identifies more Pareto-optimal, cost-efficient solutions with reduced navigation time under the same exploration budget. Experimental results also demonstrate the efficiency of the adoption of the LLM-based approach for DSE. We believe that this framework paves the way to the design automation of autonomous driving systems.
☆ Using reinforcement learning to probe the role of feedback in skill acquisition
Many high-performance human activities are executed with little or no external feedback: think of a figure skater landing a triple jump, a pitcher throwing a curveball for a strike, or a barista pouring latte art. To study the process of skill acquisition under fully controlled conditions, we bypass human subjects. Instead, we directly interface a generalist reinforcement learning agent with a spinning cylinder in a tabletop circulating water channel to maximize or minimize drag. This setup has several desirable properties. First, it is a physical system, with the rich interactions and complex dynamics that only the physical world has: the flow is highly chaotic and extremely difficult, if not impossible, to model or simulate accurately. Second, the objective -- drag minimization or maximization -- is easy to state and can be captured directly in the reward, yet good strategies are not obvious beforehand. Third, decades-old experimental studies provide recipes for simple, high-performance open-loop policies. Finally, the setup is inexpensive and far easier to reproduce than human studies. In our experiments we find that high-dimensional flow feedback lets the agent discover high-performance drag-control strategies with only minutes of real-world interaction. When we later replay the same action sequences without any feedback, we obtain almost identical performance. This shows that feedback, and in particular flow feedback, is not needed to execute the learned policy. Surprisingly, without flow feedback during training the agent fails to discover any well-performing policy in drag maximization, but still succeeds in drag minimization, albeit more slowly and less reliably. Our studies show that learning a high-performance skill can require richer information than executing it, and learning conditions can be kind or wicked depending solely on the goal, not on dynamics or policy complexity.
comment: Website: https://antonioterpin.com/fluids-control
☆ SDT-6D: Fully Sparse Depth-Transformer for Staged End-to-End 6D Pose Estimation in Industrial Multi-View Bin Picking WACV 2026
Accurately recovering 6D poses in densely packed industrial bin-picking environments remain a serious challenge, owing to occlusions, reflections, and textureless parts. We introduce a holistic depth-only 6D pose estimation approach that fuses multi-view depth maps into either a fine-grained 3D point cloud in its vanilla version, or a sparse Truncated Signed Distance Field (TSDF). At the core of our framework lies a staged heatmap mechanism that yields scene-adaptive attention priors across different resolutions, steering computation toward foreground regions, thus keeping memory requirements at high resolutions feasible. Along, we propose a density-aware sparse transformer block that dynamically attends to (self-) occlusions and the non-uniform distribution of 3D data. While sparse 3D approaches has proven effective for long-range perception, its potential in close-range robotic applications remains underexplored. Our framework operates fully sparse, enabling high-resolution volumetric representations to capture fine geometric details crucial for accurate pose estimation in clutter. Our method processes the entire scene integrally, predicting the 6D pose via a novel per-voxel voting strategy, allowing simultaneous pose predictions for an arbitrary number of target objects. We validate our method on the recently published IPD and MV-YCB multi-view datasets, demonstrating competitive performance in heavily cluttered industrial and household bin picking scenarios.
comment: Accepted to WACV 2026. Preprint version
☆ Prismatic World Model: Learning Compositional Dynamics for Planning in Hybrid Systems
Model-based planning in robotic domains is fundamentally challenged by the hybrid nature of physical dynamics, where continuous motion is punctuated by discrete events such as contacts and impacts. Conventional latent world models typically employ monolithic neural networks that enforce global continuity, inevitably over-smoothing the distinct dynamic modes (e.g., sticking vs. sliding, flight vs. stance). For a planner, this smoothing results in catastrophic compounding errors during long-horizon lookaheads, rendering the search process unreliable at physical boundaries. To address this, we introduce the Prismatic World Model (PRISM-WM), a structured architecture designed to decompose complex hybrid dynamics into composable primitives. PRISM-WM leverages a context-aware Mixture-of-Experts (MoE) framework where a gating mechanism implicitly identifies the current physical mode, and specialized experts predict the associated transition dynamics. We further introduce a latent orthogonalization objective to ensure expert diversity, effectively preventing mode collapse. By accurately modeling the sharp mode transitions in system dynamics, PRISM-WM significantly reduces rollout drift. Extensive experiments on challenging continuous control benchmarks, including high-dimensional humanoids and diverse multi-task settings, demonstrate that PRISM-WM provides a superior high-fidelity substrate for trajectory optimization algorithms (e.g., TD-MPC), proving its potential as a powerful foundational model for next-generation model-based agents.
☆ Learning Robot Manipulation from Audio World Models
World models have demonstrated impressive performance on robotic learning tasks. Many such tasks inherently demand multimodal reasoning; for example, filling a bottle with water will lead to visual information alone being ambiguous or incomplete, thereby requiring reasoning over the temporal evolution of audio, accounting for its underlying physical properties and pitch patterns. In this paper, we propose a generative latent flow matching model to anticipate future audio observations, enabling the system to reason about long-term consequences when integrated into a robot policy. We demonstrate the superior capabilities of our system through two manipulation tasks that require perceiving in-the-wild audio or music signals, compared to methods without future lookahead. We further emphasize that successful robot action learning for these tasks relies not merely on multi-modal input, but critically on the accurate prediction of future audio states that embody intrinsic rhythmic patterns.
☆ Robust Finetuning of Vision-Language-Action Robot Policies via Parameter Merging
Generalist robot policies, trained on large and diverse datasets, have demonstrated the ability to generalize across a wide spectrum of behaviors, enabling a single policy to act in varied real-world environments. However, they still fall short on new tasks not covered in the training data. When finetuned on limited demonstrations of a new task, these policies often overfit to the specific demonstrations--not only losing their prior abilities to solve a wide variety of generalist tasks but also failing to generalize within the new task itself. In this work, we aim to develop a method that preserves the generalization capabilities of the generalist policy during finetuning, allowing a single policy to robustly incorporate a new skill into its repertoire. Our goal is a single policy that both learns to generalize to variations of the new task and retains the broad competencies gained from pretraining. We show that this can be achieved through a simple yet effective strategy: interpolating the weights of a finetuned model with that of the pretrained model. We show, across extensive simulated and real-world experiments, that such model merging produces a single model that inherits the generalist abilities of the base model and learns to solve the new task robustly, outperforming both the pretrained and finetuned model on out-of-distribution variations of the new task. Moreover, we show that model merging enables continual acquisition of new skills in a lifelong learning setting, without sacrificing previously learned generalist abilities.
☆ Model-Based Diffusion Sampling for Predictive Control in Offline Decision Making
Offline decision-making requires synthesizing reliable behaviors from fixed datasets without further interaction, yet existing generative approaches often yield trajectories that are dynamically infeasible. We propose Model Predictive Diffuser (MPDiffuser), a compositional model-based diffusion framework consisting of: (i) a planner that generates diverse, task-aligned trajectories; (ii) a dynamics model that enforces consistency with the underlying system dynamics; and (iii) a ranker module that selects behaviors aligned with the task objectives. MPDiffuser employs an alternating diffusion sampling scheme, where planner and dynamics updates are interleaved to progressively refine trajectories for both task alignment and feasibility during the sampling process. We also provide a theoretical rationale for this procedure, showing how it balances fidelity to data priors with dynamics consistency. Empirically, the compositional design improves sample efficiency, as it leverages even low-quality data for dynamics learning and adapts seamlessly to novel dynamics. We evaluate MPDiffuser on both unconstrained (D4RL) and constrained (DSRL) offline decision-making benchmarks, demonstrating consistent gains over existing approaches. Furthermore, we present a preliminary study extending MPDiffuser to vision-based control tasks, showing its potential to scale to high-dimensional sensory inputs. Finally, we deploy our method on a real quadrupedal robot, showcasing its practicality for real-world control.
☆ Zero-Splat TeleAssist: A Zero-Shot Pose Estimation Framework for Semantic Teleoperation ICRA 2025
We introduce Zero-Splat TeleAssist, a zero-shot sensor-fusion pipeline that transforms commodity CCTV streams into a shared, 6-DoF world model for multilateral teleoperation. By integrating vision-language segmentation, monocular depth, weighted-PCA pose extraction, and 3D Gaussian Splatting (3DGS), TeleAssist provides every operator with real-time global positions and orientations of multiple robots without fiducials or depth sensors in an interaction-centric teleoperation setup.
comment: Published and Presented at 3rd Workshop on Human-Centric Multilateral Teleoperation in ICRA 2025
☆ RLCNet: An end-to-end deep learning framework for simultaneous online calibration of LiDAR, RADAR, and Camera
Accurate extrinsic calibration of LiDAR, RADAR, and camera sensors is essential for reliable perception in autonomous vehicles. Still, it remains challenging due to factors such as mechanical vibrations and cumulative sensor drift in dynamic environments. This paper presents RLCNet, a novel end-to-end trainable deep learning framework for the simultaneous online calibration of these multimodal sensors. Validated on real-world datasets, RLCNet is designed for practical deployment and demonstrates robust performance under diverse conditions. To support real-time operation, an online calibration framework is introduced that incorporates a weighted moving average and outlier rejection, enabling dynamic adjustment of calibration parameters with reduced prediction noise and improved resilience to drift. An ablation study highlights the significance of architectural choices, while comparisons with existing methods demonstrate the superior accuracy and robustness of the proposed approach.
☆ Learning Spatiotemporal Tubes for Temporal Reach-Avoid-Stay Tasks using Physics-Informed Neural Networks
This paper presents a Spatiotemporal Tube (STT)-based control framework for general control-affine MIMO nonlinear pure-feedback systems with unknown dynamics to satisfy prescribed time reach-avoid-stay tasks under external disturbances. The STT is defined as a time-varying ball, whose center and radius are jointly approximated by a Physics-Informed Neural Network (PINN). The constraints governing the STT are first formulated as loss functions of the PINN, and a training algorithm is proposed to minimize the overall violation. The PINN being trained on certain collocation points, we propose a Lipschitz-based validity condition to formally verify that the learned PINN satisfies the conditions over the continuous time horizon. Building on the learned STT representation, an approximation-free closed-form controller is defined to guarantee satisfaction of the T-RAS specification. Finally, the effectiveness and scalability of the framework are validated through two case studies involving a mobile robot and an aerial vehicle navigating through cluttered environments.
☆ Semantic-Metric Bayesian Risk Fields: Learning Robot Safety from Human Videos with a VLM Prior
Humans interpret safety not as a binary signal but as a continuous, context- and spatially-dependent notion of risk. While risk is subjective, humans form rational mental models that guide action selection in dynamic environments. This work proposes a framework for extracting implicit human risk models by introducing a novel, semantically-conditioned and spatially-varying parametrization of risk, supervised directly from safe human demonstration videos and VLM common sense. Notably, we define risk through a Bayesian formulation. The prior is furnished by a pretrained vision-language model. In order to encourage the risk estimate to be more human aligned, a likelihood function modulates the prior to produce a relative metric of risk. Specifically, the likelihood is a learned ViT that maps pretrained features, to pixel-aligned risk values. Our pipeline ingests RGB images and a query object string, producing pixel-dense risk images. These images that can then be used as value-predictors in robot planning tasks or be projected into 3D for use in conventional trajectory optimization to produce human-like motion. This learned mapping enables generalization to novel objects and contexts, and has the potential to scale to much larger training datasets. In particular, the Bayesian framework that is introduced enables fast adaptation of our model to additional observations or common sense rules. We demonstrate that our proposed framework produces contextual risk that aligns with human preferences. Additionally, we illustrate several downstream applications of the model; as a value learner for visuomotor planners or in conjunction with a classical trajectory optimization algorithm. Our results suggest that our framework is a significant step toward enabling autonomous systems to internalize human-like risk. Code and results can be found at https://riskbayesian.github.io/bayesian_risk/.
☆ Geometry-Aware Sparse Depth Sampling for High-Fidelity RGB-D Depth Completion in Robotic Systems
Accurate three-dimensional perception is essential for modern industrial robotic systems that perform manipulation, inspection, and navigation tasks. RGB-D and stereo vision sensors are widely used for this purpose, but the depth maps they produce are often noisy, incomplete, or biased due to sensor limitations and environmental conditions. Depth completion methods aim to generate dense, reliable depth maps from RGB images and sparse depth input. However, a key limitation in current depth completion pipelines is the unrealistic generation of sparse depth: sparse pixels are typically selected uniformly at random from dense ground-truth depth, ignoring the fact that real sensors exhibit geometry-dependent and spatially nonuniform reliability. In this work, we propose a normal-guided sparse depth sampling strategy that leverages PCA-based surface normal estimation on the RGB-D point cloud to compute a per-pixel depth reliability measure. The sparse depth samples are then drawn according to this reliability distribution. We integrate this sampling method with the Marigold-DC diffusion-based depth completion model and evaluate it on NYU Depth v2 using the standard metrics. Experiments show that our geometry-aware sparse depth improves accuracy, reduces artifacts near edges and discontinuities, and produces more realistic training conditions that better reflect real sensor behavior.
☆ High-Performance Dual-Arm Task and Motion Planning for Tabletop Rearrangement ICRA 2026
We propose Synchronous Dual-Arm Rearrangement Planner (SDAR), a task and motion planning (TAMP) framework for tabletop rearrangement, where two robot arms equipped with 2-finger grippers must work together in close proximity to rearrange objects whose start and goal configurations are strongly entangled. To tackle such challenges, SDAR tightly knit together its dependency-driven task planner (SDAR-T) and synchronous dual-arm motion planner (SDAR-M), to intelligently sift through a large number of possible task and motion plans. Specifically, SDAR-T applies a simple yet effective strategy to decompose the global object dependency graph induced by the rearrangement task, to produce more optimal dual-arm task plans than solutions derived from optimal task plans for a single arm. Leveraging state-of-the-art GPU SIMD-based motion planning tools, SDAR-M employs a layered motion planning strategy to sift through many task plans for the best synchronous dual-arm motion plan while ensuring high levels of success rate. Comprehensive evaluation demonstrates that SDAR delivers a 100% success rate in solving complex, non-monotone, long-horizon tabletop rearrangement tasks with solution quality far exceeding the previous state-of-the-art. Experiments on two UR-5e arms further confirm SDAR directly and reliably transfers to robot hardware.
comment: ICRA 2026 Submission
☆ Embodied Tree of Thoughts: Deliberate Manipulation Planning with Embodied World Model
World models have emerged as a pivotal component in robot manipulation planning, enabling agents to predict future environmental states and reason about the consequences of actions before execution. While video-generation models are increasingly adopted, they often lack rigorous physical grounding, leading to hallucinations and a failure to maintain consistency in long-horizon physical constraints. To address these limitations, we propose Embodied Tree of Thoughts (EToT), a novel Real2Sim2Real planning framework that leverages a physics-based interactive digital twin as an embodied world model. EToT formulates manipulation planning as a tree search expanded through two synergistic mechanisms: (1) Priori Branching, which generates diverse candidate execution paths based on semantic and spatial analysis; and (2) Reflective Branching, which utilizes VLMs to diagnose execution failures within the simulator and iteratively refine the planning tree with corrective actions. By grounding high-level reasoning in a physics simulator, our framework ensures that generated plans adhere to rigid-body dynamics and collision constraints. We validate EToT on a suite of short- and long-horizon manipulation tasks, where it consistently outperforms baselines by effectively predicting physical dynamics and adapting to potential failures. Website at https://embodied-tree-of-thoughts.github.io .
comment: Website at https://embodied-tree-of-thoughts.github.io
☆ Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-and-Language Navigation
While recent large vision-language models (VLMs) have improved generalization in vision-language navigation (VLN), existing methods typically rely on end-to-end pipelines that map vision-language inputs directly to short-horizon discrete actions. Such designs often produce fragmented motions, incur high latency, and struggle with real-world challenges like dynamic obstacle avoidance. We propose DualVLN, the first dual-system VLN foundation model that synergistically integrates high-level reasoning with low-level action execution. System 2, a VLM-based global planner, "grounds slowly" by predicting mid-term waypoint goals via image-grounded reasoning. System 1, a lightweight, multi-modal conditioning Diffusion Transformer policy, "moves fast" by leveraging both explicit pixel goals and latent features from System 2 to generate smooth and accurate trajectories. The dual-system design enables robust real-time control and adaptive local decision-making in complex, dynamic environments. By decoupling training, the VLM retains its generalization, while System 1 achieves interpretable and effective local navigation. DualVLN outperforms prior methods across all VLN benchmarks and real-world experiments demonstrate robust long-horizon planning and real-time adaptability in dynamic environments.
☆ RAVES-Calib: Robust, Accurate and Versatile Extrinsic Self Calibration Using Optimal Geometric Features
In this paper, we present a user-friendly LiDAR-camera calibration toolkit that is compatible with various LiDAR and camera sensors and requires only a single pair of laser points and a camera image in targetless environments. Our approach eliminates the need for an initial transform and remains robust even with large positional and rotational LiDAR-camera extrinsic parameters. We employ the Gluestick pipeline to establish 2D-3D point and line feature correspondences for a robust and automatic initial guess. To enhance accuracy, we quantitatively analyze the impact of feature distribution on calibration results and adaptively weight the cost of each feature based on these metrics. As a result, extrinsic parameters are optimized by filtering out the adverse effects of inferior features. We validated our method through extensive experiments across various LiDAR-camera sensors in both indoor and outdoor settings. The results demonstrate that our method provides superior robustness and accuracy compared to SOTA techniques. Our code is open-sourced on GitHub to benefit the community.
☆ Chat with UAV -- Human-UAV Interaction Based on Large Language Models
The future of UAV interaction systems is evolving from engineer-driven to user-driven, aiming to replace traditional predefined Human-UAV Interaction designs. This shift focuses on enabling more personalized task planning and design, thereby achieving a higher quality of interaction experience and greater flexibility, which can be used in many fileds, such as agriculture, aerial photography, logistics, and environmental monitoring. However, due to the lack of a common language between users and the UAVs, such interactions are often difficult to be achieved. The developments of Large Language Models possess the ability to understand nature languages and Robots' (UAVs') behaviors, marking the possibility of personalized Human-UAV Interaction. Recently, some HUI frameworks based on LLMs have been proposed, but they commonly suffer from difficulties in mixed task planning and execution, leading to low adaptability in complex scenarios. In this paper, we propose a novel dual-agent HUI framework. This framework constructs two independent LLM agents (a task planning agent, and an execution agent) and applies different Prompt Engineering to separately handle the understanding, planning, and execution of tasks. To verify the effectiveness and performance of the framework, we have built a task database covering four typical application scenarios of UAVs and quantified the performance of the HUI framework using three independent metrics. Meanwhile different LLM models are selected to control the UAVs with compared performance. Our user study experimental results demonstrate that the framework improves the smoothness of HUI and the flexibility of task execution in the tasks scenario we set up, effectively meeting users' personalized needs.
☆ Semantic Trajectory Generation for Goal-Oriented Spacecraft Rendezvous SC
Reliable real-time trajectory generation is essential for future autonomous spacecraft. While recent progress in nonconvex guidance and control is paving the way for onboard autonomous trajectory optimization, these methods still rely on extensive expert input (e.g., waypoints, constraints, mission timelines, etc.), which limits the operational scalability in real rendezvous missions.This paper introduces SAGES (Semantic Autonomous Guidance Engine for Space), a trajectory-generation framework that translates natural-language commands into spacecraft trajectories that reflect high-level intent while respecting nonconvex constraints. Experiments in two settings -- fault-tolerant proximity operations with continuous-time constraint enforcement and a free-flying robotic platform -- demonstrate that SAGES reliably produces trajectories aligned with human commands, achieving over 90\% semantic-behavioral consistency across diverse behavior modes. Ultimately, this work marks an initial step toward language-conditioned, constraint-aware spacecraft trajectory generation, enabling operators to interactively guide both safety and behavior through intuitive natural-language commands with reduced expert burden.
comment: 28 pages, 12 figures. Submitted to AIAA SCITECH 2026
☆ Cognitive Trust in HRI: "Pay Attention to Me and I'll Trust You Even if You are Wrong"
Cognitive trust and the belief that a robot is capable of accurately performing tasks, are recognized as central factors in fostering high-quality human-robot interactions. It is well established that performance factors such as the robot's competence and its reliability shape cognitive trust. Recent studies suggest that affective factors, such as robotic attentiveness, also play a role in building cognitive trust. This work explores the interplay between these two factors that shape cognitive trust. Specifically, we evaluated whether different combinations of robotic competence and attentiveness introduce a compensatory mechanism, where one factor compensates for the lack of the other. In the experiment, participants performed a search task with a robotic dog in a 2x2 experimental design that included two factors: competence (high or low) and attentiveness (high or low). The results revealed that high attentiveness can compensate for low competence. Participants who collaborated with a highly attentive robot that performed poorly reported trust levels comparable to those working with a highly competent robot. When the robot did not demonstrate attentiveness, low competence resulted in a substantial decrease in cognitive trust. The findings indicate that building cognitive trust in human-robot interaction may be more complex than previously believed, involving emotional processes that are typically overlooked. We highlight an affective compensatory mechanism that adds a layer to consider alongside traditional competence-based models of cognitive trust.
comment: Confrence paper
☆ Masked Generative Policy for Robotic Control
We present Masked Generative Policy (MGP), a novel framework for visuomotor imitation learning. We represent actions as discrete tokens, and train a conditional masked transformer that generates tokens in parallel and then rapidly refines only low-confidence tokens. We further propose two new sampling paradigms: MGP-Short, which performs parallel masked generation with score-based refinement for Markovian tasks, and MGP-Long, which predicts full trajectories in a single pass and dynamically refines low-confidence action tokens based on new observations. With globally coherent prediction and robust adaptive execution capabilities, MGP-Long enables reliable control on complex and non-Markovian tasks that prior methods struggle with. Extensive evaluations on 150 robotic manipulation tasks spanning the Meta-World and LIBERO benchmarks show that MGP achieves both rapid inference and superior success rates compared to state-of-the-art diffusion and autoregressive policies. Specifically, MGP increases the average success rate by 9% across 150 tasks while cutting per-sequence inference time by up to 35x. It further improves the average success rate by 60% in dynamic and missing-observation environments, and solves two non-Markovian scenarios where other state-of-the-art methods fail.
☆ Characterizing Human Feedback-Based Control in Naturalistic Driving Interactions via Gaussian Process Regression with Linear Feedback
Understanding driver interactions is critical to designing autonomous vehicles to interoperate safely with human-driven cars. We consider the impact of these interactions on the policies drivers employ when navigating unsigned intersections in a driving simulator. The simulator allows the collection of naturalistic decision-making and behavior data in a controlled environment. Using these data, we model the human driver responses as state-based feedback controllers learned via Gaussian Process regression methods. We compute the feedback gain of the controller using a weighted combination of linear and nonlinear priors. We then analyze how the individual gains are reflected in driver behavior. We also assess differences in these controllers across populations of drivers. Our work in data-driven analyses of how drivers determine their policies can facilitate future work in the design of socially responsive autonomy for vehicles.
☆ Inferring Operator Emotions from a Motion-Controlled Robotic Arm
A remote robot operator's affective state can significantly impact the resulting robot's motions leading to unexpected consequences, even when the user follows protocol and performs permitted tasks. The recognition of a user operator's affective states in remote robot control scenarios is, however, underexplored. Current emotion recognition methods rely on reading the user's vital signs or body language, but the devices and user participation these measures require would add limitations to remote robot control. We demonstrate that the functional movements of a remote-controlled robotic avatar, which was not designed for emotional expression, can be used to infer the emotional state of the human operator via a machine-learning system. Specifically, our system achieved 83.3$\%$ accuracy in recognizing the user's emotional state expressed by robot movements, as a result of their hand motions. We discuss the implications of this system on prominent current and future remote robot operation and affective robotic contexts.
☆ Adaptive Thresholding for Visual Place Recognition using Negative Gaussian Mixture Statistics
Visual place recognition (VPR) is an important component technology for camera-based mapping and navigation applications. This is a challenging problem because images of the same place may appear quite different for reasons including seasonal changes, weather illumination, structural changes to the environment, as well as transient pedestrian or vehicle traffic. Papers focusing on generating image descriptors for VPR report their results using metrics such as recall@K and ROC curves. However, for a robot implementation, determining which matches are sufficiently good is often reduced to a manually set threshold. And it is difficult to manually select a threshold that will work for a variety of visual scenarios. This paper addresses the problem of automatically selecting a threshold for VPR by looking at the 'negative' Gaussian mixture statistics for a place - image statistics indicating not this place. We show that this approach can be used to select thresholds that work well for a variety of image databases and image descriptors.
comment: Accepted and presented at IEEE RoboticCC 2025. 4 pages short paper
☆ ShelfAware: Real-Time Visual-Inertial Semantic Localization in Quasi-Static Environments with Low-Cost Sensors
Many indoor workspaces are quasi-static: global layout is stable but local semantics change continually, producing repetitive geometry, dynamic clutter, and perceptual noise that defeat vision-based localization. We present ShelfAware, a semantic particle filter for robust global localization that treats scene semantics as statistical evidence over object categories rather than fixed landmarks. ShelfAware fuses a depth likelihood with a category-centric semantic similarity and uses a precomputed bank of semantic viewpoints to perform inverse semantic proposals inside MCL, yielding fast, targeted hypothesis generation on low-cost, vision-only hardware. Across 100 global-localization trials spanning four conditions (cart-mounted, wearable, dynamic obstacles, and sparse semantics) in a semantically dense, retail environment, ShelfAware achieves a 96% success rate (vs. 22% MCL and 10% AMCL) with a mean time-to-convergence of 1.91s, attains the lowest translational RMSE in all conditions, and maintains stable tracking in 80% of tested sequences, all while running in real time on a consumer laptop-class platform. By modeling semantics distributionally at the category level and leveraging inverse proposals, ShelfAware resolves geometric aliasing and semantic drift common to quasi-static domains. Because the method requires only vision sensors and VIO, it integrates as an infrastructure-free building block for mobile robots in warehouses, labs, and retail settings; as a representative application, it also supports the creation of assistive devices providing start-anytime, shared-control assistive navigation for people with visual impairments.
comment: 8 pages
♻ ☆ Botany Meets Robotics in Alpine Scree Monitoring
According to the European Union's Habitat Directive, habitat monitoring plays a critical role in response to the escalating problems posed by biodiversity loss and environmental degradation. Scree habitats, hosting unique and often endangered species, face severe threats from climate change due to their high-altitude nature. Traditionally, their monitoring has required highly skilled scientists to conduct extensive fieldwork in remote, potentially hazardous locations, making the process resource-intensive and time-consuming. This paper presents a novel approach for scree habitat monitoring using a legged robot to assist botanists in data collection and species identification. Specifically, we deployed the ANYmal C robot in the Italian Alpine bio-region in two field campaigns spanning two years and leveraged deep learning to detect and classify key plant species of interest. Our results demonstrate that agile legged robots can navigate challenging terrains and increase the frequency and efficiency of scree monitoring. When paired with traditional phytosociological surveys performed by botanists, this robotics-assisted protocol not only streamlines field operations but also enhances data acquisition, storage, and usage. The outcomes of this research contribute to the evolving landscape of robotics in environmental science, paving the way for a more comprehensive and sustainable approach to habitat monitoring and preservation.
comment: 19 pages, 13 figures
♻ ☆ Obstacle Avoidance of UAV in Dynamic Environments Using Direction and Velocity-Adaptive Artificial Potential Field
The conventional Artificial Potential Field (APF) is fundamentally limited by the local minima issue and its inability to account for the kinematics of moving obstacles. This paper addresses the critical challenge of autonomous collision avoidance for Unmanned Aerial Vehicles (UAVs) operating in dynamic and cluttered airspace by proposing a novel Direction and Relative Velocity Weighted Artificial Potential Field (APF). In this approach, a bounded weighting function, $ω(θ,v_{e})$, is introduced to dynamically scale the repulsive potential based on the direction and velocity of the obstacle relative to the UAV. This robust APF formulation is integrated within a Model Predictive Control (MPC) framework to generate collision-free trajectories while adhering to kinematic constraints. Simulation results demonstrate that the proposed method effectively resolves local minima and significantly enhances safety by enabling smooth, predictive avoidance maneuvers. The system ensures superior path integrity and reliable performance, confirming its viability for autonomous navigation in complex environments.
♻ ☆ Towards Task-Oriented Flying: Framework, Infrastructure, and Principles
Deploying robot learning methods to aerial robots in unstructured environments remains both challenging and promising. While recent advances in deep reinforcement learning (DRL) have enabled end-to-end flight control, the field still lacks systematic design guidelines and a unified infrastructure to support reproducible training and real-world deployment. We present a task-oriented framework for end-to-end DRL in quadrotors that integrates design principles for complex task specification and reveals the interdependencies among simulated task definition, training design principles, and physical deployment. Our framework involves software infrastructure, hardware platforms, and open-source firmware to support a full-stack learning infrastructure and workflow. Extensive empirical results demonstrate robust flight and sim-to-real generalization under real-world disturbances. By reducing the entry barrier for deploying learning-based controllers on aerial robots, our work lays a practical foundation for advancing autonomous flight in dynamic and unstructured environments.
♻ ☆ DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving
Most end-to-end autonomous driving methods rely on imitation learning from single expert demonstrations, often leading to conservative and homogeneous behaviors that limit generalization in complex real-world scenarios. In this work, we propose DIVER, an end-to-end driving framework that integrates reinforcement learning with diffusion-based generation to produce diverse and feasible trajectories. At the core of DIVER lies a reinforced diffusion-based generation mechanism. First, the model conditions on map elements and surrounding agents to generate multiple reference trajectories from a single ground-truth trajectory, alleviating the limitations of imitation learning that arise from relying solely on single expert demonstrations. Second, reinforcement learning is employed to guide the diffusion process, where reward-based supervision enforces safety and diversity constraints on the generated trajectories, thereby enhancing their practicality and generalization capability. Furthermore, to address the limitations of L2-based open-loop metrics in capturing trajectory diversity, we propose a novel Diversity metric to evaluate the diversity of multi-mode predictions.Extensive experiments on the closed-loop NAVSIM and Bench2Drive benchmarks, as well as the open-loop nuScenes dataset, demonstrate that DIVER significantly improves trajectory diversity, effectively addressing the mode collapse problem inherent in imitation learning.
comment: 18 pages, 8 figures
♻ ☆ RoboBPP: Benchmarking Robotic Online Bin Packing with Physics-based Simulation
Physical feasibility in 3D bin packing is a key requirement in modern industrial logistics and robotic automation. With the growing adoption of industrial automation, online bin packing has gained increasing attention. However, inconsistencies in problem settings, test datasets, and evaluation metrics have hindered progress in the field, and there is a lack of a comprehensive benchmarking system. Direct testing on real hardware is costly, and building a realistic simulation environment is also challenging. To address these limitations, we introduce RoboBPP, a benchmarking system designed for robotic online bin packing. RoboBPP integrates a physics-based simulator to assess physical feasibility. In our simulation environment, we introduce a robotic arm and boxes at real-world scales to replicate real industrial packing workflows. By simulating conditions that arise in real industrial applications, we ensure that evaluated algorithms are practically deployable. In addition, prior studies often rely on synthetic datasets whose distributions differ from real-world industrial data. To address this issue, we collect three datasets from real industrial workflows, including assembly-line production, logistics packing, and furniture manufacturing. The benchmark comprises three carefully designed test settings and extends existing evaluation metrics with new metrics for structural stability and operational safety. We design a scoring system and derive a range of insights from the evaluation results. RoboBPP is fully open-source and is equipped with visualization tools and an online leaderboard, providing a reproducible and extensible foundation for future research and industrial applications (https://robot-bin-packing-benchmark.github.io).
comment: Under review at the International Journal of Robotics Research (IJRR)
♻ ☆ Control Your Robot: A Unified System for Robot Control and Policy Deployment
Cross-platform robot control remains difficult because hardware interfaces, data formats, and control paradigms vary widely, which fragments toolchains and slows deployment. To address this, we present Control Your Robot, a modular, general-purpose framework that unifies data collection and policy deployment across diverse platforms. The system reduces fragmentation through a standardized workflow with modular design, unified APIs, and a closed-loop architecture. It supports flexible robot registration, dual-mode control with teleoperation and trajectory playback, and seamless integration from multimodal data acquisition to inference. Experiments on single-arm and dual-arm systems show efficient, low-latency data collection and effective support for policy learning with imitation learning and vision-language-action models. Policies trained on data gathered by Control Your Robot match expert demonstrations closely, indicating that the framework enables scalable and reproducible robot learning across platforms.
comment: Code: https://github.com/Tian-Nian/control_your_robot
♻ ☆ Humanoid Whole-Body Badminton via Multi-Stage Reinforcement Learning
Humanoid robots have demonstrated strong capabilities for interacting with static scenes across locomotion, manipulation, and more challenging loco-manipulation tasks. Yet the real world is dynamic, and quasi-static interactions are insufficient to cope with diverse environmental conditions. As a step toward more dynamic interaction scenarios, we present a reinforcement-learning-based training pipeline that produces a unified whole-body controller for humanoid badminton, enabling coordinated lower-body footwork and upper-body striking without motion priors or expert demonstrations. Training follows a three-stage curriculum: first footwork acquisition, then precision-guided racket swing generation, and finally task-focused refinement, yielding motions in which both legs and arms serve the hitting objective. For deployment, we incorporate an Extended Kalman Filter (EKF) to estimate and predict shuttlecock trajectories for target striking. We also introduce a prediction-free variant that dispenses with EKF and explicit trajectory prediction. To validate the framework, we conduct five sets of experiments in both simulation and the real world. In simulation, two robots sustain a rally of 21 consecutive hits. Moreover, the prediction-free variant achieves successful hits with comparable performance relative to the target-known policy. In real-world tests, both prediction and controller modules exhibit high accuracy, and on-court hitting achieves an outgoing shuttle speed up to 19.1 m/s with a mean return landing distance of 4 m. These experimental results show that our proposed training scheme can deliver highly dynamic while precise goal striking in badminton, and can be adapted to more dynamics-critical domains.
comment: Project Page: https://humanoid-badminton.github.io/Humanoid-Whole-Body-Badminton-via-Multi-Stage-Reinforcement-Learning
♻ ☆ db-LaCAM: Fast and Scalable Multi-Robot Kinodynamic Motion Planning with Discontinuity-Bounded Search and Lightweight MAPF
State-of-the-art multi-robot kinodynamic motion planners struggle to handle more than a few robots due to high computational burden, which limits their scalability and results in slow planning time. In this work, we combine the scalability and speed of modern multi-agent path finding (MAPF) algorithms with the dynamic-awareness of kinodynamic planners to address these limitations. To this end, we propose discontinuity-Bounded LaCAM (db-LaCAM), a planner that utilizes a precomputed set of motion primitives that respect robot dynamics to generate horizon-length motion sequences, while allowing a user-defined discontinuity between successive motions. The planner db-LaCAM is resolution-complete with respect to motion primitives and supports arbitrary robot dynamics. Extensive experiments demonstrate that db-LaCAM scales efficiently to scenarios with up to 50 robots, achieving up to ten times faster runtime compared to state-of-the-art planners, while maintaining comparable solution quality. The approach is validated in both 2D and 3D environments with dynamics such as the unicycle and 3D double integrator. We demonstrate the safe execution of trajectories planned with db-LaCAM in two distinct physical experiments involving teams of flying robots and car-with-trailer robots.
♻ ☆ Unifying Entropy Regularization in Optimal Control: From and Back to Classical Objectives via Iterated Soft Policies and Path Integral Solutions
This paper develops a unified perspective on several stochastic optimal control formulations through the lens of Kullback-Leibler regularization. We propose a central problem that separates the KL penalties on policies and transitions, assigning them independent weights, thereby generalizing the standard trajectory-level KL-regularization commonly used in probabilistic and KL-regularized control. This generalized formulation acts as a generative structure allowing to recover various control problems. These include the classical Stochastic Optimal Control (SOC), Risk-Sensitive Optimal Control (RSOC), and their policy-based KL-regularized counterparts. The latter we refer to as soft-policy SOC and RSOC, facilitating alternative problems with tractable solutions. Beyond serving as regularized variants, we show that these soft-policy formulations majorize the original SOC and RSOC problem. This means that the regularized solution can be iterated to retrieve the original solution. Furthermore, we identify a structurally synchronized case of the risk-seeking soft-policy RSOC formulation, wherein the policy and transition KL-regularization weights coincide. Remarkably, this specific setting gives rise to several powerful properties such as a linear Bellman equation, path integral solution, and, compositionality, thereby extending these computationally favourable properties to a broad class of control problems.
comment: Corrected "DRO" to "DRC" and fixed theorem numbering throughout paper
♻ ☆ Fitts' List Revisited: An Empirical Study on Function Allocation in a Two-Agent Physical Human-Robot Collaborative Position/Force Task
In this letter, we investigate whether classical function allocation-the principle of assigning tasks to either a human or a machine-holds for physical Human-Robot Collaboration, which is important for providing insights for Industry 5.0 to guide how to best augment rather than replace workers. This study empirically tests the applicability of Fitts' List within physical Human-Robot Collaboration, by conducting a user study (N=26, within-subject design) to evaluate four distinct allocations of position/force control between human and robot in an abstract blending task. We hypothesize that the function in which humans control the position achieves better performance and receives higher user ratings. When allocating position control to the human and force control to the robot, compared to the opposite case, we observed a significant improvement in preventing overblending. This was also perceived better in terms of physical demand and overall system acceptance, while participants experienced greater autonomy, more engagement and less frustration. An interesting insight was that the supervisory role (when the robot controls both position and force) was rated second best in terms of subjective acceptance. Another surprising insight was that if position control was delegated to the robot, the participants perceived much lower autonomy than when the force control was delegated to the robot. These findings empirically support applying Fitts' principles to static function allocation for physical collaboration, while also revealing important nuanced user experience trade-offs, particularly regarding perceived autonomy when delegating position control.
comment: 8 pages, 6 figures, published in IEEE Robotics and Automation Letters, col. 11, no. 1, January 2026
♻ ☆ Haptic-based Complementary Filter for Rigid Body Rotations
The non-commutative nature of 3D rotations poses well-known challenges in generalizing planar problems to three-dimensional ones, even more so in contact-rich tasks where haptic information (i.e., forces/torques) is involved. In this sense, not all learning-based algorithms that are currently available generalize to 3D orientation estimation. Non-linear filters defined on $\mathbf{\mathbb{SO}(3)}$ are widely used with inertial measurement sensors; however, none of them have been used with haptic measurements. This paper presents a unique complementary filtering framework that interprets the geometric shape of objects in the form of superquadrics, exploits the symmetry of $\mathbf{\mathbb{SO}(3)}$, and uses force and vision sensors as measurements to provide an estimate of orientation. The framework's robustness and almost global stability are substantiated by a set of experiments on a dual-arm robotic setup.
comment: 7 pages, 7 figures; Updated filter design; Submitted to IFAC for possible publication
♻ ☆ Capability-Driven Skill Generation with LLMs: A RAG-Based Approach for Reusing Existing Libraries and Interfaces
Modern automation systems increasingly rely on modular architectures, with capabilities and skills as one solution approach. Capabilities define the functions of resources in a machine-readable form and skills provide the concrete implementations that realize those capabilities. However, the development of a skill implementation conforming to a corresponding capability remains a time-consuming and challenging task. In this paper, we present a method that treats capabilities as contracts for skill implementations and leverages large language models to generate executable code based on natural language user input. A key feature of our approach is the integration of existing software libraries and interface technologies, enabling the generation of skill implementations across different target languages. We introduce a framework that allows users to incorporate their own libraries and resource interfaces into the code generation process through a retrieval-augmented generation architecture. The proposed method is evaluated using an autonomous mobile robot controlled via Python and ROS 2, demonstrating the feasibility and flexibility of the approach.
comment: \c{opyright} 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
♻ ☆ 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.
♻ ☆ MARL Warehouse Robots
We present a comparative study of multi-agent reinforcement learning (MARL) algorithms for cooperative warehouse robotics. We evaluate QMIX and IPPO on the Robotic Warehouse (RWARE) environment and a custom Unity 3D simulation. Our experiments reveal that QMIX's value decomposition significantly outperforms independent learning approaches (achieving 3.25 mean return vs. 0.38 for advanced IPPO), but requires extensive hyperparameter tuning -- particularly extended epsilon annealing (5M+ steps) for sparse reward discovery. We demonstrate successful deployment in Unity ML-Agents, achieving consistent package delivery after 1M training steps. While MARL shows promise for small-scale deployments (2-4 robots), significant scaling challenges remain. Code and analyses: https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/
comment: 5 pages.Project documentation: https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/
♻ ☆ CDKFormer: Contextual Deviation Knowledge-Based Transformer for Long-Tail Trajectory Prediction
Predicting the future movements of surrounding vehicles is essential for ensuring the safe operation and efficient navigation of autonomous vehicles (AVs) in urban traffic environments. Existing vehicle trajectory prediction methods primarily focus on improving overall performance, yet they struggle to address long-tail scenarios effectively. This limitation often leads to poor predictions in rare cases, significantly increasing the risk of safety incidents. Taking Argoverse 2 motion forecasting dataset as an example, we first investigate the long-tail characteristics in trajectory samples from two perspectives, individual motion and group interaction, and deriving deviation features to distinguish abnormal from regular scenarios. On this basis, we propose CDKFormer, a Contextual Deviation Knowledge-based Transformer model for long-tail trajectory prediction. CDKFormer integrates an attention-based scene context fusion module to encode spatiotemporal interaction and road topology. An additional deviation feature fusion module is proposed to capture the dynamic deviations in the target vehicle status. We further introduce a dual query-based decoder, supported by a multi-stream decoder block, to sequentially decode heterogeneous scene deviation features and generate multimodal trajectory predictions. Extensive experiments demonstrate that CDKFormer achieves state-of-the-art performance, significantly enhancing prediction accuracy and robustness for long-tailed trajectories compared to existing methods, thus advancing the reliability of AVs in complex real-world environments.
♻ ☆ Khalasi: Energy-Efficient Navigation for Surface Vehicles in Vortical Flow Fields ICRA 2026
For centuries, khalasi (Gujarati for sailor) have skillfully harnessed ocean currents to navigate vast waters with minimal effort. Emulating this intuition in autonomous systems remains a significant challenge, particularly for Autonomous Surface Vehicles tasked with long duration missions under strict energy budgets. In this work, we present a learning-based approach for energy-efficient surface vehicle navigation in vortical flow fields, where partial observability often undermines traditional path-planning methods. We present an end to end reinforcement learning framework based on Soft Actor Critic that learns flow-aware navigation policies using only local velocity measurements. Through extensive evaluation across diverse and dynamically rich scenarios, our method demonstrates substantial energy savings and robust generalization to previously unseen flow conditions, offering a promising path toward long term autonomy in ocean environments. The navigation paths generated by our proposed approach show an improvement in energy conservation 30 to 50 percent compared to the existing state of the art techniques.
comment: Under Review for International Conference on Robotics and Automation (ICRA 2026)
♻ ☆ Incremental Generalized Hybrid A*
We address the problem of efficiently organizing search over very large trees, which arises in many applications ranging from autonomous driving to aerial vehicles. Here, we are motivated by off-road autonomy, where real-time planning is essential. Classical approaches use graphs of motion primitives and exploit dominance to mitigate the curse of dimensionality and prune expansions efficiently. However, for complex dynamics, repeatedly solving two-point boundary-value problems makes graph construction too slow for fast kinodynamic planning. Hybrid A* (HA*) addressed this challenge by searching over a tree of motion primitives and introducing approximate pruning using a grid-based dominance check. However, choosing the grid resolution is difficult: too coarse risks failure, while too fine leads to excessive expansions and slow planning. We propose Incremental Generalized Hybrid A* (IGHA*), an anytime tree-search framework that dynamically organizes vertex expansions without rigid pruning. IGHA* provably matches or outperforms HA*. For both on-road kinematic and off-road kinodynamic planning queries for a car-like robot, variants of IGHA* use 6x fewer expansions to the best solution compared to an optimized version of HA* (HA*M, an internal baseline). In simulated off-road experiments in a high-fidelity simulator, IGHA* outperforms HA*M when both are used in the loop with a model predictive controller. We demonstrate real-time performance both in simulation and on a small-scale off-road vehicle, enabling fast, robust planning under complex dynamics. Website: https://personalrobotics.github.io/IGHAStar/
comment: 8 pages, 7 figures, Accepted to IEEE RA-L, Nov 2025
♻ ☆ Training-Time Action Conditioning for Efficient Real-Time Chunking
Real-time chunking (RTC) enables vision-language-action models (VLAs) to generate smooth, reactive robot trajectories by asynchronously predicting action chunks and conditioning on previously committed actions via inference-time inpainting. However, this inpainting method introduces computational overhead that increases inference latency. In this work, we propose a simple alternative: simulating inference delay at training time and conditioning on action prefixes directly, eliminating any inference-time overhead. Our method requires no modifications to the model architecture or robot runtime, and can be implemented with only a few additional lines of code. In simulated experiments, we find that training-time RTC outperforms inference-time RTC at higher inference delays. In real-world experiments on box building and espresso making tasks with the $π_{0.6}$ VLA, we demonstrate that training-time RTC maintains both task performance and speed parity with inference-time RTC while being computationally cheaper. Our results suggest that training-time action conditioning is a practical drop-in replacement for inference-time inpainting in real-time robot control.
♻ ☆ Enhancing the NAO: Extending Capabilities of Legacy Robots for Long-Term Research
Legacy (unsupported) robotic platforms often lose research utility when manufacturer support ends, preventing integration of modern sensing, speech, and interaction capabilities. We present the Enhanced NAO, a revitalized version of Aldebaran's NAO robot featuring upgraded beamforming microphones, RGB-D and thermal cameras, and additional compute resources in a fully self-contained package. This system combines cloud-based and local models for perception and dialogue, while preserving the NAO's expressive body and behaviors. In a pilot user study validating conversational performance, the Enhanced NAO delivered significantly higher conversational quality and elicited stronger user preference compared to the NAO AI Edition, without increasing response latency. The added visual and thermal sensing modalities established a foundation for future perception-driven interaction. Beyond this implementation, our framework provides a platform-agnostic strategy for extending the lifespan and research utility of legacy robots, ensuring they remain valuable tools for human-robot interaction.
♻ ☆ Connectivity-Preserving Multi-Agent Area Coverage via Optimal-Transport-Based Density-Driven Optimal Control (D2OC)
Multi-agent systems play a central role in area coverage tasks across search-and-rescue, environmental monitoring, and precision agriculture. Achieving non-uniform coverage, where spatial priorities vary across the domain, requires coordinating agents while respecting dynamic and communication constraints. Density-driven approaches can distribute agents according to a prescribed reference density, but existing methods do not ensure connectivity. This limitation often leads to communication loss, reduced coordination, and degraded coverage performance. This letter introduces a connectivity-preserving extension of the Density-Driven Optimal Control (D2OC) framework. The coverage objective, defined using the Wasserstein distance between the agent distribution and the reference density, admits a convex quadratic program formulation. Communication constraints are incorporated through a smooth connectivity penalty, which maintains strict convexity, supports distributed implementation, and preserves inter-agent communication without imposing rigid formations. Simulation studies show that the proposed method consistently maintains connectivity, improves convergence speed, and enhances non-uniform coverage quality compared with density-driven schemes that do not incorporate explicit connectivity considerations.
comment: Under review in IEEE Control Systems Letters (LCSS). 6 pages
Robotics 53
☆ Efficient and Compliant Control Framework for Versatile Human-Humanoid Collaborative Transportation
We present a control framework that enables humanoid robots to perform collaborative transportation tasks with a human partner. The framework supports both translational and rotational motions, which are fundamental to co-transport scenarios. It comprises three components: a high-level planner, a low-level controller, and a stiffness modulation mechanism. At the planning level, we introduce the Interaction Linear Inverted Pendulum (I-LIP), which, combined with an admittance model and an MPC formulation, generates dynamically feasible footstep plans. These are executed by a QP-based whole-body controller that accounts for the coupled humanoid-object dynamics. Stiffness modulation regulates robot-object interaction, ensuring convergence to the desired relative configuration defined by the distance between the object and the robot's center of mass. We validate the effectiveness of the framework through real-world experiments conducted on the Digit humanoid platform. To quantify collaboration quality, we propose an efficiency metric that captures both task performance and inter-agent coordination. We show that this metric highlights the role of compliance in collaborative tasks and offers insights into desirable trajectory characteristics across both high- and low-level control layers. Finally, we showcase experimental results on collaborative behaviors, including translation, turning, and combined motions such as semi circular trajectories, representative of naturally occurring co-transportation tasks.
☆ Inchworm-Inspired Soft Robot with Groove-Guided Locomotion
Soft robots require directional control to navigate complex terrains. However, achieving such control often requires multiple actuators, which increases mechanical complexity, complicates control systems, and raises energy consumption. Here, we introduce an inchworm-inspired soft robot whose locomotion direction is controlled passively by patterned substrates. The robot employs a single rolled dielectric elastomer actuator, while groove patterns on a 3D-printed substrate guide its alignment and trajectory. Through systematic experiments, we demonstrate that varying groove angles enables precise control of locomotion direction without the need for complex actuation strategies. This groove-guided approach reduces energy consumption, simplifies robot design, and expands the applicability of bio-inspired soft robots in fields such as search and rescue, pipe inspection, and planetary exploration.
☆ OptMap: Geometric Map Distillation via Submodular Maximization
Autonomous robots rely on geometric maps to inform a diverse set of perception and decision-making algorithms. As autonomy requires reasoning and planning on multiple scales of the environment, each algorithm may require a different map for optimal performance. Light Detection And Ranging (LiDAR) sensors generate an abundance of geometric data to satisfy these diverse requirements, but selecting informative, size-constrained maps is computationally challenging as it requires solving an NP-hard combinatorial optimization. In this work we present OptMap: a geometric map distillation algorithm which achieves real-time, application-specific map generation via multiple theoretical and algorithmic innovations. A central feature is the maximization of set functions that exhibit diminishing returns, i.e., submodularity, using polynomial-time algorithms with provably near-optimal solutions. We formulate a novel submodular reward function which quantifies informativeness, reduces input set sizes, and minimizes bias in sequentially collected datasets. Further, we propose a dynamically reordered streaming submodular algorithm which improves empirical solution quality and addresses input order bias via an online approximation of the value of all scans. Testing was conducted on open-source and custom datasets with an emphasis on long-duration mapping sessions, highlighting OptMap's minimal computation requirements. Open-source ROS1 and ROS2 packages are available and can be used alongside any LiDAR SLAM algorithm.
☆ Toward Seamless Physical Human-Humanoid Interaction: Insights from Control, Intent, and Modeling with a Vision for What Comes Next
Physical Human-Humanoid Interaction (pHHI) is a rapidly advancing field with significant implications for deploying robots in unstructured, human-centric environments. In this review, we examine the current state of the art in pHHI through three core pillars: (i) humanoid modeling and control, (ii) human intent estimation, and (iii) computational human models. For each pillar, we survey representative approaches, identify open challenges, and analyze current limitations that hinder robust, scalable, and adaptive interaction. These include the need for whole-body control strategies capable of handling uncertain human dynamics, real-time intent inference under limited sensing, and modeling techniques that account for variability in human physical states. Although significant progress has been made within each domain, integration across pillars remains limited. We propose pathways for unifying methods across these areas to enable cohesive interaction frameworks. This structure enables us not only to map the current landscape but also to propose concrete directions for future research that aim to bridge these domains. Additionally, we introduce a unified taxonomy of interaction types based on modality, distinguishing between direct interactions (e.g., physical contact) and indirect interactions (e.g., object-mediated), and on the level of robot engagement, ranging from assistance to cooperation and collaboration. For each category in this taxonomy, we provide the three core pillars that highlight opportunities for cross-pillar unification. Our goal is to suggest avenues to advance robust, safe, and intuitive physical interaction, providing a roadmap for future research that will allow humanoid systems to effectively understand, anticipate, and collaborate with human partners in diverse real-world settings.
comment: 60 pages, 5 figures, 3 tables
☆ UltrasODM: A Dual Stream Optical Flow Mamba Network for 3D Freehand Ultrasound Reconstruction
Clinical ultrasound acquisition is highly operator-dependent, where rapid probe motion and brightness fluctuations often lead to reconstruction errors that reduce trust and clinical utility. We present UltrasODM, a dual-stream framework that assists sonographers during acquisition through calibrated per-frame uncertainty, saliency-based diagnostics, and actionable prompts. UltrasODM integrates (i) a contrastive ranking module that groups frames by motion similarity, (ii) an optical-flow stream fused with Dual-Mamba temporal modules for robust 6-DoF pose estimation, and (iii) a Human-in-the-Loop (HITL) layer combining Bayesian uncertainty, clinician-calibrated thresholds, and saliency maps highlighting regions of low confidence. When uncertainty exceeds the threshold, the system issues unobtrusive alerts suggesting corrective actions such as re-scanning highlighted regions or slowing the sweep. Evaluated on a clinical freehand ultrasound dataset, UltrasODM reduces drift by 15.2%, distance error by 12.1%, and Hausdorff distance by 10.1% relative to UltrasOM, while producing per-frame uncertainty and saliency outputs. By emphasizing transparency and clinician feedback, UltrasODM improves reconstruction reliability and supports safer, more trustworthy clinical workflows. Our code is publicly available at https://github.com/AnandMayank/UltrasODM.
☆ sim2art: Accurate Articulated Object Modeling from a Single Video using Synthetic Training Data Only
Understanding articulated objects is a fundamental challenge in robotics and digital twin creation. To effectively model such objects, it is essential to recover both part segmentation and the underlying joint parameters. Despite the importance of this task, previous work has largely focused on setups like multi-view systems, object scanning, or static cameras. In this paper, we present the first data-driven approach that jointly predicts part segmentation and joint parameters from monocular video captured with a freely moving camera. Trained solely on synthetic data, our method demonstrates strong generalization to real-world objects, offering a scalable and practical solution for articulated object understanding. Our approach operates directly on casually recorded video, making it suitable for real-time applications in dynamic environments. Project webpage: https://aartykov.github.io/sim2art/
☆ Delay-Aware Diffusion Policy: Bridging the Observation-Execution Gap in Dynamic Tasks
As a robot senses and selects actions, the world keeps changing. This inference delay creates a gap of tens to hundreds of milliseconds between the observed state and the state at execution. In this work, we take the natural generalization from zero delay to measured delay during training and inference. We introduce Delay-Aware Diffusion Policy (DA-DP), a framework for explicitly incorporating inference delays into policy learning. DA-DP corrects zero-delay trajectories to their delay-compensated counterparts, and augments the policy with delay conditioning. We empirically validate DA-DP on a variety of tasks, robots, and delays and find its success rate more robust to delay than delay-unaware methods. DA-DP is architecture agnostic and transfers beyond diffusion policies, offering a general pattern for delay-aware imitation learning. More broadly, DA-DP encourages evaluation protocols that report performance as a function of measured latency, not just task difficulty.
☆ AMBER: Aerial deployable gripping crawler with compliant microspine 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 degrees of body roll and inclination, while effective climbing on branches inclined up to 67.5 degrees, achieving a maximum speed of 0.55 body lengths per second on horizontal branches. The compliant tracks allow yaw steering of up to 10 degrees, 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. Integrated within a drone-tether deployment system, the crawler provides a robust, low-power platform for environmental sampling and in-canopy sensing, bridging the gap between aerial and surface-based ecological robotics.
☆ Multi-Domain Motion Embedding: Expressive Real-Time Mimicry for Legged Robots
Effective motion representation is crucial for enabling robots to imitate expressive behaviors in real time, yet existing motion controllers often ignore inherent patterns in motion. Previous efforts in representation learning do not attempt to jointly capture structured periodic patterns and irregular variations in human and animal movement. To address this, we present Multi-Domain Motion Embedding (MDME), a motion representation that unifies the embedding of structured and unstructured features using a wavelet-based encoder and a probabilistic embedding in parallel. This produces a rich representation of reference motions from a minimal input set, enabling improved generalization across diverse motion styles and morphologies. We evaluate MDME on retargeting-free real-time motion imitation by conditioning robot control policies on the learned embeddings, demonstrating accurate reproduction of complex trajectories on both humanoid and quadruped platforms. Our comparative studies confirm that MDME outperforms prior approaches in reconstruction fidelity and generalizability to unseen motions. Furthermore, we demonstrate that MDME can reproduce novel motion styles in real-time through zero-shot deployment, eliminating the need for task-specific tuning or online retargeting. These results position MDME as a generalizable and structure-aware foundation for scalable real-time robot imitation.
comment: 15 pages
☆ Obstacle Avoidance of UAV in Dynamic Environments Using Direction and Velocity-Adaptive Artificial Potential Field
The conventional Artificial Potential Field (APF) is fundamentally limited by the local minima issue and its inability to account for the kinematics of moving obstacles. This paper addresses the critical challenge of autonomous collision avoidance for Unmanned Aerial Vehicles (UAVs) operating in dynamic and cluttered airspace by proposing a novel Direction and Relative Velocity Weighted Artificial Potential Field (APF). In this approach, a bounded weighting function, $ω(θ,v_{e})$, is introduced to dynamically scale the repulsive potential based on the direction and velocity of the obstacle relative to the UAV. This robust APF formulation is integrated within a Model Predictive Control (MPC) framework to generate collision-free trajectories while adhering to kinematic constraints. Simulation results demonstrate that the proposed method effectively resolves local minima and significantly enhances safety by enabling smooth, predictive avoidance maneuvers. The system ensures superior path integrity and reliable performance, confirming its viability for autonomous navigation in complex environments.
☆ More than Segmentation: Benchmarking SAM 3 for Segmentation, 3D Perception, and Reconstruction in Robotic Surgery
The recent Segment Anything Model (SAM) 3 has introduced significant advancements over its predecessor, SAM 2, particularly with the integration of language-based segmentation and enhanced 3D perception capabilities. SAM 3 supports zero-shot segmentation across a wide range of prompts, including point, bounding box, and language-based prompts, allowing for more flexible and intuitive interactions with the model. In this empirical evaluation, we assess the performance of SAM 3 in robot-assisted surgery, benchmarking its zero-shot segmentation with point and bounding box prompts and exploring its effectiveness in dynamic video tracking, alongside its newly introduced language prompt segmentation. While language prompts show potential, their performance in the surgical domain is currently suboptimal, highlighting the need for further domain-specific training. Additionally, we investigate SAM 3's 3D reconstruction abilities, demonstrating its capacity to process surgical scene data and reconstruct 3D anatomical structures from 2D images. Through comprehensive testing on the MICCAI EndoVis 2017 and EndoVis 2018 benchmarks, SAM 3 shows clear improvements over SAM and SAM 2 in both image and video segmentation under spatial prompts, while zero-shot evaluations on SCARED, StereoMIS, and EndoNeRF indicate strong monocular depth estimation and realistic 3D instrument reconstruction, yet also reveal remaining limitations in complex, highly dynamic surgical scenes.
comment: Technical Report
☆ See Once, Then Act: Vision-Language-Action Model with Task Learning from One-Shot Video Demonstrations
Developing robust and general-purpose manipulation policies represents a fundamental objective in robotics research. While Vision-Language-Action (VLA) models have demonstrated promising capabilities for end-to-end robot control, existing approaches still exhibit limited generalization to tasks beyond their training distributions. In contrast, humans possess remarkable proficiency in acquiring novel skills by simply observing others performing them once. Inspired by this capability, we propose ViVLA, a generalist robotic manipulation policy that achieves efficient task learning from a single expert demonstration video at test time. Our approach jointly processes an expert demonstration video alongside the robot's visual observations to predict both the demonstrated action sequences and subsequent robot actions, effectively distilling fine-grained manipulation knowledge from expert behavior and transferring it seamlessly to the agent. To enhance the performance of ViVLA, we develop a scalable expert-agent pair data generation pipeline capable of synthesizing paired trajectories from easily accessible human videos, further augmented by curated pairs from publicly available datasets. This pipeline produces a total of 892,911 expert-agent samples for training ViVLA. Experimental results demonstrate that our ViVLA is able to acquire novel manipulation skills from only a single expert demonstration video at test time. Our approach achieves over 30% improvement on unseen LIBERO tasks and maintains above 35% gains with cross-embodiment videos. Real-world experiments demonstrate effective learning from human videos, yielding more than 38% improvement on unseen tasks.
☆ VP-AutoTest: A Virtual-Physical Fusion Autonomous Driving Testing Platform
The rapid development of autonomous vehicles has led to a surge in testing demand. Traditional testing methods, such as virtual simulation, closed-course, and public road testing, face several challenges, including unrealistic vehicle states, limited testing capabilities, and high costs. These issues have prompted increasing interest in virtual-physical fusion testing. However, despite its potential, virtual-physical fusion testing still faces challenges, such as limited element types, narrow testing scope, and fixed evaluation metrics. To address these challenges, we propose the Virtual-Physical Testing Platform for Autonomous Vehicles (VP-AutoTest), which integrates over ten types of virtual and physical elements, including vehicles, pedestrians, and roadside infrastructure, to replicate the diversity of real-world traffic participants. The platform also supports both single-vehicle interaction and multi-vehicle cooperation testing, employing adversarial testing and parallel deduction to accelerate fault detection and explore algorithmic limits, while OBU and Redis communication enable seamless vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) cooperation across all levels of cooperative automation. Furthermore, VP-AutoTest incorporates a multidimensional evaluation framework and AI-driven expert systems to conduct comprehensive performance assessment and defect diagnosis. Finally, by comparing virtual-physical fusion test results with real-world experiments, the platform performs credibility self-evaluation to ensure both the fidelity and efficiency of autonomous driving testing. Please refer to the website for the full testing functionalities on the autonomous driving public service platform OnSite:https://www.onsite.com.cn.
☆ From Real-World Traffic Data to Relevant Critical Scenarios
The reliable operation of autonomous vehicles, automated driving functions, and advanced driver assistance systems across a wide range of relevant scenarios is critical for their development and deployment. Identifying a near-complete set of relevant driving scenarios for such functionalities is challenging due to numerous degrees of freedom involved, each affecting the outcomes of the driving scenario differently. Moreover, with increasing technical complexity of new functionalities, the number of potentially relevant, particularly "unknown unsafe" scenarios is increasing. To enhance validation efficiency, it is essential to identify relevant scenarios in advance, starting with simpler domains like highways before moving to more complex environments such as urban traffic. To address this, this paper focuses on analyzing lane change scenarios in highway traffic, which involve multiple degrees of freedom and present numerous safetyrelevant scenarios. We describe the process of data acquisition and processing of real-world data from public highway traffic, followed by the application of criticality measures on trajectory data to evaluate scenarios, as conducted within the AVEAS project (www.aveas.org). By linking the calculated measures to specific lane change driving scenarios and the conditions under which the data was collected, we facilitate the identification of safetyrelevant driving scenarios for various applications. Further, to tackle the extensive range of "unknown unsafe" scenarios, we propose a way to generate relevant scenarios by creating synthetic scenarios based on recorded ones. Consequently, we demonstrate and evaluate a processing chain that enables the identification of safety-relevant scenarios, the development of data-driven methods for extracting these scenarios, and the generation of synthetic critical scenarios via sampling on highways.
comment: 8 pages, 8 figures
☆ Affordance Field Intervention: Enabling VLAs to Escape Memory Traps in Robotic Manipulation
Vision-Language-Action (VLA) models have shown great performance in robotic manipulation by mapping visual observations and language instructions directly to actions. However, they remain brittle under distribution shifts: when test scenarios change, VLAs often reproduce memorized trajectories instead of adapting to the updated scene, which is a failure mode we refer to as the "Memory Trap". This limitation stems from the end-to-end design, which lacks explicit 3D spatial reasoning and prevents reliable identification of actionable regions in unfamiliar environments. To compensate for this missing spatial understanding, 3D Spatial Affordance Fields (SAFs) can provide a geometric representation that highlights where interactions are physically feasible, offering explicit cues about regions the robot should approach or avoid. We therefore introduce Affordance Field Intervention (AFI), a lightweight hybrid framework that uses SAFs as an on-demand plug-in to guide VLA behavior. Our system detects memory traps through proprioception, repositions the robot to recent high-affordance regions, and proposes affordance-driven waypoints that anchor VLA-generated actions. A SAF-based scorer then selects trajectories with the highest cumulative affordance. Extensive experiments demonstrate that our method achieves an average improvement of 23.5% across different VLA backbones ($π_{0}$ and $π_{0.5}$) under out-of-distribution scenarios on real-world robotic platforms, and 20.2% on the LIBERO-Pro benchmark, validating its effectiveness in enhancing VLA robustness to distribution shifts.
☆ Gait-Adaptive Perceptive Humanoid Locomotion with Real-Time Under-Base Terrain Reconstruction
For full-size humanoid robots, even with recent advances in reinforcement learning-based control, achieving reliable locomotion on complex terrains, such as long staircases, remains challenging. In such settings, limited perception, ambiguous terrain cues, and insufficient adaptation of gait timing can cause even a single misplaced or mistimed step to result in rapid loss of balance. We introduce a perceptive locomotion framework that merges terrain sensing, gait regulation, and whole-body control into a single reinforcement learning policy. A downward-facing depth camera mounted under the base observes the support region around the feet, and a compact U-Net reconstructs a dense egocentric height map from each frame in real time, operating at the same frequency as the control loop. The perceptual height map, together with proprioceptive observations, is processed by a unified policy that produces joint commands and a global stepping-phase signal, allowing gait timing and whole-body posture to be adapted jointly to the commanded motion and local terrain geometry. We further adopt a single-stage successive teacher-student training scheme for efficient policy learning and knowledge transfer. Experiments conducted on a 31-DoF, 1.65 m humanoid robot demonstrate robust locomotion in both simulation and real-world settings, including forward and backward stair ascent and descent, as well as crossing a 46 cm gap. Project Page:https://ga-phl.github.io/
☆ KAN-Dreamer: Benchmarking Kolmogorov-Arnold Networks as Function Approximators in World Models
DreamerV3 is a state-of-the-art online model-based reinforcement learning (MBRL) algorithm known for remarkable sample efficiency. Concurrently, Kolmogorov-Arnold Networks (KANs) have emerged as a promising alternative to Multi-Layer Perceptrons (MLPs), offering superior parameter efficiency and interpretability. To mitigate KANs' computational overhead, variants like FastKAN leverage Radial Basis Functions (RBFs) to accelerate inference. In this work, we investigate integrating KAN architectures into the DreamerV3 framework. We introduce KAN-Dreamer, replacing specific MLP and convolutional components of DreamerV3 with KAN and FastKAN layers. To ensure efficiency within the JAX-based World Model, we implement a tailored, fully vectorized version with simplified grid management. We structure our investigation into three subsystems: Visual Perception, Latent Prediction, and Behavior Learning. Empirical evaluations on the DeepMind Control Suite (walker_walk) analyze sample efficiency, training time, and asymptotic performance. Experimental results demonstrate that utilizing our adapted FastKAN as a drop-in replacement for the Reward and Continue predictors yields performance on par with the original MLP-based architecture, maintaining parity in both sample efficiency and training speed. This report serves as a preliminary study for future developments in KAN-based world models.
comment: 23 pages, 8 figures, 3 tables
☆ ESPADA: Execution Speedup via Semantics Aware Demonstration Data Downsampling for Imitation Learning
Behavior-cloning based visuomotor policies enable precise manipulation but often inherit the slow, cautious tempo of human demonstrations, limiting practical deployment. However, prior studies on acceleration methods mainly rely on statistical or heuristic cues that ignore task semantics and can fail across diverse manipulation settings. We present ESPADA, a semantic and spatially aware framework that segments demonstrations using a VLM-LLM pipeline with 3D gripper-object relations, enabling aggressive downsampling only in non-critical segments while preserving precision-critical phases, without requiring extra data or architectural modifications, or any form of retraining. To scale from a single annotated episode to the full dataset, ESPADA propagates segment labels via Dynamic Time Warping (DTW) on dynamics-only features. Across both simulation and real-world experiments with ACT and DP baselines, ESPADA achieves approximately a 2x speed-up while maintaining success rates, narrowing the gap between human demonstrations and efficient robot control.
comment: project page: https://project-espada.github.io/espada/
☆ Multi-Rigid-Body Approximation of Human Hands with Application to Digital Twin
Human hand simulation plays a critical role in digital twin applications, requiring models that balance anatomical fidelity with computational efficiency. We present a complete pipeline for constructing multi-rigid-body approximations of human hands that preserve realistic appearance while enabling real-time physics simulation. Starting from optical motion capture of a specific human hand, we construct a personalized MANO (Multi-Abstracted hand model with Neural Operations) model and convert it to a URDF (Unified Robot Description Format) representation with anatomically consistent joint axes. The key technical challenge is projecting MANO's unconstrained SO(3) joint rotations onto the kinematically constrained joints of the rigid-body model. We derive closed-form solutions for single degree-of-freedom joints and introduce a Baker-Campbell-Hausdorff (BCH)-corrected iterative method for two degree-of-freedom joints that properly handles the non-commutativity of rotations. We validate our approach through digital twin experiments where reinforcement learning policies control the multi-rigid-body hand to replay captured human demonstrations. Quantitative evaluation shows sub-centimeter reconstruction error and successful grasp execution across diverse manipulation tasks.
comment: 10 pages, 4 figures. Accepted at ICBSR'25 (International Conference on Biomechanical Systems and Robotics)
☆ Model Predictive Control for Cooperative Docking Between Autonomous Surface Vehicles with Disturbance Rejection
Uncrewed Surface Vehicles (USVs) are a popular and efficient type of marine craft that find application in a large number of water-based tasks. When multiple USVs operate in the same area, they may be required to dock to each other to perform a shared task. Existing approaches for the docking between autonomous USVs generally consider one USV as a stationary target, while the second one is tasked to reach the required docking pose. In this work, we propose a cooperative approach for USV-USV docking, where two USVs work together to dock at an agreed location. We use a centralized Model Predictive Control (MPC) approach to solve the control problem, obtaining feasible trajectories that also guarantee constraint satisfaction. Owing to its model-based nature, this approach allows the rejection of disturbances, inclusive of exogenous inputs, by anticipating their effect on the USVs through the MPC prediction model. This is particularly effective in case of almost-stationary disturbances such as water currents. In simulations, we demonstrate how the proposed approach allows for a faster and more efficient docking with respect to existing approaches.
comment: 7 pages, 4 figures, submitted to IFAC World Congress 2026
☆ Efficient Computation of a Continuous Topological Model of the Configuration Space of Tethered Mobile Robots
Despite the attention that the problem of path planning for tethered robots has garnered in the past few decades, the approaches proposed to solve it typically rely on a discrete representation of the configuration space and do not exploit a model that can simultaneously capture the topological information of the tether and the continuous location of the robot. In this work, we explicitly build a topological model of the configuration space of a tethered robot starting from a polygonal representation of the workspace where the robot moves. To do so, we first establish a link between the configuration space of the tethered robot and the universal covering space of the workspace, and then we exploit this link to develop an algorithm to compute a simplicial complex model of the configuration space. We show how this approach improves the performances of existing algorithms that build other types of representations of the configuration space. The proposed model can be computed in a fraction of the time required to build traditional homotopy-augmented graphs, and is continuous, allowing to solve the path planning task for tethered robots using a broad set of path planning algorithms.
comment: 7 pages, 3 figures, submitted to IFAC World Congress 2026
☆ SINRL: Socially Integrated Navigation with Reinforcement Learning using Spiking Neural Networks
Integrating autonomous mobile robots into human environments requires human-like decision-making and energy-efficient, event-based computation. Despite progress, neuromorphic methods are rarely applied to Deep Reinforcement Learning (DRL) navigation approaches due to unstable training. We address this gap with a hybrid socially integrated DRL actor-critic approach that combines Spiking Neural Networks (SNNs) in the actor with Artificial Neural Networks (ANNs) in the critic and a neuromorphic feature extractor to capture temporal crowd dynamics and human-robot interactions. Our approach enhances social navigation performance and reduces estimated energy consumption by approximately 1.69 orders of magnitude.
comment: 8 pages, 6 figures
☆ Spatiotemporal Calibration and Ground Truth Estimation for High-Precision SLAM Benchmarking in Extended Reality
Simultaneous localization and mapping (SLAM) plays a fundamental role in extended reality (XR) applications. As the standards for immersion in XR continue to increase, the demands for SLAM benchmarking have become more stringent. Trajectory accuracy is the key metric, and marker-based optical motion capture (MoCap) systems are widely used to generate ground truth (GT) because of their drift-free and relatively accurate measurements. However, the precision of MoCap-based GT is limited by two factors: the spatiotemporal calibration with the device under test (DUT) and the inherent jitter in the MoCap measurements. These limitations hinder accurate SLAM benchmarking, particularly for key metrics like rotation error and inter-frame jitter, which are critical for immersive XR experiences. This paper presents a novel continuous-time maximum likelihood estimator to address these challenges. The proposed method integrates auxiliary inertial measurement unit (IMU) data to compensate for MoCap jitter. Additionally, a variable time synchronization method and a pose residual based on screw congruence constraints are proposed, enabling precise spatiotemporal calibration across multiple sensors and the DUT. Experimental results demonstrate that our approach outperforms existing methods, achieving the precision necessary for comprehensive benchmarking of state-of-the-art SLAM algorithms in XR applications. Furthermore, we thoroughly validate the practicality of our method by benchmarking several leading XR devices and open-source SLAM algorithms. The code is publicly available at https://github.com/ylab-xrpg/xr-hpgt.
☆ Characterizing Lane-Changing Behavior in Mixed Traffic
Characterizing and understanding lane-changing behavior in the presence of automated vehicles (AVs) is crucial to ensuring safety and efficiency in mixed traffic. Accordingly, this study aims to characterize the interactions between the lane-changing vehicle (active vehicle) and the vehicle directly impacted by the maneuver in the target lane (passive vehicle). Utilizing real-world trajectory data from the Waymo Open Motion Dataset (WOMD), this study explores patterns in lane-changing behavior and provides insight into how these behaviors evolve under different AV market penetration rates (MPRs). In particular, we propose a game-theoretic framework to analyze cooperative and defective behaviors in mixed traffic, applied to the 7,636 observed lane-changing events in the WOMD. First, we utilize k-means clustering to classify vehicles as cooperative or defective, revealing that the proportions of cooperative AVs are higher than those of HDVs in both active and passive roles. Next, we jointly estimate the utilities of active and passive vehicles to model their behaviors using the quantal response equilibrium framework. Empirical payoff tables are then constructed based on these utilities. Using these payoffs, we analyze the presence of social dilemmas and examine the evolution of cooperative behaviors using evolutionary game theory. Our results reveal the presence of social dilemmas in approximately 4% and 11% of lane-changing events for active and passive vehicles, respectively, with most classified as Stag Hunt or Prisoner's Dilemma (Chicken Game rarely observed). Moreover, the Monte Carlo simulation results show that repeated lane-changing interactions consistently lead to increased cooperative behavior over time, regardless of the AV penetration rate.
☆ Using Vision-Language Models as Proxies for Social Intelligence in Human-Robot Interaction
Robots operating in everyday environments must often decide when and whether to engage with people, yet such decisions often hinge on subtle nonverbal cues that unfold over time and are difficult to model explicitly. Drawing on a five-day Wizard-of-Oz deployment of a mobile service robot in a university cafe, we analyze how people signal interaction readiness through nonverbal behaviors and how expert wizards use these cues to guide engagement. Motivated by these observations, we propose a two-stage pipeline in which lightweight perceptual detectors (gaze shifts and proxemics) are used to selectively trigger heavier video-based vision-language model (VLM) queries at socially meaningful moments. We evaluate this pipeline on replayed field interactions and compare two prompting strategies. Our findings suggest that selectively using VLMs as proxies for social reasoning enables socially responsive robot behavior, allowing robots to act appropriately by attending to the cues people naturally provide in real-world interactions.
☆ Time-Varying Formation Tracking Control of Wheeled Mobile Robots With Region Constraint: A Generalized Udwadia-Kalaba Framework
In this paper, the time-varying formation tracking control of wheeled mobile robots with region constraint is investigated from a generalized Udwadia-Kalaba framework. The communication topology is directed, weighted and has a spanning tree with the leader being the root. By reformulating the time-varying formation tracking control objective as a constrained equation and transforming the region constraint by a diffeomorphism, the time-varying formation tracking controller with the region constraint is designed under the generalized Udwadia-Kalaba framework. Compared with the existing works on time-varying formation tracking control, the region constraint is takeninto account in this paper, which ensures the safety of the robots.Finally, some numerical simulations are presented to illustrate the effectiveness of the proposed control strategy.
comment: 10 pages,9 figures
☆ Mimir: Hierarchical Goal-Driven Diffusion with Uncertainty Propagation for End-to-End Autonomous Driving
End-to-end autonomous driving has emerged as a pivotal direction in the field of autonomous systems. Recent works have demonstrated impressive performance by incorporating high-level guidance signals to steer low-level trajectory planners. However, their potential is often constrained by inaccurate high-level guidance and the computational overhead of complex guidance modules. To address these limitations, we propose Mimir, a novel hierarchical dual-system framework capable of generating robust trajectories relying on goal points with uncertainty estimation: (1) Unlike previous approaches that deterministically model, we estimate goal point uncertainty with a Laplace distribution to enhance robustness; (2) To overcome the slow inference speed of the guidance system, we introduce a multi-rate guidance mechanism that predicts extended goal points in advance. Validated on challenging Navhard and Navtest benchmarks, Mimir surpasses previous state-of-the-art methods with a 20% improvement in the driving score EPDMS, while achieving 1.6 times improvement in high-level module inference speed without compromising accuracy. The code and models will be released soon to promote reproducibility and further development. The code is available at https://github.com/ZebinX/Mimir-Uncertainty-Driving
☆ Surrogate compliance modeling enables reinforcement learned locomotion gaits for soft robots
Adaptive morphogenetic robots adapt their morphology and control policies to meet changing tasks and environmental conditions. Many such systems leverage soft components, which enable shape morphing but also introduce simulation and control challenges. Soft-body simulators remain limited in accuracy and computational tractability, while rigid-body simulators cannot capture soft-material dynamics. Here, we present a surrogate compliance modeling approach: rather than explicitly modeling soft-body physics, we introduce indirect variables representing soft-material deformation within a rigid-body simulator. We validate this approach using our amphibious robotic turtle, a quadruped with soft morphing limbs designed for multi-environment locomotion. By capturing deformation effects as changes in effective limb length and limb center of mass, and by applying reinforcement learning with extensive randomization of these indirect variables, we achieve reliable policy learning entirely in a rigid-body simulation. The resulting gaits transfer directly to hardware, demonstrating high-fidelity sim-to-real performance on hard, flat substrates and robust, though lower-fidelity, transfer on rheologically complex terrains. The learned closed-loop gaits exhibit unprecedented terrestrial maneuverability and achieve an order-of-magnitude reduction in cost of transport compared to open-loop baselines. Field experiments with the robot further demonstrate stable, multi-gait locomotion across diverse natural terrains, including gravel, grass, and mud.
☆ A Flexible Funnel-Shaped Robotic Hand with an Integrated Single-Sheet Valve for Milligram-Scale Powder Handling
Laboratory Automation (LA) has the potential to accelerate solid-state materials discovery by enabling continuous robotic operation without human intervention. While robotic systems have been developed for tasks such as powder grinding and X-ray diffraction (XRD) analysis, fully automating powder handling at the milligram scale remains a significant challenge due to the complex flow dynamics of powders and the diversity of laboratory tasks. To address this challenge, this study proposes a novel, funnel-shaped, flexible robotic hand that preserves the softness and conical sheet designs in prior work while incorporating a controllable valve at the cone apex to enable precise, incremental dispensing of milligram-scale powder quantities. The hand is integrated with an external balance through a feedback control system based on a model of powder flow and online parameter identification. Experimental evaluations with glass beads, monosodium glutamate, and titanium dioxide demonstrated that 80% of the trials achieved an error within 2 mg, and the maximum error observed was approximately 20 mg across a target range of 20 mg to 3 g. In addition, by incorporating flow prediction models commonly used for hoppers and performing online parameter identification, the system is able to adapt to variations in powder dynamics. Compared to direct PID control, the proposed model-based control significantly improved both accuracy and convergence speed. These results highlight the potential of the proposed system to enable efficient and flexible powder weighing, with scalability toward larger quantities and applicability to a broad range of laboratory automation tasks.
comment: 9 pages, 8 figures
☆ An Introduction to Deep Reinforcement and Imitation Learning
Embodied agents, such as robots and virtual characters, must continuously select actions to execute tasks effectively, solving complex sequential decision-making problems. Given the difficulty of designing such controllers manually, learning-based approaches have emerged as promising alternatives, most notably Deep Reinforcement Learning (DRL) and Deep Imitation Learning (DIL). DRL leverages reward signals to optimize behavior, while DIL uses expert demonstrations to guide learning. This document introduces DRL and DIL in the context of embodied agents, adopting a concise, depth-first approach to the literature. It is self-contained, presenting all necessary mathematical and machine learning concepts as they are needed. It is not intended as a survey of the field; rather, it focuses on a small set of foundational algorithms and techniques, prioritizing in-depth understanding over broad coverage. The material ranges from Markov Decision Processes to REINFORCE and Proximal Policy Optimization (PPO) for DRL, and from Behavioral Cloning to Dataset Aggregation (DAgger) and Generative Adversarial Imitation Learning (GAIL) for DIL.
☆ Optimized Area Coverage in Disaster Response Utilizing Autonomous UAV Swarm Formations
This paper presents a UAV swarm system designed to assist first responders in disaster scenarios like wildfires. By distributing sensors across multiple agents, the system extends flight duration and enhances data availability, reducing the risk of mission failure due to collisions. To mitigate this risk further, we introduce an autonomous navigation framework that utilizes a local Euclidean Signed Distance Field (ESDF) map for obstacle avoidance while maintaining swarm formation with minimal path deviation. Additionally, we incorporate a Traveling Salesman Problem (TSP) variant to optimize area coverage, prioritizing Points of Interest (POIs) based on preassigned values derived from environmental behavior and critical infrastructure. The proposed system is validated through simulations with varying swarm sizes, demonstrating its ability to maximize coverage while ensuring collision avoidance between UAVs and obstacles.
☆ DIJIT: A Robotic Head for an Active Observer
We present DIJIT, a novel binocular robotic head expressly designed for mobile agents that behave as active observers. DIJIT's unique breadth of functionality enables active vision research and the study of human-like eye and head-neck motions, their interrelationships, and how each contributes to visual ability. DIJIT is also being used to explore the differences between how human vision employs eye/head movements to solve visual tasks and current computer vision methods. DIJIT's design features nine mechanical degrees of freedom, while the cameras and lenses provide an additional four optical degrees of freedom. The ranges and speeds of the mechanical design are comparable to human performance. Our design includes the ranges of motion required for convergent stereo, namely, vergence, version, and cyclotorsion. The exploration of the utility of these to both human and machine vision is ongoing. Here, we present the design of DIJIT and evaluate aspects of its performance. We present a new method for saccadic camera movements. In this method, a direct relationship between camera orientation and motor values is developed. The resulting saccadic camera movements are close to human movements in terms of their accuracy.
☆ VLD: Visual Language Goal Distance for Reinforcement Learning Navigation
Training end-to-end policies from image data to directly predict navigation actions for robotic systems has proven inherently difficult. Existing approaches often suffer from either the sim-to-real gap during policy transfer or a limited amount of training data with action labels. To address this problem, we introduce Vision-Language Distance (VLD) learning, a scalable framework for goal-conditioned navigation that decouples perception learning from policy learning. Instead of relying on raw sensory inputs during policy training, we first train a self-supervised distance-to-goal predictor on internet-scale video data. This predictor generalizes across both image- and text-based goals, providing a distance signal that can be minimized by a reinforcement learning (RL) policy. The RL policy can be trained entirely in simulation using privileged geometric distance signals, with injected noise to mimic the uncertainty of the trained distance predictor. At deployment, the policy consumes VLD predictions, inheriting semantic goal information-"where to go"-from large-scale visual training while retaining the robust low-level navigation behaviors learned in simulation. We propose using ordinal consistency to assess distance functions directly and demonstrate that VLD outperforms prior temporal distance approaches, such as ViNT and VIP. Experiments show that our decoupled design achieves competitive navigation performance in simulation while supporting flexible goal modalities, providing an alternative and, most importantly, scalable path toward reliable, multimodal navigation policies.
☆ Sparse Variable Projection in Robotic Perception: Exploiting Separable Structure for Efficient Nonlinear Optimization
Robotic perception often requires solving large nonlinear least-squares (NLS) problems. While sparsity has been well-exploited to scale solvers, a complementary and underexploited structure is \emph{separability} -- where some variables (e.g., visual landmarks) appear linearly in the residuals and, for any estimate of the remaining variables (e.g., poses), have a closed-form solution. Variable projection (VarPro) methods are a family of techniques that exploit this structure by analytically eliminating the linear variables and presenting a reduced problem in the remaining variables that has favorable properties. However, VarPro has seen limited use in robotic perception; a major challenge arises from gauge symmetries (e.g., cost invariance to global shifts and rotations), which are common in perception and induce specific computational challenges in standard VarPro approaches. We present a VarPro scheme designed for problems with gauge symmetries that jointly exploits separability and sparsity. Our method can be applied as a one-time preprocessing step to construct a \emph{matrix-free Schur complement operator}. This operator allows efficient evaluation of costs, gradients, and Hessian-vector products of the reduced problem and readily integrates with standard iterative NLS solvers. We provide precise conditions under which our method applies, and describe extensions when these conditions are only partially met. Across synthetic and real benchmarks in SLAM, SNL, and SfM, our approach achieves up to \textbf{2$\times$--35$\times$ faster runtimes} than state-of-the-art methods while maintaining accuracy. We release an open-source C++ implementation and all datasets from our experiments.
comment: 8 pages, submitted for review
♻ ☆ MAPLE: Encoding Dexterous Robotic Manipulation Priors Learned From Egocentric Videos
Large-scale egocentric video datasets capture diverse human activities across a wide range of scenarios, offering rich and detailed insights into how humans interact with objects, especially those that require fine-grained dexterous control. Such complex, dexterous skills with precise controls are crucial for many robotic manipulation tasks, yet are often insufficiently addressed by traditional data-driven approaches to robotic manipulation. To address this gap, we leverage manipulation priors learned from large-scale egocentric video datasets to improve policy learning for dexterous robotic manipulation tasks. We present MAPLE, a novel method for dexterous robotic manipulation that learns features to predict object contact points and detailed hand poses at the moment of contact from egocentric images. We then use the learned features to train policies for downstream manipulation tasks. Experimental results demonstrate the effectiveness of MAPLE across 4 existing simulation benchmarks, as well as a newly designed set of 4 challenging simulation tasks requiring fine-grained object control and complex dexterous skills. The benefits of MAPLE are further highlighted in real-world experiments using a 17 DoF dexterous robotic hand, whereas the simultaneous evaluation across both simulation and real-world experiments has remained underexplored in prior work. We additionally showcase the efficacy of our model on an egocentric contact point prediction task, validating its usefulness beyond dexterous manipulation policy learning.
♻ ☆ Incremental Generalized Hybrid A*
We address the problem of efficiently organizing search over very large trees, which arises in many applications ranging from autonomous driving to aerial vehicles. Here, we are motivated by off-road autonomy, where real-time planning is essential. Classical approaches use graphs of motion primitives and exploit dominance to mitigate the curse of dimensionality and prune expansions efficiently. However, for complex dynamics, repeatedly solving two-point boundary-value problems makes graph construction too slow for fast kinodynamic planning. Hybrid A* (HA*) addressed this challenge by searching over a tree of motion primitives and introducing approximate pruning using a grid-based dominance check. However, choosing the grid resolution is difficult: too coarse risks failure, while too fine leads to excessive expansions and slow planning. We propose Incremental Generalized Hybrid A* (IGHA*), an anytime tree-search framework that dynamically organizes vertex expansions without rigid pruning. IGHA* provably matches or outperforms HA*. For both on-road kinematic and off-road kinodynamic planning queries for a car-like robot, variants of IGHA* use 6x fewer expansions to the best solution compared to an optimized version of HA* (HA*M, an internal baseline). In simulated off- road experiments in a high-fidelity simulator, IGHA* outper- forms HA*M when both are used in the loop with a model predictive controller. We demonstrate real-time performance both in simulation and on a small-scale off-road vehicle, enabling fast, robust planning under complex dynamics. Website: https: //personalrobotics.github.io/IGHAStar/
comment: 8 pages, 7 figures, Accepted to IEEE RA-L, Nov 2025
♻ ☆ PosA-VLA: Enhancing Action Generation via Pose-Conditioned Anchor Attention
The Vision-Language-Action (VLA) models have demonstrated remarkable performance on embodied tasks and shown promising potential for real-world applications. However, current VLAs still struggle to produce consistent and precise target-oriented actions, as they often generate redundant or unstable motions along trajectories, limiting their applicability in time-sensitive scenarios.In this work, we attribute these redundant actions to the spatially uniform perception field of existing VLAs, which causes them to be distracted by target-irrelevant objects, especially in complex environments.To address this issue, we propose an efficient PosA-VLA framework that anchors visual attention via pose-conditioned supervision, consistently guiding the model's perception toward task-relevant regions. The pose-conditioned anchor attention mechanism enables the model to better align instruction semantics with actionable visual cues, thereby improving action generation precision and efficiency. Moreover, our framework adopts a lightweight architecture and requires no auxiliary perception modules (e.g., segmentation or grounding networks), ensuring efficient inference. Extensive experiments verify that our method executes embodied tasks with precise and time-efficient behavior across diverse robotic manipulation benchmarks and shows robust generalization in a variety of challenging environments.
♻ ☆ MM-ACT: Learn from Multimodal Parallel Generation to Act
A generalist robotic policy needs both semantic understanding for task planning and the ability to interact with the environment through predictive capabilities. To tackle this, we present MM-ACT, a unified Vision-Language-Action (VLA) model that integrates text, image, and action in shared token space and performs generation across all three modalities. MM-ACT adopts a re-mask parallel decoding strategy for text and image generation, and employs a one-step parallel decoding strategy for action generation to improve efficiency. We introduce Context-Shared Multimodal Learning, a unified training paradigm that supervises generation in all three modalities from a shared context, enhancing action generation through cross-modal learning. Experiments were conducted on the LIBERO simulation and Franka real-robot setups as well as RoboTwin2.0 to assess in-domain and out-of-domain performances respectively. Our approach achieves a success rate of 96.3% on LIBERO, 72.0% across three tasks of real Franka, and 52.38% across eight bimanual tasks of RoboTwin2.0 with an additional gain of 9.25% from cross-modal learning. We release our codes, models and data at https://github.com/HHYHRHY/MM-ACT.
comment: 17 pages
♻ ☆ Exploring Adversarial Obstacle Attacks in Search-based Path Planning for Autonomous Mobile Robots
Path planning algorithms, such as the search-based A*, are a critical component of autonomous mobile robotics, enabling robots to navigate from a starting point to a destination efficiently and safely. We investigated the resilience of the A* algorithm in the face of potential adversarial interventions known as obstacle attacks. The adversary's goal is to delay the robot's timely arrival at its destination by introducing obstacles along its original path. We developed malicious software to execute the attacks and conducted experiments to assess their impact, both in simulation using TurtleBot in Gazebo and in real-world deployment with the Unitree Go1 robot. In simulation, the attacks resulted in an average delay of 36\%, with the most significant delays occurring in scenarios where the robot was forced to take substantially longer alternative paths. In real-world experiments, the delays were even more pronounced, with all attacks successfully rerouting the robot and causing measurable disruptions. These results highlight that the algorithm's robustness is not solely an attribute of its design but is significantly influenced by the operational environment. For example, in constrained environments like tunnels, the delays were maximized due to the limited availability of alternative routes.
♻ ☆ Quantization-Free Autoregressive Action Transformer
Current transformer-based imitation learning approaches introduce discrete action representations and train an autoregressive transformer decoder on the resulting latent code. However, the initial quantization breaks the continuous structure of the action space thereby limiting the capabilities of the generative model. We propose a quantization-free method instead that leverages Generative Infinite-Vocabulary Transformers (GIVT) as a direct, continuous policy parametrization for autoregressive transformers. This simplifies the imitation learning pipeline while achieving state-of-the-art performance on a variety of popular simulated robotics tasks. We enhance our policy roll-outs by carefully studying sampling algorithms, further improving the results.
♻ ☆ SwarmDiffusion: End-To-End Traversability-Guided Diffusion for Embodiment-Agnostic Navigation of Heterogeneous Robots
Visual traversability estimation is critical for autonomous navigation, but existing VLM-based methods rely on hand-crafted prompts, generalize poorly across embodiments, and output only traversability maps, leaving trajectory generation to slow external planners. We propose SwarmDiffusion, a lightweight end-to-end diffusion model that jointly predicts traversability and generates a feasible trajectory from a single RGB image. To remove the need for annotated or planner-produced paths, we introduce a planner-free trajectory construction pipeline based on randomized waypoint sampling, Bezier smoothing, and regularization enforcing connectivity, safety, directionality, and path thinness. This enables learning stable motion priors without demonstrations. SwarmDiffusion leverages VLM-derived supervision without prompt engineering and conditions the diffusion process on a compact embodiment state, producing physically consistent, traversable paths that transfer across different robot platforms. Across indoor environments and two embodiments (quadruped and aerial), the method achieves 80-100% navigation success and 0.09s inference, and adapts to a new robot using only-500 additional visual samples. It generalizes reliably to unseen environments in simulation and real-world trials, offering a scalable, prompt-free approach to unified traversability reasoning and trajectory generation.
comment: This work has been submitted for publication and is currently under review
♻ ☆ RealD$^2$iff: Bridging Real-World Gap in Robot Manipulation via Depth Diffusion
Robot manipulation in the real world is fundamentally constrained by the visual sim2real gap, where depth observations collected in simulation fail to reflect the complex noise patterns inherent to real sensors. In this work, inspired by the denoising capability of diffusion models, we invert the conventional perspective and propose a clean-to-noisy paradigm that learns to synthesize noisy depth, thereby bridging the visual sim2real gap through purely simulation-driven robotic learning. Building on this idea, we introduce RealD$^2$iff, a hierarchical coarse-to-fine diffusion framework that decomposes depth noise into global structural distortions and fine-grained local perturbations. To enable progressive learning of these components, we further develop two complementary strategies: Frequency-Guided Supervision (FGS) for global structure modeling and Discrepancy-Guided Optimization (DGO) for localized refinement. To integrate RealD$^2$iff seamlessly into imitation learning, we construct a pipeline that spans six stages. We provide comprehensive empirical and experimental validation demonstrating the effectiveness of this paradigm. RealD$^2$iff enables two key applications: (1) generating real-world-like depth to construct clean-noisy paired datasets without manual sensor data collection. (2) Achieving zero-shot sim2real robot manipulation, substantially improving real-world performance without additional fine-tuning.
comment: We are the author team of the paper "RealD$^2$iff: Bridging Real-World Gap in Robot Manipulation via Depth Diffusion". After self-examination, our team discovered inappropriate wording in the citation of related work, the introduction, and the contribution statement, which may affect the contribution of other related works. Therefore, we have decided to revise the paper and request its withdrawal
♻ ☆ LOG-Nav: Efficient Layout-Aware Object-Goal Navigation with Hierarchical Planning
We introduce LOG-Nav, an efficient layout-aware object-goal navigation approach designed for complex multi-room indoor environments. By planning hierarchically leveraging a global topologigal map with layout information and local imperative approach with detailed scene representation memory, LOG-Nav achieves both efficient and effective navigation. The process is managed by an LLM-powered agent, ensuring seamless effective planning and navigation, without the need for human interaction, complex rewards, or costly training. Our experimental results on the MP3D benchmark achieves 85\% object navigation success rate (SR) and 79\% success rate weighted by path length (SPL) (over 40\% point improvement in SR and 60\% improvement in SPL compared to exsisting methods). Furthermore, we validate the robustness of our approach through virtual agent and real-world robotic deployment, showcasing its capability in practical scenarios.
♻ ☆ Enhanced Spatiotemporal Consistency for Image-to-LiDAR Data Pretraining
LiDAR representation learning has emerged as a promising approach to reducing reliance on costly and labor-intensive human annotations. While existing methods primarily focus on spatial alignment between LiDAR and camera sensors, they often overlook the temporal dynamics critical for capturing motion and scene continuity in driving scenarios. To address this limitation, we propose SuperFlow++, a novel framework that integrates spatiotemporal cues in both pretraining and downstream tasks using consecutive LiDAR-camera pairs. SuperFlow++ introduces four key components: (1) a view consistency alignment module to unify semantic information across camera views, (2) a dense-to-sparse consistency regularization mechanism to enhance feature robustness across varying point cloud densities, (3) a flow-based contrastive learning approach that models temporal relationships for improved scene understanding, and (4) a temporal voting strategy that propagates semantic information across LiDAR scans to improve prediction consistency. Extensive evaluations on 11 heterogeneous LiDAR datasets demonstrate that SuperFlow++ outperforms state-of-the-art methods across diverse tasks and driving conditions. Furthermore, by scaling both 2D and 3D backbones during pretraining, we uncover emergent properties that provide deeper insights into developing scalable 3D foundation models. With strong generalizability and computational efficiency, SuperFlow++ establishes a new benchmark for data-efficient LiDAR-based perception in autonomous driving. The code is publicly available at https://github.com/Xiangxu-0103/SuperFlow
comment: IEEE Transactions on Pattern Analysis and Machine Intelligence
♻ ☆ Building Gradient by Gradient: Decentralised Energy Functions for Bimanual Robot Assembly
There are many challenges in bimanual assembly, including high-level sequencing, multi-robot coordination, and low-level, contact-rich operations such as component mating. Task and motion planning (TAMP) methods, while effective in this domain, may be prohibitively slow to converge when adapting to disturbances that require new task sequencing and optimisation. These events are common during tight-tolerance assembly, where difficult-to-model dynamics such as friction or deformation require rapid replanning and reattempts. Moreover, defining explicit task sequences for assembly can be cumbersome, limiting flexibility when task replanning is required. To simplify this planning, we introduce a decentralised gradient-based framework that uses a piecewise continuous energy function through the automatic composition of adaptive potential functions. This approach generates sub-goals using only myopic optimisation, rather than long-horizon planning. It demonstrates effectiveness at solving long-horizon tasks due to the structure and adaptivity of the energy function. We show that our approach scales to physical bimanual assembly tasks for constructing tight-tolerance assemblies. In these experiments, we discover that our gradient-based rapid replanning framework generates automatic retries, coordinated motions and autonomous handovers in an emergent fashion.
comment: 8 pages, 7 figures, 1 table
♻ ☆ Switch-JustDance: Benchmarking Whole Body Motion Tracking Policies Using a Commercial Console Game
Recent advances in whole-body robot control have enabled humanoid and legged robots to perform increasingly agile and coordinated motions. However, standardized benchmarks for evaluating these capabilities in real-world settings, and in direct comparison to humans, remain scarce. Existing evaluations often rely on pre-collected human motion datasets or simulation-based experiments, which limit reproducibility, overlook hardware factors, and hinder fair human-robot comparisons. We present Switch-JustDance, a low-cost and reproducible benchmarking pipeline that leverages motion-sensing console games, Just Dance on the Nintendo Switch, to evaluate robot whole-body control. Using Just Dance on the Nintendo Switch as a representative platform, Switch-JustDance converts in-game choreography into robot-executable motions through streaming, motion reconstruction, and motion retargeting modules and enables users to evaluate controller performance through the game's built-in scoring system. We first validate the evaluation properties of Just Dance, analyzing its reliability, validity, sensitivity, and potential sources of bias. Our results show that the platform provides consistent and interpretable performance measures, making it a suitable tool for benchmarking embodied AI. Building on this foundation, we benchmark three state-of-the-art humanoid whole-body controllers on hardware and provide insights into their relative strengths and limitations.
♻ ☆ FASTer: Toward Efficient Autoregressive Vision Language Action Modeling via Neural Action Tokenization
Autoregressive vision-language-action (VLA) models have recently demonstrated strong capabilities in robotic manipulation. However, their core process of action tokenization often involves a trade-off between reconstruction fidelity and inference efficiency. We introduce FASTer, a unified framework for efficient and generalizable robot learning that integrates a learnable tokenizer with an autoregressive policy built upon it. FASTerVQ encodes action chunks as single-channel images, capturing global spatio-temporal dependencies while maintaining a high compression ratio. FASTerVLA builds on this tokenizer with block-wise autoregressive decoding and a lightweight action expert, achieving both faster inference and higher task performance. Extensive experiments across simulated and real-world benchmarks show that FASTerVQ delivers superior reconstruction quality, high token utilization, and strong cross-task and cross-embodiment generalization, while FASTerVLA further improves overall capability, surpassing previous state-of-the-art VLA models in both inference speed and task performance.
♻ ☆ Pretraining in Actor-Critic Reinforcement Learning for Robot Locomotion
The pretraining-finetuning paradigm has facilitated numerous transformative advancements in artificial intelligence research in recent years. However, in the domain of reinforcement learning (RL) for robot locomotion, individual skills are often learned from scratch despite the high likelihood that some generalizable knowledge is shared across all task-specific policies belonging to the same robot embodiment. This work aims to define a paradigm for pretraining neural network models that encapsulate such knowledge and can subsequently serve as a basis for warm-starting the RL process in classic actor-critic algorithms, such as Proximal Policy Optimization (PPO). We begin with a task-agnostic exploration-based data collection algorithm to gather diverse, dynamic transition data, which is then used to train a Proprioceptive Inverse Dynamics Model (PIDM) through supervised learning. The pretrained weights are then loaded into both the actor and critic networks to warm-start the policy optimization of actual tasks. We systematically validated our proposed method with 9 distinct robot locomotion RL environments comprising 3 different robot embodiments, showing significant benefits of this initialization strategy. Our proposed approach on average improves sample efficiency by 36.9% and task performance by 7.3% compared to random initialization. We further present key ablation studies and empirical analyses that shed light on the mechanisms behind the effectiveness of this method.
♻ ☆ An Open-Source Soft Robotic Platform for Autonomous Aerial Manipulation in the Wild
Aerial manipulation combines the versatility and speed of flying platforms with the functional capabilities of mobile manipulation, which presents significant challenges due to the need for precise localization and control. Traditionally, researchers have relied on offboard perception systems, which are limited to expensive and impractical specially equipped indoor environments. In this work, we introduce a novel platform for autonomous aerial manipulation that exclusively utilizes onboard perception systems. Our platform can perform aerial manipulation in various indoor and outdoor environments without depending on external perception systems. Our experimental results demonstrate the platform's ability to autonomously grasp various objects in diverse settings. This advancement significantly improves the scalability and practicality of aerial manipulation applications by eliminating the need for costly tracking solutions. To accelerate future research, we open source our ROS 2 software stack and custom hardware design, making our contributions accessible to the broader research community.
comment: Project website: https://sites.google.com/view/open-source-soft-platform/open-source-soft-robotic-platform GitHub: https://github.com/raptor-ethz
♻ ☆ Disturbance Compensation for Safe Kinematic Control of Robotic Systems with Closed Architecture
In commercial robotic systems, it is common to encounter a closed inner-loop torque controller that is not user-modifiable. However, the outer-loop controller, which sends kinematic commands such as position or velocity for the inner-loop controller to track, is typically exposed to users. In this work, we focus on the development of an easily integrated add-on at the outer-loop layer by combining disturbance rejection control and robust control barrier function for high-performance tracking and safe control of the whole dynamic system of an industrial manipulator. This is particularly beneficial when 1) the inner-loop controller is imperfect, unmodifiable, and uncertain; and 2) the dynamic model exhibits significant uncertainty. Stability analysis, formal safety guarantee proof, and hardware experiments with a PUMA robotic manipulator are presented. Our solution demonstrates superior performance in terms of simplicity of implementation, robustness, tracking precision, and safety compared to the state of the art. Video: https://youtu.be/zw1tanvrV8Q
comment: Extended version of the paper submitted for publication. This document contains detailed mathematical derivations and additional experimental results omitted from the submission due to page constraints
♻ ☆ BEDI: A Comprehensive Benchmark for Evaluating Embodied Agents on UAVs
With the rapid advancement of low-altitude remote sensing and Vision-Language Models (VLMs), Embodied Agents based on Unmanned Aerial Vehicles (UAVs) have shown significant potential in autonomous tasks. However, current evaluation methods for UAV-Embodied Agents (UAV-EAs) remain constrained by the lack of standardized benchmarks, diverse testing scenarios and open system interfaces. To address these challenges, we propose BEDI (Benchmark for Embodied Drone Intelligence), a systematic and standardized benchmark designed for evaluating UAV-EAs. Specifically, we introduce a novel Dynamic Chain-of-Embodied-Task paradigm based on the perception-decision-action loop, which decomposes complex UAV tasks into standardized, measurable subtasks. Building on this paradigm, we design a unified evaluation framework encompassing six core sub-skills: semantic perception, spatial perception, motion control, tool utilization, task planning and action generation. Furthermore, we develop a hybrid testing platform that incorporates a wide range of both virtual and real-world scenarios, enabling a comprehensive evaluation of UAV-EAs across diverse contexts. The platform also offers open and standardized interfaces, allowing researchers to customize tasks and extend scenarios, thereby enhancing flexibility and scalability in the evaluation process. Finally, through empirical evaluations of several state-of-the-art (SOTA) VLMs, we reveal their limitations in embodied UAV tasks, underscoring the critical role of the BEDI benchmark in advancing embodied intelligence research and model optimization. By filling the gap in systematic and standardized evaluation within this field, BEDI facilitates objective model comparison and lays a robust foundation for future development in this field. Our benchmark is now publicly available at https://github.com/lostwolves/BEDI.
♻ ☆ First Responders' Perceptions of Semantic Information for Situational Awareness in Robot-Assisted Emergency Response
This study investigates First Responders' (FRs) attitudes toward the use of semantic information and Situational Awareness (SA) in robotic systems during emergency operations. A structured questionnaire was administered to 22 FRs across eight countries, capturing their demographic profiles, general attitudes toward robots, and experiences with semantics-enhanced SA. Results show that most FRs expressed positive attitudes toward robots, and rated the usefulness of semantic information for building SA at an average of 3.6 out of 5. Semantic information was also valued for its role in predicting unforeseen emergencies (mean 3.9). Participants reported requiring an average of 74.6\% accuracy to trust semantic outputs and 67.8\% for them to be considered useful, revealing a willingness to use imperfect but informative AI support tools. To the best of our knowledge, this study offers novel insights by being one of the first to directly survey FRs on semantic-based SA in a cross-national context. It reveals the types of semantic information most valued in the field, such as object identity, spatial relationships, and risk context-and connects these preferences to the respondents' roles, experience, and education levels. The findings also expose a critical gap between lab-based robotics capabilities and the realities of field deployment, highlighting the need for more meaningful collaboration between FRs and robotics researchers. These insights contribute to the development of more user-aligned and situationally aware robotic systems for emergency response.
♻ ☆ 2Fast-2Lamaa: Large-Scale Lidar-Inertial Localization and Mapping with Continuous Distance Fields
This paper introduces 2Fast-2Lamaa, a lidar-inertial state estimation framework for odometry, mapping, and localization. Its first key component is the optimization-based undistortion of lidar scans, which uses continuous IMU preintegration to model the system's pose at every lidar point timestamp. The continuous trajectory over 100-200ms is parameterized only by the initial scan conditions (linear velocity and gravity orientation) and IMU biases, yielding eleven state variables. These are estimated by minimizing point-to-line and point-to-plane distances between lidar-extracted features without relying on previous estimates, resulting in a prior-less motion-distortion correction strategy. Because the method performs local state estimation, it directly provides scan-to-scan odometry. To maintain geometric consistency over longer periods, undistorted scans are used for scan-to-map registration. The map representation employs Gaussian Processes to form a continuous distance field, enabling point-to-surface distance queries anywhere in space. Poses of the undistorted scans are refined by minimizing these distances through non-linear least-squares optimization. For odometry and mapping, the map is built incrementally in real time; for pure localization, existing maps are reused. The incremental map construction also includes mechanisms for removing dynamic objects. We benchmark 2Fast-2Lamaa on 250km (over 10h) of public and self-collected datasets from both automotive and handheld systems. The framework achieves state-of-the-art performance across diverse and challenging scenarios, reaching odometry and localization errors as low as 0.27% and 0.06 m, respectively. The real-time implementation is publicly available at https://github.com/clegenti/2fast2lamaa.
Robotics 34
☆ CERNet: Class-Embedding Predictive-Coding RNN for Unified Robot Motion, Recognition, and Confidence Estimation
Robots interacting with humans must not only generate learned movements in real-time, but also infer the intent behind observed behaviors and estimate the confidence of their own inferences. This paper proposes a unified model that achieves all three capabilities within a single hierarchical predictive-coding recurrent neural network (PC-RNN) equipped with a class embedding vector, CERNet, which leverages a dynamically updated class embedding vector to unify motor generation and recognition. The model operates in two modes: generation and inference. In the generation mode, the class embedding constrains the hidden state dynamics to a class-specific subspace; in the inference mode, it is optimized online to minimize prediction error, enabling real-time recognition. Validated on a humanoid robot across 26 kinesthetically taught alphabets, our hierarchical model achieves 76% lower trajectory reproduction error than a parameter-matched single-layer baseline, maintains motion fidelity under external perturbations, and infers the demonstrated trajectory class online with 68% Top-1 and 81% Top-2 accuracy. Furthermore, internal prediction errors naturally reflect the model's confidence in its recognition. This integration of robust generation, real-time recognition, and intrinsic uncertainty estimation within a compact PC-RNN framework offers a compact and extensible approach to motor memory in physical robots, with potential applications in intent-sensitive human-robot collaboration.
☆ A Hetero-Associative Sequential Memory Model Utilizing Neuromorphic Signals: Validated on a Mobile Manipulator
This paper presents a hetero-associative sequential memory system for mobile manipulators that learns compact, neuromorphic bindings between robot joint states and tactile observations to produce step-wise action decisions with low compute and memory cost. The method encodes joint angles via population place coding and converts skin-measured forces into spike-rate features using an Izhikevich neuron model; both signals are transformed into bipolar binary vectors and bound element-wise to create associations stored in a large-capacity sequential memory. To improve separability in binary space and inject geometry from touch, we introduce 3D rotary positional embeddings that rotate subspaces as a function of sensed force direction, enabling fuzzy retrieval through a softmax weighted recall over temporally shifted action patterns. On a Toyota Human Support Robot covered by robot skin, the hetero-associative sequential memory system realizes a pseudocompliance controller that moves the link under touch in the direction and with speed correlating to the amplitude of applied force, and it retrieves multi-joint grasp sequences by continuing tactile input. The system sets up quickly, trains from synchronized streams of states and observations, and exhibits a degree of generalization while remaining economical. Results demonstrate single-joint and full-arm behaviors executed via associative recall, and suggest extensions to imitation learning, motion planning, and multi-modal integration.
☆ Parametric Design of a Cable-Driven Coaxial Spherical Parallel Mechanism for Ultrasound Scans
Haptic interfaces play a critical role in medical teleoperation by enabling surgeons to interact with remote environments through realistic force and motion feedback. Achieving high fidelity in such systems requires balancing performance trade-off among workspace, dexterity, stiffness, inertia, and bandwidth, particularly in applications demanding pure rotational motion. This paper presents the design methodology and kinematic analysis of a Cable-Driven Coaxial Spherical Parallel Mechanism (CDC-SPM) developed to address these challenges. The proposed cable-driven interface design allows for reducing the mass placed at the robot arm end-effector, thereby minimizing inertial loads, enhancing stiffness, and improving dynamic responsiveness. Through parallel and coaxial actuation, the mechanism achieves decoupled rotational degrees of freedom with isotropic force and torque transmission. Simulation and analysis demonstrate that the CDC-SPM provides accurate, responsive, and safe motion characteristics suitable for high-precision haptic applications. These results highlight the mechanism's potential for medical teleoperation tasks such as ultrasound imaging, where precise and intuitive manipulation is essential.
☆ VideoVLA: Video Generators Can Be Generalizable Robot Manipulators
Generalization in robot manipulation is essential for deploying robots in open-world environments and advancing toward artificial general intelligence. While recent Vision-Language-Action (VLA) models leverage large pre-trained understanding models for perception and instruction following, their ability to generalize to novel tasks, objects, and settings remains limited. In this work, we present VideoVLA, a simple approach that explores the potential of transforming large video generation models into robotic VLA manipulators. Given a language instruction and an image, VideoVLA predicts an action sequence as well as the future visual outcomes. Built on a multi-modal Diffusion Transformer, VideoVLA jointly models video, language, and action modalities, using pre-trained video generative models for joint visual and action forecasting. Our experiments show that high-quality imagined futures correlate with reliable action predictions and task success, highlighting the importance of visual imagination in manipulation. VideoVLA demonstrates strong generalization, including imitating other embodiments' skills and handling novel objects. This dual-prediction strategy - forecasting both actions and their visual consequences - explores a paradigm shift in robot learning and unlocks generalization capabilities in manipulation systems.
comment: Project page: https://videovla-nips2025.github.io
☆ Task adaptation of Vision-Language-Action model: 1st Place Solution for the 2025 BEHAVIOR Challenge NeurIPS
We present a vision-action policy that won 1st place in the 2025 BEHAVIOR Challenge - a large-scale benchmark featuring 50 diverse long-horizon household tasks in photo-realistic simulation, requiring bimanual manipulation, navigation, and context-aware decision making. Building on the Pi0.5 architecture, we introduce several innovations. Our primary contribution is correlated noise for flow matching, which improves training efficiency and enables correlation-aware inpainting for smooth action sequences. We also apply learnable mixed-layer attention and System 2 stage tracking for ambiguity resolution. Training employs multi-sample flow matching to reduce variance, while inference uses action compression and challenge-specific correction rules. Our approach achieves 26% q-score across all 50 tasks on both public and private leaderboards.
comment: 2025 NeurIPS Behavior Challenge 1st place solution
☆ Interconnection and Damping Assignment Passivity-Based Control using Sparse Neural ODEs
Interconnection and Damping Assignment Passivity-Based Control (IDA-PBC) is a nonlinear control technique that assigns a port-Hamiltonian (pH) structure to a controlled system using a state-feedback law. While IDA-PBC has been extensively studied and applied to many systems, its practical implementation often remains confined to academic examples and, almost exclusively, to stabilization tasks. The main limitation of IDA-PBC stems from the complexity of analytically solving a set of partial differential equations (PDEs), referred to as the matching conditions, which enforce the pH structure of the closed-loop system. However, this is extremely challenging, especially for complex physical systems and tasks. In this work, we propose a novel numerical approach for designing IDA-PBC controllers without solving the matching PDEs exactly. We cast the IDA-PBC problem as the learning of a neural ordinary differential equation. In particular, we rely on sparse dictionary learning to parametrize the desired closed-loop system as a sparse linear combination of nonlinear state-dependent functions. Optimization of the controller parameters is achieved by solving a multi-objective optimization problem whose cost function is composed of a generic task-dependent cost and a matching condition-dependent cost. Our numerical results show that the proposed method enables (i) IDA-PBC to be applicable to complex tasks beyond stabilization, such as the discovery of periodic oscillatory behaviors, (ii) the derivation of closed-form expressions of the controlled system, including residual terms
☆ Energy-Efficient Navigation for Surface Vehicles in Vortical Flow Fields ICRA 2026
For centuries, khalasi have skillfully harnessed ocean currents to navigate vast waters with minimal effort. Emulating this intuition in autonomous systems remains a significant challenge, particularly for Autonomous Surface Vehicles tasked with long duration missions under strict energy budgets. In this work, we present a learning-based approach for energy-efficient surface vehicle navigation in vortical flow fields, where partial observability often undermines traditional path-planning methods. We present an end to end reinforcement learning framework based on Soft Actor Critic that learns flow-aware navigation policies using only local velocity measurements. Through extensive evaluation across diverse and dynamically rich scenarios, our method demonstrates substantial energy savings and robust generalization to previously unseen flow conditions, offering a promising path toward long term autonomy in ocean environments. The navigation paths generated by our proposed approach show an improvement in energy conservation 30 to 50 percent compared to the existing state of the art techniques.
comment: Under Review for International Conference on Robotics and Automation (ICRA 2026)
☆ Ground Compliance Improves Retention of Visual Feedback-Based Propulsion Training for Gait Rehabilitation
This study investigates whether adding ground compliance to visual feedback (VF) gait training is more effective at increasing push-off force (POF) compared to using VF alone, with implications for gait rehabilitation. Ten healthy participants walked on a custom split-belt treadmill. All participants received real-time visual feedback of their ground reaction forces. One group also experienced changes in ground compliance, while a control group received only visual feedback. Intentional increases in propulsive ground reaction forces (POF) were successfully achieved and sustained post-intervention, especially in the group that experienced ground compliance. This group also demonstrated lasting after-effects in muscle activity and joint kinematics, indicating a more robust learning of natural strategies to increase propulsion. This work demonstrates how visual and proprioceptive systems coordinate during gait adaptation. It uniquely shows that combining ground compliance with visual feedback enhances the learning of propulsive forces, supporting the potential use of compliant terrain in long-term rehabilitation targeting propulsion deficits, such as those following a stroke.
☆ Control of Powered Ankle-Foot Prostheses on Compliant Terrain: A Quantitative Approach to Stability Enhancement
Walking on compliant terrain presents a substantial challenge for individuals with lower-limb amputation, further elevating their already high risk of falling. While powered ankle-foot prostheses have demonstrated adaptability across speeds and rigid terrains, control strategies optimized for soft or compliant surfaces remain underexplored. This work experimentally validates an admittance-based control strategy that dynamically adjusts the quasi-stiffness of powered prostheses to enhance gait stability on compliant ground. Human subject experiments were conducted with three healthy individuals walking on two bilaterally compliant surfaces with ground stiffness values of 63 and 25 kN/m, representative of real-world soft environments. Controller performance was quantified using phase portraits and two walking stability metrics, offering a direct assessment of fall risk. Compared to a standard phase-variable controller developed for rigid terrain, the proposed admittance controller consistently improved gait stability across all compliant conditions. These results demonstrate the potential of adaptive, stability-aware prosthesis control to reduce fall risk in real-world environments and advance the robustness of human-prosthesis interaction in rehabilitation robotics.
☆ From Zero to High-Speed Racing: An Autonomous Racing Stack
High-speed, head-to-head autonomous racing presents substantial technical and logistical challenges, including precise localization, rapid perception, dynamic planning, and real-time control-compounded by limited track access and costly hardware. This paper introduces the Autonomous Race Stack (ARS), developed by the IU Luddy Autonomous Racing team for the Indy Autonomous Challenge (IAC). We present three iterations of our ARS, each validated on different tracks and achieving speeds up to 260 km/h. Our contributions include: (i) the modular architecture and evolution of the ARS across ARS1, ARS2, and ARS3; (ii) a detailed performance evaluation that contrasts control, perception, and estimation across oval and road-course environments; and (iii) the release of a high-speed, multi-sensor dataset collected from oval and road-course tracks. Our findings highlight the unique challenges and insights from real-world high-speed full-scale autonomous racing.
☆ AQUILA: A QUIC-Based Link Architecture for Resilient Long-Range UAV Communication
The proliferation of autonomous Unmanned Aerial Vehicles (UAVs) in Beyond Visual Line of Sight (BVLOS) applications is critically dependent on resilient, high-bandwidth, and low-latency communication links. Existing solutions face critical limitations: TCP's head-of-line blocking stalls time-sensitive data, UDP lacks reliability and congestion control, and cellular networks designed for terrestrial users degrade severely for aerial platforms. This paper introduces AQUILA, a cross-layer communication architecture built on QUIC to address these challenges. AQUILA contributes three key innovations: (1) a unified transport layer using QUIC's reliable streams for MAVLink Command and Control (C2) and unreliable datagrams for video, eliminating head-of-line blocking under unified congestion control; (2) a priority scheduling mechanism that structurally ensures C2 latency remains bounded and independent of video traffic intensity; (3) a UAV-adapted congestion control algorithm extending SCReAM with altitude-adaptive delay targeting and telemetry headroom reservation. AQUILA further implements 0-RTT connection resumption to minimize handover blackouts with application-layer replay protection, deployed over an IP-native architecture enabling global operation. Experimental validation demonstrates that AQUILA significantly outperforms TCP- and UDP-based approaches in C2 latency, video quality, and link resilience under realistic conditions, providing a robust foundation for autonomous BVLOS missions.
comment: 13 pages, 10 figures
☆ Dynamic Visual SLAM using a General 3D Prior
Reliable incremental estimation of camera poses and 3D reconstruction is key to enable various applications including robotics, interactive visualization, and augmented reality. However, this task is particularly challenging in dynamic natural environments, where scene dynamics can severely deteriorate camera pose estimation accuracy. In this work, we propose a novel monocular visual SLAM system that can robustly estimate camera poses in dynamic scenes. To this end, we leverage the complementary strengths of geometric patch-based online bundle adjustment and recent feed-forward reconstruction models. Specifically, we propose a feed-forward reconstruction model to precisely filter out dynamic regions, while also utilizing its depth prediction to enhance the robustness of the patch-based visual SLAM. By aligning depth prediction with estimated patches from bundle adjustment, we robustly handle the inherent scale ambiguities of the batch-wise application of the feed-forward reconstruction model.
comment: 8 pages
☆ MagicSkin: Balancing Marker and Markerless Modes in Vision-Based Tactile Sensors with a Translucent Skin ICRA2026
Vision-based tactile sensors (VBTS) face a fundamental trade-off in marker and markerless design on the tactile skin: opaque ink markers enable measurement of force and tangential displacement but completely occlude geometric features necessary for object and texture classification, while markerless skin preserves surface details but struggles in measuring tangential displacements effectively. Current practice to solve the above problem via UV lighting or virtual transfer using learning-based models introduces hardware complexity or computing burdens. This paper introduces MagicSkin, a novel tactile skin with translucent, tinted markers balancing the modes of marker and markerless for VBTS. It enables simultaneous tangential displacement tracking, force prediction, and surface detail preservation. This skin is easy to plug into GelSight-family sensors without requiring additional hardware or software tools. We comprehensively evaluate MagicSkin in downstream tasks. The translucent markers impressively enhance rather than degrade sensing performance compared with traditional markerless and inked marker design: it achieves best performance in object classification (99.17\%), texture classification (93.51\%), tangential displacement tracking (97\% point retention) and force prediction (66\% improvement in total force error). These experimental results demonstrate that translucent skin eliminates the traditional performance trade-off in marker or markerless modes, paving the way for multimodal tactile sensing essential in tactile robotics. See videos at this \href{https://zhuochenn.github.io/MagicSkin_project/}{link}.
comment: Submitted to ICRA2026
☆ db-LaCAM: Fast and Scalable Multi-Robot Kinodynamic Motion Planning with Discontinuity-Bounded Search and Lightweight MAPF
State-of-the-art multi-robot kinodynamic motion planners struggle to handle more than a few robots due to high computational burden, which limits their scalability and results in slow planning time. In this work, we combine the scalability and speed of modern multi-agent path finding (MAPF) algorithms with the dynamic-awareness of kinodynamic planners to address these limitations. To this end, we propose discontinuity-Bounded LaCAM (db-LaCAM), a planner that utilizes a precomputed set of motion primitives that respect robot dynamics to generate horizon-length motion sequences, while allowing a user-defined discontinuity between successive motions. The planner db-LaCAM is resolution-complete with respect to motion primitives and supports arbitrary robot dynamics. Extensive experiments demonstrate that db-LaCAM scales efficiently to scenarios with up to 50 robots, achieving up to ten times faster runtime compared to state-of-the-art planners, while maintaining comparable solution quality. The approach is validated in both 2D and 3D environments with dynamics such as the unicycle and 3D double integrator. We demonstrate the safe execution of trajectories planned with db-LaCAM in two distinct physical experiments involving teams of flying robots and car-with-trailer robots.
☆ Model-Less Feedback Control of Space-based Continuum Manipulators using Backbone Tension Optimization
Continuum manipulators offer intrinsic dexterity and safe geometric compliance for navigation within confined and obstacle-rich environments. However, their infinite-dimensional backbone deformation, unmodeled internal friction, and configuration-dependent stiffness fundamentally limit the reliability of model-based kinematic formulations, resulting in inaccurate Jacobian predictions, artificial singularities, and unstable actuation behavior. Motivated by these limitations, this work presents a complete model-less control framework that bypasses kinematic modeling by using an empirically initialized Jacobian refined online through differential convex updates. Tip motion is generated via a real-time quadratic program that computes actuator increments while enforcing tendon slack avoidance and geometric limits. A backbone tension optimization term is introduced in this paper to regulate axial loading and suppress co-activation compression. The framework is validated across circular, pentagonal, and square trajectories, demonstrating smooth convergence, stable tension evolution, and sub-millimeter steady-state accuracy without any model calibration or parameter identification. These results establish the proposed controller as a scalable alternative to model-dependent continuum manipulation in a constrained environment.
☆ A Novel Deep Neural Network Architecture for Real-Time Water Demand Forecasting
Short-term water demand forecasting (StWDF) is the foundation stone in the derivation of an optimal plan for controlling water supply systems. Deep learning (DL) approaches provide the most accurate solutions for this purpose. However, they suffer from complexity problem due to the massive number of parameters, in addition to the high forecasting error at the extreme points. In this work, an effective method to alleviate the error at these points is proposed. It is based on extending the data by inserting virtual data within the actual data to relieve the nonlinearity around them. To our knowledge, this is the first work that considers the problem related to the extreme points. Moreover, the water demand forecasting model proposed in this work is a novel DL model with relatively low complexity. The basic model uses the gated recurrent unit (GRU) to handle the sequential relationship in the historical demand data, while an unsupervised classification method, K-means, is introduced for the creation of new features to enhance the prediction accuracy with less number of parameters. Real data obtained from two different water plants in China are used to train and verify the model proposed. The prediction results and the comparison with the state-of-the-art illustrate that the method proposed reduces the complexity of the model six times of what achieved in the literature while conserving the same accuracy. Furthermore, it is found that extending the data set significantly reduces the error by about 30%. However, it increases the training time.
☆ FedDSR: Federated Deep Supervision and Regularization Towards Autonomous Driving
Federated Learning (FL) enables collaborative training of autonomous driving (AD) models across distributed vehicles while preserving data privacy. However, FL encounters critical challenges such as poor generalization and slow convergence due to non-independent and identically distributed (non-IID) data from diverse driving environments. To overcome these obstacles, we introduce Federated Deep Supervision and Regularization (FedDSR), a paradigm that incorporates multi-access intermediate layer supervision and regularization within federated AD system. Specifically, FedDSR comprises following integral strategies: (I) to select multiple intermediate layers based on predefined architecture-agnostic standards. (II) to compute mutual information (MI) and negative entropy (NE) on those selected layers to serve as intermediate loss and regularizer. These terms are integrated into the output-layer loss to form a unified optimization objective, enabling comprehensive optimization across the network hierarchy. (III) to aggregate models from vehicles trained based on aforementioned rules of (I) and (II) to generate the global model on central server. By guiding and penalizing the learning of feature representations at intermediate stages, FedDSR enhances the model generalization and accelerates model convergence for federated AD. We then take the semantic segmentation task as an example to assess FedDSR and apply FedDSR to multiple model architectures and FL algorithms. Extensive experiments demonstrate that FedDSR achieves up to 8.93% improvement in mIoU and 28.57% reduction in training rounds, compared to other FL baselines, making it highly suitable for practical deployment in federated AD ecosystems.
comment: 9 pages
☆ Statistic-Augmented, Decoupled MoE Routing and Aggregating in Autonomous Driving
Autonomous driving (AD) scenarios are inherently complex and diverse, posing significant challenges for a single deep learning model to effectively cover all possible conditions, such as varying weather, traffic densities, and road types. Large Model (LM)-Driven Mixture of Experts (MoE) paradigm offers a promising solution, where LM serves as the backbone to extract latent features while MoE serves as the downstream head to dynamically select and aggregate specialized experts to adapt to different scenarios. However, routing and aggregating in MoE face intrinsic challenges, including imprecise expert selection due to flawed routing strategy and inefficient expert aggregation leading to suboptimal prediction. To address these issues, we propose a statistic-augmented, decoupled MoE }outing and Aggregating Mechanism (MoE-RAM) driven by LM. Specifically, on the one hand, MoE-RAM enhances expert routing by incorporating statistical retrieval mechanism to match LM-extracted latent features with cached prototypical features of the most relevant experts; on the other hand, MoE-RAM adaptively reweights experts' outputs in fusion by measuring statistical distances of experts' instant features against LM-extracted latent features. Benefiting from the synergy of the statistic-augmented MoE's routing and aggregating, MoE-RAM ultimately improves the prediction performance. We take the AD semantic segmentation task as an example to assess the proposed MoE-RAM. Extensive experiments on AD datasets demonstrate the superiority of MoE-RAM compared to other MoE baselines and conventional single-model approaches.
comment: 9 pages
☆ MIND-V: Hierarchical Video Generation for Long-Horizon Robotic Manipulation with RL-based Physical Alignment
Embodied imitation learning is constrained by the scarcity of diverse, long-horizon robotic manipulation data. Existing video generation models for this domain are limited to synthesizing short clips of simple actions and often rely on manually defined trajectories. To this end, we introduce MIND-V, a hierarchical framework designed to synthesize physically plausible and logically coherent videos of long-horizon robotic manipulation. Inspired by cognitive science, MIND-V bridges high-level reasoning with pixel-level synthesis through three core components: a Semantic Reasoning Hub (SRH) that leverages a pre-trained vision-language model for task planning; a Behavioral Semantic Bridge (BSB) that translates abstract instructions into domain-invariant representations; and a Motor Video Generator (MVG) for conditional video rendering. MIND-V employs Staged Visual Future Rollouts, a test-time optimization strategy to enhance long-horizon robustness. To align the generated videos with physical laws, we introduce a GRPO reinforcement learning post-training phase guided by a novel Physical Foresight Coherence (PFC) reward. PFC leverages the V-JEPA world model to enforce physical plausibility by aligning the predicted and actual dynamic evolutions in the feature space. MIND-V demonstrates state-of-the-art performance in long-horizon robotic manipulation video generation, establishing a scalable and controllable paradigm for embodied data synthesis.
☆ Robust Optimization-based Autonomous Dynamic Soaring with a Fixed-Wing UAV
Dynamic soaring is a flying technique to exploit the energy available in wind shear layers, enabling potentially unlimited flight without the need for internal energy sources. We propose a framework for autonomous dynamic soaring with a fixed-wing unmanned aerial vehicle (UAV). The framework makes use of an explicit representation of the wind field and a classical approach for guidance and control of the UAV. Robustness to wind field estimation error is achieved by constructing point-wise robust reference paths for dynamic soaring and the development of a robust path following controller for the fixed-wing UAV. The framework is evaluated in dynamic soaring scenarios in simulation and real flight tests. In simulation, we demonstrate robust dynamic soaring flight subject to varied wind conditions, estimation errors and disturbances. Critical components of the framework, including energy predictions and path-following robustness, are further validated in real flights to assure small sim-to-real gap. Together, our results strongly indicate the ability of the proposed framework to achieve autonomous dynamic soaring flight in wind shear.
comment: This work has been submitted to the IEEE for possible publication
☆ A New Trajectory-Oriented Approach to Enhancing Comprehensive Crowd Navigation Performance
Crowd navigation has garnered considerable research interest in recent years, especially with the proliferating application of deep reinforcement learning (DRL) techniques. Many studies, however, do not sufficiently analyze the relative priorities among evaluation metrics, which compromises the fair assessment of methods with divergent objectives. Furthermore, trajectory-continuity metrics, specifically those requiring $C^2$ smoothness, are rarely incorporated. Current DRL approaches generally prioritize efficiency and proximal comfort, often neglecting trajectory optimization or addressing it only through simplistic, unvalidated smoothness reward. Nevertheless, effective trajectory optimization is essential to ensure naturalness, enhance comfort, and maximize the energy efficiency of any navigation system. To address these gaps, this paper proposes a unified framework that enables the fair and transparent assessment of navigation methods by examining the prioritization and joint evaluation of multiple optimization objectives. We further propose a novel reward-shaping strategy that explicitly emphasizes trajectory-curvature optimization. The resulting trajectory quality and adaptability are significantly enhanced across multi-scale scenarios. Through extensive 2D and 3D experiments, we demonstrate that the proposed method achieves superior performance compared to state-of-the-art approaches.
comment: 8 pages, 6 figures
♻ ☆ Safety-Aware Reinforcement Learning for Control via Risk-Sensitive Action-Value Iteration and Quantile Regression
Mainstream approximate action-value iteration reinforcement learning (RL) algorithms suffer from overestimation bias, leading to suboptimal policies in high-variance stochastic environments. Quantile-based action-value iteration methods reduce this bias by learning a distribution of the expected cost-to-go using quantile regression. However, ensuring that the learned policy satisfies safety constraints remains a challenge when these constraints are not explicitly integrated into the RL framework. Existing methods often require complex neural architectures or manual tradeoffs due to combined cost functions. To address this, we propose a risk-regularized quantile-based algorithm integrating Conditional Value-at-Risk (CVaR) to enforce safety without complex architectures. We also provide theoretical guarantees on the contraction properties of the risk-sensitive distributional Bellman operator in Wasserstein space, ensuring convergence to a unique cost distribution. Simulations of a mobile robot in a dynamic reach-avoid task show that our approach leads to more goal successes, fewer collisions, and better safety-performance trade-offs than risk-neutral methods.
comment: 13 pages, 4 figures. Expanded version of a paper to appear in the Proceedings of the 2025 IEEE Conference on Decision and Control (CDC)
♻ ☆ MeshA*: Efficient Path Planning With Motion Primitives AAAI-2026
We study a path planning problem where the possible move actions are represented as a finite set of motion primitives aligned with the grid representation of the environment. That is, each primitive corresponds to a short kinodynamically-feasible motion of an agent and is represented as a sequence of the swept cells of a grid. Typically, heuristic search, i.e. A*, is conducted over the lattice induced by these primitives (lattice-based planning) to find a path. However, due to the large branching factor, such search may be inefficient in practice. To this end, we suggest a novel technique rooted in the idea of searching over the grid cells (as in vanilla A*) simultaneously fitting the possible sequences of the motion primitives into these cells. The resultant algorithm, MeshA*, provably preserves the guarantees on completeness and optimality, on the one hand, and is shown to notably outperform conventional lattice-based planning (x1.5-x2 decrease in the runtime), on the other hand.
comment: Accepted to AAAI-2026
♻ ☆ Unveiling the Impact of Data and Model Scaling on High-Level Control for Humanoid Robots
Data scaling has long remained a critical bottleneck in robot learning. For humanoid robots, human videos and motion data are abundant and widely available, offering a free and large-scale data source. Besides, the semantics related to the motions enable modality alignment and high-level robot control learning. However, how to effectively mine raw video, extract robot-learnable representations, and leverage them for scalable learning remains an open problem. To address this, we introduce Humanoid-Union, a large-scale dataset generated through an autonomous pipeline, comprising over 260 hours of diverse, high-quality humanoid robot motion data with semantic annotations derived from human motion videos. The dataset can be further expanded via the same pipeline. Building on this data resource, we propose SCHUR, a scalable learning framework designed to explore the impact of large-scale data on high-level control in humanoid robots. Experimental results demonstrate that SCHUR achieves high robot motion generation quality and strong text-motion alignment under data and model scaling, with 37\% reconstruction improvement under MPJPE and 25\% alignment improvement under FID comparing with previous methods. Its effectiveness is further validated through deployment in real-world humanoid robot.
♻ ☆ Concerns and Values in Human-Robot Interactions: A Focus on Social Robotics
Robots, as AI with physical instantiation, inhabit our social and physical world, where their actions have both social and physical consequences, posing challenges for researchers when designing social robots. This study starts with a scoping review to identify discussions and potential concerns arising from interactions with robotic systems in the context of healthcare, education, and private homes. Two focus groups of technology ethics experts then validated a comprehensive list of key topics and values in human-robot interaction (HRI) literature in these contexts. These insights were integrated into the HRI Value Compass web tool, to help HRI researchers identify these values in robot design. The tool was evaluated in a pilot study. This work benefits the HRI community by highlighting key concerns in human-robot interactions and providing an instrument to help researchers design robots that align with human values, ensuring future robotic systems adhere to these values in social applications.
comment: 31 pages, 7 figures, 6 tables; 4 appendices
♻ ☆ Dribble Master: Learning Agile Humanoid Dribbling Through Legged Locomotion
Humanoid soccer dribbling is a highly challenging task that demands dexterous ball manipulation while maintaining dynamic balance. Traditional rule-based methods often struggle to achieve accurate ball control due to their reliance on fixed walking patterns and limited adaptability to real-time ball dynamics. To address these challenges, we propose a two-stage curriculum learning framework that enables a humanoid robot to acquire dribbling skills without explicit dynamics or predefined trajectories. In the first stage, the robot learns basic locomotion skills; in the second stage, we fine-tune the policy for agile dribbling maneuvers. We further introduce a virtual camera model in simulation that simulates the field of view and perception constraints of the real robot, enabling realistic ball perception during training. We also design heuristic rewards to encourage active sensing, promoting a broader visual range for continuous ball perception. The policy is trained in simulation and successfully transferred to a physical humanoid robot. Experiment results demonstrate that our method enables effective ball manipulation, achieving flexible and visually appealing dribbling behaviors across multiple environments. This work highlights the potential of reinforcement learning in developing agile humanoid soccer robots. Additional details and videos are available at https://zhuoheng0910.github.io/dribble-master/.
♻ ☆ Griffin: Aerial-Ground Cooperative Detection and Tracking Dataset and Benchmark AAAI 2026
While cooperative perception can overcome the limitations of single-vehicle systems, the practical implementation of vehicle-to-vehicle and vehicle-to-infrastructure systems is often impeded by significant economic barriers. Aerial-ground cooperation (AGC), which pairs ground vehicles with drones, presents a more economically viable and rapidly deployable alternative. However, this emerging field has been held back by a critical lack of high-quality public datasets and benchmarks. To bridge this gap, we present \textit{Griffin}, a comprehensive AGC 3D perception dataset, featuring over 250 dynamic scenes (37k+ frames). It incorporates varied drone altitudes (20-60m), diverse weather conditions, realistic drone dynamics via CARLA-AirSim co-simulation, and critical occlusion-aware 3D annotations. Accompanying the dataset is a unified benchmarking framework for cooperative detection and tracking, with protocols to evaluate communication efficiency, altitude adaptability, and robustness to communication latency, data loss and localization noise. By experiments through different cooperative paradigms, we demonstrate the effectiveness and limitations of current methods and provide crucial insights for future research. The dataset and codes are available at https://github.com/wang-jh18-SVM/Griffin.
comment: Accepted by AAAI 2026
♻ ☆ Kinodynamic Motion Planning for Collaborative Object Transportation by Multiple Mobile Manipulators
This work proposes a kinodynamic motion planning technique for collaborative object transportation by multiple mobile manipulators in dynamic environments. A global path planner computes a linear piecewise path from start to goal. A novel algorithm detects the narrow regions between the static obstacles and aids in defining the obstacle-free region to enhance the feasibility of the global path. We then formulate a local online motion planning technique for trajectory generation that minimizes the control efforts in a receding horizon manner. It plans the trajectory for finite time horizons, considering the kinodynamic constraints and the static and dynamic obstacles. The planning technique jointly plans for the mobile bases and the arms to utilize the locomotion capability of the mobile base and the manipulation capability of the arm efficiently. We use a convex cone approach to avoid self-collision of the formation by modifying the mobile manipulators admissible state without imposing additional constraints. Numerical simulations and hardware experiments showcase the efficiency of the proposed approach.
comment: Video: https://youtu.be/LhE_HcK4g-s
♻ ☆ Dejavu: Towards Experience Feedback Learning for Embodied Intelligence
Embodied agents face a fundamental limitation: once deployed in real-world environments to perform specific tasks, they are unable to acquire additional knowledge to enhance task performance. In this paper, we propose a general post-deployment learning framework Dejavu, which employs an Experience Feedback Network (EFN) and augments the frozen Vision-Language-Action (VLA) policy with retrieved execution memories. EFN identifies contextually prior action experiences and conditions action prediction on this retrieved guidance. We adopt reinforcement learning with semantic similarity rewards to train EFN, ensuring that the predicted actions align with past behaviors under current observations. During deployment, EFN continually enriches its memory with new trajectories, enabling the agent to exhibit "learning from experience". Experiments across diverse embodied tasks show that EFN improves adaptability, robustness, and success rates over frozen baselines. We provide code and demo in our supplementary material.
♻ ☆ NeuralRemaster: Phase-Preserving Diffusion for Structure-Aligned Generation
Standard diffusion corrupts data using Gaussian noise whose Fourier coefficients have random magnitudes and random phases. While effective for unconditional or text-to-image generation, corrupting phase components destroys spatial structure, making it ill-suited for tasks requiring geometric consistency, such as re-rendering, simulation enhancement, and image-to-image translation. We introduce Phase-Preserving Diffusion φ-PD, a model-agnostic reformulation of the diffusion process that preserves input phase while randomizing magnitude, enabling structure-aligned generation without architectural changes or additional parameters. We further propose Frequency-Selective Structured (FSS) noise, which provides continuous control over structural rigidity via a single frequency-cutoff parameter. φ-PD adds no inference-time cost and is compatible with any diffusion model for images or videos. Across photorealistic and stylized re-rendering, as well as sim-to-real enhancement for driving planners, φ-PD produces controllable, spatially aligned results. When applied to the CARLA simulator, φ-PD improves CARLA-to-Waymo planner performance by 50\%. The method is complementary to existing conditioning approaches and broadly applicable to image-to-image and video-to-video generation. Videos, additional examples, and code are available on our \href{https://yuzeng-at-tri.github.io/ppd-page/}{project page}.
♻ ☆ LabUtopia: High-Fidelity Simulation and Hierarchical Benchmark for Scientific Embodied Agents NeurIPS 2025
Scientific embodied agents play a crucial role in modern laboratories by automating complex experimental workflows. Compared to typical household environments, laboratory settings impose significantly higher demands on perception of physical-chemical transformations and long-horizon planning, making them an ideal testbed for advancing embodied intelligence. However, its development has been long hampered by the lack of suitable simulator and benchmarks. In this paper, we address this gap by introducing LabUtopia, a comprehensive simulation and benchmarking suite designed to facilitate the development of generalizable, reasoning-capable embodied agents in laboratory settings. Specifically, it integrates i) LabSim, a high-fidelity simulator supporting multi-physics and chemically meaningful interactions; ii) LabScene, a scalable procedural generator for diverse scientific scenes; and iii) LabBench, a hierarchical benchmark spanning five levels of complexity from atomic actions to long-horizon mobile manipulation. LabUtopia supports 30 distinct tasks and includes more than 200 scene and instrument assets, enabling large-scale training and principled evaluation in high-complexity environments. We demonstrate that LabUtopia offers a powerful platform for advancing the integration of perception, planning, and control in scientific-purpose agents and provides a rigorous testbed for exploring the practical capabilities and generalization limits of embodied intelligence in future research.
comment: Accepted by NeurIPS 2025 Dataset and Benchmark Track
♻ ☆ Safe MPC Alignment with Human Directional Feedback
In safety-critical robot planning or control, manually specifying safety constraints or learning them from demonstrations can be challenging. In this article, we propose a certifiable alignment method for a robot to learn a safety constraint in its model predictive control (MPC) policy from human online directional feedback. To our knowledge, it is the first method to learn safety constraints from human feedback. The proposed method is based on an empirical observation: human directional feedback, when available, tends to guide the robot toward safer regions. The method only requires the direction of human feedback to update the learning hypothesis space. It is certifiable, providing an upper bound on the total number of human feedback in the case of successful learning, or declaring the hypothesis misspecification, i.e., the true safety constraint cannot be found within the specified hypothesis space. We evaluated the proposed method in numerical examples and user studies with two simulation games. Additionally, we tested the proposed method on a real-world Franka robot arm performing mobile water-pouring tasks. The results demonstrate the efficacy and efficiency of our method, showing that it enables a robot to successfully learn safety constraints with a small handful (tens) of human directional corrections.
comment: 20 pages, pre-print, submitted to TRO
♻ ☆ Energy-Aware Lane Planning for Connected Electric Vehicles in Urban Traffic: Design and Vehicle-in-the-Loop Validation
Urban driving with connected and automated vehicles (CAVs) offers potential for energy savings, yet most eco-driving strategies focus solely on longitudinal speed control within a single lane. This neglects the significant impact of lateral decisions, such as lane changes, on overall energy efficiency, especially in environments with traffic signals and heterogeneous traffic flow. To address this gap, we propose a novel energy-aware motion planning framework that jointly optimizes longitudinal speed and lateral lane-change decisions using vehicle-to-infrastructure (V2I) communication. Our approach estimates long-term energy costs using a graph-based approximation and solves short-horizon optimal control problems under traffic constraints. Using a data-driven energy model calibrated to an actual battery electric vehicle, we demonstrate with vehicle-in-the-loop experiments that our method reduces motion energy consumption by up to 24 percent compared to a human driver, highlighting the potential of connectivity-enabled planning for sustainable urban autonomy.
comment: Accepted at 2025 IEEE Conference on Decision and Control (CDC25')
♻ ☆ Inversely Learning Transferable Rewards via Abstracted States
Inverse reinforcement learning (IRL) has progressed significantly toward accurately learning the underlying rewards in both discrete and continuous domains from behavior data. The next advance is to learn {\em intrinsic} preferences in ways that produce useful behavior in settings or tasks which are different but aligned with the observed ones. In the context of robotic applications, this helps integrate robots into processing lines involving new tasks (with shared intrinsic preferences) without programming from scratch. We introduce a method to inversely learn an abstract reward function from behavior trajectories in two or more differing instances of a domain. The abstract reward function is then used to learn task behavior in another separate instance of the domain. This step offers evidence of its transferability and validates its correctness. We evaluate the method on trajectories in tasks from multiple domains in OpenAI's Gym testbed and AssistiveGym and show that the learned abstract reward functions can successfully learn task behaviors in instances of the respective domains, which have not been seen previously.
Robotics 22
☆ Error-Centric PID Untrained Neural-Net (EC-PIDUNN) For Nonlinear Robotics Control
Classical Proportional-Integral-Derivative (PID) control has been widely successful across various industrial systems such as chemical processes, robotics, and power systems. However, as these systems evolved, the increase in the nonlinear dynamics and the complexity of interconnected variables have posed challenges that classical PID cannot effectively handle, often leading to instability, overshooting, or prolonged settling times. Researchers have proposed PIDNN models that combine the function approximation capabilities of neural networks with PID control to tackle these nonlinear challenges. However, these models require extensive, highly refined training data and have significant computational costs, making them less favorable for real-world applications. In this paper, We propose a novel EC-PIDUNN architecture, which integrates an untrained neural network with an improved PID controller, incorporating a stabilizing factor (\(τ\)) to generate the control signal. Like classical PID, our architecture uses the steady-state error \(e_t\) as input bypassing the need for explicit knowledge of the systems dynamics. By forming an input vector from \(e_t\) within the neural network, we increase the dimensionality of input allowing for richer data representation. Additionally, we introduce a vector of parameters \( ρ_t \) to shape the output trajectory and a \textit{dynamic compute} function to adjust the PID coefficients from predefined values. We validate the effectiveness of EC-PIDUNN on multiple nonlinear robotics applications: (1) nonlinear unmanned ground vehicle systems that represent the Ackermann steering mechanism and kinematics control, (2) Pan-Tilt movement system. In both tests, it outperforms classical PID in convergence and stability achieving a nearly critically damped response.
comment: Under review at SoftComputing
☆ Deep Neural Network-Based Aerial Transport in the Presence of Cooperative and Uncooperative UAS
We present a resilient deep neural network (DNN) framework for decentralized transport and coverage using uncrewed aerial systems (UAS) operating in $\mathbb{R}^n$. The proposed DNN-based mass-transport architecture constructs a layered inter-UAS communication graph from an initial formation, assigns time-varying communication weights through a forward scheduling mechanism that guides the team from the initial to the final configuration, and ensures stability and convergence of the resulting multi-agent transport dynamics. The framework is explicitly designed to remain robust in the presence of uncooperative agents that deviate from or refuse to follow the prescribed protocol. Our method preserves a fixed feed-forward topology but dynamically prunes edges to uncooperative agents, maintains convex, feedforward mentoring among cooperative agents, and computes global desired set points through a sparse linear relation consistent with leader references. The target set is abstracted by $N$ points that become final desired positions, enabling coverage-optimal transport while keeping computation low and guarantees intact. Extensive simulations demonstrate that, under full cooperation, all agents converge rapidly to the target zone with a 10\% boundary margin and under partial cooperation with uncooperative agents, the system maintains high convergence among cooperative agents with performance degradation localized near the disruptions, evidencing graceful resilience and scalability. These results confirm that forward-weight scheduling, hierarchical mentor--mentee coordination, and on-the-fly DNN restructuring yield robust, provably stable UAS transport in realistic fault scenarios.
☆ Learning Agile Striker Skills for Humanoid Soccer Robots from Noisy Sensory Input
Learning fast and robust ball-kicking skills is a critical capability for humanoid soccer robots, yet it remains a challenging problem due to the need for rapid leg swings, postural stability on a single support foot, and robustness under noisy sensory input and external perturbations (e.g., opponents). This paper presents a reinforcement learning (RL)-based system that enables humanoid robots to execute robust continual ball-kicking with adaptability to different ball-goal configurations. The system extends a typical teacher-student training framework -- in which a "teacher" policy is trained with ground truth state information and the "student" learns to mimic it with noisy, imperfect sensing -- by including four training stages: (1) long-distance ball chasing (teacher); (2) directional kicking (teacher); (3) teacher policy distillation (student); and (4) student adaptation and refinement (student). Key design elements -- including tailored reward functions, realistic noise modeling, and online constrained RL for adaptation and refinement -- are critical for closing the sim-to-real gap and sustaining performance under perceptual uncertainty. Extensive evaluations in both simulation and on a real robot demonstrate strong kicking accuracy and goal-scoring success across diverse ball-goal configurations. Ablation studies further highlight the necessity of the constrained RL, noise modeling, and the adaptation stage. This work presents a system for learning robust continual humanoid ball-kicking under imperfect perception, establishing a benchmark task for visuomotor skill learning in humanoid whole-body control.
☆ Embodied Referring Expression Comprehension in Human-Robot Interaction
As robots enter human workspaces, there is a crucial need for them to comprehend embodied human instructions, enabling intuitive and fluent human-robot interaction (HRI). However, accurate comprehension is challenging due to a lack of large-scale datasets that capture natural embodied interactions in diverse HRI settings. Existing datasets suffer from perspective bias, single-view collection, inadequate coverage of nonverbal gestures, and a predominant focus on indoor environments. To address these issues, we present the Refer360 dataset, a large-scale dataset of embodied verbal and nonverbal interactions collected across diverse viewpoints in both indoor and outdoor settings. Additionally, we introduce MuRes, a multimodal guided residual module designed to improve embodied referring expression comprehension. MuRes acts as an information bottleneck, extracting salient modality-specific signals and reinforcing them into pre-trained representations to form complementary features for downstream tasks. We conduct extensive experiments on four HRI datasets, including the Refer360 dataset, and demonstrate that current multimodal models fail to capture embodied interactions comprehensively; however, augmenting them with MuRes consistently improves performance. These findings establish Refer360 as a valuable benchmark and exhibit the potential of guided residual learning to advance embodied referring expression comprehension in robots operating within human environments.
comment: 14 pages, 7 figures, accepted at the ACM/IEEE International Conference on Human-Robot Interaction (HRI) 2026
☆ TacFinRay: Soft Tactile Fin-Ray Finger with Indirect Tactile Sensing for Robust Grasping
We present a tactile-sensorized Fin-Ray finger that enables simultaneous detection of contact location and indentation depth through an indirect sensing approach. A hinge mechanism is integrated between the soft Fin-Ray structure and a rigid sensing module, allowing deformation and translation information to be transferred to a bottom crossbeam upon which are an array of marker-tipped pins based on the biomimetic structure of the TacTip vision-based tactile sensor. Deformation patterns captured by an internal camera are processed using a convolutional neural network to infer contact conditions without directly sensing the finger surface. The finger design was optimized by varying pin configurations and hinge orientations, achieving 0.1\,mm depth and 2mm location-sensing accuracies. The perception demonstrated robust generalization to various indenter shapes and sizes, which was applied to a pick-and-place task under uncertain picking positions, where the tactile feedback significantly improved placement accuracy. Overall, this work provides a lightweight, flexible, and scalable tactile sensing solution suitable for soft robotic structures where the sensing needs situating away from the contact interface.
comment: Accepted in IEEE Robotics Automation Letters. S. Nam, B. Deng co-first authors
☆ Vision-Guided Grasp Planning for Prosthetic Hands in Unstructured Environments
Recent advancements in prosthetic technology have increasingly focused on enhancing dexterity and autonomy through intelligent control systems. Vision-based approaches offer promising results for enabling prosthetic hands to interact more naturally with diverse objects in dynamic environments. Building on this foundation, the paper presents a vision-guided grasping algorithm for a prosthetic hand, integrating perception, planning, and control for dexterous manipulation. A camera mounted on the set up captures the scene, and a Bounding Volume Hierarchy (BVH)-based vision algorithm is employed to segment an object for grasping and define its bounding box. Grasp contact points are then computed by generating candidate trajectories using Rapidly-exploring Random Tree Star algorithm, and selecting fingertip end poses based on the minimum Euclidean distance between these trajectories and the objects point cloud. Each finger grasp pose is determined independently, enabling adaptive, object-specific configurations. Damped Least Square (DLS) based Inverse kinematics solver is used to compute the corresponding joint angles, which are subsequently transmitted to the finger actuators for execution. This modular pipeline enables per-finger grasp planning and supports real-time adaptability in unstructured environments. The proposed method is validated in simulation, and experimental integration on a Linker Hand O7 platform.
☆ Method of UAV Inspection of Photovoltaic Modules Using Thermal and RGB Data Fusion
The subject of this research is the development of an intelligent, integrated framework for the automated inspection of photovoltaic (PV) infrastructure that addresses the critical shortcomings of conventional methods, including thermal palette bias, data redundancy, and high communication bandwidth requirements. The goal of this study is to design, develop, and validate a comprehensive, multi-modal system that fully automates the monitoring workflow, from data acquisition to the generation of actionable, geo-located maintenance alerts, thereby enhancing plant safety and operational efficiency. The methods employed involve a synergistic architecture that begins with a palette-invariant thermal embedding, learned by enforcing representational consistency, which is fused with a contrast-normalized RGB stream via a gated mechanism. This is supplemented by a closed-loop, adaptive re-acquisition controller that uses Rodrigues-based updates for targeted confirmation of ambiguous anomalies and a geospatial deduplication module that clusters redundant alerts using DBSCAN over the haversine distance. In conclusion, this study establishes a powerful new paradigm for proactive PV inspection, with the proposed system achieving a mean Average Precision (mAP@0.5) of 0.903 on the public PVF-10 benchmark, a significant 12-15% improvement over single-modality baselines. Field validation confirmed the system's readiness, achieving 96% recall, while the de-duplication process reduced duplicate-induced false positives by 15-20%, and relevance-only telemetry cut airborne data transmission by 60-70%.
☆ Entropy-Controlled Intrinsic Motivation Reinforcement Learning for Quadruped Robot Locomotion in Complex Terrains
Learning is the basis of both biological and artificial systems when it comes to mimicking intelligent behaviors. From the classical PPO (Proximal Policy Optimization), there is a series of deep reinforcement learning algorithms which are widely used in training locomotion policies for quadrupedal robots because of their stability and sample efficiency. However, among all these variants, experiments and simulations often converge prematurely, leading to suboptimal locomotion and reduced task performance. Therefore, in this paper, we introduce Entropy-Controlled Intrinsic Motivation (ECIM), an entropy-based reinforcement learning algorithm in contrast with the PPO series, that can reduce premature convergence by combining intrinsic motivation with adaptive exploration. For experiments, in order to parallel with other baselines, we chose to apply it in Isaac Gym across six terrain categories: upward slopes, downward slopes, uneven rough terrain, ascending stairs, descending stairs, and flat ground as widely used. For comparison, our experiments consistently achieve better performance: task rewards increase by 4--12%, peak body pitch oscillation is reduced by 23--29%, joint acceleration decreases by 20--32%, and joint torque consumption declines by 11--20%. Overall, our model ECIM, by combining entropy control and intrinsic motivation control, achieves better results in stability across different terrains for quadrupedal locomotion, and at the same time reduces energetic cost and makes it a practical choice for complex robotic control tasks.
☆ Fault Tolerant Control of Mecanum Wheeled Mobile Robots
Mecanum wheeled mobile robots (MWMRs) are highly susceptible to actuator faults that degrade performance and risk mission failure. Current fault tolerant control (FTC) schemes for MWMRs target complete actuator failures like motor stall, ignoring partial faults e.g., in torque degradation. We propose an FTC strategy handling both fault types, where we adopt posterior probability to learn real-time fault parameters. We derive the FTC law by aggregating probability-weighed control laws corresponding to predefined faults. This ensures the robustness and safety of MWMR control despite varying levels of fault occurrence. Simulation results demonstrate the effectiveness of our FTC under diverse scenarios.
☆ Leveraging Port-Hamiltonian Theory for Impedance Control Benchmarking
This work proposes PH-based metrics for benchmarking impedance control. A causality-consistent PH model is introduced for mass-spring-damper impedance in Cartesian space. Based on this model, a differentiable, force-torque sensing-independent, n-DoF passivity condition is derived, valid for time-varying references. An impedance fidelity metric is also defined from step-response power in free motion, capturing dynamic decoupling. The proposed metrics are validated in Gazebo simulations with a six-DoF manipulator and a quadruped leg. Results demonstrate the suitability of the PH framework for standardized impedance control benchmarking.
comment: This is the author's version of the paper accepted for publication in the 2025 International Conference on Advanced Robotics (ICAR). The final version will be available at IEEE Xplore
☆ Beyond Model Jailbreak: Systematic Dissection of the "Ten DeadlySins" in Embodied Intelligence
Embodied AI systems integrate language models with real world sensing, mobility, and cloud connected mobile apps. Yet while model jailbreaks have drawn significant attention, the broader system stack of embodied intelligence remains largely unexplored. In this work, we conduct the first holistic security analysis of the Unitree Go2 platform and uncover ten cross layer vulnerabilities the "Ten Sins of Embodied AI Security." Using BLE sniffing, traffic interception, APK reverse engineering, cloud API testing, and hardware probing, we identify systemic weaknesses across three architectural layers: wireless provisioning, core modules, and external interfaces. These include hard coded keys, predictable handshake tokens, WiFi credential leakage, missing TLS validation, static SSH password, multilingual safety bypass behavior, insecure local relay channels, weak binding logic, and unrestricted firmware access. Together, they allow adversaries to hijack devices, inject arbitrary commands, extract sensitive information, or gain full physical control.Our findings show that securing embodied AI requires far more than aligning the model itself. We conclude with system level lessons learned and recommendations for building embodied platforms that remain robust across their entire software hardware ecosystem.
☆ Proportional integral derivative booster for neural networks-based time-series prediction: Case of water demand prediction
Multi-step time-series prediction is an essential supportive step for decision-makers in several industrial areas. Artificial intelligence techniques, which use a neural network component in various forms, have recently frequently been used to accomplish this step. However, the complexity of the neural network structure still stands up as a critical problem against prediction accuracy. In this paper, a method inspired by the proportional-integral-derivative (PID) control approach is investigated to enhance the performance of neural network models used for multi-step ahead prediction of periodic time-series information while maintaining a negligible impact on the complexity of the system. The PID-based method is applied to the predicted value at each time step to bring that value closer to the real value. The water demand forecasting problem is considered as a case study, where two deep neural network models from the literature are used to prove the effectiveness of the proposed boosting method. Furthermore, to prove the applicability of this PID-based booster to other types of periodic time-series prediction problems, it is applied to enhance the accuracy of a neural network model used for multi-step forecasting of hourly energy consumption. The comparison between the results of the original prediction models and the results after using the proposed technique demonstrates the superiority of the proposed method in terms of prediction accuracy and system complexity.
comment: Engineering Applications of Artificial Intelligence 2022
☆ Safe Model Predictive Diffusion with Shielding
Generating safe, kinodynamically feasible, and optimal trajectories for complex robotic systems is a central challenge in robotics. This paper presents Safe Model Predictive Diffusion (Safe MPD), a training-free diffusion planner that unifies a model-based diffusion framework with a safety shield to generate trajectories that are both kinodynamically feasible and safe by construction. By enforcing feasibility and safety on all samples during the denoising process, our method avoids the common pitfalls of post-processing corrections, such as computational intractability and loss of feasibility. We validate our approach on challenging non-convex planning problems, including kinematic and acceleration-controlled tractor-trailer systems. The results show that it substantially outperforms existing safety strategies in success rate and safety, while achieving sub-second computation times.
comment: Project page: https://www.taekyung.me/safe-mpd
♻ ☆ Variational Shape Inference for Grasp Diffusion on SE(3)
Grasp synthesis is a fundamental task in robotic manipulation which usually has multiple feasible solutions. Multimodal grasp synthesis seeks to generate diverse sets of stable grasps conditioned on object geometry, making the robust learning of geometric features crucial for success. To address this challenge, we propose a framework for learning multimodal grasp distributions that leverages variational shape inference to enhance robustness against shape noise and measurement sparsity. Our approach first trains a variational autoencoder for shape inference using implicit neural representations, and then uses these learned geometric features to guide a diffusion model for grasp synthesis on the SE(3) manifold. Additionally, we introduce a test-time grasp optimization technique that can be integrated as a plugin to further enhance grasping performance. Experimental results demonstrate that our shape inference for grasp synthesis formulation outperforms state-of-the-art multimodal grasp synthesis methods on the ACRONYM dataset by 6.3%, while demonstrating robustness to deterioration in point cloud density compared to other approaches. Furthermore, our trained model achieves zero-shot transfer to real-world manipulation of household objects, generating 34% more successful grasps than baselines despite measurement noise and point cloud calibration errors.
comment: This work has been submitted to the IEEE for possible publication
♻ ☆ Grasping a Handful: Sequential Multi-Object Dexterous Grasp Generation
We introduce the sequential multi-object robotic grasp sampling algorithm SeqGrasp that can robustly synthesize stable grasps on diverse objects using the robotic hand's partial Degrees of Freedom (DoF). We use SeqGrasp to construct the large-scale Allegro Hand sequential grasping dataset SeqDataset and use it for training the diffusion-based sequential grasp generator SeqDiffuser. We experimentally evaluate SeqGrasp and SeqDiffuser against the state-of-the-art non-sequential multi-object grasp generation method MultiGrasp in simulation and on a real robot. The experimental results demonstrate that SeqGrasp and SeqDiffuser reach an 8.71%-43.33% higher grasp success rate than MultiGrasp. Furthermore, SeqDiffuser is approximately 1000 times faster at generating grasps than SeqGrasp and MultiGrasp. Project page: https://yulihn.github.io/SeqGrasp/.
comment: We replace the sets in Section II with an odered sequences
♻ ☆ Collaborative Drill Alignment in Surgical Robotics
Robotic assistance allows surgeries to be reliably and accurately executed while still under direct supervision of the surgeon, combining the strengths of robotic technology with the surgeon's expertise. This paper describes a robotic system designed to assist in surgical procedures by implementing a virtual drill guide. The system integrates virtual-fixture functionality using a novel virtual-mechanism controller with additional visual feedback. The controller constrains the tool to the desired axis, while allowing axial motion to remain under the surgeon's control. Compared to prior virtual-fixture approaches -- which primarily perform pure energy-shaping and damping injection with linear springs and dampers -- our controller uses a virtual prismatic joint to which the robot is constrained by nonlinear springs, allowing us to easily shape the dynamics of the system. We detail the calibration procedures required to achieve sufficient precision, and describe the implementation of the controller. We apply this system to a veterinary procedure: drilling for transcondylar screw placement in dogs. The results of the trials on 3D-printed bone models demonstrate sufficient precision to perform the procedure and suggest improved angular accuracy and reduced exit translation errors compared to patient specific guides (PSG). Discussion and future improvements follow.
comment: 14 pages, 12 figures, accepted to IEEE Transactions on Control Systems Technology
♻ ☆ Optimal Virtual Model Control for Robotics: Design and Tuning of Passivity-Based Controllers
Passivity-based control is a cornerstone of control theory and an established design approach in robotics. Its strength is based on the passivity theorem, which provides a powerful interconnection framework for robotics. However, the design of passivity-based controllers and their optimal tuning remain challenging. We propose here an intuitive design approach for fully actuated robots, where the control action is determined by a `virtual-mechanism' as in classical virtual model control. The result is a robot whose controlled behavior can be understood in terms of physics. We achieve optimal tuning by applying algorithmic differentiation to ODE simulations of the rigid body dynamics. Overall, this leads to a flexible design and optimization approach: stability is proven by passivity of the virtual mechanism, while performance is obtained by optimization using algorithmic differentiation.
comment: 16 pages, 19 figures
♻ ☆ Primal-Dual iLQR for GPU-Accelerated Learning and Control in Legged Robots
This paper introduces a novel Model Predictive Control (MPC) implementation for legged robot locomotion that leverages GPU parallelization. Our approach enables both temporal and state-space parallelization by incorporating a parallel associative scan to solve the primal-dual Karush-Kuhn-Tucker (KKT) system. In this way, the optimal control problem is solved in $\mathcal{O}(\log^2(n)\log{N} + \log^2(m))$ complexity, instead of $\mathcal{O}(N(n + m)^3)$, where $n$, $m$, and $N$ are the dimension of the system state, control vector, and the length of the prediction horizon. We demonstrate the advantages of this implementation over two state-of-the-art solvers (acados and crocoddyl), achieving up to a 60\% improvement in runtime for Whole Body Dynamics (WB)-MPC and a 700\% improvement for Single Rigid Body Dynamics (SRBD)-MPC when varying the prediction horizon length. The presented formulation scales efficiently with the problem state dimensions as well, enabling the definition of a centralized controller for up to 16 legged robots that can be computed in less than 25 ms. Furthermore, thanks to the JAX implementation, the solver supports large-scale parallelization across multiple environments, allowing the possibility of performing learning with the MPC in the loop directly in GPU. The code associated with this work can be found at https://github.com/iit-DLSLab/mpx.
♻ ☆ SIGN: Safety-Aware Image-Goal Navigation for Autonomous Drones via Reinforcement Learning
Image-goal navigation (ImageNav) tasks a robot with autonomously exploring an unknown environment and reaching a location that visually matches a given target image. While prior works primarily study ImageNav for ground robots, enabling this capability for autonomous drones is substantially more challenging due to their need for high-frequency feedback control and global localization for stable flight. In this paper, we propose a novel sim-to-real framework that leverages reinforcement learning (RL) to achieve ImageNav for drones. To enhance visual representation ability, our approach trains the vision backbone with auxiliary tasks, including image perturbations and future transition prediction, which results in more effective policy training. The proposed algorithm enables end-to-end ImageNav with direct velocity control, eliminating the need for external localization. Furthermore, we integrate a depth-based safety module for real-time obstacle avoidance, allowing the drone to safely navigate in cluttered environments. Unlike most existing drone navigation methods that focus solely on reference tracking or obstacle avoidance, our framework supports comprehensive navigation behaviors, including autonomous exploration, obstacle avoidance, and image-goal seeking, without requiring explicit global mapping. Code and model checkpoints are available at https://github.com/Zichen-Yan/SIGN.
comment: Accepted to IEEE Robotics and Automation Letters (RA-L)
♻ ☆ Certifying Stability of Reinforcement Learning Policies using Generalized Lyapunov Functions NeurIPS 2025
Establishing stability certificates for closed-loop systems under reinforcement learning (RL) policies is essential to move beyond empirical performance and offer guarantees of system behavior. Classical Lyapunov methods require a strict stepwise decrease in the Lyapunov function but such certificates are difficult to construct for learned policies. The RL value function is a natural candidate but it is not well understood how it can be adapted for this purpose. To gain intuition, we first study the linear quadratic regulator (LQR) problem and make two key observations. First, a Lyapunov function can be obtained from the value function of an LQR policy by augmenting it with a residual term related to the system dynamics and stage cost. Second, the classical Lyapunov decrease requirement can be relaxed to a generalized Lyapunov condition requiring only decrease on average over multiple time steps. Using this intuition, we consider the nonlinear setting and formulate an approach to learn generalized Lyapunov functions by augmenting RL value functions with neural network residual terms. Our approach successfully certifies the stability of RL policies trained on Gymnasium and DeepMind Control benchmarks. We also extend our method to jointly train neural controllers and stability certificates using a multi-step Lyapunov loss, resulting in larger certified inner approximations of the region of attraction compared to the classical Lyapunov approach. Overall, our formulation enables stability certification for a broad class of systems with learned policies by making certificates easier to construct, thereby bridging classical control theory and modern learning-based methods.
comment: NeurIPS 2025
♻ ☆ UMI-on-Air: Embodiment-Aware Guidance for Embodiment-Agnostic Visuomotor Policies
We introduce UMI-on-Air, a framework for embodiment-aware deployment of embodiment-agnostic manipulation policies. Our approach leverages diverse, unconstrained human demonstrations collected with a handheld gripper (UMI) to train generalizable visuomotor policies. A central challenge in transferring these policies to constrained robotic embodiments-such as aerial manipulators-is the mismatch in control and robot dynamics, which often leads to out-of-distribution behaviors and poor execution. To address this, we propose Embodiment-Aware Diffusion Policy (EADP), which couples a high-level UMI policy with a low-level embodiment-specific controller at inference time. By integrating gradient feedback from the controller's tracking cost into the diffusion sampling process, our method steers trajectory generation towards dynamically feasible modes tailored to the deployment embodiment. This enables plug-and-play, embodiment-aware trajectory adaptation at test time. We validate our approach on multiple long-horizon and high-precision aerial manipulation tasks, showing improved success rates, efficiency, and robustness under disturbances compared to unguided diffusion baselines. Finally, we demonstrate deployment in previously unseen environments, using UMI demonstrations collected in the wild, highlighting a practical pathway for scaling generalizable manipulation skills across diverse-and even highly constrained-embodiments. All code, data, and checkpoints will be publicly released after acceptance. Result videos can be found at umi-on-air.github.io.
comment: Result videos can be found at umi-on-air.github.io
♻ ☆ High Torque Density PCB Axial Flux Permanent Magnet Motor for Micro Robots
Quasi-direct-drive (QDD) actuation is transforming legged and manipulator robots by eliminating high-ratio gearboxes, yet it demands motors that deliver very high torque at low speed within a thin, disc-shaped joint envelope. Axial-flux permanent-magnet (AFPM) machines meet these geometric and torque requirements, but scaling them below a 20mm outer diameter is hampered by poor copper fill in conventional wound stators, inflating resistance and throttling continuous torque. This paper introduces a micro-scale AFPM motor that overcomes these limitations through printed-circuit-board (PCB) windings fabricated with advanced IC-substrate high-density interconnect (HDI) technology. The resulting 48-layer stator-formed by stacking four 12-layer HDI modules-achieves a record 45\% copper fill in a package only 5mm thick and 19mm in diameter. We perform comprehensive electromagnetic and thermal analyses to inform the motor design, then fabricate a prototype whose performance characteristics are experimentally verified.
Robotics 51
☆ Training-Time Action Conditioning for Efficient Real-Time Chunking
Real-time chunking (RTC) enables vision-language-action models (VLAs) to generate smooth, reactive robot trajectories by asynchronously predicting action chunks and conditioning on previously committed actions via inference-time inpainting. However, this inpainting method introduces computational overhead that increases inference latency. In this work, we propose a simple alternative: simulating inference delay at training time and conditioning on action prefixes directly, eliminating any inference-time overhead. Our method requires no modifications to the model architecture or robot runtime, and can be implemented with only a few additional lines of code. In simulated experiments, we find that training-time RTC outperforms inference-time RTC at higher inference delays. In real-world experiments on box building and espresso making tasks with the $π_{0.6}$ VLA, we demonstrate that training-time RTC maintains both task performance and speed parity with inference-time RTC while being computationally cheaper. Our results suggest that training-time action conditioning is a practical drop-in replacement for inference-time inpainting in real-time robot control.
☆ SIMPACT: Simulation-Enabled Action Planning using Vision-Language Models
Vision-Language Models (VLMs) exhibit remarkable common-sense and semantic reasoning capabilities. However, they lack a grounded understanding of physical dynamics. This limitation arises from training VLMs on static internet-scale visual-language data that contain no causal interactions or action-conditioned changes. Consequently, it remains challenging to leverage VLMs for fine-grained robotic manipulation tasks that require physical understanding, reasoning, and corresponding action planning. To overcome this, we present SIMPACT, a test-time, SIMulation-enabled ACTion Planning framework that equips VLMs with physical reasoning through simulation-in-the-loop world modeling, without requiring any additional training. From a single RGB-D observation, SIMPACT efficiently constructs physics simulations, enabling the VLM to propose informed actions, observe simulated rollouts, and iteratively refine its reasoning. By integrating language reasoning with physics prediction, our simulation-enabled VLM can understand contact dynamics and action outcomes in a physically grounded way. Our method demonstrates state-of-the-art performance on five challenging, real-world rigid-body and deformable manipulation tasks that require fine-grained physical reasoning, outperforming existing general-purpose robotic manipulation models. Our results demonstrate that embedding physics understanding via efficient simulation into VLM reasoning at test time offers a promising path towards generalizable embodied intelligence. Project webpage can be found at https://simpact-bot.github.io
☆ Correspondence-Oriented Imitation Learning: Flexible Visuomotor Control with 3D Conditioning
We introduce Correspondence-Oriented Imitation Learning (COIL), a conditional policy learning framework for visuomotor control with a flexible task representation in 3D. At the core of our approach, each task is defined by the intended motion of keypoints selected on objects in the scene. Instead of assuming a fixed number of keypoints or uniformly spaced time intervals, COIL supports task specifications with variable spatial and temporal granularity, adapting to different user intents and task requirements. To robustly ground this correspondence-oriented task representation into actions, we design a conditional policy with a spatio-temporal attention mechanism that effectively fuses information across multiple input modalities. The policy is trained via a scalable self-supervised pipeline using demonstrations collected in simulation, with correspondence labels automatically generated in hindsight. COIL generalizes across tasks, objects, and motion patterns, achieving superior performance compared to prior methods on real-world manipulation tasks under both sparse and dense specifications.
☆ Measuring the Effect of Background on Classification and Feature Importance in Deep Learning for AV Perception
Common approaches to explainable AI (XAI) for deep learning focus on analyzing the importance of input features on the classification task in a given model: saliency methods like SHAP and GradCAM are used to measure the impact of spatial regions of the input image on the classification result. Combined with ground truth information about the location of the object in the input image (e.g., a binary mask), it is determined whether object pixels had a high impact on the classification result, or whether the classification focused on background pixels. The former is considered to be a sign of a healthy classifier, whereas the latter is assumed to suggest overfitting on spurious correlations. A major challenge, however, is that these intuitive interpretations are difficult to test quantitatively, and hence the output of such explanations lacks an explanation itself. One particular reason is that correlations in real-world data are difficult to avoid, and whether they are spurious or legitimate is debatable. Synthetic data in turn can facilitate to actively enable or disable correlations where desired but often lack a sufficient quantification of realism and stochastic properties. [...] Therefore, we systematically generate six synthetic datasets for the task of traffic sign recognition, which differ only in their degree of camera variation and background correlation [...] to quantify the isolated influence of background correlation, different levels of camera variation, and considered traffic sign shapes on the classification performance, as well as background feature importance. [...] Results include a quantification of when and how much background features gain importance to support the classification task based on changes in the training domain [...]. Download: synset.de/datasets/synset-signset-ger/background-effect
comment: 8 pages, 2 figures, 7 tables
☆ Synset Signset Germany: a Synthetic Dataset for German Traffic Sign Recognition
In this paper, we present a synthesis pipeline and dataset for training / testing data in the task of traffic sign recognition that combines the advantages of data-driven and analytical modeling: GAN-based texture generation enables data-driven dirt and wear artifacts, rendering unique and realistic traffic sign surfaces, while the analytical scene modulation achieves physically correct lighting and allows detailed parameterization. In particular, the latter opens up applications in the context of explainable AI (XAI) and robustness tests due to the possibility of evaluating the sensitivity to parameter changes, which we demonstrate with experiments. Our resulting synthetic traffic sign recognition dataset Synset Signset Germany contains a total of 105500 images of 211 different German traffic sign classes, including newly published (2020) and thus comparatively rare traffic signs. In addition to a mask and a segmentation image, we also provide extensive metadata including the stochastically selected environment and imaging effect parameters for each image. We evaluate the degree of realism of Synset Signset Germany on the real-world German Traffic Sign Recognition Benchmark (GTSRB) and in comparison to CATERED, a state-of-the-art synthetic traffic sign recognition dataset.
comment: 8 pages, 8 figures, 3 tables
☆ Physically-Based Simulation of Automotive LiDAR
We present an analytic model for simulating automotive time-of-flight (ToF) LiDAR that includes blooming, echo pulse width, and ambient light, along with steps to determine model parameters systematically through optical laboratory measurements. The model uses physically based rendering (PBR) in the near-infrared domain. It assumes single-bounce reflections and retroreflections over rasterized rendered images from shading or ray tracing, including light emitted from the sensor as well as stray light from other, non-correlated sources such as sunlight. Beams from the sensor and sensitivity of the receiving diodes are modeled with flexible beam steering patterns and with non-vanishing diameter. Different (all non-real time) computational approaches can be chosen based on system properties, computing capabilities, and desired output properties. Model parameters include system-specific properties, namely the physical spread of the LiDAR beam, combined with the sensitivity of the receiving diode; the intensity of the emitted light; the conversion between the intensity of reflected light and the echo pulse width; and scenario parameters such as environment lighting, positioning, and surface properties of the target(s) in the relevant infrared domain. System-specific properties of the model are determined from laboratory measurements of the photometric luminance on different target surfaces aligned with a goniometer at 0.01° resolution, which marks the best available resolution for measuring the beam pattern. The approach is calibrated for and tested on two automotive LiDAR systems, the Valeo Scala Gen. 2 and the Blickfeld Cube 1. Both systems differ notably in their properties and available interfaces, but the relevant model parameters could be extracted successfully.
☆ World Models That Know When They Don't Know: Controllable Video Generation with Calibrated Uncertainty
Recent advances in generative video models have led to significant breakthroughs in high-fidelity video synthesis, specifically in controllable video generation where the generated video is conditioned on text and action inputs, e.g., in instruction-guided video editing and world modeling in robotics. Despite these exceptional capabilities, controllable video models often hallucinate - generating future video frames that are misaligned with physical reality - which raises serious concerns in many tasks such as robot policy evaluation and planning. However, state-of-the-art video models lack the ability to assess and express their confidence, impeding hallucination mitigation. To rigorously address this challenge, we propose C3, an uncertainty quantification (UQ) method for training continuous-scale calibrated controllable video models for dense confidence estimation at the subpatch level, precisely localizing the uncertainty in each generated video frame. Our UQ method introduces three core innovations to empower video models to estimate their uncertainty. First, our method develops a novel framework that trains video models for correctness and calibration via strictly proper scoring rules. Second, we estimate the video model's uncertainty in latent space, avoiding training instability and prohibitive training costs associated with pixel-space approaches. Third, we map the dense latent-space uncertainty to interpretable pixel-level uncertainty in the RGB space for intuitive visualization, providing high-resolution uncertainty heatmaps that identify untrustworthy regions. Through extensive experiments on large-scale robot learning datasets (Bridge and DROID) and real-world evaluations, we demonstrate that our method not only provides calibrated uncertainty estimates within the training distribution, but also enables effective out-of-distribution detection.
☆ A Residual Variance Matching Recursive Least Squares Filter for Real-time UAV Terrain Following
Accurate real-time waypoints estimation for the UAV-based online Terrain Following during wildfire patrol missions is critical to ensuring flight safety and enabling wildfire detection. However, existing real-time filtering algorithms struggle to maintain accurate waypoints under measurement noise in nonlinear and time-varying systems, posing risks of flight instability and missed wildfire detections during UAV-based terrain following. To address this issue, a Residual Variance Matching Recursive Least Squares (RVM-RLS) filter, guided by a Residual Variance Matching Estimation (RVME) criterion, is proposed to adaptively estimate the real-time waypoints of nonlinear, time-varying UAV-based terrain following systems. The proposed method is validated using a UAV-based online terrain following system within a simulated terrain environment. Experimental results show that the RVM-RLS filter improves waypoints estimation accuracy by approximately 88$\%$ compared with benchmark algorithms across multiple evaluation metrics. These findings demonstrate both the methodological advances in real-time filtering and the practical potential of the RVM-RLS filter for UAV-based online wildfire patrol.
☆ Optimal Safety-Aware Scheduling for Multi-Agent Aerial 3D Printing with Utility Maximization under Dependency Constraints
This article presents a novel coordination and task-planning framework to enable the simultaneous conflict-free collaboration of multiple unmanned aerial vehicles (UAVs) for aerial 3D printing. The proposed framework formulates an optimization problem that takes a construction mission divided into sub-tasks and a team of autonomous UAVs, along with limited volume and battery. It generates an optimal mission plan comprising task assignments and scheduling while accounting for task dependencies arising from the geometric and structural requirements of the 3D design, inter-UAV safety constraints, material usage, and total flight time of each UAV. The potential conflicts occurring during the simultaneous operation of the UAVs are addressed at a segment level by dynamically selecting the starting time and location of each task to guarantee collision-free parallel execution. An importance prioritization is proposed to accelerate the computation by guiding the solution toward more important tasks. Additionally, a utility maximization formulation is proposed to dynamically determine the optimal number of UAVs required for a given mission, balancing the trade-off between minimizing makespan and the deployment of excess agents. The proposed framework's effectiveness is evaluated through a Gazebo-based simulation setup, where agents are coordinated by a mission control module allocating the printing tasks based on the generated optimal scheduling plan while remaining within the material and battery constraints of each UAV.
☆ Toward Efficient and Robust Behavior Models for Multi-Agent Driving Simulation
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: This work has been submitted to the IEEE for possible publication
☆ Real-time Remote Tracking and Autonomous Planning for Whale Rendezvous using Robots
We introduce a system for real-time sperm whale rendezvous at sea using an autonomous uncrewed aerial vehicle. Our system employs model-based reinforcement learning that combines in situ sensor data with an empirical whale dive model to guide navigation decisions. Key challenges include (i) real-time acoustic tracking in the presence of multiple whales, (ii) distributed communication and decision-making for robot deployments, and (iii) on-board signal processing and long-range detection from fish-trackers. We evaluate our system by conducting rendezvous with sperm whales at sea in Dominica, performing hardware experiments on land, and running simulations using whale trajectories interpolated from marine biologists' surface observations.
☆ Global stability of vehicle-with-driver dynamics via Sum-of-Squares programming
This work estimates safe invariant subsets of the Region of Attraction (ROA) for a seven-state vehicle-with-driver system, capturing both asymptotic stability and the influence of state-safety bounds along the system trajectory. Safe sets are computed by optimizing Lyapunov functions through an original iterative Sum-of-Squares (SOS) procedure. The method is first demonstrated on a two-state benchmark, where it accurately recovers a prescribed safe region as the 1-level set of a polynomial Lyapunov function. We then describe the distinguishing characteristics of the studied vehicle-with-driver system: the control dynamics mimic human driver behavior through a delayed preview-tracking model that, with suitable parameter choices, can also emulate digital controllers. To enable SOS optimization, a polynomial approximation of the nonlinear vehicle model is derived, together with its operating-envelope constraints. The framework is then applied to understeering and oversteering scenarios, and the estimated safe sets are compared with reference boundaries obtained from exhaustive simulations. The results show that SOS techniques can efficiently deliver Lyapunov-defined safe regions, supporting their potential use for real-time safety assessment, for example as a supervisory layer for active vehicle control.
comment: 20 pages, 7 figures, 2 tables
☆ 3D Path Planning for Robot-assisted Vertebroplasty from Arbitrary Bi-plane X-ray via Differentiable Rendering
Robotic systems are transforming image-guided interventions by enhancing accuracy and minimizing radiation exposure. A significant challenge in robotic assistance lies in surgical path planning, which often relies on the registration of intraoperative 2D images with preoperative 3D CT scans. This requirement can be burdensome and costly, particularly in procedures like vertebroplasty, where preoperative CT scans are not routinely performed. To address this issue, we introduce a differentiable rendering-based framework for 3D transpedicular path planning utilizing bi-planar 2D X-rays. Our method integrates differentiable rendering with a vertebral atlas generated through a Statistical Shape Model (SSM) and employs a learned similarity loss to refine the SSM shape and pose dynamically, independent of fixed imaging geometries. We evaluated our framework in two stages: first, through vertebral reconstruction from orthogonal X-rays for benchmarking, and second, via clinician-in-the-loop path planning using arbitrary-view X-rays. Our results indicate that our method outperformed a normalized cross-correlation baseline in reconstruction metrics (DICE: 0.75 vs. 0.65) and achieved comparable performance to the state-of-the-art model ReVerteR (DICE: 0.77), while maintaining generalization to arbitrary views. Success rates for bipedicular planning reached 82% with synthetic data and 75% with cadaver data, exceeding the 66% and 31% rates of a 2D-to-3D baseline, respectively. In conclusion, our framework facilitates versatile, CT-free 3D path planning for robot-assisted vertebroplasty, effectively accommodating real-world imaging diversity without the need for preoperative CT scans.
☆ Label-Efficient Point Cloud Segmentation with Active Learning
Semantic segmentation of 3D point cloud data often comes with high annotation costs. Active learning automates the process of selecting which data to annotate, reducing the total amount of annotation needed to achieve satisfactory performance. Recent approaches to active learning for 3D point clouds are often based on sophisticated heuristics for both, splitting point clouds into annotatable regions and selecting the most beneficial for further neural network training. In this work, we propose a novel and easy-to-implement strategy to separate the point cloud into annotatable regions. In our approach, we utilize a 2D grid to subdivide the point cloud into columns. To identify the next data to be annotated, we employ a network ensemble to estimate the uncertainty in the network output. We evaluate our method on the S3DIS dataset, the Toronto-3D dataset, and a large-scale urban 3D point cloud of the city of Freiburg, which we labeled in parts manually. The extensive evaluation shows that our method yields performance on par with, or even better than, complex state-of-the-art methods on all datasets. Furthermore, we provide results suggesting that in the context of point clouds the annotated area can be a more meaningful measure for active learning algorithms than the number of annotated points.
☆ Bayesian Active Inference for Intelligent UAV Anti-Jamming and Adaptive Trajectory Planning
This paper proposes a hierarchical trajectory planning framework for UAVs operating under adversarial jamming conditions. Leveraging Bayesian Active Inference, the approach combines expert-generated demonstrations with probabilistic generative modeling to encode high-level symbolic planning, low-level motion policies, and wireless signal feedback. During deployment, the UAV performs online inference to anticipate interference, localize jammers, and adapt its trajectory accordingly, without prior knowledge of jammer locations. Simulation results demonstrate that the proposed method achieves near-expert performance, significantly reducing communication interference and mission cost compared to model-free reinforcement learning baselines, while maintaining robust generalization in dynamic environments.
comment: This paper has been accepted for the 2026 IEEE Consumer Communications & Networking Conference (IEEE CCNC 2026)
☆ HiMoE-VLA: Hierarchical Mixture-of-Experts for Generalist Vision-Language-Action Policies
The development of foundation models for embodied intelligence critically depends on access to large-scale, high-quality robot demonstration data. Recent approaches have sought to address this challenge by training on large collections of heterogeneous robotic datasets. However, unlike vision or language data, robotic demonstrations exhibit substantial heterogeneity across embodiments and action spaces as well as other prominent variations such as senor configurations and action control frequencies. The lack of explicit designs for handling such heterogeneity causes existing methods to struggle with integrating diverse factors, thereby limiting their generalization and leading to degraded performance when transferred to new settings. In this paper, we present HiMoE-VLA, a novel vision-language-action (VLA) framework tailored to effectively handle diverse robotic data with heterogeneity. Specifically, we introduce a Hierarchical Mixture-of-Experts (HiMoE) architecture for the action module which adaptively handles multiple sources of heterogeneity across layers and gradually abstracts them into shared knowledge representations. Through extensive experimentation with simulation benchmarks and real-world robotic platforms, HiMoE-VLA demonstrates a consistent performance boost over existing VLA baselines, achieving higher accuracy and robust generalization across diverse robots and action spaces. The code and models are publicly available at https://github.com/ZhiyingDu/HiMoE-VLA.
☆ Scenario-aware Uncertainty Quantification for Trajectory Prediction with Statistical Guarantees
Reliable uncertainty quantification in trajectory prediction is crucial for safety-critical autonomous driving systems, yet existing deep learning predictors lack uncertainty-aware frameworks adaptable to heterogeneous real-world scenarios. To bridge this gap, we propose a novel scenario-aware uncertainty quantification framework to provide the predicted trajectories with prediction intervals and reliability assessment. To begin with, predicted trajectories from the trained predictor and their ground truth are projected onto the map-derived reference routes within the Frenet coordinate system. We then employ CopulaCPTS as the conformal calibration method to generate temporal prediction intervals for distinct scenarios as the uncertainty measure. Building upon this, within the proposed trajectory reliability discriminator (TRD), mean error and calibrated confidence intervals are synergistically analyzed to establish reliability models for different scenarios. Subsequently, the risk-aware discriminator leverages a joint risk model that integrates longitudinal and lateral prediction intervals within the Frenet coordinate to identify critical points. This enables segmentation of trajectories into reliable and unreliable segments, holding the advantage of informing downstream planning modules with actionable reliability results. We evaluated our framework using the real-world nuPlan dataset, demonstrating its effectiveness in scenario-aware uncertainty quantification and reliability assessment across diverse driving contexts.
☆ An Integrated System for WEEE Sorting Employing X-ray Imaging, AI-based Object Detection and Segmentation, and Delta Robot Manipulation
Battery recycling is becoming increasingly critical due to the rapid growth in battery usage and the limited availability of natural resources. Moreover, as battery energy densities continue to rise, improper handling during recycling poses significant safety hazards, including potential fires at recycling facilities. Numerous systems have been proposed for battery detection and removal from WEEE recycling lines, including X-ray and RGB-based visual inspection methods, typically driven by AI-powered object detection models (e.g., Mask R-CNN, YOLO, ResNets). Despite advances in optimizing detection techniques and model modifications, a fully autonomous solution capable of accurately identifying and sorting batteries across diverse WEEEs types has yet to be realized. In response to these challenges, we present our novel approach which integrates a specialized X-ray transmission dual energy imaging subsystem with advanced pre-processing algorithms, enabling high-contrast image reconstruction for effective differentiation of dense and thin materials in WEEE. Devices move along a conveyor belt through a high-resolution X-ray imaging system, where YOLO and U-Net models precisely detect and segment battery-containing items. An intelligent tracking and position estimation algorithm then guides a Delta robot equipped with a suction gripper to selectively extract and properly discard the targeted devices. The approach is validated in a photorealistic simulation environment developed in NVIDIA Isaac Sim and on the real setup.
☆ A Comprehensive Framework for Automated Quality Control in the Automotive Industry
This paper presents a cutting-edge robotic inspection solution designed to automate quality control in automotive manufacturing. The system integrates a pair of collaborative robots, each equipped with a high-resolution camera-based vision system to accurately detect and localize surface and thread defects in aluminum high-pressure die casting (HPDC) automotive components. In addition, specialized lenses and optimized lighting configurations are employed to ensure consistent and high-quality image acquisition. The YOLO11n deep learning model is utilized, incorporating additional enhancements such as image slicing, ensemble learning, and bounding-box merging to significantly improve performance and minimize false detections. Furthermore, image processing techniques are applied to estimate the extent of the detected defects. Experimental results demonstrate real-time performance with high accuracy across a wide variety of defects, while minimizing false detections. The proposed solution is promising and highly scalable, providing the flexibility to adapt to various production environments and meet the evolving demands of the automotive industry.
☆ A Hyperspectral Imaging Guided Robotic Grasping System
Hyperspectral imaging is an advanced technique for precisely identifying and analyzing materials or objects. However, its integration with robotic grasping systems has so far been explored due to the deployment complexities and prohibitive costs. Within this paper, we introduce a novel hyperspectral imaging-guided robotic grasping system. The system consists of PRISM (Polyhedral Reflective Imaging Scanning Mechanism) and the SpectralGrasp framework. PRISM is designed to enable high-precision, distortion-free hyperspectral imaging while simplifying system integration and costs. SpectralGrasp generates robotic grasping strategies by effectively leveraging both the spatial and spectral information from hyperspectral images. The proposed system demonstrates substantial improvements in both textile recognition compared to human performance and sorting success rate compared to RGB-based methods. Additionally, a series of comparative experiments further validates the effectiveness of our system. The study highlights the potential benefits of integrating hyperspectral imaging with robotic grasping systems, showcasing enhanced recognition and grasping capabilities in complex and dynamic environments. The project is available at: https://zainzh.github.io/PRISM.
comment: 8 pages, 7 figures, Accepted to IEEE Robotics and Automation Letters (RA-L) 2025
☆ Spatiotemporal Tubes for Differential Drive Robots with Model Uncertainty
This paper presents a Spatiotemporal Tube (STT)-based control framework for differential-drive mobile robots with dynamic uncertainties and external disturbances, guaranteeing the satisfaction of Temporal Reach-Avoid-Stay (T-RAS) specifications. The approach employs circular STT, characterized by smoothly time-varying center and radius, to define dynamic safe corridors that guide the robot from the start region to the goal while avoiding obstacles. In particular, we first develop a sampling-based synthesis algorithm to construct a feasible STT that satisfies the prescribed timing and safety constraints with formal guarantees. To ensure that the robot remains confined within this tube, we then design analytically a closed-form, approximation-free control law. The resulting controller is computationally efficient, robust to disturbances and {model uncertainties}, and requires no model approximations or online optimization. The proposed framework is validated through simulation studies on a differential-drive robot and benchmarked against state-of-the-art methods, demonstrating superior robustness, accuracy, and computational efficiency.
☆ State-Conditional Adversarial Learning: An Off-Policy Visual Domain Transfer Method for End-to-End Imitation Learning
We study visual domain transfer for end-to-end imitation learning in a realistic and challenging setting where target-domain data are strictly off-policy, expert-free, and scarce. We first provide a theoretical analysis showing that the target-domain imitation loss can be upper bounded by the source-domain loss plus a state-conditional latent KL divergence between source and target observation models. Guided by this result, we propose State- Conditional Adversarial Learning, an off-policy adversarial framework that aligns latent distributions conditioned on system state using a discriminator-based estimator of the conditional KL term. Experiments on visually diverse autonomous driving environments built on the BARC-CARLA simulator demonstrate that SCAL achieves robust transfer and strong sample efficiency.
☆ Where to Fly, What to Send: Communication-Aware Aerial Support for Ground Robots
In this work we consider a multi-robot team operating in an unknown environment where one aerial agent is tasked to map the environment and transmit (a portion of) the mapped environment to a group of ground agents that are trying to reach their goals. The entire operation takes place over a bandwidth-limited communication channel, which motivates the problem of determining what and how much information the assisting agent should transmit and when while simultaneously performing exploration/mapping. The proposed framework enables the assisting aerial agent to decide what information to transmit based on the Value-of-Information (VoI), how much to transmit using a Mixed-Integer Linear Programming (MILP), and how to acquire additional information through an utility score-based environment exploration strategy. We perform a communication-motion trade-off analysis between the total amount of map data communicated by the aerial agent and the navigation cost incurred by the ground agents.
comment: Submitted to conference
☆ Cascaded Tightly-Coupled Observer Design for Single-Range-Aided Inertial Navigation
This work introduces a single-range-aided navigation observer that reconstructs the full state of a rigid body using only an Inertial Measurement Unit (IMU), a body-frame vector measurement (e.g., magnetometer), and a distance measurement from a fixed anchor point. The design first formulates an extended linear time-varying (LTV) system to estimate body-frame position, body-frame velocity, and the gravity direction. The recovered gravity direction, combined with the body-frame vector measurement, is then used to reconstruct the full orientation on $\mathrm{SO}(3)$, resulting in a cascaded observer architecture. Almost Global Asymptotic Stability (AGAS) of the cascaded design is established under a uniform observability condition, ensuring robustness to sensor noise and trajectory variations. Simulation studies on three-dimensional trajectories demonstrate accurate estimation of position, velocity, and orientation, highlighting single-range aiding as a lightweight and effective modality for autonomous navigation.
comment: 8 pages
☆ REWW-ARM -- Remote Wire-Driven Mobile Robot: Design, Control, and Experimental Validation
Electronic devices are essential for robots but limit their usable environments. To overcome this, methods excluding electronics from the operating environment while retaining advanced electronic control and actuation have been explored. These include the remote hydraulic drive of electronics-free mobile robots, which offer high reachability, and long wire-driven robot arms with motors consolidated at the base, which offer high environmental resistance. To combine the advantages of both, this study proposes a new system, "Remote Wire Drive." As a proof-of-concept, we designed and developed the Remote Wire-Driven robot "REWW-ARM", which consists of the following components: 1) a novel power transmission mechanism, the "Remote Wire Transmission Mechanism" (RWTM), the key technology of the Remote Wire Drive; 2) an electronics-free distal mobile robot driven by it; and 3) a motor-unit that generates power and provides electronic closed-loop control based on state estimation via the RWTM. In this study, we evaluated the mechanical and control performance of REWW-ARM through several experiments, demonstrating its capability for locomotion, posture control, and object manipulation both on land and underwater. This suggests the potential for applying the Remote Wire-Driven system to various types of robots, thereby expanding their operational range.
comment: Accepted on Advanced Intelligent Systems
☆ Situation-Aware Interactive MPC Switching for Autonomous Driving
To enable autonomous driving in interactive traffic scenarios, various model predictive control (MPC) formulations have been proposed, each employing different interaction models. While higher-fidelity models enable more intelligent behavior, they incur increased computational cost. Since strong interactions are relatively infrequent in traffic, a practical strategy for balancing performance and computational overhead is to invoke an appropriate controller based on situational demands. To achieve this approach, we first conduct a comparative study to assess and hierarchize the interactive capabilities of different MPC formulations. Furthermore, we develop a neural network-based classifier to enable situation-aware switching among controllers with different levels of interactive capability. We demonstrate that this situation-aware switching can both substantially improve overall performance by activating the most advanced interactive MPC in rare but critical situations, and significantly reduce computational load by using a basic MPC in the majority of scenarios.
☆ Real-Time Spatiotemporal Tubes for Dynamic Unsafe Sets
This paper presents a real-time control framework for nonlinear pure-feedback systems with unknown dynamics to satisfy reach-avoid-stay tasks within a prescribed time in dynamic environments. To achieve this, we introduce a real-time spatiotemporal tube (STT) framework. An STT is defined as a time-varying ball in the state space whose center and radius adapt online using only real-time sensory input. A closed-form, approximation-free control law is then derived to constrain the system output within the STT, ensuring safety and task satisfaction. We provide formal guarantees for obstacle avoidance and on-time task completion. The effectiveness and scalability of the framework are demonstrated through simulations and hardware experiments on a mobile robot and an aerial vehicle, navigating in cluttered dynamic environments.
☆ GuideNav: User-Informed Development of a Vision-Only Robotic Navigation Assistant For Blind Travelers
While commendable progress has been made in user-centric research on mobile assistive systems for blind and low-vision (BLV) individuals, references that directly inform robot navigation design remain rare. To bridge this gap, we conducted a comprehensive human study involving interviews with 26 guide dog handlers, four white cane users, nine guide dog trainers, and one O\&M trainer, along with 15+ hours of observing guide dog-assisted walking. After de-identification, we open-sourced the dataset to promote human-centered development and informed decision-making for assistive systems for BLV people. Building on insights from this formative study, we developed GuideNav, a vision-only, teach-and-repeat navigation system. Inspired by how guide dogs are trained and assist their handlers, GuideNav autonomously repeats a path demonstrated by a sighted person using a robot. Specifically, the system constructs a topological representation of the taught route, integrates visual place recognition with temporal filtering, and employs a relative pose estimator to compute navigation actions - all without relying on costly, heavy, power-hungry sensors such as LiDAR. In field tests, GuideNav consistently achieved kilometer-scale route following across five outdoor environments, maintaining reliability despite noticeable scene variations between teach and repeat runs. A user study with 3 guide dog handlers and 1 guide dog trainer further confirmed the system's feasibility, marking (to our knowledge) the first demonstration of a quadruped mobile system retrieving a path in a manner comparable to guide dogs.
☆ Probabilistic Weapon Engagement Zones for a Turn Constrained Pursuer
Curve-straight probabilistic engagement zones (CSPEZ) quantify the spatial regions an evader should avoid to reduce capture risk from a turn-rate-limited pursuer following a curve-straight path with uncertain parameters including position, heading, velocity, range, and maximum turn rate. This paper presents methods for generating evader trajectories that minimize capture risk under such uncertainty. We first derive an analytic solution for the deterministic curve-straight basic engagement zone (CSBEZ), then extend this formulation to a probabilistic framework using four uncertainty-propagation approaches: Monte Carlo sampling, linearization, quadratic approximation, and neural-network regression. We evaluate the accuracy and computational cost of each approximation method and demonstrate how CSPEZ constraints can be integrated into a trajectory-optimization algorithm to produce safe paths that explicitly account for pursuer uncertainty.
comment: Accepted for presentation at AIAA SciTech 2026. 17 pages, 7 figures
☆ WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving
We introduce WAM-Flow, a vision-language-action (VLA) model that casts ego-trajectory planning as discrete flow matching over a structured token space. In contrast to autoregressive decoders, WAM-Flow performs fully parallel, bidirectional denoising, enabling coarse-to-fine refinement with a tunable compute-accuracy trade-off. Specifically, the approach combines a metric-aligned numerical tokenizer that preserves scalar geometry via triplet-margin learning, a geometry-aware flow objective and a simulator-guided GRPO alignment that integrates safety, ego progress, and comfort rewards while retaining parallel generation. A multi-stage adaptation converts a pre-trained auto-regressive backbone (Janus-1.5B) from causal decoding to non-causal flow model and strengthens road-scene competence through continued multimodal pretraining. Thanks to the inherent nature of consistency model training and parallel decoding inference, WAM-Flow achieves superior closed-loop performance against autoregressive and diffusion-based VLA baselines, with 1-step inference attaining 89.1 PDMS and 5-step inference reaching 90.3 PDMS on NAVSIM v1 benchmark. These results establish discrete flow matching as a new promising paradigm for end-to-end autonomous driving. The code will be publicly available soon.
comment: 18 pages, 11 figures
☆ Unifying Entropy Regularization in Optimal Control: From and Back to Classical Objectives via Iterated Soft Policies and Path Integral Solutions
This paper develops a unified perspective on several stochastic optimal control formulations through the lens of Kullback-Leibler regularization. We propose a central problem that separates the KL penalties on policies and transitions, assigning them independent weights, thereby generalizing the standard trajectory-level KL-regularization commonly used in probabilistic and KL-regularized control. This generalized formulation acts as a generative structure allowing to recover various control problems. These include the classical Stochastic Optimal Control (SOC), Risk-Sensitive Optimal Control (RSOC), and their policy-based KL-regularized counterparts. The latter we refer to as soft-policy SOC and RSOC, facilitating alternative problems with tractable solutions. Beyond serving as regularized variants, we show that these soft-policy formulations majorize the original SOC and RSOC problem. This means that the regularized solution can be iterated to retrieve the original solution. Furthermore, we identify a structurally synchronized case of the risk-seeking soft-policy RSOC formulation, wherein the policy and transition KL-regularization weights coincide. Remarkably, this specific setting gives rise to several powerful properties such as a linear Bellman equation, path integral solution, and, compositionality, thereby extending these computationally favourable properties to a broad class of control problems.
♻ ☆ STATE-NAV: Stability-Aware Traversability Estimation for Bipedal Navigation on Rough Terrain
Bipedal robots have advantages in maneuvering human-centered environments, but face greater failure risk compared to other stable mobile platforms such as wheeled or quadrupedal robots. While learning-based traversability has been widely studied for these platforms, bipedal traversability has instead relied on manually designed rules with limited consideration of locomotion stability on rough terrain. In this work, we present the first learning-based traversability estimation and risk-sensitive navigation framework for bipedal robots operating in diverse, uneven environments. TravFormer, a transformer-based neural network, is trained to predict bipedal instability with uncertainty, enabling risk-aware and adaptive planning. Based on the network, we define traversability as stability-aware command velocity-the fastest command velocity that keeps instability below a user-defined limit. This velocity-based traversability is integrated into a hierarchical planner that combines traversability-informed Rapid Random Tree Star (TravRRT*) for time-efficient planning and Model Predictive Control (MPC) for safe execution. We validate our method in MuJoCo simulation and the real world, demonstrating improved navigation performance, with enhanced robustness and time efficiency across varying terrains compared to existing methods.
comment: Accepted to IEEE Robotics and Automation Letters (RA-L)
♻ ☆ Actor-Critic Model Predictive Control: Differentiable Optimization meets Reinforcement Learning for Agile Flight
A key open challenge in agile quadrotor flight is how to combine the flexibility and task-level generality of model-free reinforcement learning (RL) with the structure and online replanning capabilities of model predictive control (MPC), aiming to leverage their complementary strengths in dynamic and uncertain environments. This paper provides an answer by introducing a new framework called Actor-Critic Model Predictive Control. The key idea is to embed a differentiable MPC within an actor-critic RL framework. This integration allows for short-term predictive optimization of control actions through MPC, while leveraging RL for end-to-end learning and exploration over longer horizons. Through various ablation studies, conducted in the context of agile quadrotor racing, we expose the benefits of the proposed approach: it achieves better out-of-distribution behaviour, better robustness to changes in the quadrotor's dynamics and improved sample efficiency. Additionally, we conduct an empirical analysis using a quadrotor platform that reveals a relationship between the critic's learned value function and the cost function of the differentiable MPC, providing a deeper understanding of the interplay between the critic's value and the MPC cost functions. Finally, we validate our method in a drone racing task on different tracks, in both simulation and the real world. Our method achieves the same superhuman performance as state-of-the-art model-free RL, showcasing speeds of up to 21 m/s. We show that the proposed architecture can achieve real-time control performance, learn complex behaviors via trial and error, and retain the predictive properties of the MPC to better handle out-of-distribution behavior.
comment: 20 pages, 13 figures, evolved paper
♻ ☆ A neural signed configuration distance function for path planning of picking manipulators
Picking manipulators are task specific robots, with fewer degrees of freedom compared to general-purpose manipulators, and are heavily used in industry. The efficiency of the picking robots is highly dependent on the path planning solution, which is commonly done using sampling-based multi-query methods. The planner is robustly able to solve the problem, but its heavy use of collision-detection limits the planning capabilities for online use. We approach this problem by presenting a novel implicit obstacle representation for path planning, a neural signed configuration distance function (nSCDF), which allows us to form collision-free balls in the configuration space. We use the ball representation to re-formulate a state of the art multi-query path planner, i.e., instead of points, we use balls in the graph. Our planner returns a collision-free corridor, which allows us to use convex programming to produce optimized paths. From our numerical experiments, we observe that our planner produces paths that are close to those from an asymptotically optimal path planner, in significantly less time.
♻ ☆ Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving
Efficient data utilization is crucial for advancing 3D scene understanding in autonomous driving, where reliance on heavily human-annotated LiDAR point clouds challenges fully supervised methods. Addressing this, our study extends into semi-supervised learning for LiDAR semantic segmentation, leveraging the intrinsic spatial priors of driving scenes and multi-sensor complements to augment the efficacy of unlabeled datasets. We introduce LaserMix++, an evolved framework that integrates laser beam manipulations from disparate LiDAR scans and incorporates LiDAR-camera correspondences to further assist data-efficient learning. Our framework is tailored to enhance 3D scene consistency regularization by incorporating multi-modality, including 1) multi-modal LaserMix operation for fine-grained cross-sensor interactions; 2) camera-to-LiDAR feature distillation that enhances LiDAR feature learning; and 3) language-driven knowledge guidance generating auxiliary supervisions using open-vocabulary models. The versatility of LaserMix++ enables applications across LiDAR representations, establishing it as a universally applicable solution. Our framework is rigorously validated through theoretical analysis and extensive experiments on popular driving perception datasets. Results demonstrate that LaserMix++ markedly outperforms fully supervised alternatives, achieving comparable accuracy with five times fewer annotations and significantly improving the supervised-only baselines. This substantial advancement underscores the potential of semi-supervised approaches in reducing the reliance on extensive labeled data in LiDAR-based 3D scene understanding systems.
comment: IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
♻ ☆ Semantic Communication and Control Co-Design for Multi-Objective Distinct Dynamics
This letter introduces a machine-learning approach to learning the semantic dynamics of correlated systems with different control rules and dynamics. By leveraging the Koopman operator in an autoencoder (AE) framework, the system's state evolution is linearized in the latent space using a dynamic semantic Koopman (DSK) model, capturing the baseline semantic dynamics. Signal temporal logic (STL) is incorporated through a logical semantic Koopman (LSK) model to encode system-specific control rules. These models form the proposed logical Koopman AE framework that reduces communication costs while improving state prediction accuracy and control performance, showing a 91.65% reduction in communication samples and significant performance gains in simulation.
♻ ☆ Learning Visually Interpretable Oscillator Networks for Soft Continuum Robots from Video
Data-driven learning of soft continuum robot (SCR) dynamics from high-dimensional observations offers flexibility but often lacks physical interpretability, while model-based approaches require prior knowledge and can be computationally expensive. We bridge this gap by introducing (1) the Attention Broadcast Decoder (ABCD), a plug-and-play module for autoencoder-based latent dynamics learning that generates pixel-accurate attention maps localizing each latent dimension's contribution while filtering static backgrounds. (2) By coupling these attention maps to 2D oscillator networks, we enable direct on-image visualization of learned dynamics (masses, stiffness, and forces) without prior knowledge. We validate our approach on single- and double-segment SCRs, demonstrating that ABCD-based models significantly improve multi-step prediction accuracy: 5.7x error reduction for Koopman operators and 3.5x for oscillator networks on the two-segment robot. The learned oscillator network autonomously discovers a chain structure of oscillators. Unlike standard methods, ABCD models enable smooth latent space extrapolation beyond training data. This fully data-driven approach yields compact, physically interpretable models suitable for control applications.
comment: Dataset available at: https://zenodo.org/records/17812071
♻ ☆ Perspective-Invariant 3D Object Detection ICCV 2025
With the rise of robotics, LiDAR-based 3D object detection has garnered significant attention in both academia and industry. However, existing datasets and methods predominantly focus on vehicle-mounted platforms, leaving other autonomous platforms underexplored. To bridge this gap, we introduce Pi3DET, the first benchmark featuring LiDAR data and 3D bounding box annotations collected from multiple platforms: vehicle, quadruped, and drone, thereby facilitating research in 3D object detection for non-vehicle platforms as well as cross-platform 3D detection. Based on Pi3DET, we propose a novel cross-platform adaptation framework that transfers knowledge from the well-studied vehicle platform to other platforms. This framework achieves perspective-invariant 3D detection through robust alignment at both geometric and feature levels. Additionally, we establish a benchmark to evaluate the resilience and robustness of current 3D detectors in cross-platform scenarios, providing valuable insights for developing adaptive 3D perception systems. Extensive experiments validate the effectiveness of our approach on challenging cross-platform tasks, demonstrating substantial gains over existing adaptation methods. We hope this work paves the way for generalizable and unified 3D perception systems across diverse and complex environments. Our Pi3DET dataset, cross-platform benchmark suite, and annotation toolkit have been made publicly available.
comment: ICCV 2025; 54 pages, 18 figures, 22 tables; Project Page at https://pi3det.github.io
♻ ☆ Point-PNG: Conditional Pseudo-Negatives Generation for Point Cloud Pre-Training
We propose Point-PNG, a novel self-supervised learning framework that generates conditional pseudo-negatives in the latent space to learn point cloud representations that are both discriminative and transformation-sensitive. Conventional self-supervised learning methods focus on achieving invariance, discarding transformation-specific information. Recent approaches incorporate transformation sensitivity by explicitly modeling relationships between original and transformed inputs. However, they often suffer from an invariant-collapse phenomenon, where the predictor degenerates into identity mappings, resulting in latent representations with limited variation across transformations. To address this, we propose Point-PNG that explicitly penalizes invariant collapse through pseudo-negatives generation, enabling the network to capture richer transformation cues while preserving discriminative representations. To this end, we introduce a parametric network, COnditional Pseudo-Negatives Embedding (COPE), which learns localized displacements induced by transformations within the latent space. A key challenge arises when jointly training COPE with the MAE, as it tends to converge to trivial identity mappings. To overcome this, we design a loss function based on pseudo-negatives conditioned on the transformation, which penalizes such trivial invariant solutions and enforces meaningful representation learning. We validate Point-PNG on shape classification and relative pose estimation tasks, showing competitive performance on ModelNet40 and ScanObjectNN under challenging evaluation protocols, and achieving superior accuracy in relative pose estimation compared to supervised baselines.
comment: Accepted for publication in IEEE ACCESS
♻ ☆ ReSem3D: Refinable 3D Spatial Constraints via Fine-Grained Semantic Grounding for Generalizable Robotic Manipulation
Semantics-driven 3D spatial constraints align highlevel semantic representations with low-level action spaces, facilitating the unification of task understanding and execution in robotic manipulation. The synergistic reasoning of Multimodal Large Language Models (MLLMs) and Vision Foundation Models (VFMs) enables cross-modal 3D spatial constraint construction. Nevertheless, existing methods have three key limitations: (1) coarse semantic granularity in constraint modeling, (2) lack of real-time closed-loop planning, (3) compromised robustness in semantically diverse environments. To address these challenges, we propose ReSem3D, a unified manipulation framework for semantically diverse environments, leveraging the synergy between VFMs and MLLMs to achieve fine-grained visual grounding and dynamically constructs hierarchical 3D spatial constraints for real-time manipulation. Specifically, the framework is driven by hierarchical recursive reasoning in MLLMs, which interact with VFMs to automatically construct 3D spatial constraints from natural language instructions and RGB-D observations in two stages: part-level extraction and region-level refinement. Subsequently, these constraints are encoded as real-time optimization objectives in joint space, enabling reactive behavior to dynamic disturbances. Extensive simulation and real-world experiments are conducted in semantically rich household and sparse chemical lab environments. The results demonstrate that ReSem3D performs diverse manipulation tasks under zero-shot conditions, exhibiting strong adaptability and generalization. Code and videos are available at https://github.com/scy-v/ReSem3D and https://resem3d.github.io.
comment: 12 pages,9 figures
♻ ☆ Momentum-constrained Hybrid Heuristic Trajectory Optimization Framework with Residual-enhanced DRL for Visually Impaired Scenarios
This paper proposes a momentum-constrained hybrid heuristic trajectory optimization framework (MHHTOF) tailored for assistive navigation in visually impaired scenarios, integrating trajectory sampling generation, optimization and evaluation with residual-enhanced deep reinforcement learning (DRL). In the first stage, heuristic trajectory sampling cluster (HTSC) is generated in the Frenet coordinate system using third-order interpolation with fifth-order polynomials and momentum-constrained trajectory optimization (MTO) constraints to ensure smoothness and feasibility. After first stage cost evaluation, the second stage leverages a residual-enhanced actor-critic network with LSTM-based temporal feature modeling to adaptively refine trajectory selection in the Cartesian coordinate system. A dual-stage cost modeling mechanism (DCMM) with weight transfer aligns semantic priorities across stages, supporting human-centered optimization. Experimental results demonstrate that the proposed LSTM-ResB-PPO achieves significantly faster convergence, attaining stable policy performance in approximately half the training iterations required by the PPO baseline, while simultaneously enhancing both reward outcomes and training stability. Compared to baseline method, the selected model reduces average cost and cost variance by 30.3% and 53.3%, and lowers ego and obstacle risks by over 77%. These findings validate the framework's effectiveness in enhancing robustness, safety, and real-time feasibility in complex assistive planning tasks.
comment: Upon further internal evaluation, we found that the current version does not adequately represent the clarity and completeness that we intend for this work. To avoid possible misunderstanding caused by this preliminary form, we request withdrawal. A refined version will be prepared privately before any further dissemination
♻ ☆ GEX: Democratizing Dexterity with Fully-Actuated Dexterous Hand and Exoskeleton Glove
This paper introduces GEX, an innovative low-cost dexterous manipulation system that combines the GX11 tri-finger anthropomorphic hand (11 DoF) with the EX12 tri-finger exoskeleton glove (12 DoF), forming a closed-loop teleoperation framework through kinematic retargeting for high-fidelity control. Both components employ modular 3D-printed finger designs, achieving ultra-low manufacturing costs while maintaining full actuation capabilities. Departing from conventional tendon-driven or underactuated approaches, our electromechanical system integrates independent joint motors across all 23 DoF, ensuring complete state observability and accurate kinematic modeling. This full-actuation architecture enables precise bidirectional kinematic calculations, substantially enhancing kinematic retargeting fidelity between the exoskeleton and robotic hand. The proposed system bridges the cost-performance gap in dexterous manipulation research, providing an accessible platform for acquiring high-quality demonstration data to advance embodied AI and dexterous robotic skill transfer learning.
♻ ☆ Real-Time Execution of Action Chunking Flow Policies NeurIPS 2025
Modern AI systems, especially those interacting with the physical world, increasingly require real-time performance. However, the high latency of state-of-the-art generalist models, including recent vision-language action models (VLAs), poses a significant challenge. While action chunking has enabled temporal consistency in high-frequency control tasks, it does not fully address the latency problem, leading to pauses or out-of-distribution jerky movements at chunk boundaries. This paper presents a novel inference-time algorithm that enables smooth asynchronous execution of action chunking policies. Our method, real-time chunking (RTC), is applicable to any diffusion- or flow-based VLA out of the box with no re-training. It generates the next action chunk while executing the current one, "freezing" actions guaranteed to execute and "inpainting" the rest. To test RTC, we introduce a new benchmark of 12 highly dynamic tasks in the Kinetix simulator, as well as evaluate 6 challenging real-world bimanual manipulation tasks. Results demonstrate that RTC is fast, performant, and uniquely robust to inference delay, significantly improving task throughput and enabling high success rates in precise tasks $\unicode{x2013}$ such as lighting a match $\unicode{x2013}$ even in the presence of significant latency. See https://pi.website/research/real_time_chunking for videos.
comment: published in NeurIPS 2025
♻ ☆ Evo-1: Lightweight Vision-Language-Action Model with Preserved Semantic Alignment
Vision-Language-Action (VLA) models have emerged as a powerful framework that unifies perception, language, and control, enabling robots to perform diverse tasks through multimodal understanding. However, current VLA models typically contain massive parameters and rely heavily on large-scale robot data pretraining, leading to high computational costs during training, as well as limited deployability for real-time inference. Moreover, most training paradigms often degrade the perceptual representations of the vision-language backbone, resulting in overfitting and poor generalization to downstream tasks. In this work, we present Evo-1, a lightweight VLA model that reduces computation and improves deployment efficiency, while maintaining strong performance without pretraining on robot data. Evo-1 builds on a native multimodal Vision-Language model (VLM), incorporating a novel cross-modulated diffusion transformer along with an optimized integration module, together forming an effective architecture. We further introduce a two-stage training paradigm that progressively aligns action with perception, preserving the representations of the VLM. Notably, with only 0.77 billion parameters, Evo-1 achieves state-of-the-art results on the Meta-World and RoboTwin suite, surpassing the previous best models by 12.4% and 6.9%, respectively, and also attains a competitive result of 94.8% on LIBERO. In real-world evaluations, Evo-1 attains a 78% success rate with high inference frequency and low memory overhead, outperforming all baseline methods. We release code, data, and model weights to facilitate future research on lightweight and efficient VLA models.
comment: Github: https://github.com/MINT-SJTU/Evo-1
♻ ☆ IS-Bench: Evaluating Interactive Safety of VLM-Driven Embodied Agents in Daily Household Tasks
Flawed planning from VLM-driven embodied agents poses significant safety hazards, hindering their deployment in real-world household tasks. However, existing static, non-interactive evaluation paradigms fail to adequately assess risks within these interactive environments, since they cannot simulate dynamic risks that emerge from an agent's actions and rely on unreliable post-hoc evaluations that ignore unsafe intermediate steps. To bridge this critical gap, we propose evaluating an agent's interactive safety: its ability to perceive emergent risks and execute mitigation steps in the correct procedural order. We thus present IS-Bench, the first multi-modal benchmark designed for interactive safety, featuring 161 challenging scenarios with 388 unique safety risks instantiated in a high-fidelity simulator. Crucially, it facilitates a novel process-oriented evaluation that verifies whether risk mitigation actions are performed before/after specific risk-prone steps. Extensive experiments on leading VLMs, including the GPT-4o and Gemini-2.5 series, reveal that current agents lack interactive safety awareness, and that while safety-aware Chain-of-Thought can improve performance, it often compromises task completion. By highlighting these critical limitations, IS-Bench provides a foundation for developing safer and more reliable embodied AI systems. Code and data are released under https://github.com/AI45Lab/IS-Bench.
♻ ☆ H-GAR: A Hierarchical Interaction Framework via Goal-Driven Observation-Action Refinement for Robotic Manipulation AAAI 2026
Unified video and action prediction models hold great potential for robotic manipulation, as future observations offer contextual cues for planning, while actions reveal how interactions shape the environment. However, most existing approaches treat observation and action generation in a monolithic and goal-agnostic manner, often leading to semantically misaligned predictions and incoherent behaviors. To this end, we propose H-GAR, a Hierarchical interaction framework via Goal-driven observation-Action Refinement.To anchor prediction to the task objective, H-GAR first produces a goal observation and a coarse action sketch that outline a high-level route toward the goal. To enable explicit interaction between observation and action under the guidance of the goal observation for more coherent decision-making, we devise two synergistic modules. (1) Goal-Conditioned Observation Synthesizer (GOS) synthesizes intermediate observations based on the coarse-grained actions and the predicted goal observation. (2) Interaction-Aware Action Refiner (IAAR) refines coarse actions into fine-grained, goal-consistent actions by leveraging feedback from the intermediate observations and a Historical Action Memory Bank that encodes prior actions to ensure temporal consistency. By integrating goal grounding with explicit action-observation interaction in a coarse-to-fine manner, H-GAR enables more accurate manipulation. Extensive experiments on both simulation and real-world robotic manipulation tasks demonstrate that H-GAR achieves state-of-the-art performance.
comment: Accepted to AAAI 2026 (Oral), Project Page: https://github.com/JiuTian-VL/H-GAR
♻ ☆ LLM-Driven Corrective Robot Operation Code Generation with Static Text-Based Simulation
Recent advances in Large language models (LLMs) have demonstrated their promising capabilities of generating robot operation code to enable LLM-driven robots. To enhance the reliability of operation code generated by LLMs, corrective designs with feedback from the observation of executing code have been increasingly adopted in existing research. However, the code execution in these designs relies on either a physical experiment or a customized simulation environment, which limits their deployment due to the high configuration effort of the environment and the potential long execution time. In this paper, we explore the possibility of directly leveraging LLM to enable static simulation of robot operation code, and then leverage it to design a new reliable LLM-driven corrective robot operation code generation framework. Our framework configures the LLM as a static simulator with enhanced capabilities that reliably simulate robot code execution by interpreting actions, reasoning over state transitions, analyzing execution outcomes, and generating semantic observations that accurately capture trajectory dynamics. To validate the performance of our framework, we performed experiments on various operation tasks for different robots, including UAVs and small ground vehicles. The experiment results not only demonstrated the high accuracy of our static text-based simulation but also the reliable code generation of our LLM-driven corrective framework, which achieves a comparable performance with state-of-the-art research while does not rely on dynamic code execution using physical experiments or simulators.
comment: 8 pages, 2 figures
♻ ☆ Uni-Hand: Universal Hand Motion Forecasting in Egocentric Views IROS'25
Forecasting how human hands move in egocentric views is critical for applications like augmented reality and human-robot policy transfer. Recently, several hand trajectory prediction (HTP) methods have been developed to generate future possible hand waypoints, which still suffer from insufficient prediction targets, inherent modality gaps, entangled hand-head motion, and limited validation in downstream tasks. To address these limitations, we present a universal hand motion forecasting framework considering multi-modal input, multi-dimensional and multi-target prediction patterns, and multi-task affordances for downstream applications. We harmonize multiple modalities by vision-language fusion, global context incorporation, and task-aware text embedding injection, to forecast hand waypoints in both 2D and 3D spaces. A novel dual-branch diffusion is proposed to concurrently predict human head and hand movements, capturing their motion synergy in egocentric vision. By introducing target indicators, the prediction model can forecast the specific joint waypoints of the wrist or the fingers, besides the widely studied hand center points. In addition, we enable Uni-Hand to additionally predict hand-object interaction states (contact/separation) to facilitate downstream tasks better. As the first work to incorporate downstream task evaluation in the literature, we build novel benchmarks to assess the real-world applicability of hand motion forecasting algorithms. The experimental results on multiple publicly available datasets and our newly proposed benchmarks demonstrate that Uni-Hand achieves the state-of-the-art performance in multi-dimensional and multi-target hand motion forecasting. Extensive validation in multiple downstream tasks also presents its impressive human-robot policy transfer to enable robotic manipulation, and effective feature enhancement for action anticipation/recognition.
comment: Extended journal version of MMTwin (IROS'25). Code and data: https://github.com/IRMVLab/UniHand
♻ ☆ Adaptive Keyframe Selection for Scalable 3D Scene Reconstruction in Dynamic Environments
In this paper, we propose an adaptive keyframe selection method for improved 3D scene reconstruction in dynamic environments. The proposed method integrates two complementary modules: an error-based selection module utilizing photometric and structural similarity (SSIM) errors, and a momentum-based update module that dynamically adjusts keyframe selection thresholds according to scene motion dynamics. By dynamically curating the most informative frames, our approach addresses a key data bottleneck in real-time perception. This allows for the creation of high-quality 3D world representations from a compressed data stream, a critical step towards scalable robot learning and deployment in complex, dynamic environments. Experimental results demonstrate significant improvements over traditional static keyframe selection strategies, such as fixed temporal intervals or uniform frame skipping. These findings highlight a meaningful advancement toward adaptive perception systems that can dynamically respond to complex and evolving visual scenes. We evaluate our proposed adaptive keyframe selection module on two recent state-of-the-art 3D reconstruction networks, Spann3r and CUT3R, and observe consistent improvements in reconstruction quality across both frameworks. Furthermore, an extensive ablation study confirms the effectiveness of each individual component in our method, underlining their contribution to the overall performance gains.
comment: Accepted at ROBOVIS 2026
♻ ☆ DexFruit: Dexterous Manipulation and Gaussian Splatting Inspection of Fruit
DexFruit is a robotic manipulation framework that enables gentle, autonomous handling of fragile fruit and precise evaluation of damage. Many fruits are fragile and prone to bruising, thus requiring humans to manually harvest them with care. In this work, we demonstrate by using optical tactile sensing, autonomous manipulation of fruit with minimal damage can be achieved. We show that our tactile informed diffusion policies outperform baselines in both reduced bruising and pick-and-place success rate across three fruits: strawberries, tomatoes, and blackberries. In addition, we introduce FruitSplat, a novel technique to represent and quantify visual damage in high-resolution 3D representation via 3D Gaussian Splatting (3DGS). Existing metrics for measuring damage lack quantitative rigor or require expensive equipment. With FruitSplat, we distill a 2D strawberry mask as well as a 2D bruise segmentation mask into the 3DGS representation. Furthermore, this representation is modular and general, compatible with any relevant 2D model. Overall, we demonstrate a 92% grasping policy success rate, up to a 20% reduction in visual bruising, and up to an 31% improvement in grasp success rate on challenging fruit compared to our baselines across our three tested fruits. We rigorously evaluate this result with over 630 trials. Please checkout our website at https://dex-fruit.github.io .
comment: 8 pages, 5 figures
♻ ☆ Much Ado About Noising: Dispelling the Myths of Generative Robotic Control
Generative models, like flows and diffusions, have recently emerged as popular and efficacious policy parameterizations in robotics. There has been much speculation as to the factors underlying their successes, ranging from capturing multi-modal action distribution to expressing more complex behaviors. In this work, we perform a comprehensive evaluation of popular generative control policies (GCPs) on common behavior cloning (BC) benchmarks. We find that GCPs do not owe their success to their ability to capture multi-modality or to express more complex observation-to-action mappings. Instead, we find that their advantage stems from iterative computation, as long as intermediate steps are supervised during training and this supervision is paired with a suitable level of stochasticity. As a validation of our findings, we show that a minimum iterative policy (MIP), a lightweight two-step regression-based policy, essentially matches the performance of flow GCPs, and often outperforms distilled shortcut models. Our results suggest that the distribution-fitting component of GCPs is less salient than commonly believed, and point toward new design spaces focusing solely on control performance. Project page: https://simchowitzlabpublic.github.io/much-ado-about-noising-project/
Robotics 57
☆ STARE-VLA: Progressive Stage-Aware Reinforcement for Fine-Tuning Vision-Language-Action Models
Recent advances in Vision-Language-Action (VLA) models, powered by large language models and reinforcement learning-based fine-tuning, have shown remarkable progress in robotic manipulation. Existing methods often treat long-horizon actions as linguistic sequences and apply trajectory-level optimization methods such as Trajectory-wise Preference Optimization (TPO) or Proximal Policy Optimization (PPO), leading to coarse credit assignment and unstable training. However, unlike language, where a unified semantic meaning is preserved despite flexible sentence order, action trajectories progress through causally chained stages with different learning difficulties. This motivates progressive stage optimization. Thereby, we present Stage-Aware Reinforcement (STARE), a module that decomposes a long-horizon action trajectory into semantically meaningful stages and provides dense, interpretable, and stage-aligned reinforcement signals. Integrating STARE into TPO and PPO, we yield Stage-Aware TPO (STA-TPO) and Stage-Aware PPO (STA-PPO) for offline stage-wise preference and online intra-stage interaction, respectively. Further building on supervised fine-tuning as initialization, we propose the Imitation -> Preference -> Interaction (IPI), a serial fine-tuning pipeline for improving action accuracy in VLA models. Experiments on SimplerEnv and ManiSkill3 demonstrate substantial gains, achieving state-of-the-art success rates of 98.0 percent on SimplerEnv and 96.4 percent on ManiSkill3 tasks.
☆ NeuralRemaster: Phase-Preserving Diffusion for Structure-Aligned Generation
Standard diffusion corrupts data using Gaussian noise whose Fourier coefficients have random magnitudes and random phases. While effective for unconditional or text-to-image generation, corrupting phase components destroys spatial structure, making it ill-suited for tasks requiring geometric consistency, such as re-rendering, simulation enhancement, and image-to-image translation. We introduce Phase-Preserving Diffusion φ-PD, a model-agnostic reformulation of the diffusion process that preserves input phase while randomizing magnitude, enabling structure-aligned generation without architectural changes or additional parameters. We further propose Frequency-Selective Structured (FSS) noise, which provides continuous control over structural rigidity via a single frequency-cutoff parameter. φ-PD adds no inference-time cost and is compatible with any diffusion model for images or videos. Across photorealistic and stylized re-rendering, as well as sim-to-real enhancement for driving planners, φ-PD produces controllable, spatially aligned results. When applied to the CARLA simulator, φ-PD improves CARLA-to-Waymo planner performance by 50\%. The method is complementary to existing conditioning approaches and broadly applicable to image-to-image and video-to-video generation. Videos, additional examples, and code are available on our \href{https://yuzeng-at-tri.github.io/ppd-page/}{project page}.
From Generated Human Videos to Physically Plausible Robot Trajectories
Video generation models are rapidly improving in their ability to synthesize human actions in novel contexts, holding the potential to serve as high-level planners for contextual robot control. To realize this potential, a key research question remains open: how can a humanoid execute the human actions from generated videos in a zero-shot manner? This challenge arises because generated videos are often noisy and exhibit morphological distortions that make direct imitation difficult compared to real video. To address this, we introduce a two-stage pipeline. First, we lift video pixels into a 4D human representation and then retarget to the humanoid morphology. Second, we propose GenMimic-a physics-aware reinforcement learning policy conditioned on 3D keypoints, and trained with symmetry regularization and keypoint-weighted tracking rewards. As a result, GenMimic can mimic human actions from noisy, generated videos. We curate GenMimicBench, a synthetic human-motion dataset generated using two video generation models across a spectrum of actions and contexts, establishing a benchmark for assessing zero-shot generalization and policy robustness. Extensive experiments demonstrate improvements over strong baselines in simulation and confirm coherent, physically stable motion tracking on a Unitree G1 humanoid robot without fine-tuning. This work offers a promising path to realizing the potential of video generation models as high-level policies for robot control.
comment: For project website, see https://genmimic.github.io
☆ Object Reconstruction under Occlusion with Generative Priors and Contact-induced Constraints
Object geometry is key information for robot manipulation. Yet, object reconstruction is a challenging task because cameras only capture partial observations of objects, especially when occlusion occurs. In this paper, we leverage two extra sources of information to reduce the ambiguity of vision signals. First, generative models learn priors of the shapes of commonly seen objects, allowing us to make reasonable guesses of the unseen part of geometry. Second, contact information, which can be obtained from videos and physical interactions, provides sparse constraints on the boundary of the geometry. We combine the two sources of information through contact-guided 3D generation. The guidance formulation is inspired by drag-based editing in generative models. Experiments on synthetic and real-world data show that our approach improves the reconstruction compared to pure 3D generation and contact-based optimization.
comment: Project page: https://contactgen3d.github.io/
☆ Contact-Implicit Modeling and Simulation of a Snake Robot on Compliant and Granular Terrain
This thesis presents a unified modeling and simulation framework for analyzing sidewinding and tumbling locomotion of the COBRA snake robot across rigid, compliant, and granular terrains. A contact-implicit formulation is used to model distributed frictional interactions during sidewinding, and validated through MATLAB Simscape simulations and physical experiments on rigid ground and loose sand. To capture terrain deformation effects, Project Chrono's Soil Contact Model (SCM) is integrated with the articulated multibody dynamics, enabling prediction of slip, sinkage, and load redistribution that reduce stride efficiency on deformable substrates. For high-energy rolling locomotion on steep slopes, the Chrono DEM Engine is used to simulate particle-resolved granular interactions, revealing soil failure, intermittent lift-off, and energy dissipation mechanisms not captured by rigid models. Together, these methods span real-time control-oriented simulation and high-fidelity granular physics. Results demonstrate that rigid-ground models provide accurate short-horizon motion prediction, while continuum and particle-based terrain modeling becomes necessary for reliable mobility analysis in soft and highly dynamic environments. This work establishes a hierarchical simulation pipeline that advances robust, terrain-aware locomotion for robots operating in challenging unstructured settings.
☆ Introducing V-Soft Pro: a Modular Platform for a Transhumeral Prosthesis with Controllable Stiffness
Current upper limb prostheses aim to enhance user independence in daily activities by incorporating basic motor functions. However, they fall short of replicating the natural movement and interaction capabilities of the human arm. In contrast, human limbs leverage intrinsic compliance and actively modulate joint stiffness, enabling adaptive responses to varying tasks, impact absorption, and efficient energy transfer during dynamic actions. Inspired by this adaptability, we developed a transhumeral prosthesis with Variable Stiffness Actuators (VSAs) to replicate the controllable compliance found in biological joints. The proposed prosthesis features a modular design, allowing customization for different residual limb shapes and accommodating a range of independent control signals derived from users' biological cues. Integrated elastic elements passively support more natural movements, facilitate safe interactions with the environment, and adapt to diverse task requirements. This paper presents a comprehensive overview of the platform and its functionalities, highlighting its potential applications in the field of prosthetics.
comment: This article has been accepted for publication in Proceedings of the International Conference On Rehabilitation Robotics (ICORR), 2025. This is the author's version, which has not been fully edited, and content may change prior to final publication. Citation information: DOI 10.1109/ICORR66766.2025.11062964
☆ Preliminary Analysis and Simulation of a Compact Variable Stiffness Wrist
Variable Stiffness Actuators prove invaluable for robotics applications in unstructured environments, fostering safe interactions and enhancing task adaptability. Nevertheless, their mechanical design inevitably results in larger and heavier structures compared to classical rigid actuators. This paper introduces a novel 3 Degrees of Freedom (DoFs) parallel wrist that achieves variable stiffness through redundant elastic actuation. Leveraging its parallel architecture, the device employs only four motors, rendering it compact and lightweight. This characteristic makes it particularly well-suited for applications in prosthetics or humanoid robotics. The manuscript delves into the theoretical model of the device and proposes a sophisticated control strategy for independent regulation of joint position and stiffness. Furthermore, it validates the proposed controller through simulation, utilizing a comprehensive analysis of the system dynamics. The reported results affirm the ability of the device to achieve high accuracy and disturbance rejection in rigid configurations while minimizing interaction forces with its compliant behavior.
comment: This article has been accepted for publication in Springer Proceedings in Advanced Robotics, vol 31. Springer, Cham. This is the author's version, which has not been fully edited, and the content may change prior to final publication. Citation information: DOI https://doi.org/10.1007/978-3-031-64057-5_9
☆ Hybrid-Diffusion Models: Combining Open-loop Routines with Visuomotor Diffusion Policies
Despite the fact that visuomotor-based policies obtained via imitation learning demonstrate good performances in complex manipulation tasks, they usually struggle to achieve the same accuracy and speed as traditional control based methods. In this work, we introduce Hybrid-Diffusion models that combine open-loop routines with visuomotor diffusion policies. We develop Teleoperation Augmentation Primitives (TAPs) that allow the operator to perform predefined routines, such as locking specific axes, moving to perching waypoints, or triggering task-specific routines seamlessly during demonstrations. Our Hybrid-Diffusion method learns to trigger such TAPs during inference. We validate the method on challenging real-world tasks: Vial Aspiration, Open-Container Liquid Transfer, and container unscrewing. All experimental videos are available on the project's website: https://hybriddiffusion.github.io/
☆ FASTer: Toward Efficient Autoregressive Vision Language Action Modeling via neural Action Tokenization
Autoregressive vision-language-action (VLA) models have recently demonstrated strong capabilities in robotic manipulation. However, their core process of action tokenization often involves a trade-off between reconstruction fidelity and inference efficiency. We introduce FASTer, a unified framework for efficient and generalizable robot learning that integrates a learnable tokenizer with an autoregressive policy built upon it. FASTerVQ encodes action chunks as single-channel images, capturing global spatio-temporal dependencies while maintaining a high compression ratio. FASTerVLA builds on this tokenizer with block-wise autoregressive decoding and a lightweight action expert, achieving both faster inference and higher task performance. Extensive experiments across simulated and real-world benchmarks show that FASTerVQ delivers superior reconstruction quality, high token utilization, and strong cross-task and cross-embodiment generalization, while FASTerVLA further improves overall capability, surpassing previous state-of-the-art VLA models in both inference speed and task performance.
☆ On Disturbance-Aware Minimum-Time Trajectory Planning: Evidence from Tests on a Dynamic Driving Simulator
This work investigates how disturbance-aware, robustness-embedded reference trajectories translate into driving performance when executed by professional drivers in a dynamic simulator. Three planned reference trajectories are compared against a free-driving baseline (NOREF) to assess trade-offs between lap time (LT) and steering effort (SE): NOM, the nominal time-optimal trajectory; TLC, a track-limit-robust trajectory obtained by tightening margins to the track edges; and FLC, a friction-limit-robust trajectory obtained by tightening against axle and tire saturation. All trajectories share the same minimum lap-time objective with a small steering-smoothness regularizer and are evaluated by two professional drivers using a high-performance car on a virtual track. The trajectories derive from a disturbance-aware minimum-lap-time framework recently proposed by the authors, where worst-case disturbance growth is propagated over a finite horizon and used to tighten tire-friction and track-limit constraints, preserving performance while providing probabilistic safety margins. LT and SE are used as performance indicators, while RMS lateral deviation, speed error, and drift angle characterize driving style. Results show a Pareto-like LT-SE trade-off: NOM yields the shortest LT but highest SE; TLC minimizes SE at the cost of longer LT; FLC lies near the efficient frontier, substantially reducing SE relative to NOM with only a small LT increase. Removing trajectory guidance (NOREF) increases both LT and SE, confirming that reference trajectories improve pace and control efficiency. Overall, the findings highlight reference-based and disturbance-aware planning, especially FLC, as effective tools for training and for achieving fast yet stable trajectories.
comment: 18 pages, 11 figures, 5 tables
☆ 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.
☆ MOVE: A Simple Motion-Based Data Collection Paradigm for Spatial Generalization in Robotic Manipulation
Imitation learning method has shown immense promise for robotic manipulation, yet its practical deployment is fundamentally constrained by the data scarcity. Despite prior work on collecting large-scale datasets, there still remains a significant gap to robust spatial generalization. We identify a key limitation: individual trajectories, regardless of their length, are typically collected from a \emph{single, static spatial configuration} of the environment. This includes fixed object and target spatial positions as well as unchanging camera viewpoints, which significantly restricts the diversity of spatial information available for learning. To address this critical bottleneck in data efficiency, we propose \textbf{MOtion-Based Variability Enhancement} (\emph{MOVE}), a simple yet effective data collection paradigm that enables the acquisition of richer spatial information from dynamic demonstrations. Our core contribution is an augmentation strategy that injects motion into any movable objects within the environment for each demonstration. This process implicitly generates a dense and diverse set of spatial configurations within a single trajectory. We conduct extensive experiments in both simulation and real-world environments to validate our approach. For example, in simulation tasks requiring strong spatial generalization, \emph{MOVE} achieves an average success rate of 39.1\%, a 76.1\% relative improvement over the static data collection paradigm (22.2\%), and yields up to 2--5$\times$ gains in data efficiency on certain tasks. Our code is available at https://github.com/lucywang720/MOVE.
comment: 9 pages, 9 figures
☆ SIMA 2: A Generalist Embodied Agent for Virtual Worlds
We introduce SIMA 2, a generalist embodied agent that understands and acts in a wide variety of 3D virtual worlds. Built upon a Gemini foundation model, SIMA 2 represents a significant step toward active, goal-directed interaction within an embodied environment. Unlike prior work (e.g., SIMA 1) limited to simple language commands, SIMA 2 acts as an interactive partner, capable of reasoning about high-level goals, conversing with the user, and handling complex instructions given through language and images. Across a diverse portfolio of games, SIMA 2 substantially closes the gap with human performance and demonstrates robust generalization to previously unseen environments, all while retaining the base model's core reasoning capabilities. Furthermore, we demonstrate a capacity for open-ended self-improvement: by leveraging Gemini to generate tasks and provide rewards, SIMA 2 can autonomously learn new skills from scratch in a new environment. This work validates a path toward creating versatile and continuously learning agents for both virtual and, eventually, physical worlds.
☆ Using Machine Learning to Take Stay-or-Go Decisions in Data-driven Drone Missions
Drones are becoming indispensable in many application domains. In data-driven missions, besides sensing, the drone must process the collected data at runtime to decide whether additional action must be taken on the spot, before moving to the next point of interest. If processing does not reveal an event or situation that requires such an action, the drone has waited in vain instead of moving to the next point. If, however, the drone starts moving to the next point and it turns out that a follow-up action is needed at the previous point, it must spend time to fly-back. To take this decision, we propose different machine-learning methods based on branch prediction and reinforcement learning. We evaluate these methods for a wide range of scenarios where the probability of event occurrence changes with time. Our results show that the proposed methods consistently outperform the regression-based method proposed in the literature and can significantly improve the worst-case mission time by up to 4.1x. Also, the achieved median mission time is very close, merely up to 2.7% higher, to that of a method with perfect knowledge of the current underlying event probability at each point of interest.
comment: 19 pages, 3 figures, to appear in the proceedings of MobiQuitous 2025
☆ TEMPO-VINE: A Multi-Temporal Sensor Fusion Dataset for Localization and Mapping in Vineyards
In recent years, precision agriculture has been introducing groundbreaking innovations in the field, with a strong focus on automation. However, research studies in robotics and autonomous navigation often rely on controlled simulations or isolated field trials. The absence of a realistic common benchmark represents a significant limitation for the diffusion of robust autonomous systems under real complex agricultural conditions. Vineyards pose significant challenges due to their dynamic nature, and they are increasingly drawing attention from both academic and industrial stakeholders interested in automation. In this context, we introduce the TEMPO-VINE dataset, a large-scale multi-temporal dataset specifically designed for evaluating sensor fusion, simultaneous localization and mapping (SLAM), and place recognition techniques within operational vineyard environments. TEMPO-VINE is the first multi-modal public dataset that brings together data from heterogeneous LiDARs of different price levels, AHRS, RTK-GPS, and cameras in real trellis and pergola vineyards, with multiple rows exceeding 100 m in length. In this work, we address a critical gap in the landscape of agricultural datasets by providing researchers with a comprehensive data collection and ground truth trajectories in different seasons, vegetation growth stages, terrain and weather conditions. The sequence paths with multiple runs and revisits will foster the development of sensor fusion, localization, mapping and place recognition solutions for agricultural fields. The dataset, the processing tools and the benchmarking results will be available at the dedicated webpage upon acceptance.
☆ Embodied Co-Design for Rapidly Evolving Agents: Taxonomy, Frontiers, and Challenges
Brain-body co-evolution enables animals to develop complex behaviors in their environments. Inspired by this biological synergy, embodied co-design (ECD) has emerged as a transformative paradigm for creating intelligent agents-from virtual creatures to physical robots-by jointly optimizing their morphologies and controllers rather than treating control in isolation. This integrated approach facilitates richer environmental interactions and robust task performance. In this survey, we provide a systematic overview of recent advances in ECD. We first formalize the concept of ECD and position it within related fields. We then introduce a hierarchical taxonomy: a lower layer that breaks down agent design into three fundamental components-controlling brain, body morphology, and task environment-and an upper layer that integrates these components into four major ECD frameworks: bi-level, single-level, generative, and open-ended. This taxonomy allows us to synthesize insights from more than one hundred recent studies. We further review notable benchmarks, datasets, and applications in both simulated and real-world scenarios. Finally, we identify significant challenges and offer insights into promising future research directions. A project associated with this survey has been created at https://github.com/Yuxing-Wang-THU/SurveyBrainBody.
☆ Bridging Simulation and Reality: Cross-Domain Transfer with Semantic 2D Gaussian Splatting
Cross-domain transfer in robotic manipulation remains a longstanding challenge due to the significant domain gap between simulated and real-world environments. Existing methods such as domain randomization, adaptation, and sim-real calibration often require extensive tuning or fail to generalize to unseen scenarios. To address this issue, we observe that if domain-invariant features are utilized during policy training in simulation, and the same features can be extracted and provided as the input to policy during real-world deployment, the domain gap can be effectively bridged, leading to significantly improved policy generalization. Accordingly, we propose Semantic 2D Gaussian Splatting (S2GS), a novel representation method that extracts object-centric, domain-invariant spatial features. S2GS constructs multi-view 2D semantic fields and projects them into a unified 3D space via feature-level Gaussian splatting. A semantic filtering mechanism removes irrelevant background content, ensuring clean and consistent inputs for policy learning. To evaluate the effectiveness of S2GS, we adopt Diffusion Policy as the downstream learning algorithm and conduct experiments in the ManiSkill simulation environment, followed by real-world deployment. Results demonstrate that S2GS significantly improves sim-to-real transferability, maintaining high and stable task performance in real-world scenarios.
☆ When Robots Should Say "I Don't Know": Benchmarking Abstention in Embodied Question Answering
Embodied Question Answering (EQA) requires an agent to interpret language, perceive its environment, and navigate within 3D scenes to produce responses. Existing EQA benchmarks assume that every question must be answered, but embodied agents should know when they do not have sufficient information to answer. In this work, we focus on a minimal requirement for EQA agents, abstention: knowing when to withhold an answer. From an initial study of 500 human queries, we find that 32.4% contain missing or underspecified context. Drawing on this initial study and cognitive theories of human communication errors, we derive five representative categories requiring abstention: actionability limitation, referential underspecification, preference dependence, information unavailability, and false presupposition. We augment OpenEQA by having annotators transform well-posed questions into ambiguous variants outlined by these categories. The resulting dataset, AbstainEQA, comprises 1,636 annotated abstention cases paired with 1,636 original OpenEQA instances for balanced evaluation. Evaluating on AbstainEQA, we find that even the best frontier model only attains 42.79% abstention recall, while humans achieve 91.17%. We also find that scaling, prompting, and reasoning only yield marginal gains, and that fine-tuned models overfit to textual cues. Together, these results position abstention as a fundamental prerequisite for reliable interaction in embodied settings and as a necessary basis for effective clarification.
☆ Gauss-Newton accelerated MPPI Control
Model Predictive Path Integral (MPPI) control is a sampling-based optimization method that has recently attracted attention, particularly in the robotics and reinforcement learning communities. MPPI has been widely applied as a GPU-accelerated random search method to deterministic direct single-shooting optimal control problems arising in model predictive control (MPC) formulations. MPPI offers several key advantages, including flexibility, robustness, ease of implementation, and inherent parallelizability. However, its performance can deteriorate in high-dimensional settings since the optimal control problem is solved via Monte Carlo sampling. To address this limitation, this paper proposes an enhanced MPPI method that incorporates a Jacobian reconstruction technique and the second-order Generalized Gauss-Newton method. This novel approach is called \textit{Gauss-Newton accelerated MPPI}. The numerical results show that the Gauss-Newton accelerated MPPI approach substantially improves MPPI scalability and computational efficiency while preserving the key benefits of the classical MPPI framework, making it a promising approach even for high-dimensional problems.
comment: 6 pages, 3 figures, submitted to the IFAC World Congress 2026
☆ One Ring to Rule Them All: Constrained Distributional Control for Massive-Scale Heterogeneous Robotic Ensemble Systems
Ensemble control aims to steer a population of dynamical systems using a shared control input. This paper introduces a constrained ensemble control framework for parameterized, heterogeneous robotic systems operating under state and environmental constraints, such as obstacle avoidance. We develop a moment kernel transform that maps the parameterized ensemble dynamics to the moment system in a kernel space, enabling the characterization of population-level behavior. The state-space constraints, such as polyhedral waypoints to be visited and obstacles to be avoided, are also transformed into the moment space, leading to a unified formulation for safe, large-scale ensemble control. Expressive signal temporal logic specifications are employed to encode complex visit-avoid tasks, which are achieved through a single shared controller synthesized from our constrained ensemble control formulation. Simulation and hardware experiments demonstrate the effectiveness of the proposed approach in safely and efficiently controlling robotic ensembles within constrained environments.
comment: 9 pages, 8 figures
☆ MARL Warehouse Robots
We present a comparative study of multi-agent reinforcement learning (MARL) algorithms for cooperative warehouse robotics. We evaluate QMIX and IPPO on the Robotic Warehouse (RWARE) environment and a custom Unity 3D simulation. Our experiments reveal that QMIX's value decomposition significantly outperforms independent learning approaches (achieving 3.25 mean return vs. 0.38 for advanced IPPO), but requires extensive hyperparameter tuning -- particularly extended epsilon annealing (5M+ steps) for sparse reward discovery. We demonstrate successful deployment in Unity ML-Agents, achieving consistent package delivery after 1M training steps. While MARL shows promise for small-scale deployments (2-4 robots), significant scaling challenges remain. Code and analyses: https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/
comment: 6 pages, 4 tables. Project documentation: https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/
☆ Open-Ended Goal Inference through Actions and Language for Human-Robot Collaboration
To collaborate with humans, robots must infer goals that are often ambiguous, difficult to articulate, or not drawn from a fixed set. Prior approaches restrict inference to a predefined goal set, rely only on observed actions, or depend exclusively on explicit instructions, making them brittle in real-world interactions. We present BALI (Bidirectional Action-Language Inference) for goal prediction, a method that integrates natural language preferences with observed human actions in a receding-horizon planning tree. BALI combines language and action cues from the human, asks clarifying questions only when the expected information gain from the answer outweighs the cost of interruption, and selects supportive actions that align with inferred goals. We evaluate the approach in collaborative cooking tasks, where goals may be novel to the robot and unbounded. Compared to baselines, BALI yields more stable goal predictions and significantly fewer mistakes.
comment: Accepted to ACM/IEEE International Conference on Human-Robot Interaction, 2026 (HRI 2026), 10 pages, 4 figures
☆ Vision-Language-Action Models for Selective Robotic Disassembly: A Case Study on Critical Component Extraction from Desktops
Automating disassembly of critical components from end-of-life (EoL) desktops, such as high-value items like RAM modules and CPUs, as well as sensitive parts like hard disk drives, remains challenging due to the inherent variability and uncertainty of these products. Moreover, their disassembly requires sequential, precise, and dexterous operations, further increasing the complexity of automation. Current robotic disassembly processes are typically divided into several stages: perception, sequence planning, task planning, motion planning, and manipulation. Each stage requires explicit modeling, which limits generalization to unfamiliar scenarios. Recent development of vision-language-action (VLA) models has presented an end-to-end approach for general robotic manipulation tasks. Although VLAs have demonstrated promising performance on simple tasks, the feasibility of applying such models to complex disassembly remains largely unexplored. In this paper, we collected a customized dataset for robotic RAM and CPU disassembly and used it to fine-tune two well-established VLA approaches, OpenVLA and OpenVLA-OFT, as a case study. We divided the whole disassembly task into several small steps, and our preliminary experimental results indicate that the fine-tuned VLA models can faithfully complete multiple early steps but struggle with certain critical subtasks, leading to task failure. However, we observed that a simple hybrid strategy that combines VLA with a rule-based controller can successfully perform the entire disassembly operation. These findings highlight the current limitations of VLA models in handling the dexterity and precision required for robotic EoL product disassembly. By offering a detailed analysis of the observed results, this study provides insights that may inform future research to address current challenges and advance end-to-end robotic automated disassembly.
☆ RoboBPP: Benchmarking Robotic Online Bin Packing with Physics-based Simulation
Physical feasibility in 3D bin packing is a key requirement in modern industrial logistics and robotic automation. With the growing adoption of industrial automation, online bin packing has gained increasing attention. However, inconsistencies in problem settings, test datasets, and evaluation metrics have hindered progress in the field, and there is a lack of a comprehensive benchmarking system. Direct testing on real hardware is costly, and building a realistic simulation environment is also challenging. To address these limitations, we introduce RoboBPP, a benchmarking system designed for robotic online bin packing. RoboBPP integrates a physics-based simulator to assess physical feasibility. In our simulation environment, we introduce a robotic arm and boxes at real-world scales to replicate real industrial packing workflows. By simulating conditions that arise in real industrial applications, we ensure that evaluated algorithms are practically deployable. In addition, prior studies often rely on synthetic datasets whose distributions differ from real-world industrial data. To address this issue, we collect three datasets from real industrial workflows, including assembly-line production, logistics packing, and furniture manufacturing. The benchmark comprises three carefully designed test settings and extends existing evaluation metrics with new metrics for structural stability and operational safety. We design a scoring system and derive a range of insights from the evaluation results. RoboBPP is fully open-source and is equipped with visualization tools and an online leaderboard, providing a reproducible and extensible foundation for future research and industrial applications (https://robot-bin-packing-benchmark.github.io).
comment: Under review at the International Journal of Robotics Research (IJRR)
☆ Bridging Probabilistic Inference and Behavior Trees: An Interactive Framework for Adaptive Multi-Robot Cooperation
This paper proposes an Interactive Inference Behavior Tree (IIBT) framework that integrates behavior trees (BTs) with active inference under the free energy principle for distributed multi-robot decision-making. The proposed IIBT node extends conventional BTs with probabilistic reasoning, enabling online joint planning and execution across multiple robots. It remains fully compatible with standard BT architectures, allowing seamless integration into existing multi-robot control systems. Within this framework, multi-robot cooperation is formulated as a free-energy minimization process, where each robot dynamically updates its preference matrix based on perceptual inputs and peer intentions, thereby achieving adaptive coordination in partially observable and dynamic environments. The proposed approach is validated through both simulation and real-world experiments, including a multi-robot maze navigation and a collaborative manipulation task, compared against traditional BTs(https://youtu.be/KX_oT3IDTf4). Experimental results demonstrate that the IIBT framework reduces BT node complexity by over 70%, while maintaining robust, interpretable, and adaptive cooperative behavior under environmental uncertainty.
comment: 34 pages, is submitted RAS Journal
☆ Development of a 15-Degree-of-Freedom Bionic Hand with Cable-Driven Transmission and Distributed Actuation
In robotic hand research, minimizing the number of actuators while maintaining human-hand-consistent dimensions and degrees of freedom constitutes a fundamental challenge. Drawing bio-inspiration from human hand kinematic configurations and muscle distribution strategies, this work proposes a novel 15-DoF dexterous robotic hand, with detailed analysis of its mechanical architecture, electrical system, and control system. The bionic hand employs a new tendon-driven mechanism, significantly reducing the number of motors required by traditional tendon-driven systems while enhancing motion performance and simplifying the mechanical structure. This design integrates five motors in the forearm to provide strong gripping force, while ten small motors are installed in the palm to support fine manipulation tasks. Additionally, a corresponding joint sensing and motor driving electrical system was developed to ensure efficient control and feedback. The entire system weighs only 1.4kg, combining lightweight and high-performance features. Through experiments, the bionic hand exhibited exceptional dexterity and robust grasping capabilities, demonstrating significant potential for robotic manipulation tasks.
☆ FALCON: Actively Decoupled Visuomotor Policies for Loco-Manipulation with Foundation-Model-Based Coordination
We present FoundAtion-model-guided decoupled LoCO-maNipulation visuomotor policies (FALCON), a framework for loco-manipulation that combines modular diffusion policies with a vision-language foundation model as the coordinator. Our approach explicitly decouples locomotion and manipulation into two specialized visuomotor policies, allowing each subsystem to rely on its own observations. This mitigates the performance degradation that arise when a single policy is forced to fuse heterogeneous, potentially mismatched observations from locomotion and manipulation. Our key innovation lies in restoring coordination between these two independent policies through a vision-language foundation model, which encodes global observations and language instructions into a shared latent embedding conditioning both diffusion policies. On top of this backbone, we introduce a phase-progress head that uses textual descriptions of task stages to infer discrete phase and continuous progress estimates without manual phase labels. To further structure the latent space, we incorporate a coordination-aware contrastive loss that explicitly encodes cross-subsystem compatibility between arm and base actions. We evaluate FALCON on two challenging loco-manipulation tasks requiring navigation, precise end-effector placement, and tight base-arm coordination. Results show that it surpasses centralized and decentralized baselines while exhibiting improved robustness and generalization to out-of-distribution scenarios.
☆ Vertical Planetary Landing on Sloped Terrain Using Optical Flow Divergence Estimates
Autonomous landing on sloped terrain poses significant challenges for small, lightweight spacecraft, such as rotorcraft and landers. These vehicles have limited processing capability and payload capacity, which makes advanced deep learning methods and heavy sensors impractical. Flying insects, such as bees, achieve remarkable landings with minimal neural and sensory resources, relying heavily on optical flow. By regulating flow divergence, a measure of vertical velocity divided by height, they perform smooth landings in which velocity and height decay exponentially together. However, adapting this bio-inspired strategy for spacecraft landings on sloped terrain presents two key challenges: global flow-divergence estimates obscure terrain inclination, and the nonlinear nature of divergence-based control can lead to instability when using conventional controllers. This paper proposes a nonlinear control strategy that leverages two distinct local flow divergence estimates to regulate both thrust and attitude during vertical landings. The control law is formulated based on Incremental Nonlinear Dynamic Inversion to handle the nonlinear flow divergence. The thrust control ensures a smooth vertical descent by keeping a constant average of the local flow divergence estimates, while the attitude control aligns the vehicle with the inclined surface at touchdown by exploiting their difference. The approach is evaluated in numerical simulations using a simplified 2D spacecraft model across varying slopes and divergence setpoints. Results show that regulating the average divergence yields stable landings with exponential decay of velocity and height, and using the divergence difference enables effective alignment with inclined terrain. Overall, the method offers a robust, low-resource landing strategy that enhances the feasibility of autonomous planetary missions with small spacecraft.
comment: This paper is accepted at International Astronautical Congress (IAC 2025)
☆ Seabed-to-Sky Mapping of Maritime Environments with a Dual Orthogonal SONAR and LiDAR Sensor Suite
Critical maritime infrastructure increasingly demands situational awareness both above and below the surface, yet existing ''seabed-to-sky'' mapping pipelines either rely on GNSS (vulnerable to shadowing/spoofing) or expensive bathymetric sonars. We present a unified, GNSS-independent mapping system that fuses LiDAR-IMU with a dual, orthogonally mounted Forward Looking Sonars (FLS) to generate consistent seabed-to-sky maps from an Autonomous Surface Vehicle. On the acoustic side, we extend orthogonal wide-aperture fusion to handle arbitrary inter-sonar translations (enabling heterogeneous, non-co-located models) and extract a leading edge from each FLS to form line-scans. On the mapping side, we modify LIO-SAM to ingest both stereo-derived 3D sonar points and leading-edge line-scans at and between keyframes via motion-interpolated poses, allowing sparse acoustic updates to contribute continuously to a single factor-graph map. We validate the system on real-world data from Belvederekanalen (Copenhagen), demonstrating real-time operation with approx. 2.65 Hz map updates and approx. 2.85 Hz odometry while producing a unified 3D model that spans air-water domains.
☆ ARCAS: An Augmented Reality Collision Avoidance System with SLAM-Based Tracking for Enhancing VRU Safety
Vulnerable road users (VRUs) face high collision risks in mixed traffic, yet most existing safety systems prioritize driver or vehicle assistance over direct VRU support. This paper presents ARCAS, a real-time augmented reality collision avoidance system that provides personalized spatial alerts to VRUs via wearable AR headsets. By fusing roadside 360-degree 3D LiDAR with SLAM-based headset tracking and an automatic 3D calibration procedure, ARCAS accurately overlays world-locked 3D bounding boxes and directional arrows onto approaching hazards in the user's passthrough view. The system also enables multi-headset coordination through shared world anchoring. Evaluated in real-world pedestrian interactions with e-scooters and vehicles (180 trials), ARCAS nearly doubled pedestrians' time-to-collision and increased counterparts' reaction margins by up to 4x compared to unaided-eye conditions. Results validate the feasibility and effectiveness of LiDAR-driven AR guidance and highlight the potential of wearable AR as a promising next-generation safety tool for urban mobility.
comment: 8 pages, 3 figures, 1 table
☆ Disturbance Compensation for Safe Kinematic Control of Robotic Systems with Closed Architecture
In commercial robotic systems, it is common to encounter a closed inner-loop torque controller that is not user-modifiable. However, the outer-loop controller, which sends kinematic commands such as position or velocity for the inner-loop controller to track, is typically exposed to users. In this work, we focus on the development of an easily integrated add-on at the outer-loop layer by combining disturbance rejection control and robust control barrier function for high-performance tracking and safe control of the whole dynamic system of an industrial manipulator. This is particularly beneficial when 1) the inner-loop controller is imperfect, unmodifiable, and uncertain; and 2) the dynamic model exhibits significant uncertainty. Stability analysis, formal safety guarantee proof, and hardware experiments with a PUMA robotic manipulator are presented. Our solution demonstrates superior performance in terms of simplicity of implementation, robustness, tracking precision, and safety compared to the state of the art. Video: https://youtu.be/zw1tanvrV8Q
☆ XR-DT: Extended Reality-Enhanced Digital Twin for Agentic Mobile Robots
As mobile robots increasingly operate alongside humans in shared workspaces, ensuring safe, efficient, and interpretable Human-Robot Interaction (HRI) has become a pressing challenge. While substantial progress has been devoted to human behavior prediction, limited attention has been paid to how humans perceive, interpret, and trust robots' inferences, impeding deployment in safety-critical and socially embedded environments. This paper presents XR-DT, an eXtended Reality-enhanced Digital Twin framework for agentic mobile robots, that bridges physical and virtual spaces to enable bi-directional understanding between humans and robots. Our hierarchical XR-DT architecture integrates virtual-, augmented-, and mixed-reality layers, fusing real-time sensor data, simulated environments in the Unity game engine, and human feedback captured through wearable AR devices. Within this framework, we design an agentic mobile robot system with a unified diffusion policy for context-aware task adaptation. We further propose a chain-of-thought prompting mechanism that allows multimodal large language models to reason over human instructions and environmental context, while leveraging an AutoGen-based multi-agent coordination layer to enhance robustness and collaboration in dynamic tasks. Initial experimental results demonstrate accurate human and robot trajectory prediction, validating the XR-DT framework's effectiveness in HRI tasks. By embedding human intention, environmental dynamics, and robot cognition into the XR-DT framework, our system enables interpretable, trustworthy, and adaptive HRI.
comment: 10 pages, 5 figures
☆ Invariance Co-training for Robot Visual Generalization
Reasoning from diverse observations is a fundamental capability for generalist robot policies to operate in a wide range of environments. Despite recent advancements, many large-scale robotic policies still remain sensitive to key sources of observational variation such as changes in camera perspective, lighting, and the presence of distractor objects. We posit that the limited generalizability of these models arises from the substantial diversity required to robustly cover these quasistatic axes, coupled with the current scarcity of large-scale robotic datasets that exhibit rich variation across them. In this work, we propose to systematically examine what robots need to generalize across these challenging axes by introducing two key auxiliary tasks, state similarity and invariance to observational perturbations, applied to both demonstration data and static visual data. We then show that via these auxiliary tasks, leveraging both more-expensive robotic demonstration data and less-expensive, visually rich synthetic images generated from non-physics-based simulation (for example, Unreal Engine) can lead to substantial increases in generalization to unseen camera viewpoints, lighting configurations, and distractor conditions. Our results demonstrate that co-training on this diverse data improves performance by 18 percent over existing generative augmentation methods. For more information and videos, please visit https://invariance-cotraining.github.io
comment: 14 pages, 10 figures
☆ Search at Scale: Improving Numerical Conditioning of Ergodic Coverage Optimization for Multi-Scale Domains
Recent methods in ergodic coverage planning have shown promise as tools that can adapt to a wide range of geometric coverage problems with general constraints, but are highly sensitive to the numerical scaling of the problem space. The underlying challenge is that the optimization formulation becomes brittle and numerically unstable with changing scales, especially under potentially nonlinear constraints that impose dynamic restrictions, due to the kernel-based formulation. This paper proposes to address this problem via the development of a scale-agnostic and adaptive ergodic coverage optimization method based on the maximum mean discrepancy metric (MMD). Our approach allows the optimizer to solve for the scale of differential constraints while annealing the hyperparameters to best suit the problem domain and ensure physical consistency. We also derive a variation of the ergodic metric in the log space, providing additional numerical conditioning without loss of performance. We compare our approach with existing coverage planning methods and demonstrate the utility of our approach on a wide range of coverage problems.
☆ Wake Vectoring for Efficient Morphing Flight
Morphing aerial robots have the potential to transform autonomous flight, enabling navigation through cluttered environments, perching, and seamless transitions between aerial and terrestrial locomotion. Yet mid-flight reconfiguration presents a critical aerodynamic challenge: tilting propulsors to achieve shape change reduces vertical thrust, undermining stability and control authority. Here, we introduce a passive wake vectoring mechanism that recovers lost thrust during morphing. Integrated into a novel robotic system, Aerially Transforming Morphobot (ATMO), internal deflectors intercept and redirect rotor wake downward, passively steering airflow momentum that would otherwise be wasted. This electronics-free solution achieves up to a 40% recovery of vertical thrust in configurations where no useful thrust would otherwise be produced, substantially extending hover and maneuvering capabilities during transformation. Our findings highlight a new direction for morphing aerial robot design, where passive aerodynamic structures, inspired by thrust vectoring in rockets and aircraft, enable efficient, agile flight without added mechanical complexity.
☆ Two-Stage Camera Calibration Method for Multi-Camera Systems Using Scene Geometry
Calibration of multi-camera systems is a key task for accurate object tracking. However, it remains a challenging problem in real-world conditions, where traditional methods are not applicable due to the lack of accurate floor plans, physical access to place calibration patterns, or synchronized video streams. This paper presents a novel two-stage calibration method that overcomes these limitations. In the first stage, partial calibration of individual cameras is performed based on an operator's annotation of natural geometric primitives (parallel, perpendicular, and vertical lines, or line segments of equal length). This allows estimating key parameters (roll, pitch, focal length) and projecting the camera's Effective Field of View (EFOV) onto the horizontal plane in a base 3D coordinate system. In the second stage, precise system calibration is achieved through interactive manipulation of the projected EFOV polygons. The operator adjusts their position, scale, and rotation to align them with the floor plan or, in its absence, using virtual calibration elements projected onto all cameras in the system. This determines the remaining extrinsic parameters (camera position and yaw). Calibration requires only a static image from each camera, eliminating the need for physical access or synchronized video. The method is implemented as a practical web service. Comparative analysis and demonstration videos confirm the method's applicability, accuracy, and flexibility, enabling the deployment of precise multi-camera tracking systems in scenarios previously considered infeasible.
♻ ☆ SO-Bench: A Structural Output Evaluation of Multimodal LLMs
Multimodal large language models (MLLMs) are increasingly deployed in real-world, agentic settings where outputs must not only be correct, but also conform to predefined data schemas. Despite recent progress in structured generation in textual domain, there is still no benchmark that systematically evaluates schema-grounded information extraction and reasoning over visual inputs. In this work, we conduct a comprehensive study of visual structural output capabilities for MLLMs with our carefully designed SO-Bench benchmark. Covering four visual domains, including UI screens, natural images, documents, and charts, SO-Bench is built from over 6.5K diverse JSON schemas and 1.8K curated image-schema pairs with human-verified quality. Benchmarking experiments on open-sourced and frontier proprietary models reveal persistent gaps in predicting accurate, schema compliant outputs, highlighting the need for better multimodal structured reasoning. Beyond benchmarking, we further conduct training experiments to largely improve the model's structured output capability. We plan to make the benchmark available to the community.
comment: v2 preprint. Fixed some typos, add a discussion about limitation, provide pseudo-codes for eval
♻ ☆ The Autonomy-Alignment Problem in Open-Ended Learning Robots: Formalising the Purpose Framework
The rapid advancement of artificial intelligence is enabling the development of increasingly autonomous robots capable of operating beyond engineered factory settings and into the unstructured environments of human life. This shift raises a critical autonomy-alignment problem: how to ensure that a robot's autonomous learning focuses on acquiring knowledge and behaviours that serve human practical objectives while remaining aligned with broader human values (e.g., safety and ethics). This problem remains largely underexplored and lacks a unifying conceptual and formal framework. Here, we address one of its most challenging instances of the problem: open-ended learning (OEL) robots, which autonomously acquire new knowledge and skills through interaction with the environment, guided by intrinsic motivations and self-generated goals. We propose a computational framework, introduced qualitatively and then formalised, to guide the design of OEL architectures that balance autonomy with human control. At its core is the novel concept of purpose, which specifies what humans (designers or users) want the robot to learn, do, or avoid, independently of specific task domains. The framework decomposes the autonomy-alignment problem into four tractable sub-problems: the alignment of robot purposes (hardwired or learnt) with human purposes; the arbitration between multiple purposes; the grounding of abstract purposes into domain-specific goals; and the acquisition of competence to achieve those goals. The framework supports formal definitions of alignment across multiple cases and proofs of necessary and sufficient conditions under which alignment holds. Illustrative hypothetical scenarios showcase the applicability of the framework for guiding the development of purpose-aligned autonomous robots.
comment: 33 pages, 5 figures
♻ ☆ HAFO: A Force-Adaptive Control Framework for Humanoid Robots in Intense Interaction Environments
Reinforcement learning (RL) controllers have made impressive progress in humanoid locomotion and light-weight object manipulation. However, achieving robust and precise motion control with intense force interaction remains a significant challenge. To address these limitations, this paper proposes HAFO, a dual-agent reinforcement learning framework that concurrently optimizes both a robust locomotion strategy and a precise upper-body manipulation strategy via coupled training in environments with external disturbances. The external pulling disturbances are explicitly modeled using a spring-damper system, allowing for fine-grained force control through manipulation of the virtual spring. In this process, the reinforcement learning policy autonomously generates a disturbance-rejection response by utilizing environmental feedback. Furthermore, HAFO employs an asymmetric Actor-Critic framework in which the Critic network's access to privileged external forces guides the actor network to acquire generalizable force adaptation for resisting external disturbances. The experimental results demonstrate that HAFO achieves whole-body control for humanoid robots across diverse force-interaction environments, delivering outstanding performance in load-bearing tasks and maintaining stable operation even under rope suspension state.
♻ ☆ GigaBrain-0: A World Model-Powered Vision-Language-Action Model
Training Vision-Language-Action (VLA) models for generalist robots typically requires large-scale real-world robot data, which is expensive and time-consuming to collect. The inefficiency of physical data collection severely limits the scalability, and generalization capacity of current VLA systems. To address this challenge, we introduce GigaBrain-0, a novel VLA foundation model empowered by world model-generated data (e.g., video generation, real2real transfer, human transfer, view transfer, sim2real transfer data). By leveraging world models to generate diverse data at scale, GigaBrain-0 significantly reduces reliance on real robot data while improving cross-task generalization. Our approach further improves policy robustness through RGBD input modeling and embodied Chain-of-Thought (CoT) supervision, enabling the model to reason about spatial geometry, object states, and long-horizon dependencies during task execution. This leads to substantial gains in real-world performance on dexterous, long-horizon, and mobile manipulation tasks. Extensive experiments demonstrate that GigaBrain-0 achieves superior generalization across variations in appearances (e.g., textures, colors), object placements, and camera viewpoints. Additionally, we present GigaBrain-0-Small, an optimized lightweight variant designed to run efficiently on devices such as the NVIDIA Jetson AGX Orin.
comment: https://gigabrain0.github.io/
♻ ☆ Surfel-LIO: Fast LiDAR-Inertial Odometry with Pre-computed Surfels and Hierarchical Z-order Voxel Hashing
LiDAR-inertial odometry (LIO) is an active research area, as it enables accurate real-time state estimation in GPS-denied environments. Recent advances in map data structures and spatial indexing have significantly improved the efficiency of LIO systems. Nevertheless, we observe that two aspects may still leave room for improvement: (1) nearest neighbor search often requires examining multiple spatial units to gather sufficient points for plane fitting, and (2) plane parameters are typically recomputed at every iteration despite unchanged map geometry. Motivated by these observations, we propose Surfel-LIO, which employs a hierarchical voxel structure (hVox) with pre-computed surfel representation. This design enables O(1) correspondence retrieval without runtime neighbor enumeration or plane fitting, combined with Z-order curve encoding for cache-friendly spatial indexing. Experimental results on the M3DGR dataset demonstrate that our method achieves significantly faster processing speed compared to recent state-of-the-art methods while maintaining comparable state estimation accuracy. Our implementation is publicly available at https://github.com/93won/lidar_inertial_odometry.
♻ ☆ Estimating the Joint Probability of Scenario Parameters with Gaussian Mixture Copula Models
This paper presents the first application of Gaussian Mixture Copula Models to the statistical modeling of driving scenarios for the safety validation of automated driving systems. Knowledge of the joint probability distribution of scenario parameters is essential for scenario-based safety assessment, where risk quantification depends on the likelihood of concrete parameter combinations. Gaussian Mixture Copula Models bring together the multimodal expressivity of Gaussian Mixture Models and the flexibility of copulas, enabling separate modeling of marginal distributions and dependencies. We benchmark Gaussian Mixture Copula Models against previously proposed approaches - Gaussian Mixture Models and Gaussian Copula Models - using real-world driving data drawn from two scenarios defined in United Nations Regulation No. 157. Our evaluation on approximately 18 million instances of these two scenarios demonstrates that Gaussian Mixture Copula Models consistently surpass Gaussian Copula Models and perform better than, or at least comparably to, Gaussian Mixture Models, as measured by both log-likelihood and Sinkhorn distance. These results are promising for the adoption of Gaussian Mixture Copula Models as a statistical foundation for future scenario-based validation frameworks.
comment: 9 pages, 4 figures; This work has been submitted to the IEEE for possible publication; Code available at: https://codeocean.com/capsule/1003615/tree
♻ ☆ WeatherPrompt: Multi-modality Representation Learning for All-Weather Drone Visual Geo-Localization
Visual geo-localization for drones faces critical degradation under weather perturbations, \eg, rain and fog, where existing methods struggle with two inherent limitations: 1) Heavy reliance on limited weather categories that constrain generalization, and 2) Suboptimal disentanglement of entangled scene-weather features through pseudo weather categories. We present WeatherPrompt, a multi-modality learning paradigm that establishes weather-invariant representations through fusing the image embedding with the text context. Our framework introduces two key contributions: First, a Training-free Weather Reasoning mechanism that employs off-the-shelf large multi-modality models to synthesize multi-weather textual descriptions through human-like reasoning. It improves the scalability to unseen or complex weather, and could reflect different weather strength. Second, to better disentangle the scene and weather feature, we propose a multi-modality framework with the dynamic gating mechanism driven by the text embedding to adaptively reweight and fuse visual features across modalities. The framework is further optimized by the cross-modal objectives, including image-text contrastive learning and image-text matching, which maps the same scene with different weather conditions closer in the respresentation space. Extensive experiments validate that, under diverse weather conditions, our method achieves competitive recall rates compared to state-of-the-art drone geo-localization methods. Notably, it improves Recall@1 by +13.37\% under night conditions and by 18.69\% under fog and snow conditions.
♻ ☆ PPL: Point Cloud Supervised Proprioceptive Locomotion Reinforcement Learning for Legged Robots in Crawl Spaces
Legged locomotion in constrained spaces (called crawl spaces) is challenging. In crawl spaces, current proprioceptive locomotion learning methods are difficult to achieve traverse because only ground features are inferred. In this study, a point cloud supervised RL framework for proprioceptive locomotion in crawl spaces is proposed. A state estimation network is designed to estimate the robot's collision states as well as ground and spatial features for locomotion. A point cloud feature extraction method is proposed to supervise the state estimation network. The method uses representation of the point cloud in polar coordinate frame and MLPs for efficient feature extraction. Experiments demonstrate that, compared with existing methods, our method exhibits faster iteration time in the training and more agile locomotion in crawl spaces. This study enhances the ability of legged robots to traverse constrained spaces without requiring exteroceptive sensors.
comment: Accepted by RA-L
♻ ☆ SkillWrapper: Generative Predicate Invention for Skill Abstraction
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.
♻ ☆ Energy-Aware Lane Planning for Connected Electric Vehicles in Urban Traffic: Design and Vehicle-in-the-Loop Validation
Urban driving with connected and automated vehicles (CAVs) offers potential for energy savings, yet most eco-driving strategies focus solely on longitudinal speed control within a single lane. This neglects the significant impact of lateral decisions, such as lane changes, on overall energy efficiency, especially in environments with traffic signals and heterogeneous traffic flow. To address this gap, we propose a novel energy-aware motion planning framework that jointly optimizes longitudinal speed and lateral lane-change decisions using vehicle-to-infrastructure (V2I) communication. Our approach estimates long-term energy costs using a graph-based approximation and solves short-horizon optimal control problems under traffic constraints. Using a data-driven energy model calibrated to an actual battery electric vehicle, we demonstrate with vehicle-in-the-loop experiments that our method reduces motion energy consumption by up to 24 percent compared to a human driver, highlighting the potential of connectivity-enabled planning for sustainable urban autonomy.
comment: Accepted at 2025 IEEE Conference on Decision and Control (CDC25')
♻ ☆ A Fast and Model Based Approach for Evaluating Task-Competence of Antagonistic Continuum Arms
Soft robot arms have made significant progress towards completing human-scale tasks, but designing arms for tasks with specific load and workspace requirements remains difficult. A key challenge is the lack of model-based design tools, forcing advancement to occur through empirical iteration and observation. Existing models are focused on control and rely on parameter fits, which means they cannot provide general conclusions about the mapping between design and performance or the influence of factors outside the fitting data.As a first step toward model-based design tools, we introduce a novel method of analyzing whether a proposed arm design can complete desired tasks. Our method is informative, interpretable, and fast; it provides novel metrics for quantifying a proposed arm design's ability to perform a task, it yields a graphical interpretation of performance through segment forces, and computing it is over 80x faster than optimization based methods.Our formulation focuses on antagonistic, pneumatically-driven soft arms. We demonstrate our approach through example analysis, and also through consideration of antagonistic vs non-antagonistic designs. Our method enables fast, direct and task-specific comparison of these two architectures, and provides a new visualization of the comparative mechanics. While only a first step, the proposed approach will support advancement of model-based design tools, leading to highly capable soft arms.
comment: Published in the 8th IEEE-RAS International Conference on Soft Robotics (RoboSoft 2025). See https://github.com/wfan19/antagonistic-task-competency for code, proofs, and supplementary information. Please note the officially published version of the paper in IEEE contains an error in Equation 7. That has been corrected here, so this is the final version of the paper. Apologies for the confusion!
♻ ☆ BOP-ASK: Object-Interaction Reasoning for Vision-Language Models
Vision Language Models (VLMs) have achieved impressive performance on spatial reasoning benchmarks, yet these evaluations mask critical weaknesses in understanding object interactions. Current benchmarks test high level relationships ('left of,' 'behind', etc.) but ignore fine-grained spatial understanding needed for real world applications: precise 3D localization, physical compatibility between objects, object affordances and multi step spatial planning. In this work, we present BOP-ASK, a novel large scale dataset for object interaction reasoning for both training and benchmarking. Our data generation pipeline leverages 6D object poses from the Benchmark for Object Pose Estimation (BOP) datasets from which we derive fine grained annotations such as grasp poses, referred object poses, path planning trajectories, relative spatial and depth relationships, and object-to-object relationships. BOP-ASK comprises over 150k images and 33M question answer pairs spanning six tasks (four novel), providing a rich resource for training and evaluating VLMs. We evaluate proprietary and open sourced VLMs, and conduct human evaluations on BOP-ASK-core, a contributed test benchmark. We also release BOP-ASK-lab, an out-of-distribution benchmark with images not sourced from BOP, enabling testing of generalization. Our experiments demonstrate that models trained on BOP-ASK outperform baselines and exhibit emergent capabilities such as precise object and grasp pose estimation, trajectory planning, and fine-grained object-centric spatial reasoning in cluttered environments. We will publicly release our datasets and dataset generation pipeline.
♻ ☆ Designing for Distributed Heterogeneous Modularity: On Software Architecture and Deployment of MoonBots SP
This paper presents the software architecture and deployment strategy behind the MoonBot platform: a modular space robotic system composed of heterogeneous components distributed across multiple computers, networks and ultimately celestial bodies. We introduce a principled approach to distributed, heterogeneous modularity, extending modular robotics beyond physical reconfiguration to software, communication and orchestration. We detail the architecture of our system that integrates component-based design, a data-oriented communication model using ROS2 and Zenoh, and a deployment orchestrator capable of managing complex multi-module assemblies. These abstractions enable dynamic reconfiguration, decentralized control, and seamless collaboration between numerous operators and modules. At the heart of this system lies our open-source Motion Stack software, validated by months of field deployment with self-assembling robots, inter-robot cooperation, and remote operation. Our architecture tackles the significant hurdles of modular robotics by significantly reducing integration and maintenance overhead, while remaining scalable and robust. Although tested with space in mind, we propose generalizable patterns for designing robotic systems that must scale across time, hardware, teams and operational environments.
comment: 6 pages, 8 figures. Accepted at ISPARO 2025
♻ ☆ Scalable Policy Evaluation with Video World Models
Training generalist policies for robotic manipulation has shown great promise, as they enable language-conditioned, multi-task behaviors across diverse scenarios. However, evaluating these policies remains difficult because real-world testing is expensive, time-consuming, and labor-intensive. It also requires frequent environment resets and carries safety risks when deploying unproven policies on physical robots. Manually creating and populating simulation environments with assets for robotic manipulation has not addressed these issues, primarily due to the significant engineering effort required and the substantial sim-to-real gap, both in terms of physics and rendering. In this paper, we explore the use of action-conditional video generation models as a scalable way to learn world models for policy evaluation. We demonstrate how to incorporate action conditioning into existing pre-trained video generation models. This allows leveraging internet-scale in-the-wild online videos during the pre-training stage and alleviates the need for a large dataset of paired video-action data, which is expensive to collect for robotic manipulation. Our paper examines the effect of dataset diversity, pre-trained weights, and common failure cases for the proposed evaluation pipeline. Our experiments demonstrate that across various metrics, including policy ranking and the correlation between actual policy values and predicted policy values, these models offer a promising approach for evaluating policies without requiring real-world interactions.
♻ ☆ Q-STAC: Q-Guided Stein Variational Model Predictive Actor-Critic
Deep reinforcement learning (DRL) often struggles with complex robotic manipulation tasks due to low sample efficiency and biased value estimation. Model-based reinforcement learning (MBRL) improves efficiency by leveraging environment dynamics, with prior work integrating Model Predictive Control (MPC) to enhance policy robustness through online trajectory optimization. However, existing MBRL approaches still suffer from high model bias, task-specific cost function design, and significant computational overhead. To address these challenges, we propose Q-guided Stein Variational Model Predictive Actor-Critic (Q-STAC)--a unified framework that bridges Bayesian MPC and Soft Actor-Critic (SAC). Q-STAC employs Stein Variational Gradient Descent (SVGD) to iteratively optimize action sequences sampled from a learned prior distribution guided by Q-values, thereby eliminating manual cost-function engineering. By performing short-horizon model-predictive rollouts, Q-STAC reduces cumulative prediction errors, improves training stability and reduces computational complexity. Experiments on simulated particle navigation, diverse robotic manipulation tasks, and a real-world fruit-picking scenario demonstrate that Q-STAC consistently achieves superior sample efficiency, stability, and overall performance compared to both model-free and model-based baselines.
comment: 9 pages, 10 figures
♻ ☆ Bootstrap Dynamic-Aware 3D Visual Representation for Scalable Robot Learning
Despite strong results on recognition and segmentation, current 3D visual pre-training methods often underperform on robotic manipulation. We attribute this gap to two factors: the lack of state-action-state dynamics modeling and the unnecessary redundancy of explicit geometric reconstruction. We introduce AFRO, a self-supervised framework that learns dynamics-aware 3D representations without action or reconstruction supervision. AFRO casts state prediction as a generative diffusion process and jointly models forward and inverse dynamics in a shared latent space to capture causal transition structure. To prevent feature leakage in action learning, we employ feature differencing and inverse-consistency supervision, improving the quality and stability of visual features. When combined with Diffusion Policy, AFRO substantially increases manipulation success rates across 16 simulated and 4 real-world tasks, outperforming existing pre-training approaches. The framework also scales favorably with data volume and task complexity. Qualitative visualizations indicate that AFRO learns semantically rich, discriminative features, offering an effective pre-training solution for 3D representation learning in robotics. Project page: https://kolakivy.github.io/AFRO/
♻ ☆ Beyond Description: Cognitively Benchmarking Fine-Grained Action for Embodied Agents
Multimodal Large Language Models (MLLMs) show promising results as decision-making engines for embodied agents operating in complex, physical environments. However, existing benchmarks often prioritize high-level planning or spatial reasoning, leaving the fine-grained action intelligence required for embodied physical interaction underexplored. To address this gap, we introduce CFG-Bench, a new benchmark designed to systematically evaluate this crucial capability. CFG-Bench consists of 1,368 curated videos paired with 19,562 three-modalities question-answer pairs targeting four cognitive abilities: 1) Physical Interaction, 2) Temporal-Causal Relation, 3) Intentional Understanding, and 4) Evaluative Judgment. Together, these dimensions provide a systematic framework for assessing a model's ability to translate visual observations into actionable knowledge, moving beyond mere surface-level recognition. Our comprehensive evaluation on CFG-Bench reveals that leading MLLMs struggle to produce detailed instructions for physical interactions and exhibit profound limitations in the higher-order reasoning of intention and evaluation. Moreover, supervised fine-tuning (SFT) on our data demonstrates that teaching an MLLMs to articulate fine-grained actions directly translates to significant performance gains on established embodied benchmarks. Our analysis highlights these limitations and offers insights for developing more capable and grounded embodied agents. Project page: \href{https://cfg-bench.github.io/}{https://cfg-bench.github.io/}.
♻ ☆ WiSER-X: Wireless Signals-based Efficient Decentralized Multi-Robot Exploration without Explicit Information Exchange
We introduce a Wireless Signal based Efficient multi-Robot eXploration (WiSER-X) algorithm applicable to a decentralized team of robots exploring an unknown environment with communication bandwidth constraints. WiSER-X relies only on local inter-robot relative position estimates, that can be obtained by exchanging signal pings from onboard sensors such as WiFi, Ultra-Wide Band, amongst others, to inform the exploration decisions of individual robots to minimize redundant coverage overlaps. Furthermore, WiSER-X also enables asynchronous termination without requiring a shared map between the robots. It also adapts to heterogeneous robot behaviors and even complete failures in unknown environment while ensuring complete coverage. Simulations show that WiSER-X leads to 58% lower overlap than a zero-information-sharing baseline algorithm-1 and only 23% more overlap than a full-information-sharing algorithm baseline algorithm-2. Hardware experiments further validate the feasibility of WiSER-X using full onboard sensing.
♻ ☆ MIGHTY: Hermite Spline-based Efficient Trajectory Planning
Hard-constraint trajectory planners often rely on commercial solvers and demand substantial computational resources. Existing soft-constraint methods achieve faster computation, but either (1) decouple spatial and temporal optimization or (2) restrict the search space. To overcome these limitations, we introduce MIGHTY, a Hermite spline-based planner that performs spatiotemporal optimization while fully leveraging the continuous search space of a spline. In simulation, MIGHTY achieves a 9.3% reduction in computation time and a 13.1% reduction in travel time over state-of-the-art baselines, with a 100% success rate. In hardware, MIGHTY completes multiple high-speed flights up to 6.7 m/s in a cluttered static environment and long-duration flights with dynamically added obstacles.
comment: 9 pages, 10 figures
♻ ☆ SAT: Dynamic Spatial Aptitude Training for Multimodal Language Models
Reasoning about motion and space is a fundamental cognitive capability that is required by multiple real-world applications. While many studies highlight that large multimodal language models (MLMs) struggle to reason about space, they only focus on static spatial relationships, and not dynamic awareness of motion and space, i.e., reasoning about the effect of egocentric and object motions on spatial relationships. Manually annotating such object and camera movements is expensive. Hence, we introduce SAT, a simulated spatial aptitude training dataset utilizing 3D simulators, comprising both static and dynamic spatial reasoning across 175K question-answer (QA) pairs and 20K scenes. Complementing this, we also construct a small (150 image-QAs) yet challenging dynamic spatial test set using real-world images. Leveraging our SAT datasets and 6 existing static spatial benchmarks, we systematically investigate what improves both static and dynamic spatial awareness. Our results reveal that simulations are surprisingly effective at imparting spatial aptitude to MLMs that translate to real images. We show that perfect annotations in simulation are more effective than existing approaches of pseudo-annotating real images. For instance, SAT training improves a LLaVA-13B model by an average 11% and a LLaVA-Video-7B model by an average 8% on multiple spatial benchmarks, including our real-image dynamic test set and spatial reasoning on long videos -- even outperforming some large proprietary models. While reasoning over static relationships improves with synthetic training data, there is still considerable room for improvement for dynamic reasoning questions.
comment: Accepted to COLM 2025. Project webpage: https://arijitray.com/SAT/
♻ ☆ SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control
Despite the rise of billion-parameter foundation models trained across thousands of GPUs, similar scaling gains have not been shown for humanoid control. Current neural controllers for humanoids remain modest in size, target a limited set of behaviors, and are trained on a handful of GPUs over several days. We show that scaling up model capacity, data, and compute yields a generalist humanoid controller capable of creating natural and robust whole-body movements. Specifically, we posit motion tracking as a natural and scalable task for humanoid control, leveraging dense supervision from diverse motion-capture data to acquire human motion priors without manual reward engineering. We build a foundation model for motion tracking by scaling along three axes: network size (from 1.2M to 42M parameters), dataset volume (over 100M frames, 700 hours of high-quality motion data), and compute (9k GPU hours). Beyond demonstrating the benefits of scale, we show the practical utility of our model through two mechanisms: (1) a real-time universal kinematic planner that bridges motion tracking to downstream task execution, enabling natural and interactive control, and (2) a unified token space that supports various motion input interfaces, such as VR teleoperation devices, human videos, and vision-language-action (VLA) models, all using the same policy. Scaling motion tracking exhibits favorable properties: performance improves steadily with increased compute and data diversity, and learned representations generalize to unseen motions, establishing motion tracking at scale as a practical foundation for humanoid control.
comment: Project page: https://nvlabs.github.io/SONIC/
Robotics 59
☆ SpaceTools: Tool-Augmented Spatial Reasoning via Double Interactive RL
Vision Language Models (VLMs) demonstrate strong qualitative visual understanding, but struggle with metrically precise spatial reasoning required for embodied applications. The agentic paradigm promises that VLMs can use a wide variety of tools that could augment these capabilities, such as depth estimators, segmentation models, and pose estimators. Yet it remains an open challenge how to realize this vision without solely relying on handcrafted prompting strategies or enforcing fixed, predefined tool pipelines that limit VLMs' ability to discover optimal tool-use patterns. Reinforcement Learning could overcome this gap, but has so far been limited to reasoning with a single visual tool due to the large search space in multi-tool reasoning. We introduce Double Interactive Reinforcement Learning (DIRL), a two-phase training framework where VLMs learn to coordinate multiple tools through interactive exploration and feedback. In the teaching phase, we combine demonstrations from a single tool specialist trained via interactive RL with traces from a frontier model using all tools. In the exploration phase, the model further refines multi-tool coordination through continued RL. Our model, SpaceTools, with tool-augmented spatial reasoning ability, achieves state-of-the-art performance on spatial understanding benchmarks (RoboSpatial-Home, BLINK, BOP-ASK) and demonstrates reliable real-world manipulation using a 7-DOF robot as a tool. DIRL provides substantial improvements over the vanilla SFT (+12% on RoboSpatial) and RL (+16% on RoboSpatial) baselines. Project page: https://spacetools.github.io/.
☆ Artificial Microsaccade Compensation: Stable Vision for an Ornithopter
Animals with foveated vision, including humans, experience microsaccades, small, rapid eye movements that they are not aware of. Inspired by this phenomenon, we develop a method for "Artificial Microsaccade Compensation". It can stabilize video captured by a tailless ornithopter that has resisted attempts to use camera-based sensing because it shakes at 12-20 Hz. Our approach minimizes changes in image intensity by optimizing over 3D rotation represented in SO(3). This results in a stabilized video, computed in real time, suitable for human viewing, and free from distortion. When adapted to hold a fixed viewing orientation, up to occasional saccades, it can dramatically reduce inter-frame motion while also benefiting from an efficient recursive update. When compared to Adobe Premier Pro's warp stabilizer, which is widely regarded as the best commercial video stabilization software available, our method achieves higher quality results while also running in real time.
comment: 29 pages, 5 figures, 2 tables, under review
☆ When to Say "Hi" - Learn to Open a Conversation with an in-the-wild Dataset
The social capabilities of socially interactive agents (SIA) are a key to successful and smooth interactions between the user and the SIA. A successful start of the interaction is one of the essential factors for satisfying SIA interactions. For a service and information task in which the SIA helps with information, e.g. about the location, it is an important skill to master the opening of the conversation and to recognize which interlocutor opens the conversation and when. We are therefore investigating the extent to which the opening of the conversation can be trained using the user's body language as an input for machine learning to ensure smooth conversation starts for the interaction. In this paper we propose the Interaction Initiation System (IIS) which we developed, trained and validated using an in-the-wild data set. In a field test at the Deutsches Museum Bonn, a Furhat robot from Furhat Robotics was used as a service and information point. Over the period of use we collected the data of \textit{N} = 201 single user interactions for the training of the algorithms. We can show that the IIS, achieves a performance that allows the conclusion that this system is able to determine the greeting period and the opener of the interaction.
comment: 6 pages, 3 figures, 5 tables. This paper has been accepted for publication at IEEE ROMAN 2025
☆ MDE-AgriVLN: Agricultural Vision-and-Language Navigation with Monocular Depth Estimation
Agricultural robots are serving as powerful assistants across a wide range of agricultural tasks, nevertheless, still heavily relying on manual operations or railway systems for movement. The AgriVLN method and the A2A benchmark pioneeringly extend Vision-and-Language Navigation (VLN) to the agricultural domain, enabling a robot to navigate to a target position following a natural language instruction. Unlike human binocular vision, most agricultural robots are only given a single camera for monocular vision, which results in limited spatial perception. To bridge this gap, we present the method of Agricultural Vision-and-Language Navigation with Monocular Depth Estimation (MDE-AgriVLN), in which we propose the MDE module generating depth features from RGB images, to assist the decision-maker on reasoning. When evaluated on the A2A benchmark, our MDE-AgriVLN method successfully increases Success Rate from 0.23 to 0.32 and decreases Navigation Error from 4.43m to 4.08m, demonstrating the state-of-the-art performance in the agricultural VLN domain. Code: https://github.com/AlexTraveling/MDE-AgriVLN.
☆ Classification of User Satisfaction in HRI with Social Signals in the Wild
Socially interactive agents (SIAs) are being used in various scenarios and are nearing productive deployment. Evaluating user satisfaction with SIAs' performance is a key factor in designing the interaction between the user and SIA. Currently, subjective user satisfaction is primarily assessed manually through questionnaires or indirectly via system metrics. This study examines the automatic classification of user satisfaction through analysis of social signals, aiming to enhance both manual and autonomous evaluation methods for SIAs. During a field trial at the Deutsches Museum Bonn, a Furhat Robotics head was employed as a service and information hub, collecting an "in-the-wild" dataset. This dataset comprises 46 single-user interactions, including questionnaire responses and video data. Our method focuses on automatically classifying user satisfaction based on time series classification. We use time series of social signal metrics derived from the body pose, time series of facial expressions, and physical distance. This study compares three feature engineering approaches on different machine learning models. The results confirm the method's effectiveness in reliably identifying interactions with low user satisfaction without the need for manually annotated datasets. This approach offers significant potential for enhancing SIA performance and user experience through automated feedback mechanisms.
comment: 15 pages, 3 figures. This paper has been accepted for publication at ICSR+AI 2025
☆ MUT3R: Motion-aware Updating Transformer for Dynamic 3D Reconstruction
Recent stateful recurrent neural networks have achieved remarkable progress on static 3D reconstruction but remain vulnerable to motion-induced artifacts, where non-rigid regions corrupt attention propagation between the spatial memory and image feature. By analyzing the internal behaviors of the state and image token updating mechanism, we find that aggregating self-attention maps across layers reveals a consistent pattern: dynamic regions are naturally down-weighted, exposing an implicit motion cue that the pretrained transformer already encodes but never explicitly uses. Motivated by this observation, we introduce MUT3R, a training-free framework that applies the attention-derived motion cue to suppress dynamic content in the early layers of the transformer during inference. Our attention-level gating module suppresses the influence of dynamic regions before their artifacts propagate through the feature hierarchy. Notably, we do not retrain or fine-tune the model; we let the pretrained transformer diagnose its own motion cues and correct itself. This early regulation stabilizes geometric reasoning in streaming scenarios and leads to improvements in temporal consistency and camera pose robustness across multiple dynamic benchmarks, offering a simple and training-free pathway toward motion-aware streaming reconstruction.
☆ Driving is a Game: Combining Planning and Prediction with Bayesian Iterative Best Response
Autonomous driving planning systems perform nearly perfectly in routine scenarios using lightweight, rule-based methods but still struggle in dense urban traffic, where lane changes and merges require anticipating and influencing other agents. Modern motion predictors offer highly accurate forecasts, yet their integration into planning is mostly rudimental: discarding unsafe plans. Similarly, end-to-end models offer a one-way integration that avoids the challenges of joint prediction and planning modeling under uncertainty. In contrast, game-theoretic formulations offer a principled alternative but have seen limited adoption in autonomous driving. We present Bayesian Iterative Best Response (BIBeR), a framework that unifies motion prediction and game-theoretic planning into a single interaction-aware process. BIBeR is the first to integrate a state-of-the-art predictor into an Iterative Best Response (IBR) loop, repeatedly refining the strategies of the ego vehicle and surrounding agents. This repeated best-response process approximates a Nash equilibrium, enabling bidirectional adaptation where the ego both reacts to and shapes the behavior of others. In addition, our proposed Bayesian confidence estimation quantifies prediction reliability and modulates update strength, more conservative under low confidence and more decisive under high confidence. BIBeR is compatible with modern predictors and planners, combining the transparency of structured planning with the flexibility of learned models. Experiments show that BIBeR achieves an 11% improvement over state-of-the-art planners on highly interactive interPlan lane-change scenarios, while also outperforming existing approaches on standard nuPlan benchmarks.
☆ Hierarchical Vision Language Action Model Using Success and Failure Demonstrations
Prior Vision-Language-Action (VLA) models are typically trained on teleoperated successful demonstrations, while discarding numerous failed attempts that occur naturally during data collection. However, these failures encode where and how policies can be fragile, information that can be exploited to improve robustness. We address this problem by leveraging mixed-quality datasets to learn failure-aware reasoning at planning time. We introduce VINE, a hierarchical vision-language-action model that separates high-level reasoning (System 2) from low-level control (System 1) under a hierarchical reinforcement learning formalism, making failures usable as a structured learning signal rather than noisy supervision. System 2 performs feasibility-guided tree search over a 2D scene-graph abstraction: it proposes subgoal transitions, predicts success probabilities from both successes and failures, and prunes brittle branches before execution, effectively casting plan evaluation as feasibility scoring. The selected subgoal sequence is then passed to System 1, which executes low-level actions without modifying the agent's core skills. Trained entirely from offline teleoperation data, VINE integrates negative experience directly into the decision loop. Across challenging manipulation tasks, this approach consistently improves success rates and robustness, demonstrating that failure data is an essential resource for converting the broad competence of VLAs into robust execution.
comment: https://vine-vla.github.io/
☆ Autonomous Reinforcement Learning Robot Control with Intel's Loihi 2 Neuromorphic Hardware
We present an end-to-end pipeline for deploying reinforcement learning (RL) trained Artificial Neural Networks (ANNs) on neuromorphic hardware by converting them into spiking Sigma-Delta Neural Networks (SDNNs). We demonstrate that an ANN policy trained entirely in simulation can be transformed into an SDNN compatible with Intel's Loihi 2 architecture, enabling low-latency and energy-efficient inference. As a test case, we use an RL policy for controlling the Astrobee free-flying robot, similar to a previously hardware in space-validated controller. The policy, trained with Rectified Linear Units (ReLUs), is converted to an SDNN and deployed on Intel's Loihi 2, then evaluated in NVIDIA's Omniverse Isaac Lab simulation environment for closed-loop control of Astrobee's motion. We compare execution performance between GPU and Loihi 2. The results highlight the feasibility of using neuromorphic platforms for robotic control and establish a pathway toward energy-efficient, real-time neuromorphic computation in future space and terrestrial robotics applications.
comment: Submitted for review at NICE 2026 (Neuro-Inspired Computational Elements) conference
☆ Digital Twin-based Control Co-Design of Full Vehicle Active Suspensions via Deep Reinforcement Learning
Active suspension systems are critical for enhancing vehicle comfort, safety, and stability, yet their performance is often limited by fixed hardware designs and control strategies that cannot adapt to uncertain and dynamic operating conditions. Recent advances in digital twins (DTs) and deep reinforcement learning (DRL) offer new opportunities for real-time, data-driven optimization across a vehicle's lifecycle. However, integrating these technologies into a unified framework remains an open challenge. This work presents a DT-based control co-design (CCD) framework for full-vehicle active suspensions using multi-generation design concepts. By integrating automatic differentiation into DRL, we jointly optimize physical suspension components and control policies under varying driver behaviors and environmental uncertainties. DRL also addresses the challenge of partial observability, where only limited states can be sensed and fed back to the controller, by learning optimal control actions directly from available sensor information. The framework incorporates model updating with quantile learning to capture data uncertainty, enabling real-time decision-making and adaptive learning from digital-physical interactions. The approach demonstrates personalized optimization of suspension systems under two distinct driving settings (mild and aggressive). Results show that the optimized systems achieve smoother trajectories and reduce control efforts by approximately 43% and 52% for mild and aggressive, respectively, while maintaining ride comfort and stability. Contributions include: developing a DT-enabled CCD framework integrating DRL and uncertainty-aware model updating for full-vehicle active suspensions, introducing a multi-generation design strategy for self-improving systems, and demonstrating personalized optimization of active suspension systems for distinct driver types.
comment: 28 pages, 17 figures
☆ A Modular Architecture Design for Autonomous Driving Racing in Controlled Environments
This paper presents an Autonomous System (AS) architecture for vehicles in a closed circuit. The AS performs precision tasks including computer vision for environment perception, positioning and mapping for accurate localization, path planning for optimal trajectory generation, and control for precise vehicle actuation. Each subsystem operates independently while connecting data through a cohesive pipeline architecture. The system implements a modular design that combines state-of-the-art technologies for real-time autonomous navigation in controlled environments.
☆ OmniDexVLG: Learning Dexterous Grasp Generation from Vision Language Model-Guided Grasp Semantics, Taxonomy and Functional Affordance
Dexterous grasp generation aims to produce grasp poses that align with task requirements and human interpretable grasp semantics. However, achieving semantically controllable dexterous grasp synthesis remains highly challenging due to the lack of unified modeling of multiple semantic dimensions, including grasp taxonomy, contact semantics, and functional affordance. To address these limitations, we present OmniDexVLG, a multimodal, semantics aware grasp generation framework capable of producing structurally diverse and semantically coherent dexterous grasps under joint language and visual guidance. Our approach begins with OmniDexDataGen, a semantic rich dexterous grasp dataset generation pipeline that integrates grasp taxonomy guided configuration sampling, functional affordance contact point sampling, taxonomy aware differential force closure grasp sampling, and physics based optimization and validation, enabling systematic coverage of diverse grasp types. We further introduce OmniDexReasoner, a multimodal grasp type semantic reasoning module that leverages multi agent collaboration, retrieval augmented generation, and chain of thought reasoning to infer grasp related semantics and generate high quality annotations that align language instructions with task specific grasp intent. Building upon these components, we develop a unified Vision Language Grasping generation model that explicitly incorporates grasp taxonomy, contact structure, and functional affordance semantics, enabling fine grained control over grasp synthesis from natural language instructions. Extensive experiments in simulation and real world object grasping and ablation studies demonstrate that our method substantially outperforms state of the art approaches in terms of grasp diversity, contact semantic diversity, functional affordance diversity, and semantic consistency.
comment: Project Website: https://sites.google.com/view/omnidexvlg, 16 pages
☆ IM HERE: Interaction Model for Human Effort Based Robot Engagement
The effectiveness of human-robot interaction often hinges on the ability to cultivate engagement - a dynamic process of cognitive involvement that supports meaningful exchanges. Many existing definitions and models of engagement are either too vague or lack the ability to generalize across different contexts. We introduce IM HERE, a novel framework that models engagement effectively in human-human, human-robot, and robot-robot interactions. By employing an effort-based description of bilateral relationships between entities, we provide an accurate breakdown of relationship patterns, simplifying them to focus placement and four key states. This framework captures mutual relationships, group behaviors, and actions conforming to social norms, translating them into specific directives for autonomous systems. By integrating both subjective perceptions and objective states, the model precisely identifies and describes miscommunication. The primary objective of this paper is to automate the analysis, modeling, and description of social behavior, and to determine how autonomous systems can behave in accordance with social norms for full social integration while simultaneously pursuing their own social goals.
comment: 8 pages, 5 figures
☆ MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
Autonomous Driving (AD) vehicles still struggle to exhibit human-like behavior in highly dynamic and interactive traffic scenarios. The key challenge lies in AD's limited ability to interact with surrounding vehicles, largely due to a lack of understanding the underlying mechanisms of social interaction. To address this issue, we introduce MPCFormer, an explainable socially-aware autonomous driving approach with physics-informed and data-driven coupled social interaction dynamics. In this model, the dynamics are formulated into a discrete space-state representation, which embeds physics priors to enhance modeling explainability. The dynamics coefficients are learned from naturalistic driving data via a Transformer-based encoder-decoder architecture. To the best of our knowledge, MPCFormer is the first approach to explicitly model the dynamics of multi-vehicle social interactions. The learned social interaction dynamics enable the planner to generate manifold, human-like behaviors when interacting with surrounding traffic. By leveraging the MPC framework, the approach mitigates the potential safety risks typically associated with purely learning-based methods. Open-looped evaluation on NGSIM dataset demonstrates that MPCFormer achieves superior social interaction awareness, yielding the lowest trajectory prediction errors compared with other state-of-the-art approach. The prediction achieves an ADE as low as 0.86 m over a long prediction horizon of 5 seconds. Close-looped experiments in highly intense interaction scenarios, where consecutive lane changes are required to exit an off-ramp, further validate the effectiveness of MPCFormer. Results show that MPCFormer achieves the highest planning success rate of 94.67%, improves driving efficiency by 15.75%, and reduces the collision rate from 21.25% to 0.5%, outperforming a frontier Reinforcement Learning (RL) based planner.
comment: 17 pages, 18 figures
☆ Safety Reinforced Model Predictive Control (SRMPC): Improving MPC with Reinforcement Learning for Motion Planning in Autonomous Driving
Model predictive control (MPC) is widely used for motion planning, particularly in autonomous driving. Real-time capability of the planner requires utilizing convex approximation of optimal control problems (OCPs) for the planner. However, such approximations confine the solution to a subspace, which might not contain the global optimum. To address this, we propose using safe reinforcement learning (SRL) to obtain a new and safe reference trajectory within MPC. By employing a learning-based approach, the MPC can explore solutions beyond the close neighborhood of the previous one, potentially finding global optima. We incorporate constrained reinforcement learning (CRL) to ensure safety in automated driving, using a handcrafted energy function-based safety index as the constraint objective to model safe and unsafe regions. Our approach utilizes a state-dependent Lagrangian multiplier, learned concurrently with the safe policy, to solve the CRL problem. Through experimentation in a highway scenario, we demonstrate the superiority of our approach over both MPC and SRL in terms of safety and performance measures.
☆ Bayesian Optimization for Automatic Tuning of Torque-Level Nonlinear Model Predictive Control
This paper presents an auto-tuning framework for torque-based Nonlinear Model Predictive Control (nMPC), where the MPC serves as a real-time controller for optimal joint torque commands. The MPC parameters, including cost function weights and low-level controller gains, are optimized using high-dimensional Bayesian Optimization (BO) techniques, specifically Sparse Axis-Aligned Subspace (SAASBO) with a digital twin (DT) to achieve precise end-effector trajectory real-time tracking on an UR10e robot arm. The simulation model allows efficient exploration of the high-dimensional parameter space, and it ensures safe transfer to hardware. Our simulation results demonstrate significant improvements in tracking performance (+41.9%) and reduction in solve times (-2.5%) compared to manually-tuned parameters. Moreover, experimental validation on the real robot follows the trend (with a +25.8% improvement), emphasizing the importance of digital twin-enabled automated parameter optimization for robotic operations.
comment: 6 pages, 7 figures, 3 tables
☆ Prediction-Driven Motion Planning: Route Integration Strategies in Attention-Based Prediction Models SC
Combining motion prediction and motion planning offers a promising framework for enhancing interactions between automated vehicles and other traffic participants. However, this introduces challenges in conditioning predictions on navigation goals and ensuring stable, kinematically feasible trajectories. Addressing the former challenge, this paper investigates the extension of attention-based motion prediction models with navigation information. By integrating the ego vehicle's intended route and goal pose into the model architecture, we bridge the gap between multi-agent motion prediction and goal-based motion planning. We propose and evaluate several architectural navigation integration strategies to our model on the nuPlan dataset. Our results demonstrate the potential of prediction-driven motion planning, highlighting how navigation information can enhance both prediction and planning tasks. Our implementation is at: https://github.com/KIT-MRT/future-motion.
comment: In Proceedings of the IEEE International Conference on Intelligent Transportation Systems (ITSC), Gold Coast, AUSTRALIA, 18-21 November 2025
☆ Cross-embodied Co-design for Dexterous Hands
Dexterous manipulation is limited by both control and design, without consensus as to what makes manipulators best for performing dexterous tasks. This raises a fundamental challenge: how should we design and control robot manipulators that are optimized for dexterity? We present a co-design framework that learns task-specific hand morphology and complementary dexterous control policies. The framework supports 1) an expansive morphology search space including joint, finger, and palm generation, 2) scalable evaluation across the wide design space via morphology-conditioned cross-embodied control, and 3) real-world fabrication with accessible components. We evaluate the approach across multiple dexterous tasks, including in-hand rotation with simulation and real deployment. Our framework enables an end-to-end pipeline that can design, train, fabricate, and deploy a new robotic hand in under 24 hours. The full framework will be open-sourced and available on our website.
☆ Crossing the Sim2Real Gap Between Simulation and Ground Testing to Space Deployment of Autonomous Free-flyer Control
Reinforcement learning (RL) offers transformative potential for robotic control in space. We present the first on-orbit demonstration of RL-based autonomous control of a free-flying robot, the NASA Astrobee, aboard the International Space Station (ISS). Using NVIDIA's Omniverse physics simulator and curriculum learning, we trained a deep neural network to replace Astrobee's standard attitude and translation control, enabling it to navigate in microgravity. Our results validate a novel training pipeline that bridges the simulation-to-reality (Sim2Real) gap, utilizing a GPU-accelerated, scientific-grade simulation environment for efficient Monte Carlo RL training. This successful deployment demonstrates the feasibility of training RL policies terrestrially and transferring them to space-based applications. This paves the way for future work in In-Space Servicing, Assembly, and Manufacturing (ISAM), enabling rapid on-orbit adaptation to dynamic mission requirements.
comment: published at iSpaRo 2025
☆ Autonomous Planning In-space Assembly Reinforcement-learning free-flYer (APIARY) International Space Station Astrobee Testing
The US Naval Research Laboratory's (NRL's) Autonomous Planning In-space Assembly Reinforcement-learning free-flYer (APIARY) experiment pioneers the use of reinforcement learning (RL) for control of free-flying robots in the zero-gravity (zero-G) environment of space. On Tuesday, May 27th 2025 the APIARY team conducted the first ever, to our knowledge, RL control of a free-flyer in space using the NASA Astrobee robot on-board the International Space Station (ISS). A robust 6-degrees of freedom (DOF) control policy was trained using an actor-critic Proximal Policy Optimization (PPO) network within the NVIDIA Isaac Lab simulation environment, randomizing over goal poses and mass distributions to enhance robustness. This paper details the simulation testing, ground testing, and flight validation of this experiment. This on-orbit demonstration validates the transformative potential of RL for improving robotic autonomy, enabling rapid development and deployment (in minutes to hours) of tailored behaviors for space exploration, logistics, and real-time mission needs.
comment: iSpaRo 2025, Best Paper Award in Orbital Robotics
☆ PosA-VLA: Enhancing Action Generation via Pose-Conditioned Anchor Attention
The Vision-Language-Action (VLA) models have demonstrated remarkable performance on embodied tasks and shown promising potential for real-world applications. However, current VLAs still struggle to produce consistent and precise target-oriented actions, as they often generate redundant or unstable motions along trajectories, limiting their applicability in time-sensitive scenarios.In this work, we attribute these redundant actions to the spatially uniform perception field of existing VLAs, which causes them to be distracted by target-irrelevant objects, especially in complex environments.To address this issue, we propose an efficient PosA-VLA framework that anchors visual attention via pose-conditioned supervision, consistently guiding the model's perception toward task-relevant regions. The pose-conditioned anchor attention mechanism enables the model to better align instruction semantics with actionable visual cues, thereby improving action generation precision and efficiency. Moreover, our framework adopts a lightweight architecture and requires no auxiliary perception modules (e.g., segmentation or grounding networks), ensuring efficient inference. Extensive experiments verify that our method executes embodied tasks with precise and time-efficient behavior across diverse robotic manipulation benchmarks and shows robust generalization in a variety of challenging environments.
☆ ContactRL: Safe Reinforcement Learning based Motion Planning for Contact based Human Robot Collaboration
In collaborative human-robot tasks, safety requires not only avoiding collisions but also ensuring safe, intentional physical contact. We present ContactRL, a reinforcement learning (RL) based framework that directly incorporates contact safety into the reward function through force feedback. This enables a robot to learn adaptive motion profiles that minimize human-robot contact forces while maintaining task efficiency. In simulation, ContactRL achieves a low safety violation rate of 0.2\% with a high task success rate of 87.7\%, outperforming state-of-the-art constrained RL baselines. In order to guarantee deployment safety, we augment the learned policy with a kinetic energy based Control Barrier Function (eCBF) shield. Real-world experiments on an UR3e robotic platform performing small object handovers from a human hand across 360 trials confirm safe contact, with measured normal forces consistently below 10N. These results demonstrate that ContactRL enables safe and efficient physical collaboration, thereby advancing the deployment of collaborative robots in contact-rich tasks.
comment: 8 pages, 7 figures
☆ A Novel Approach to Tomato Harvesting Using a Hybrid Gripper with Semantic Segmentation and Keypoint Detection
This paper presents an autonomous tomato-harvesting system built around a hybrid robotic gripper that combines six soft auxetic fingers with a rigid exoskeleton and a latex basket to achieve gentle, cage-like grasping. The gripper is driven by a servo-actuated Scotch--yoke mechanism, and includes separator leaves that form a conical frustum for fruit isolation, with an integrated micro-servo cutter for pedicel cutting. For perception, an RGB--D camera and a Detectron2-based pipeline perform semantic segmentation of ripe/unripe tomatoes and keypoint localization of the pedicel and fruit center under occlusion and variable illumination. An analytical model derived using the principle of virtual work relates servo torque to grasp force, enabling design-level reasoning about actuation requirements. During execution, closed-loop grasp-force regulation is achieved using a proportional--integral--derivative controller with feedback from force-sensitive resistors mounted on selected fingers to prevent slip and bruising. Motion execution is supported by Particle Swarm Optimization (PSO)--based trajectory planning for a 5-DOF manipulator. Experiments demonstrate complete picking cycles (approach, separation, cutting, grasping, transport, release) with an average cycle time of 24.34~s and an overall success rate of approximately 80\%, while maintaining low grasp forces (0.20--0.50~N). These results validate the proposed hybrid gripper and integrated vision--control pipeline for reliable harvesting in cluttered environments.
☆ Context-Triggered Contingency Games for Strategic Multi-Agent Interaction
We address the challenge of reliable and efficient interaction in autonomous multi-agent systems, where agents must balance long-term strategic objectives with short-term dynamic adaptation. We propose context-triggered contingency games, a novel integration of strategic games derived from temporal logic specifications with dynamic contingency games solved in real time. Our two-layered architecture leverages strategy templates to guarantee satisfaction of high-level objectives, while a new factor-graph-based solver enables scalable, real-time model predictive control of dynamic interactions. The resulting framework ensures both safety and progress in uncertain, interactive environments. We validate our approach through simulations and hardware experiments in autonomous driving and robotic navigation, demonstrating efficient, reliable, and adaptive multi-agent interaction.
☆ Multimodal Control of Manipulators: Coupling Kinematics and Vision for Self-Driving Laboratory Operations
Motion planning schemes are used for planning motions of a manipulator from an initial pose to a final pose during a task execution. A motion planning scheme generally comprises of a trajectory planning method and an inverse kinematic solver to determine trajectories and joints solutions respectively. In this paper, 3 motion planning schemes developed based on Jacobian methods are implemented to traverse a redundant manipulator with a coupled finger gripper through given trajectories. RRT* algorithm is used for planning trajectories and screw theory based forward kinematic equations are solved for determining joint solutions of the manipulator and gripper. Inverse solutions are computed separately using 3 Jacobian based methods such as Jacobian Transpose (JT), Pseudo Inverse (PI), and Damped Least Square (DLS) methods. Space Jacobian and manipulability measurements of the manipulator and gripper are obtained using screw theory formulations. Smoothness and RMSE error of generated trajectories and velocity continuity, acceleration profile, jerk, and snap values of joint motions are analysed for determining an efficient motion planning method for a given task. Advantages and disadvantages of the proposed motion planning schemes mentioned above are analysed using simulation studies to determine a suitable inverse solution technique for the tasks.
☆ RoboScape-R: Unified Reward-Observation World Models for Generalizable Robotics Training via RL
Achieving generalizable embodied policies remains a key challenge. Traditional policy learning paradigms, including both Imitation Learning (IL) and Reinforcement Learning (RL), struggle to cultivate generalizability across diverse scenarios. While IL policies often overfit to specific expert trajectories, RL suffers from the inherent lack of a unified and general reward signal necessary for effective multi-scene generalization. We posit that the world model is uniquely capable of serving as a universal environment proxy to address this limitation. However, current world models primarily focus on their ability to predict observations and still rely on task-specific, handcrafted reward functions, thereby failing to provide a truly general training environment. Toward this problem, we propose RoboScape-R, a framework leveraging the world model to serve as a versatile, general-purpose proxy for the embodied environment within the RL paradigm. We introduce a novel world model-based general reward mechanism that generates ''endogenous'' rewards derived from the model's intrinsic understanding of real-world state transition dynamics. Extensive experiments demonstrate that RoboScape-R effectively addresses the limitations of traditional RL methods by providing an efficient and general training environment that substantially enhances the generalization capability of embodied policies. Our approach offers critical insights into utilizing the world model as an online training strategy and achieves an average 37.5% performance improvement over baselines under out-of-domain scenarios.
☆ A Learning-based Control Methodology for Transitioning VTOL UAVs
Transition control poses a critical challenge in Vertical Take-Off and Landing Unmanned Aerial Vehicle (VTOL UAV) development due to the tilting rotor mechanism, which shifts the center of gravity and thrust direction during transitions. Current control methods' decoupled control of altitude and position leads to significant vibration, and limits interaction consideration and adaptability. In this study, we propose a novel coupled transition control methodology based on reinforcement learning (RL) driven controller. Besides, contrasting to the conventional phase-transition approach, the ST3M method demonstrates a new perspective by treating cruise mode as a special case of hover. We validate the feasibility of applying our method in simulation and real-world environments, demonstrating efficient controller development and migration while accurately controlling UAV position and attitude, exhibiting outstanding trajectory tracking and reduced vibrations during the transition process.
☆ AdaPower: Specializing World Foundation Models for Predictive Manipulation
World Foundation Models (WFMs) offer remarkable visual dynamics simulation capabilities, yet their application to precise robotic control remains limited by the gap between generative realism and control-oriented precision. While existing approaches use WFMs as synthetic data generators, they suffer from high computational costs and underutilization of pre-trained VLA policies. We introduce \textbf{AdaPower} (\textbf{Ada}pt and Em\textbf{power}), a lightweight adaptation framework that transforms general-purpose WFMs into specialist world models through two novel components: Temporal-Spatial Test-Time Training (TS-TTT) for inference-time adaptation and Memory Persistence (MP) for long-horizon consistency. Integrated within a Model Predictive Control framework, our adapted world model empowers pre-trained VLAs, achieving over 41\% improvement in task success rates on LIBERO benchmarks without policy retraining, while preserving computational efficiency and generalist capabilities.
☆ Mobility Induced Sensitivity of UAV based Nodes to Jamming in Private 5G Airfield Networks An Experimental Study
This work presents an experimental performance evaluation of a private 5G airfield network under controlled directional SDR jamming attacks targeting UAV-based UE nodes. Using a QualiPoc Android UE, mounted as a payload on a quadcopter UAV, we conducted a series of experiments to evaluate signal degradation, handover performance, and ser-vice stability in the presence of constant directional jamming. The conducted experiments aimed to examine the effects of varying travel speeds, altitudes, and moving patterns of a UAV-based UE to record and analyze the key physical-layer and network-layer metrics such as CQI, MCS, RSRP, SINR, BLER, Net PDSCH Throughput and RLF. The re-sults of this work describe the link stability and signal degradation dependencies, caused by the level of mobility of the UAV-based UE nodes during autonomous and automatic operation in private 5G Airfield networks
comment: 4 pages, 4 figures
☆ MSG-Loc: Multi-Label Likelihood-based Semantic Graph Matching for Object-Level Global Localization
Robots are often required to localize in environments with unknown object classes and semantic ambiguity. However, when performing global localization using semantic objects, high semantic ambiguity intensifies object misclassification and increases the likelihood of incorrect associations, which in turn can cause significant errors in the estimated pose. Thus, in this letter, we propose a multi-label likelihood-based semantic graph matching framework for object-level global localization. The key idea is to exploit multi-label graph representations, rather than single-label alternatives, to capture and leverage the inherent semantic context of object observations. Based on these representations, our approach enhances semantic correspondence across graphs by combining the likelihood of each node with the maximum likelihood of its neighbors via context-aware likelihood propagation. For rigorous validation, data association and pose estimation performance are evaluated under both closed-set and open-set detection configurations. In addition, we demonstrate the scalability of our approach to large-vocabulary object categories in both real-world indoor scenes and synthetic environments.
comment: Accepted in IEEE Robotics and Automation Letters (2025)
☆ CSMapping: Scalable Crowdsourced Semantic Mapping and Topology Inference for Autonomous Driving
Crowdsourcing enables scalable autonomous driving map construction, but low-cost sensor noise hinders quality from improving with data volume. We propose CSMapping, a system that produces accurate semantic maps and topological road centerlines whose quality consistently increases with more crowdsourced data. For semantic mapping, we train a latent diffusion model on HD maps (optionally conditioned on SD maps) to learn a generative prior of real-world map structure, without requiring paired crowdsourced/HD-map supervision. This prior is incorporated via constrained MAP optimization in latent space, ensuring robustness to severe noise and plausible completion in unobserved areas. Initialization uses a robust vectorized mapping module followed by diffusion inversion; optimization employs efficient Gaussian-basis reparameterization, projected gradient descent zobracket multi-start, and latent-space factor-graph for global consistency. For topological mapping, we apply confidence-weighted k-medoids clustering and kinematic refinement to trajectories, yielding smooth, human-like centerlines robust to trajectory variation. Experiments on nuScenes, Argoverse 2, and a large proprietary dataset achieve state-of-the-art semantic and topological mapping performance, with thorough ablation and scalability studies.
☆ Variable-Impedance Muscle Coordination under Slow-Rate Control Frequencies and Limited Observation Conditions Evaluated through Legged Locomotion
Human motor control remains agile and robust despite limited sensory information for feedback, a property attributed to the body's ability to perform morphological computation through muscle coordination with variable impedance. However, it remains unclear how such low-level mechanical computation reduces the control requirements of the high-level controller. In this study, we implement a hierarchical controller consisting of a high-level neural network trained by reinforcement learning and a low-level variable-impedance muscle coor dination model with mono- and biarticular muscles in monoped locomotion task. We systematically restrict the high-level controller by varying the control frequency and by introducing biologically inspired observation conditions: delayed, partial, and substituted observation. Under these conditions, we evaluate how the low-level variable-impedance muscle coordination contributes to learning process of high-level neural network. The results show that variable-impedance muscle coordination enables stable locomotion even under slow-rate control frequency and limited observation conditions. These findings demonstrate that the morphological computation of muscle coordination effectively offloads high-frequency feedback of the high-level controller and provide a design principle for the controller in motor control.
comment: 12 pages, 11 figures. Submitted to IEEE Transactions on Systems, Man, and Cybernetics: Systems
☆ PerFACT: Motion Policy with LLM-Powered Dataset Synthesis and Fusion Action-Chunking Transformers
Deep learning methods have significantly enhanced motion planning for robotic manipulators by leveraging prior experiences within planning datasets. However, state-of-the-art neural motion planners are primarily trained on small datasets collected in manually generated workspaces, limiting their generalizability to out-of-distribution scenarios. Additionally, these planners often rely on monolithic network architectures that struggle to encode critical planning information. To address these challenges, we introduce Motion Policy with Dataset Synthesis powered by large language models (LLMs) and Fusion Action-Chunking Transformers (PerFACT), which incorporates two key components. Firstly, a novel LLM-powered workspace generation method, MotionGeneralizer, enables large-scale planning data collection by producing a diverse set of semantically feasible workspaces. Secondly, we introduce Fusion Motion Policy Networks (MpiNetsFusion), a generalist neural motion planner that uses a fusion action-chunking transformer to better encode planning signals and attend to multiple feature modalities. Leveraging MotionGeneralizer, we collect 3.5M trajectories to train and evaluate MpiNetsFusion against state-of-the-art planners, which shows that the proposed MpiNetsFusion can plan several times faster on the evaluated tasks.
☆ World Models for Autonomous Navigation of Terrestrial Robots from LIDAR Observations
Autonomous navigation of terrestrial robots using Reinforcement Learning (RL) from LIDAR observations remains challenging due to the high dimensionality of sensor data and the sample inefficiency of model-free approaches. Conventional policy networks struggle to process full-resolution LIDAR inputs, forcing prior works to rely on simplified observations that reduce spatial awareness and navigation robustness. This paper presents a novel model-based RL framework built on top of the DreamerV3 algorithm, integrating a Multi-Layer Perceptron Variational Autoencoder (MLP-VAE) within a world model to encode high-dimensional LIDAR readings into compact latent representations. These latent features, combined with a learned dynamics predictor, enable efficient imagination-based policy optimization. Experiments on simulated TurtleBot3 navigation tasks demonstrate that the proposed architecture achieves faster convergence and higher success rate compared to model-free baselines such as SAC, DDPG, and TD3. It is worth emphasizing that the DreamerV3-based agent attains a 100% success rate across all evaluated environments when using the full dataset of the Turtlebot3 LIDAR (360 readings), while model-free methods plateaued below 85%. These findings demonstrate that integrating predictive world models with learned latent representations enables more efficient and robust navigation from high-dimensional sensory data.
comment: Accepted for publication in the Journal of Intelligent and Fuzzy Systems
☆ What Is The Best 3D Scene Representation for Robotics? From Geometric to Foundation Models
In this paper, we provide a comprehensive overview of existing scene representation methods for robotics, covering traditional representations such as point clouds, voxels, signed distance functions (SDF), and scene graphs, as well as more recent neural representations like Neural Radiance Fields (NeRF), 3D Gaussian Splatting (3DGS), and the emerging Foundation Models. While current SLAM and localization systems predominantly rely on sparse representations like point clouds and voxels, dense scene representations are expected to play a critical role in downstream tasks such as navigation and obstacle avoidance. Moreover, neural representations such as NeRF, 3DGS, and foundation models are well-suited for integrating high-level semantic features and language-based priors, enabling more comprehensive 3D scene understanding and embodied intelligence. In this paper, we categorized the core modules of robotics into five parts (Perception, Mapping, Localization, Navigation, Manipulation). We start by presenting the standard formulation of different scene representation methods and comparing the advantages and disadvantages of scene representation across different modules. This survey is centered around the question: What is the best 3D scene representation for robotics? We then discuss the future development trends of 3D scene representations, with a particular focus on how the 3D Foundation Model could replace current methods as the unified solution for future robotic applications. The remaining challenges in fully realizing this model are also explored. We aim to offer a valuable resource for both newcomers and experienced researchers to explore the future of 3D scene representations and their application in robotics. We have published an open-source project on GitHub and will continue to add new works and technologies to this project.
☆ Surfel-LIO: Fast LiDAR-Inertial Odometry with Pre-computed Surfels and Hierarchical Z-order Voxel Hashing
LiDAR-inertial odometry (LIO) is an active research area, as it enables accurate real-time state estimation in GPS-denied environments. Recent advances in map data structures and spatial indexing have significantly improved the efficiency of LIO systems. Nevertheless, we observe that two aspects may still leave room for improvement: (1) nearest neighbor search often requires examining multiple spatial units to gather sufficient points for plane fitting, and (2) plane parameters are typically recomputed at every iteration despite unchanged map geometry. Motivated by these observations, we propose Surfel-LIO, which employs a hierarchical voxel structure (hVox) with pre-computed surfel representation. This design enables O(1) correspondence retrieval without runtime neighbor enumeration or plane fitting, combined with Z-order curve encoding for cache-friendly spatial indexing. Experimental results on the M3DGR dataset demonstrate that our method achieves significantly faster processing speed compared to recent state-of-the-art methods while maintaining comparable state estimation accuracy. Our implementation is publicly available at https://github.com/93won/lidar_inertial_odometry.
☆ GOMP: Grasped Object Manifold Projection for Multimodal Imitation Learning of Manipulation
Imitation Learning (IL) holds great potential for learning repetitive manipulation tasks, such as those in industrial assembly. However, its effectiveness is often limited by insufficient trajectory precision due to compounding errors. In this paper, we introduce Grasped Object Manifold Projection (GOMP), an interactive method that mitigates these errors by constraining a non-rigidly grasped object to a lower-dimensional manifold. GOMP assumes a precise task in which a manipulator holds an object that may shift within the grasp in an observable manner and must be mated with a grounded part. Crucially, all GOMP enhancements are learned from the same expert dataset used to train the base IL policy, and are adjusted with an n-arm bandit-based interactive component. We propose a theoretical basis for GOMP's improvement upon the well-known compounding error bound in IL literature. We demonstrate the framework on four precise assembly tasks using tactile feedback, and note that the approach remains modality-agnostic. Data and videos are available at williamvdb.github.io/GOMPsite.
comment: 8 pages, 8 figures, 2 tables
☆ NavMapFusion: Diffusion-based Fusion of Navigation Maps for Online Vectorized HD Map Construction WACV 2026
Accurate environmental representations are essential for autonomous driving, providing the foundation for safe and efficient navigation. Traditionally, high-definition (HD) maps are providing this representation of the static road infrastructure to the autonomous system a priori. However, because the real world is constantly changing, such maps must be constructed online from on-board sensor data. Navigation-grade standard-definition (SD) maps are widely available, but their resolution is insufficient for direct deployment. Instead, they can be used as coarse prior to guide the online map construction process. We propose NavMapFusion, a diffusion-based framework that performs iterative denoising conditioned on high-fidelity sensor data and on low-fidelity navigation maps. This paper strives to answer: (1) How can coarse, potentially outdated navigation maps guide online map construction? (2) What advantages do diffusion models offer for map fusion? We demonstrate that diffusion-based map construction provides a robust framework for map fusion. Our key insight is that discrepancies between the prior map and online perception naturally correspond to noise within the diffusion process; consistent regions reinforce the map construction, whereas outdated segments are suppressed. On the nuScenes benchmark, NavMapFusion conditioned on coarse road lines from OpenStreetMap data reaches a 21.4% relative improvement on 100 m, and even stronger improvements on larger perception ranges, while maintaining real-time capabilities. By fusing low-fidelity priors with high-fidelity sensor data, the proposed method generates accurate and up-to-date environment representations, guiding towards safer and more reliable autonomous driving. The code is available at https://github.com/tmonnin/navmapfusion
comment: Accepted to 2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2026)
☆ ResponsibleRobotBench: Benchmarking Responsible Robot Manipulation using Multi-modal Large Language Models
Recent advances in large multimodal models have enabled new opportunities in embodied AI, particularly in robotic manipulation. These models have shown strong potential in generalization and reasoning, but achieving reliable and responsible robotic behavior in real-world settings remains an open challenge. In high-stakes environments, robotic agents must go beyond basic task execution to perform risk-aware reasoning, moral decision-making, and physically grounded planning. We introduce ResponsibleRobotBench, a systematic benchmark designed to evaluate and accelerate progress in responsible robotic manipulation from simulation to real world. This benchmark consists of 23 multi-stage tasks spanning diverse risk types, including electrical, chemical, and human-related hazards, and varying levels of physical and planning complexity. These tasks require agents to detect and mitigate risks, reason about safety, plan sequences of actions, and engage human assistance when necessary. Our benchmark includes a general-purpose evaluation framework that supports multimodal model-based agents with various action representation modalities. The framework integrates visual perception, context learning, prompt construction, hazard detection, reasoning and planning, and physical execution. It also provides a rich multimodal dataset, supports reproducible experiments, and includes standardized metrics such as success rate, safety rate, and safe success rate. Through extensive experimental setups, ResponsibleRobotBench enables analysis across risk categories, task types, and agent configurations. By emphasizing physical reliability, generalization, and safety in decision-making, this benchmark provides a foundation for advancing the development of trustworthy, real-world responsible dexterous robotic systems. https://sites.google.com/view/responsible-robotbench
comment: https://sites.google.com/view/responsible-robotbench
☆ Driving Beyond Privilege: Distilling Dense-Reward Knowledge into Sparse-Reward Policies
We study how to exploit dense simulator-defined rewards in vision-based autonomous driving without inheriting their misalignment with deployment metrics. In realistic simulators such as CARLA, privileged state (e.g., lane geometry, infractions, time-to-collision) can be converted into dense rewards that stabilize and accelerate model-based reinforcement learning, but policies trained directly on these signals often overfit and fail to generalize when evaluated on sparse objectives such as route completion and collision-free overtaking. We propose reward-privileged world model distillation, a two-stage framework in which a teacher DreamerV3-style agent is first trained with a dense privileged reward, and only its latent dynamics are distilled into a student trained solely on sparse task rewards. Teacher and student share the same observation space (semantic bird's-eye-view images); privileged information enters only through the teacher's reward, and the student does not imitate the teacher's actions or value estimates. Instead, the student's world model is regularized to match the teacher's latent dynamics while its policy is learned from scratch on sparse success/failure signals. In CARLA lane-following and overtaking benchmarks, sparse-reward students outperform both dense-reward teachers and sparse-from-scratch baselines. On unseen lane-following routes, reward-privileged distillation improves success by about 23 percent relative to the dense teacher while maintaining comparable or better safety. On overtaking, students retain near-perfect performance on training routes and achieve up to a 27x improvement in success on unseen routes, with improved lane keeping. These results show that dense rewards can be leveraged to learn richer dynamics models while keeping the deployed policy optimized strictly for sparse, deployment-aligned objectives.
☆ Sliding Mode Control and Subspace Stabilization Methodology for the Orbital Stabilization of Periodic Trajectories
This paper presents a combined sliding-mode control and subspace stabilization methodology for orbital stabilization of periodic trajectories in underactuated mechanical systems with one degree of underactuation. The approach starts with partial feedback linearization and stabilization. Then, transverse linearization along the reference orbit is computed, resulting in a periodic linear time-varying system with a stable subspace. Sliding-mode control drives trajectories toward this subspace. The proposed design avoids solving computationally intensive periodic LQR problems and improves robustness to matched disturbances. The methodology is validated through experiments on the Butterfly robot.
☆ CRAFT-E: A Neuro-Symbolic Framework for Embodied Affordance Grounding
Assistive robots operating in unstructured environments must understand not only what objects are, but what they can be used for. This requires grounding language-based action queries to objects that both afford the requested function and can be physically retrieved. Existing approaches often rely on black-box models or fixed affordance labels, limiting transparency, controllability, and reliability for human-facing applications. We introduce CRAFT-E, a modular neuro-symbolic framework that composes a structured verb-property-object knowledge graph with visual-language alignment and energy-based grasp reasoning. The system generates interpretable grounding paths that expose the factors influencing object selection and incorporates grasp feasibility as an integral part of affordance inference. We further construct a benchmark dataset with unified annotations for verb-object compatibility, segmentation, and grasp candidates, and deploy the full pipeline on a physical robot. CRAFT-E achieves competitive performance in static scenes, ImageNet-based functional retrieval, and real-world trials involving 20 verbs and 39 objects. The framework remains robust under perceptual noise and provides transparent, component-level diagnostics. By coupling symbolic reasoning with embodied perception, CRAFT-E offers an interpretable and customizable alternative to end-to-end models for affordance-grounded object selection, supporting trustworthy decision-making in assistive robotic systems.
comment: 20 pages. 3 figures, 4 tables. Under Review
♻ ☆ Supercomputing for High-speed Avoidance and Reactive Planning in Robots
This paper presents SHARP (Supercomputing for High-speed Avoidance and Reactive Planning), a proof-of-concept study demonstrating how high-performance computing (HPC) can enable millisecond-scale responsiveness in robotic control. While modern robots face increasing demands for reactivity in human--robot shared workspaces, onboard processors are constrained by size, power, and cost. Offloading to HPC offers massive parallelism for trajectory planning, but its feasibility for real-time robotics remains uncertain due to network latency and jitter. We evaluate SHARP in a stress-test scenario where a 7-DOF manipulator must dodge high-speed foam projectiles. Using a parallelized multi-goal A* search implemented with MPI on both local and remote HPC clusters, the system achieves mean planning latencies of 22.9 ms (local) and 30.0 ms (remote, ~300 km away), with avoidance success rates of 84% and 88%, respectively. These results show that when round-trip latency remains within the tens-of-milliseconds regime, HPC-side computation is no longer the bottleneck, enabling avoidance well below human reaction times. The SHARP results motivate hybrid control architectures: low-level reflexes remain onboard for safety, while bursty, high-throughput planning tasks are offloaded to HPC for scalability. By reporting per-stage timing and success rates, this study provides a reproducible template for assessing real-time feasibility of HPC-driven robotics. Collectively, SHARP reframes HPC offloading as a viable pathway toward dependable, reactive robots in dynamic environments.
comment: Error in the graph calculation
♻ ☆ SMP: Reusable Score-Matching Motion Priors for Physics-Based Character Control
Data-driven motion priors that can guide agents toward producing naturalistic behaviors play a pivotal role in creating life-like virtual characters. Adversarial imitation learning has been a highly effective method for learning motion priors from reference motion data. However, adversarial priors, with few exceptions, need to be retrained for each new controller, thereby limiting their reusability and necessitating the retention of the reference motion data when training on downstream tasks. In this work, we present Score-Matching Motion Priors (SMP), which leverages pre-trained motion diffusion models and score distillation sampling (SDS) to create reusable task-agnostic motion priors. SMPs can be pre-trained on a motion dataset, independent of any control policy or task. Once trained, SMPs can be kept frozen and reused as general-purpose reward functions to train policies to produce naturalistic behaviors for downstream tasks. We show that a general motion prior trained on large-scale datasets can be repurposed into a variety of style-specific priors. Furthermore SMP can compose different styles to synthesize new styles not present in the original dataset. Our method produces high-quality motion comparable to state-of-the-art adversarial imitation learning methods through reusable and modular motion priors. We demonstrate the effectiveness of SMP across a diverse suite of control tasks with physically simulated humanoid characters. Video demo available at https://youtu.be/ravlZJteS20
comment: 14 pages, 9 figures
♻ ☆ MP1: MeanFlow Tames Policy Learning in 1-step for Robotic Manipulation AAAI 2026
In robot manipulation, robot learning has become a prevailing approach. However, generative models within this field face a fundamental trade-off between the slow, iterative sampling of diffusion models and the architectural constraints of faster Flow-based methods, which often rely on explicit consistency losses. To address these limitations, we introduce MP1, which pairs 3D point-cloud inputs with the MeanFlow paradigm to generate action trajectories in one network function evaluation (1-NFE). By directly learning the interval-averaged velocity via the "MeanFlow Identity", our policy avoids any additional consistency constraints. This formulation eliminates numerical ODE-solver errors during inference, yielding more precise trajectories. MP1 further incorporates CFG for improved trajectory controllability while retaining 1-NFE inference without reintroducing structural constraints. Because subtle scene-context variations are critical for robot learning, especially in few-shot learning, we introduce a lightweight Dispersive Loss that repels state embeddings during training, boosting generalization without slowing inference. We validate our method on the Adroit and Meta-World benchmarks, as well as in real-world scenarios. Experimental results show MP1 achieves superior average task success rates, outperforming DP3 by 10.2% and FlowPolicy by 7.3%. Its average inference time is only 6.8 ms-19x faster than DP3 and nearly 2x faster than FlowPolicy. Our project page is available at https://mp1-2254.github.io/, and the code can be accessed at https://github.com/LogSSim/MP1.
comment: This paper has been accepted by AAAI 2026
♻ ☆ DGFusion: Depth-Guided Sensor Fusion for Robust Semantic Perception
Robust semantic perception for autonomous vehicles relies on effectively combining multiple sensors with complementary strengths and weaknesses. State-of-the-art sensor fusion approaches to semantic perception often treat sensor data uniformly across the spatial extent of the input, which hinders performance when faced with challenging conditions. By contrast, we propose a novel depth-guided multimodal fusion method that upgrades condition-aware fusion by integrating depth information. Our network, DGFusion, poses multimodal segmentation as a multi-task problem, utilizing the lidar measurements, which are typically available in outdoor sensor suites, both as one of the model's inputs and as ground truth for learning depth. Our corresponding auxiliary depth head helps to learn depth-aware features, which are encoded into spatially varying local depth tokens that condition our attentive cross-modal fusion. Together with a global condition token, these local depth tokens dynamically adapt sensor fusion to the spatially varying reliability of each sensor across the scene, which largely depends on depth. In addition, we propose a robust loss for our depth, which is essential for learning from lidar inputs that are typically sparse and noisy in adverse conditions. Our method achieves state-of-the-art panoptic and semantic segmentation performance on the challenging MUSES and DeLiVER datasets. Code and models will be available at https://github.com/timbroed/DGFusion
comment: Code and models will be available at https://github.com/timbroed/DGFusion
♻ ☆ Anti-bullying Adaptive Cruise Control: A proactive right-of-way protection approach
Adaptive Cruise Control (ACC) systems have been widely commercialized in recent years. However, existing ACC systems remain vulnerable to close-range cut-ins, a behavior that resembles "road bullying". To address this issue, this research proposes an Anti-bullying Adaptive Cruise Control (AACC) approach, which is capable of proactively protecting right-of-way against such "road bullying" cut-ins. To handle diverse "road bullying" cut-in scenarios smoothly, the proposed approach first leverages an online Inverse Optimal Control (IOC) based algorithm for individual driving style identification. Then, based on Stackelberg competition, a game-theoretic-based motion planning framework is presented in which the identified individual driving styles are utilized to formulate cut-in vehicles' reaction functions. By integrating such reaction functions into the ego vehicle's motion planning, the ego vehicle could consider cut-in vehicles' all possible reactions to find its optimal right-of-way protection maneuver. To the best of our knowledge, this research is the first to model vehicles' interaction dynamics and develop an interactive planner that adapts cut-in vehicle's various driving styles. Simulation results show that the proposed approach can prevent "road bullying" cut-ins and be adaptive to different cut-in vehicles' driving styles. It can improve safety and comfort by up to 79.8% and 20.4%. The driving efficiency has benefits by up to 19.33% in traffic flow. The proposed approach can also adopt more flexible driving strategies. Furthermore, the proposed approach can support real-time field implementation by ensuring less than 50 milliseconds computation time.
comment: 15 pages, 19 figures
♻ ☆ HybridWorldSim: A Scalable and Controllable High-fidelity Simulator for Autonomous Driving
Realistic and controllable simulation is critical for advancing end-to-end autonomous driving, yet existing approaches often struggle to support novel view synthesis under large viewpoint changes or to ensure geometric consistency. We introduce HybridWorldSim, a hybrid simulation framework that integrates multi-traversal neural reconstruction for static backgrounds with generative modeling for dynamic agents. This unified design addresses key limitations of previous methods, enabling the creation of diverse and high-fidelity driving scenarios with reliable visual and spatial consistency. To facilitate robust benchmarking, we further release a new multi-traversal dataset MIRROR that captures a wide range of routes and environmental conditions across different cities. Extensive experiments demonstrate that HybridWorldSim surpasses previous state-of-the-art methods, providing a practical and scalable solution for high-fidelity simulation and a valuable resource for research and development in autonomous driving.
comment: Project page: https://hybridworldsim.github.io/
♻ ☆ SwarmDiffusion: End-To-End Traversability-Guided Diffusion for Embodiment-Agnostic Navigation of Heterogeneous Robots
Visual traversability estimation is critical for autonomous navigation, but existing VLM-based methods rely on hand-crafted prompts, generalize poorly across embodiments, and output only traversability maps, leaving trajectory generation to slow external planners. We propose SwarmDiffusion, a lightweight end-to-end diffusion model that jointly predicts traversability and generates a feasible trajectory from a single RGB image. To remove the need for annotated or planner-produced paths, we introduce a planner-free trajectory construction pipeline based on randomized waypoint sampling, Bezier smoothing, and regularization enforcing connectivity, safety, directionality, and path thinness. This enables learning stable motion priors without demonstrations. SwarmDiffusion leverages VLM-derived supervision without prompt engineering and conditions the diffusion process on a compact embodiment state, producing physically consistent, traversable paths that transfer across different robot platforms. Across indoor environments and two embodiments (quadruped and aerial), the method achieves 80-100% navigation success and 0.09s inference, and adapts to a new robot using only-500 additional visual samples. It generalizes reliably to unseen environments in simulation and real-world trials, offering a scalable, prompt-free approach to unified traversability reasoning and trajectory generation.
comment: This work has been submitted for publication and is currently under review
♻ ☆ DynamicCity: Large-Scale 4D Occupancy Generation from Dynamic Scenes ICLR 2025
Urban scene generation has been developing rapidly recently. However, existing methods primarily focus on generating static and single-frame scenes, overlooking the inherently dynamic nature of real-world driving environments. In this work, we introduce DynamicCity, a novel 4D occupancy generation framework capable of generating large-scale, high-quality dynamic 4D scenes with semantics. DynamicCity mainly consists of two key models. 1) A VAE model for learning HexPlane as the compact 4D representation. Instead of using naive averaging operations, DynamicCity employs a novel Projection Module to effectively compress 4D features into six 2D feature maps for HexPlane construction, which significantly enhances HexPlane fitting quality (up to 12.56 mIoU gain). Furthermore, we utilize an Expansion & Squeeze Strategy to reconstruct 3D feature volumes in parallel, which improves both network training efficiency and reconstruction accuracy than naively querying each 3D point (up to 7.05 mIoU gain, 2.06x training speedup, and 70.84% memory reduction). 2) A DiT-based diffusion model for HexPlane generation. To make HexPlane feasible for DiT generation, a Padded Rollout Operation is proposed to reorganize all six feature planes of the HexPlane as a squared 2D feature map. In particular, various conditions could be introduced in the diffusion or sampling process, supporting versatile 4D generation applications, such as trajectory- and command-driven generation, inpainting, and layout-conditioned generation. Extensive experiments on the CarlaSC and Waymo datasets demonstrate that DynamicCity significantly outperforms existing state-of-the-art 4D occupancy generation methods across multiple metrics. The code and models have been released to facilitate future research.
comment: ICLR 2025 Spotlight; 35 pages, 18 figures, 15 tables; Project Page at https://dynamic-city.github.io/
♻ ☆ Nonlinear Oscillatory Response of Automated Vehicle Car-following: Theoretical Analysis with Traffic State and Control Input Limits
This paper presents a framework grounded in the theory of describing function (DF) and incremental-input DF to theoretically analyze the nonlinear oscillatory response of automated vehicles (AVs) car-following (CF) amidst traffic oscillations, considering the limits of traffic state and control input. While prevailing approaches largely ignore these limits (i.e., saturation of acceleration/deceleration and speed) and focus on linear string stability analysis, this framework establishes a basis for theoretically analyzing the frequency response of AV systems with nonlinearities imposed by these limits. To this end, trajectories of CF pairs are decomposed into nominal and oscillatory trajectories, subsequently, the controlled AV system is repositioned within the oscillatory trajectory coordinates. Built on this base, DFs are employed to approximate the frequency responses of nonlinear saturation components by using their first harmonic output, thereby capturing the associated amplification ratio and phase shift. Considering the closed-loop nature of AV control systems, where system states and control input mutually influence each other, amplification ratios and phase shifts are balanced within the loop to ensure consistency. This balancing process may render multiple solutions, hence the incremental-input DF is further applied to identify the reasonable ones. The proposed method is validated by estimations from Simulink, and further comparisons with prevailing methods are conducted. Results confirm the alignment of our framework with Simulink results and exhibit its superior accuracy in analysis compared to the prevailing methods. Furthermore, the framework proves valuable in string stability analysis, especially when conventional linear methods offer misleading insights.
♻ ☆ LargeAD: Large-Scale Cross-Sensor Data Pretraining for Autonomous Driving
Recent advancements in vision foundation models (VFMs) have revolutionized visual perception in 2D, yet their potential for 3D scene understanding, particularly in autonomous driving applications, remains underexplored. In this paper, we introduce LargeAD, a versatile and scalable framework designed for large-scale 3D pretraining across diverse real-world driving datasets. Our framework leverages VFMs to extract semantically rich superpixels from 2D images, which are aligned with LiDAR point clouds to generate high-quality contrastive samples. This alignment facilitates cross-modal representation learning, enhancing the semantic consistency between 2D and 3D data. We introduce several key innovations: (i) VFM-driven superpixel generation for detailed semantic representation, (ii) a VFM-assisted contrastive learning strategy to align multimodal features, (iii) superpoint temporal consistency to maintain stable representations across time, and (iv) multi-source data pretraining to generalize across various LiDAR configurations. Our approach achieves substantial gains over state-of-the-art methods in linear probing and fine-tuning for LiDAR-based segmentation and object detection. Extensive experiments on 11 large-scale multi-sensor datasets highlight our superior performance, demonstrating adaptability, efficiency, and robustness in real-world autonomous driving scenarios.
comment: IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
♻ ☆ 3D and 4D World Modeling: A Survey
World modeling has become a cornerstone in AI research, enabling agents to understand, represent, and predict the dynamic environments they inhabit. While prior work largely emphasizes generative methods for 2D image and video data, they overlook the rapidly growing body of work that leverages native 3D and 4D representations such as RGB-D imagery, occupancy grids, and LiDAR point clouds for large-scale scene modeling. At the same time, the absence of a standardized definition and taxonomy for ``world models'' has led to fragmented and sometimes inconsistent claims in the literature. This survey addresses these gaps by presenting the first comprehensive review explicitly dedicated to 3D and 4D world modeling and generation. We establish precise definitions, introduce a structured taxonomy spanning video-based (VideoGen), occupancy-based (OccGen), and LiDAR-based (LiDARGen) approaches, and systematically summarize datasets and evaluation metrics tailored to 3D/4D settings. We further discuss practical applications, identify open challenges, and highlight promising research directions, aiming to provide a coherent and foundational reference for advancing the field. A systematic summary of existing literature is available at https://github.com/worldbench/awesome-3d-4d-world-models
comment: Survey; 50 pages, 10 figures, 14 tables; GitHub Repo at https://github.com/worldbench/awesome-3d-4d-world-models
♻ ☆ Robust Deterministic Policy Gradient for Disturbance Attenuation and Its Application to Quadrotor Control
This paper presents a robust reinforcement learning algorithm called robust deterministic policy gradient (RDPG), which reformulates the H-infinity control problem as a two-player zero-sum dynamic game between a user and an adversary. The method combines deterministic policy gradients with deep reinforcement learning to train a robust policy that attenuates disturbances efficiently. A practical variant, robust deep deterministic policy gradient (RDDPG), integrates twin-delayed updates for stability and sample efficiency. Experiments on an unmanned aerial vehicle demonstrate superior robustness and tracking accuracy under severe disturbance conditions.
comment: 8 pages
♻ ☆ Diagnose, Correct, and Learn from Manipulation Failures via Visual Symbols
Vision-Language-Action (VLA) models have recently achieved remarkable progress in robotic manipulation, yet they remain limited in failure diagnosis and learning from failures. Additionally, existing failure datasets are mostly generated programmatically in simulation, which limits their generalization to the real world. In light of these, we introduce ViFailback, a framework designed to diagnose robotic manipulation failures and provide both textual and visual correction guidance. Our framework utilizes explicit visual symbols to enhance annotation efficiency. We further release the ViFailback dataset, a large-scale collection of 58,126 Visual Question Answering (VQA) pairs along with their corresponding 5,202 real-world manipulation trajectories. Based on the dataset, we establish ViFailback-Bench, a benchmark of 11 fine-grained VQA tasks designed to assess the failure diagnosis and correction abilities of Vision-Language Models (VLMs), featuring ViFailback-Bench Lite for closed-ended and ViFailback-Bench Hard for open-ended evaluation. To demonstrate the effectiveness of our framework, we built the ViFailback-8B VLM, which not only achieves significant overall performance improvement on ViFailback-Bench but also generates visual symbols for corrective action guidance. Finally, by integrating ViFailback-8B with a VLA model, we conduct real-world robotic experiments demonstrating its ability to assist the VLA model in recovering from failures. Project Website: https://x1nyuzhou.github.io/vifailback.github.io/
♻ ☆ FPC-VLA: A Vision-Language-Action Framework with a Supervisor for Failure Prediction and Correction
Robotic manipulation is a fundamental component of automation. However, traditional perception-planning pipelines often fall short in open-ended tasks due to limited flexibility, while the architecture of a single end-to-end Vision-Language-Action (VLA) offers promising capabilities but lacks crucial mechanisms for anticipating and recovering from failure. To address these challenges, we propose FPC-VLA, a dual-model framework that integrates VLA with a supervisor for failure prediction and correction. The supervisor evaluates action viability through vision-language queries and generates corrective strategies when risks arise, trained efficiently without manual labeling. A dual-stream fusion module further refines actions by leveraging past predictions. Evaluation results on multiple simulation platforms (SIMPLER and LIBERO) and robot embodiments (WidowX, Google Robot, Franka) show that FPC-VLA outperforms state-of-the-art models in both zero-shot and fine-tuned settings. Successful real-world deployments on diverse, long-horizon tasks confirm FPC-VLA's strong generalization and practical utility for building more reliable autonomous systems.
♻ ☆ Quaternion-Based Sliding Mode Control for Six Degrees of Freedom Flight Control of Quadrotors
Despite extensive research on sliding mode control (SMC) design for quadrotors, the existing approaches suffer from certain limitations. Euler angle-based SMC formulations suffer from poor performance in high-pitch or -roll maneuvers. Quaternion-based SMC approaches have unwinding issues and complex architecture. Coordinate-free methods are slow and only almost globally stable. This paper presents a new six degrees of freedom SMC flight controller to address the above limitations. We use a cascaded architecture with a position controller in the outer loop and a quaternion-based attitude controller in the inner loop. The position controller generates the desired trajectory for the attitude controller using a coordinate-free approach. The quaternion-based attitude controller uses the natural characteristics of the quaternion hypersphere, featuring a simple structure while providing global stability and avoiding unwinding issues. We compare our controller with three other common control methods conducting challenging maneuvers like flip-over and high-speed trajectory tracking in the presence of model uncertainties and disturbances. Our controller consistently outperforms the benchmark approaches with less control effort and actuator saturation, offering highly effective and efficient flight control.
♻ ☆ AugMapNet: Improving Spatial Latent Structure via BEV Grid Augmentation for Enhanced Vectorized Online HD Map Construction WACV 2026
Autonomous driving requires understanding infrastructure elements, such as lanes and crosswalks. To navigate safely, this understanding must be derived from sensor data in real-time and needs to be represented in vectorized form. Learned Bird's-Eye View (BEV) encoders are commonly used to combine a set of camera images from multiple views into one joint latent BEV grid. Traditionally, from this latent space, an intermediate raster map is predicted, providing dense spatial supervision but requiring post-processing into the desired vectorized form. More recent models directly derive infrastructure elements as polylines using vectorized map decoders, providing instance-level information. Our approach, Augmentation Map Network (AugMapNet), proposes latent BEV feature grid augmentation, a novel technique that significantly enhances the latent BEV representation. AugMapNet combines vector decoding and dense spatial supervision more effectively than existing architectures while remaining easy to integrate compared to other hybrid approaches. It additionally benefits from extra processing on its latent BEV features. Experiments on nuScenes and Argoverse2 datasets demonstrate significant improvements on vectorized map prediction of up to 13.3% over the StreamMapNet baseline on 60 m range and greater improvements on larger ranges. We confirm transferability by applying our method to another baseline, SQD-MapNet, and find similar improvements. A detailed analysis of the latent BEV grid confirms a more structured latent space of AugMapNet and shows the value of our novel concept beyond pure performance improvement. The code can be found at https://github.com/tmonnin/augmapnet
comment: Accepted to 2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2026)
♻ ☆ Efficient Preference-Based Reinforcement Learning: Randomized Exploration Meets Experimental Design
We study reinforcement learning from human feedback in general Markov decision processes, where agents learn from trajectory-level preference comparisons. A central challenge in this setting is to design algorithms that select informative preference queries to identify the underlying reward while ensuring theoretical guarantees. We propose a meta-algorithm based on randomized exploration, which avoids the computational challenges associated with optimistic approaches and remains tractable. We establish both regret and last-iterate guarantees under mild reinforcement learning oracle assumptions. To improve query complexity, we introduce and analyze an improved algorithm that collects batches of trajectory pairs and applies optimal experimental design to select informative comparison queries. The batch structure also enables parallelization of preference queries, which is relevant in practical deployment as feedback can be gathered concurrently. Empirical evaluation confirms that the proposed method is competitive with reward-based reinforcement learning while requiring a small number of preference queries.
Robotics 51
☆ Video2Act: A Dual-System Video Diffusion Policy with Robotic Spatio-Motional Modeling
Robust perception and dynamics modeling are fundamental to real-world robotic policy learning. Recent methods employ video diffusion models (VDMs) to enhance robotic policies, improving their understanding and modeling of the physical world. However, existing approaches overlook the coherent and physically consistent motion representations inherently encoded across frames in VDMs. To this end, we propose Video2Act, a framework that efficiently guides robotic action learning by explicitly integrating spatial and motion-aware representations. Building on the inherent representations of VDMs, we extract foreground boundaries and inter-frame motion variations while filtering out background noise and task-irrelevant biases. These refined representations are then used as additional conditioning inputs to a diffusion transformer (DiT) action head, enabling it to reason about what to manipulate and how to move. To mitigate inference inefficiency, we propose an asynchronous dual-system design, where the VDM functions as the slow System 2 and the DiT head as the fast System 1, working collaboratively to generate adaptive actions. By providing motion-aware conditions to System 1, Video2Act maintains stable manipulation even with low-frequency updates from the VDM. For evaluation, Video2Act surpasses previous state-of-the-art VLA methods by 7.7% in simulation and 21.7% in real-world tasks in terms of average success rate, further exhibiting strong generalization capabilities.
☆ SMP: Reusable Score-Matching Motion Priors for Physics-Based Character Control
Data-driven motion priors that can guide agents toward producing naturalistic behaviors play a pivotal role in creating life-like virtual characters. Adversarial imitation learning has been a highly effective method for learning motion priors from reference motion data. However, adversarial priors, with few exceptions, need to be retrained for each new controller, thereby limiting their reusability and necessitating the retention of the reference motion data when training on downstream tasks. In this work, we present Score-Matching Motion Priors (SMP), which leverages pre-trained motion diffusion models and score distillation sampling (SDS) to create reusable task-agnostic motion priors. SMPs can be pre-trained on a motion dataset, independent of any control policy or task. Once trained, SMPs can be kept frozen and reused as general-purpose reward functions to train policies to produce naturalistic behaviors for downstream tasks. We show that a general motion prior trained on large-scale datasets can be repurposed into a variety of style-specific priors. Furthermore SMP can compose different styles to synthesize new styles not present in the original dataset. Our method produces high-quality motion comparable to state-of-the-art adversarial imitation learning methods through reusable and modular motion priors. We demonstrate the effectiveness of SMP across a diverse suite of control tasks with physically simulated humanoid characters. Video demo available at https://youtu.be/ravlZJteS20
comment: 14 pages, 9 figures
☆ SurfFill: Completion of LiDAR Point Clouds via Gaussian Surfel Splatting
LiDAR-captured point clouds are often considered the gold standard in active 3D reconstruction. While their accuracy is exceptional in flat regions, the capturing is susceptible to miss small geometric structures and may fail with dark, absorbent materials. Alternatively, capturing multiple photos of the scene and applying 3D photogrammetry can infer these details as they often represent feature-rich regions. However, the accuracy of LiDAR for featureless regions is rarely reached. Therefore, we suggest combining the strengths of LiDAR and camera-based capture by introducing SurfFill: a Gaussian surfel-based LiDAR completion scheme. We analyze LiDAR capturings and attribute LiDAR beam divergence as a main factor for artifacts, manifesting mostly at thin structures and edges. We use this insight to introduce an ambiguity heuristic for completed scans by evaluating the change in density in the point cloud. This allows us to identify points close to missed areas, which we can then use to grow additional points from to complete the scan. For this point growing, we constrain Gaussian surfel reconstruction [Huang et al. 2024] to focus optimization and densification on these ambiguous areas. Finally, Gaussian primitives of the reconstruction in ambiguous areas are extracted and sampled for points to complete the point cloud. To address the challenges of large-scale reconstruction, we extend this pipeline with a divide-and-conquer scheme for building-sized point cloud completion. We evaluate on the task of LiDAR point cloud completion of synthetic and real-world scenes and find that our method outperforms previous reconstruction methods.
comment: Project page: https://lfranke.github.io/surffill
☆ U4D: Uncertainty-Aware 4D World Modeling from LiDAR Sequences
Modeling dynamic 3D environments from LiDAR sequences is central to building reliable 4D worlds for autonomous driving and embodied AI. Existing generative frameworks, however, often treat all spatial regions uniformly, overlooking the varying uncertainty across real-world scenes. This uniform generation leads to artifacts in complex or ambiguous regions, limiting realism and temporal stability. In this work, we present U4D, an uncertainty-aware framework for 4D LiDAR world modeling. Our approach first estimates spatial uncertainty maps from a pretrained segmentation model to localize semantically challenging regions. It then performs generation in a "hard-to-easy" manner through two sequential stages: (1) uncertainty-region modeling, which reconstructs high-entropy regions with fine geometric fidelity, and (2) uncertainty-conditioned completion, which synthesizes the remaining areas under learned structural priors. To further ensure temporal coherence, U4D incorporates a mixture of spatio-temporal (MoST) block that adaptively fuses spatial and temporal representations during diffusion. Extensive experiments show that U4D produces geometrically faithful and temporally consistent LiDAR sequences, advancing the reliability of 4D world modeling for autonomous perception and simulation.
comment: Preprint; 19 pages, 7 figures, 8 tables
☆ BEVDilation: LiDAR-Centric Multi-Modal Fusion for 3D Object Detection AAAI26
Integrating LiDAR and camera information in the bird's eye view (BEV) representation has demonstrated its effectiveness in 3D object detection. However, because of the fundamental disparity in geometric accuracy between these sensors, indiscriminate fusion in previous methods often leads to degraded performance. In this paper, we propose BEVDilation, a novel LiDAR-centric framework that prioritizes LiDAR information in the fusion. By formulating image BEV features as implicit guidance rather than naive concatenation, our strategy effectively alleviates the spatial misalignment caused by image depth estimation errors. Furthermore, the image guidance can effectively help the LiDAR-centric paradigm to address the sparsity and semantic limitations of point clouds. Specifically, we propose a Sparse Voxel Dilation Block that mitigates the inherent point sparsity by densifying foreground voxels through image priors. Moreover, we introduce a Semantic-Guided BEV Dilation Block to enhance the LiDAR feature diffusion processing with image semantic guidance and long-range context capture. On the challenging nuScenes benchmark, BEVDilation achieves better performance than state-of-the-art methods while maintaining competitive computational efficiency. Importantly, our LiDAR-centric strategy demonstrates greater robustness to depth noise compared to naive fusion. The source code is available at https://github.com/gwenzhang/BEVDilation.
comment: Accept by AAAI26
☆ Experimental Characterization of Fingertip Trajectory following for a 3-DoF Series-Parallel Hybrid Robotic Finger
Task-space control of robotic fingers is a critical enabler of dexterous manipulation, as manipulation objectives are most naturally specified in terms of fingertip motions and applied forces rather than individual joint angles. While task-space planning and control have been extensively studied for larger, arm-scale manipulators, demonstrations of precise task-space trajectory tracking in compact, multi-DoF robotic fingers remain scarce. In this paper, we present the physical prototyping and experimental characterization of a three-degree-of-freedom, linkage-driven, series-parallel robotic finger with analytic forward kinematics and a closed-form Jacobian. A resolved motion rate control (RMRC) scheme is implemented to achieve closed-loop task-space trajectory tracking. We experimentally evaluate the fingertip tracking performance across a variety of trajectories, including straight lines, circles, and more complex curves, and report millimeter-level accuracy. To the best of our knowledge, this work provides one of the first systematic experimental demonstrations of precise task-space trajectory tracking in a linkage-driven robotic finger, thereby establishing a benchmark for future designs aimed at dexterous in-hand manipulation.
☆ VLA Models Are More Generalizable Than You Think: Revisiting Physical and Spatial Modeling
Vision-language-action (VLA) models achieve strong in-distribution performance but degrade sharply under novel camera viewpoints and visual perturbations. We show that this brittleness primarily arises from misalignment in Spatial Modeling, rather than Physical Modeling. To address this, we propose a one-shot adaptation framework that recalibrates visual representations through lightweight, learnable updates. Our first method, Feature Token Modulation (FTM), applies a global affine transformation to visual tokens and improves Libero viewpoint accuracy from 48.5% to 87.1% with only 4K parameters. Building on this, Feature Linear Adaptation (FLA) introduces low-rank updates to the ViT encoder, achieving 90.8% success with 4.7M parameters -- matching LoRA-scale finetuning at far lower cost. Together, these results reveal substantial untapped robustness in pretrained VLA models and demonstrate that targeted, minimal visual adaptation is sufficient to restore viewpoint generalization.
☆ Polar Perspectives: Evaluating 2-D LiDAR Projections for Robust Place Recognition with Visual Foundation Models
This work presents a systematic investigation into how alternative LiDAR-to-image projections affect metric place recognition when coupled with a state-of-the-art vision foundation model. We introduce a modular retrieval pipeline that controls for backbone, aggregation, and evaluation protocol, thereby isolating the influence of the 2-D projection itself. Using consistent geometric and structural channels across multiple datasets and deployment scenarios, we identify the projection characteristics that most strongly determine discriminative power, robustness to environmental variation, and suitability for real-time autonomy. Experiments with different datasets, including integration into an operational place recognition policy, validate the practical relevance of these findings and demonstrate that carefully designed projections can serve as an effective surrogate for end-to-end 3-D learning in LiDAR place recognition.
comment: 13 Pages, 5 Figures, 2 Tables Under Review
☆ SwarmDiffusion: End-To-End Traversability-Guided Diffusion for Embodiment-Agnostic Navigation of Heterogeneous Robots
Visual traversability estimation is critical for autonomous navigation, but existing VLM-based methods rely on hand-crafted prompts, generalize poorly across embodiments, and output only traversability maps, leaving trajectory generation to slow external planners. We propose SwarmDiffusion, a lightweight end-to-end diffusion model that jointly predicts traversability and generates a feasible trajectory from a single RGB image. To remove the need for annotated or planner-produced paths, we introduce a planner-free trajectory construction pipeline based on randomized waypoint sampling, Bezier smoothing, and regularization enforcing connectivity, safety, directionality, and path thinness. This enables learning stable motion priors without demonstrations. SwarmDiffusion leverages VLM-derived supervision without prompt engineering and conditions the diffusion process on a compact embodiment state, producing physically consistent, traversable paths that transfer across different robot platforms. Across indoor environments and two embodiments (quadruped and aerial), the method achieves 80-100\% navigation success and 0.09 s inference, and adapts to a new robot using only-500 additional visual samples. It generalizes reliably to unseen environments in simulation and real-world trials, offering a scalable, prompt-free approach to unified traversability reasoning and trajectory generation.
comment: This work has been submitted for publication and is currently under review
☆ VLM as Strategist: Adaptive Generation of Safety-critical Testing Scenarios via Guided Diffusion
The safe deployment of autonomous driving systems (ADSs) relies on comprehensive testing and evaluation. However, safety-critical scenarios that can effectively expose system vulnerabilities are extremely sparse in the real world. Existing scenario generation methods face challenges in efficiently constructing long-tail scenarios that ensure fidelity, criticality, and interactivity, while particularly lacking real-time dynamic response capabilities to the vehicle under test (VUT). To address these challenges, this paper proposes a safety-critical testing scenario generation framework that integrates the high-level semantic understanding capabilities of Vision Language Models (VLMs) with the fine-grained generation capabilities of adaptive guided diffusion models. The framework establishes a three-layer hierarchical architecture comprising a strategic layer for VLM-directed scenario generation objective determination, a tactical layer for guidance function formulation, and an operational layer for guided diffusion execution. We first establish a high-quality fundamental diffusion model that learns the data distribution of real driving scenarios. Next, we design an adaptive guided diffusion method that enables real-time, precise control of background vehicles (BVs) in closed-loop simulation. The VLM is then incorporated to autonomously generate scenario generation objectives and guidance functions through deep scenario understanding and risk reasoning, ultimately guiding the diffusion model to achieve VLM-directed scenario generation. Experimental results demonstrate that the proposed method can efficiently generate realistic, diverse, and highly interactive safety-critical testing scenarios. Furthermore, case studies validate the adaptability and VLM-directed generation performance of the proposed method.
comment: 25 pages, 9 figures
☆ Steering Vision-Language-Action Models as Anti-Exploration: A Test-Time Scaling Approach
Vision-Language-Action (VLA) models, trained via flow-matching or diffusion objectives, excel at learning complex behaviors from large-scale, multi-modal datasets (e.g., human teleoperation, scripted policies). However, since VLAs incorporate diverse data modes in the pre-training stage, and the finetuning dataset often contains demonstration data collected in a kinematically suboptimal or undesirable way, it exists redundant action modes that are irrelevant to the success action modes of the downstream task. Specifically, we observe a critical inference-time fragility among various sampled noises after supervised finetuning of pre-trained VLAs. In this paper, we attribute this instability to the distribution shift between the VLA policy and the policy induced by stable success modes of the downstream task dataset. Thus, we propose \textbf{TACO}, a test-time-scaling (TTS) framework that applies a lightweight pseudo-count estimator as a high-fidelity verifier of action chunks. The VLA models integrated with TACO can execute the actions with maximum pseudo-count from all sampled action chunks, thereby preventing distribution shifts while preserving the generalization ability of VLAs since the constraint is applied only during inference. Our method resembles the classical anti-exploration principle in offline reinforcement learning (RL), and being gradient-free, it incurs significant computational benefits compared to RL update, especially for flow or diffusion-based VLAs which are difficult to perform RL update due to denoising process. Extensive experiments across four simulation benchmarks (RoboTwin2.0, Robotwin, LIBERO, SimplerEnv) and a dual-arm platform demonstrate that our method significantly improves the inference stability and success rates in downstream-task adaptations.
comment: The first two authors contributed equally. Yang Zhang leads the whole project
☆ Phase-Adaptive LLM Framework with Multi-Stage Validation for Construction Robot Task Allocation: A Systematic Benchmark Against Traditional Optimization Algorithms
Multi-robot task allocation in construction automation has traditionally relied on optimization methods such as Dynamic Programming and Reinforcement Learning. This research introduces the LangGraph-based Task Allocation Agent (LTAA), an LLM-driven framework that integrates phase-adaptive allocation strategies, multi-stage validation with hierarchical retries, and dynamic prompting for efficient robot coordination. Although recent LLM approaches show potential for construction robotics, they largely lack rigorous validation and benchmarking against established algorithms. This paper presents the first systematic comparison of LLM-based task allocation with traditional methods in construction scenarios.The study validates LLM feasibility through SMART-LLM replication and addresses implementation challenges using a Self-Corrective Agent Architecture. LTAA leverages natural-language reasoning combined with structured validation mechanisms, achieving major computational gains reducing token usage by 94.6% and allocation time by 86% through dynamic prompting. The framework adjusts its strategy across phases: emphasizing execution feasibility early and workload balance in later allocations.The authors evaluate LTAA against Dynamic Programming, Q-learning, and Deep Q-Network (DQN) baselines using construction operations from the TEACh human-robot collaboration dataset. In the Heavy Excels setting, where robots have strong task specializations, LTAA achieves 77% task completion with superior workload balance, outperforming all traditional methods. These findings show that LLM-based reasoning with structured validation can match established optimization algorithms while offering additional advantages such as interpretability, adaptability, and the ability to update task logic without retraining.
☆ Diagnose, Correct, and Learn from Manipulation Failures via Visual Symbols
Vision-Language-Action (VLA) models have recently achieved remarkable progress in robotic manipulation, yet they remain limited in failure diagnosis and learning from failures. Additionally, existing failure datasets are mostly generated programmatically in simulation, which limits their generalization to the real world. In light of these, we introduce ViFailback, a framework designed to diagnose robotic manipulation failures and provide both textual and visual correction guidance. Our framework utilizes explicit visual symbols to enhance annotation efficiency. We further release the ViFailback dataset, a large-scale collection of 58,126 Visual Question Answering (VQA) pairs along with their corresponding 5,202 real-world manipulation trajectories. Based on the dataset, we establish ViFailback-Bench, a benchmark of 11 fine-grained VQA tasks designed to assess the failure diagnosis and correction abilities of Vision-Language Models (VLMs), featuring ViFailback-Bench Lite for closed-ended and ViFailback-Bench Hard for open-ended evaluation. To demonstrate the effectiveness of our framework, we built the ViFailback-8B VLM, which not only achieves significant overall performance improvement on ViFailback-Bench but also generates visual symbols for corrective action guidance. Finally, by integrating ViFailback-8B with a VLA model, we conduct real-world robotic experiments demonstrating its ability to assist the VLA model in recovering from failures. Project Website: https://x1nyuzhou.github.io/vifailback.github.io/
☆ CogDrive: Cognition-Driven Multimodal Prediction-Planning Fusion for Safe Autonomy
Safe autonomous driving in mixed traffic requires a unified understanding of multimodal interactions and dynamic planning under uncertainty. Existing learning based approaches struggle to capture rare but safety critical behaviors, while rule based systems often lack adaptability in complex interactions. To address these limitations, CogDrive introduces a cognition driven multimodal prediction and planning framework that integrates explicit modal reasoning with safety aware trajectory optimization. The prediction module adopts cognitive representations of interaction modes based on topological motion semantics and nearest neighbor relational encoding. With a differentiable modal loss and multimodal Gaussian decoding, CogDrive learns sparse and unbalanced interaction behaviors and improves long horizon trajectory prediction. The planning module incorporates an emergency response concept and optimizes safety stabilized trajectories, where short term consistent branches ensure safety during replanning cycles and long term branches support smooth and collision free motion under low probability switching modes. Experiments on Argoverse2 and INTERACTION datasets show that CogDrive achieves strong performance in trajectory accuracy and miss rate, while closed loop simulations confirm adaptive behavior in merge and intersection scenarios. By combining cognitive multimodal prediction with safety oriented planning, CogDrive offers an interpretable and reliable paradigm for safe autonomy in complex traffic.
comment: 25 pages, 6 figures
☆ RoboWheel: A Data Engine from Real-World Human Demonstrations for Cross-Embodiment Robotic Learning
We introduce Robowheel, a data engine that converts human hand object interaction (HOI) videos into training-ready supervision for cross morphology robotic learning. From monocular RGB or RGB-D inputs, we perform high precision HOI reconstruction and enforce physical plausibility via a reinforcement learning (RL) optimizer that refines hand object relative poses under contact and penetration constraints. The reconstructed, contact rich trajectories are then retargeted to cross-embodiments, robot arms with simple end effectors, dexterous hands, and humanoids, yielding executable actions and rollouts. To scale coverage, we build a simulation-augmented framework on Isaac Sim with diverse domain randomization (embodiments, trajectories, object retrieval, background textures, hand motion mirroring), which enriches the distributions of trajectories and observations while preserving spatial relationships and physical plausibility. The entire data pipeline forms an end to end pipeline from video,reconstruction,retargeting,augmentation data acquisition. We validate the data on mainstream vision language action (VLA) and imitation learning architectures, demonstrating that trajectories produced by our pipeline are as stable as those from teleoperation and yield comparable continual performance gains. To our knowledge, this provides the first quantitative evidence that HOI modalities can serve as effective supervision for robotic learning. Compared with teleoperation, Robowheel is lightweight, a single monocular RGB(D) camera is sufficient to extract a universal, embodiment agnostic motion representation that could be flexibly retargeted across embodiments. We further assemble a large scale multimodal dataset combining multi-camera captures, monocular videos, and public HOI corpora for training and evaluating embodied models.
comment: 27 Pages, 21 figures
☆ SAM2Grasp: Resolve Multi-modal Grasping via Prompt-conditioned Temporal Action Prediction
Imitation learning for robotic grasping is often plagued by the multimodal problem: when a scene contains multiple valid targets, demonstrations of grasping different objects create conflicting training signals. Standard imitation learning policies fail by averaging these distinct actions into a single, invalid action. In this paper, we introduce SAM2Grasp, a novel framework that resolves this issue by reformulating the task as a uni-modal, prompt-conditioned prediction problem. Our method leverages the frozen SAM2 model to use its powerful visual temporal tracking capability and introduces a lightweight, trainable action head that operates in parallel with its native segmentation head. This design allows for training only the small action head on pre-computed temporal-visual features from SAM2. During inference, an initial prompt, such as a bounding box provided by an upstream object detection model, designates the specific object to be grasped. This prompt conditions the action head to predict a unique, unambiguous grasp trajectory for that object alone. In all subsequent video frames, SAM2's built-in temporal tracking capability automatically maintains stable tracking of the selected object, enabling our model to continuously predict the grasp trajectory from the video stream without further external guidance. This temporal-prompted approach effectively eliminates ambiguity from the visuomotor policy. We demonstrate through extensive experiments that SAM2Grasp achieves state-of-the-art performance in cluttered, multi-object grasping tasks.
☆ Reframing Human-Robot Interaction Through Extended Reality: Unlocking Safer, Smarter, and More Empathic Interactions with Virtual Robots and Foundation Models
This perspective reframes human-robot interaction (HRI) through extended reality (XR), arguing that virtual robots powered by large foundation models (FMs) can serve as cognitively grounded, empathic agents. Unlike physical robots, XR-native agents are unbound by hardware constraints and can be instantiated, adapted, and scaled on demand, while still affording embodiment and co-presence. We synthesize work across XR, HRI, and cognitive AI to show how such agents can support safety-critical scenarios, socially and cognitively empathic interaction across domains, and outreaching physical capabilities with XR and AI integration. We then discuss how multimodal large FMs (e.g., large language model, large vision model, and vision-language model) enable context-aware reasoning, affect-sensitive situations, and long-term adaptation, positioning virtual robots as cognitive and empathic mediators rather than mere simulation assets. At the same time, we highlight challenges and potential risks, including overtrust, cultural and representational bias, privacy concerns around biometric sensing, and data governance and transparency. The paper concludes by outlining a research agenda for human-centered, ethically grounded XR agents - emphasizing multi-layered evaluation frameworks, multi-user ecosystems, mixed virtual-physical embodiment, and societal and ethical design practices to envision XR-based virtual agents powered by FMs as reshaping future HRI into a more efficient and adaptive paradigm.
comment: This paper is under review
☆ Robotic capabilities framework: A boundary object and intermediate-level knowledge artifact for co-designing robotic processes
As robots become more adaptable, responsive, and capable of interacting with humans, the design of effective human-robot collaboration becomes critical. Yet, this design process is typically led by monodisciplinary approaches, often overlooking interdisciplinary knowledge and the experiential knowledge of workers who will ultimately share tasks with these systems. To address this gap, we introduce the robotic capabilities framework, a vocabulary that enables transdisciplinary collaborations to meaningfully shape the future of work when robotic systems are integrated into the workplace. Rather than focusing on the internal workings of robots, the framework centers discussion on high-level capabilities, supporting dialogue around which elements of a task should remain human-led and which can be delegated to robots. We developed the framework through reflexive and iterative processes, and applied it in two distinct settings: by engaging roboticists in describing existing commercial robots using its vocabulary, and through a design activity with students working on robotics-related projects. The framework emerges as an intermediate-level knowledge artifact and a boundary object that bridges technical and experiential domains, guiding designers, empowering workers, and contributing to more just and collaborative futures of work.
☆ AID: Agent Intent from Diffusion for Multi-Agent Informative Path Planning
Information gathering in large-scale or time-critical scenarios (e.g., environmental monitoring, search and rescue) requires broad coverage within limited time budgets, motivating the use of multi-agent systems. These scenarios are commonly formulated as multi-agent informative path planning (MAIPP), where multiple agents must coordinate to maximize information gain while operating under budget constraints. A central challenge in MAIPP is ensuring effective coordination while the belief over the environment evolves with incoming measurements. Recent learning-based approaches address this by using distributions over future positions as "intent" to support coordination. However, these autoregressive intent predictors are computationally expensive and prone to compounding errors. Inspired by the effectiveness of diffusion models as expressive, long-horizon policies, we propose AID, a fully decentralized MAIPP framework that leverages diffusion models to generate long-term trajectories in a non-autoregressive manner. AID first performs behavior cloning on trajectories produced by existing MAIPP planners and then fine-tunes the policy using reinforcement learning via Diffusion Policy Policy Optimization (DPPO). This two-stage pipeline enables the policy to inherit expert behavior while learning improved coordination through online reward feedback. Experiments demonstrate that AID consistently improves upon the MAIPP planners it is trained from, achieving up to 4x faster execution and 17% increased information gain, while scaling effectively to larger numbers of agents. Our implementation is publicly available at https://github.com/marmotlab/AID.
☆ nuScenes Revisited: Progress and Challenges in Autonomous Driving
Autonomous Vehicles (AV) and Advanced Driver Assistance Systems (ADAS) have been revolutionized by Deep Learning. As a data-driven approach, Deep Learning relies on vast amounts of driving data, typically labeled in great detail. As a result, datasets, alongside hardware and algorithms, are foundational building blocks for the development of AVs. In this work we revisit one of the most widely used autonomous driving datasets: the nuScenes dataset. nuScenes exemplifies key trends in AV development, being the first dataset to include radar data, to feature diverse urban driving scenes from two continents, and to be collected using a fully autonomous vehicle operating on public roads, while also promoting multi-modal sensor fusion, standardized benchmarks, and a broad range of tasks including perception, localization \& mapping, prediction and planning. We provide an unprecedented look into the creation of nuScenes, as well as its extensions nuImages and Panoptic nuScenes, summarizing many technical details that have hitherto not been revealed in academic publications. Furthermore, we trace how the influence of nuScenes impacted a large number of other datasets that were released later and how it defined numerous standards that are used by the community to this day. Finally, we present an overview of both official and unofficial tasks using the nuScenes dataset and review major methodological developments, thereby offering a comprehensive survey of the autonomous driving literature, with a particular focus on nuScenes.
comment: 18 pages, 17 figures
☆ Vehicle Dynamics Embedded World Models for Autonomous Driving
World models have gained significant attention as a promising approach for autonomous driving. By emulating human-like perception and decision-making processes, these models can predict and adapt to dynamic environments. Existing methods typically map high-dimensional observations into compact latent spaces and learn optimal policies within these latent representations. However, prior work usually jointly learns ego-vehicle dynamics and environmental transition dynamics from the image input, leading to inefficiencies and a lack of robustness to variations in vehicle dynamics. To address these issues, we propose the Vehicle Dynamics embedded Dreamer (VDD) method, which decouples the modeling of ego-vehicle dynamics from environmental transition dynamics. This separation allows the world model to generalize effectively across vehicles with diverse parameters. Additionally, we introduce two strategies to further enhance the robustness of the learned policy: Policy Adjustment during Deployment (PAD) and Policy Augmentation during Training (PAT). Comprehensive experiments in simulated environments demonstrate that the proposed model significantly improves both driving performance and robustness to variations in vehicle dynamics, outperforming existing approaches.
☆ On the Convergence of Density-Based Predictive Control for Multi-Agent Non-Uniform Area Coverage
This paper presents Density-based Predictive Control (DPC), a novel multi-agent control strategy for efficient non-uniform area coverage, grounded in optimal transport theory. In large-scale scenarios such as search and rescue or environmental monitoring, traditional uniform coverage fails to account for varying regional priorities. DPC leverages a pre-constructed reference distribution to allocate agents' coverage efforts, spending more time in high-priority or densely sampled regions. We analyze convergence conditions using the Wasserstein distance, derive an analytic optimal control law for unconstrained cases, and propose a numerical method for constrained scenarios. Simulations on first-order dynamics and linearized quadrotor models demonstrate that DPC achieves trajectories closely matching the non-uniform reference distribution, outperforming existing coverage methods.
comment: Accepted for publication in ASME JDSMC
☆ VIGS-SLAM: Visual Inertial Gaussian Splatting SLAM
We present VIGS-SLAM, a visual-inertial 3D Gaussian Splatting SLAM system that achieves robust real-time tracking and high-fidelity reconstruction. Although recent 3DGS-based SLAM methods achieve dense and photorealistic mapping, their purely visual design degrades under motion blur, low texture, and exposure variations. Our method tightly couples visual and inertial cues within a unified optimization framework, jointly refining camera poses, depths, and IMU states. It features robust IMU initialization, time-varying bias modeling, and loop closure with consistent Gaussian updates. Experiments on four challenging datasets demonstrate our superiority over state-of-the-art methods. Project page: https://vigs-slam.github.io
comment: Project page: https://vigs-slam.github.io
☆ KALIKO: Kalman-Implicit Koopman Operator Learning For Prediction of Nonlinear Dynamical Systems
Long-horizon dynamical prediction is fundamental in robotics and control, underpinning canonical methods like model predictive control. Yet, many systems and disturbance phenomena are difficult to model due to effects like nonlinearity, chaos, and high-dimensionality. Koopman theory addresses this by modeling the linear evolution of embeddings of the state under an infinite-dimensional linear operator that can be approximated with a suitable finite basis of embedding functions, effectively trading model nonlinearity for representational complexity. However, explicitly computing a good choice of basis is nontrivial, and poor choices may cause inaccurate forecasts or overfitting. To address this, we present Kalman-Implicit Koopman Operator (KALIKO) Learning, a method that leverages the Kalman filter to implicitly learn embeddings corresponding to latent dynamics without requiring an explicit encoder. KALIKO produces interpretable representations consistent with both theory and prior works, yielding high-quality reconstructions and inducing a globally linear latent dynamics. Evaluated on wave data generated by a high-dimensional PDE, KALIKO surpasses several baselines in open-loop prediction and in a demanding closed-loop simulated control task: stabilizing an underactuated manipulator's payload by predicting and compensating for strong wave disturbances.
☆ Flux4D: Flow-based Unsupervised 4D Reconstruction NeurIPS 2025
Reconstructing large-scale dynamic scenes from visual observations is a fundamental challenge in computer vision, with critical implications for robotics and autonomous systems. While recent differentiable rendering methods such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have achieved impressive photorealistic reconstruction, they suffer from scalability limitations and require annotations to decouple actor motion. Existing self-supervised methods attempt to eliminate explicit annotations by leveraging motion cues and geometric priors, yet they remain constrained by per-scene optimization and sensitivity to hyperparameter tuning. In this paper, we introduce Flux4D, a simple and scalable framework for 4D reconstruction of large-scale dynamic scenes. Flux4D directly predicts 3D Gaussians and their motion dynamics to reconstruct sensor observations in a fully unsupervised manner. By adopting only photometric losses and enforcing an "as static as possible" regularization, Flux4D learns to decompose dynamic elements directly from raw data without requiring pre-trained supervised models or foundational priors simply by training across many scenes. Our approach enables efficient reconstruction of dynamic scenes within seconds, scales effectively to large datasets, and generalizes well to unseen environments, including rare and unknown objects. Experiments on outdoor driving datasets show Flux4D significantly outperforms existing methods in scalability, generalization, and reconstruction quality.
comment: NeurIPS 2025. Project page: https://waabi.ai/flux4d/
☆ 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 (LRR) 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.
☆ Multi-Agent Reinforcement Learning and Real-Time Decision-Making in Robotic Soccer for Virtual Environments
The deployment of multi-agent systems in dynamic, adversarial environments like robotic soccer necessitates real-time decision-making, sophisticated cooperation, and scalable algorithms to avoid the curse of dimensionality. While Reinforcement Learning (RL) offers a promising framework, existing methods often struggle with the multi-granularity of tasks (long-term strategy vs. instant actions) and the complexity of large-scale agent interactions. This paper presents a unified Multi-Agent Reinforcement Learning (MARL) framework that addresses these challenges. First, we establish a baseline using Proximal Policy Optimization (PPO) within a client-server architecture for real-time action scheduling, with PPO demonstrating superior performance (4.32 avg. goals, 82.9% ball control). Second, we introduce a Hierarchical RL (HRL) structure based on the options framework to decompose the problem into a high-level trajectory planning layer (modeled as a Semi-Markov Decision Process) and a low-level action execution layer, improving global strategy (avg. goals increased to 5.26). Finally, to ensure scalability, we integrate mean-field theory into the HRL framework, simplifying many-agent interactions into a single agent vs. the population average. Our mean-field actor-critic method achieves a significant performance boost (5.93 avg. goals, 89.1% ball control, 92.3% passing accuracy) and enhanced training stability. Extensive simulations of 4v4 matches in the Webots environment validate our approach, demonstrating its potential for robust, scalable, and cooperative behavior in complex multi-agent domains.
♻ ☆ Forecasting in Offline Reinforcement Learning for Non-stationary Environments NeurIPS 2025
Offline Reinforcement Learning (RL) provides a promising avenue for training policies from pre-collected datasets when gathering additional interaction data is infeasible. However, existing offline RL methods often assume stationarity or only consider synthetic perturbations at test time, assumptions that often fail in real-world scenarios characterized by abrupt, time-varying offsets. These offsets can lead to partial observability, causing agents to misperceive their true state and degrade performance. To overcome this challenge, we introduce Forecasting in Non-stationary Offline RL (FORL), a framework that unifies (i) conditional diffusion-based candidate state generation, trained without presupposing any specific pattern of future non-stationarity, and (ii) zero-shot time-series foundation models. FORL targets environments prone to unexpected, potentially non-Markovian offsets, requiring robust agent performance from the onset of each episode. Empirical evaluations on offline RL benchmarks, augmented with real-world time-series data to simulate realistic non-stationarity, demonstrate that FORL consistently improves performance compared to competitive baselines. By integrating zero-shot forecasting with the agent's experience, we aim to bridge the gap between offline RL and the complexities of real-world, non-stationary environments.
comment: The Thirty-Ninth Annual Conference on Neural Information Processing Systems, NeurIPS 2025
♻ ☆ Guardian: Detecting Robotic Planning and Execution Errors with Vision-Language Models
Robust robotic manipulation requires reliable failure detection and recovery. Although current Vision-Language Models (VLMs) show promise, their accuracy and generalization are limited by the scarcity of failure data. To address this data gap, we propose an automatic robot failure synthesis approach that procedurally perturbs successful trajectories to generate diverse planning and execution failures. This method produces not only binary classification labels but also fine-grained failure categories and step-by-step reasoning traces in both simulation and the real world. With it, we construct three new failure detection benchmarks: RLBench-Fail, BridgeDataV2-Fail, and UR5-Fail, substantially expanding the diversity and scale of existing failure datasets. We then train Guardian, a VLM with multi-view images for detailed failure reasoning and detection. Guardian achieves state-of-the-art performance on both existing and newly introduced benchmarks. It also effectively improves task success rates when integrated into a state-of-the-art manipulation system in simulation and real robots, demonstrating the impact of our generated failure data. Code, Data, and Models available at https://www.di.ens.fr/willow/research/guardian/.
comment: Code, Data, and Models available at https://www.di.ens.fr/willow/research/guardian/. The paper contains 8 pages, 9 figures, 6 tables
♻ ☆ DYNEMO-SLAM: Dynamic Entity and Motion-Aware 3D Scene Graph SLAM
Robots operating in dynamic environments face significant challenges due to the presence of moving agents and displaced objects. Traditional SLAM systems typically assume a static world or treat dynamic as outliers, discarding their information to preserve map consistency. As a result, they cannot exploit dynamic entities as persistent landmarks, do not model and exploit their motion over time, and therefore quickly degrade in highly cluttered environments with few reliable static features. This paper presents a novel 3D scene graph-based SLAM framework that addresses the challenge of modeling and estimating the pose of dynamic entities into the SLAM backend. Our framework incorporates semantic motion priors and dynamic entity-aware constraints to jointly optimize the robot trajectory, dynamic entity poses, and the surrounding environment structure within a unified graph formulation. In parallel, a dynamic keyframe selection policy and a semantic loop-closure prefiltering step enable the system to remain robust and effective in highly dynamic environments by continuously adapting to scene changes and filtering inconsistent observations. The simulation and real-world experimental results show a 49.97% reduction in ATE compared to the baseline method employed, demonstrating the effectiveness of incorporating dynamic entities and estimating their poses for improved robustness and richer scene representation in complex scenarios while maintaining real-time performance.
comment: 8 pages, 4 figures, 5 tables
♻ ☆ Is Image-based Object Pose Estimation Ready to Support Grasping?
We present a framework for evaluating 6-DoF instance-level object pose estimators, focusing on those that require a single RGB (not RGB-D) image as input. Besides gaining intuition about how accurate these estimators are, we are interested in the degree to which they can serve as the sole perception mechanism for robotic grasping. To assess this, we perform grasping trials in a physics-based simulator, using image-based pose estimates to guide a parallel gripper and an underactuated robotic hand in picking up 3D models of objects. Our experiments on a subset of the BOP (Benchmark for 6D Object Pose Estimation) dataset compare five open-source object pose estimators and provide insights that were missing from the literature.
♻ ☆ GR-RL: Going Dexterous and Precise for Long-Horizon Robotic Manipulation
We present GR-RL, a robotic learning framework that turns a generalist vision-language-action (VLA) policy into a highly capable specialist for long-horizon dexterous manipulation. Assuming the optimality of human demonstrations is core to existing VLA policies. However, we claim that in highly dexterous and precise manipulation tasks, human demonstrations are noisy and suboptimal. GR-RL proposes a multi-stage training pipeline that filters, augments, and reinforces the demonstrations by reinforcement learning. First, GR-RL learns a vision-language-conditioned task progress, filters the demonstration trajectories, and only keeps the transitions that contribute positively to the progress. Specifically, we show that by directly applying offline RL with sparse reward, the resulting $Q$-values can be treated as a robust progress function. Next, we introduce morphological symmetry augmentation that greatly improves the generalization and performance of GR-RL. Lastly, to better align the VLA policy with its deployment behaviors for high-precision control, we perform online RL by learning a latent space noise predictor. With this pipeline, GR-RL is, to our knowledge, the first learning-based policy that can autonomously lace up a shoe by threading shoelaces through multiple eyelets with an 83.3% success rate, a task requiring long-horizon reasoning, millimeter-level precision, and compliant soft-body interaction. We hope GR-RL provides a step toward enabling generalist robot foundations models to specialize into reliable real-world experts.
♻ ☆ 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 $γ$.
♻ ☆ LAP: Fast LAtent Diffusion Planner with Fine-Grained Feature Distillation for Autonomous Driving
Diffusion models have demonstrated strong capabilities for modeling human-like driving behaviors in autonomous driving, but their iterative sampling process induces substantial latency, and operating directly on raw trajectory points forces the model to spend capacity on low-level kinematics, rather than high-level multi-modal semantics. To address these limitations, we propose LAtent Planner (LAP), a framework that plans in a VAE-learned latent space that disentangles high-level intents from low-level kinematics, enabling our planner to capture rich, multi-modal driving strategies. We further introduce a fine-grained feature distillation mechanism to guide a better interaction and fusion between the high-level semantic planning space and the vectorized scene context. Notably, LAP can produce high-quality plans in one single denoising step, substantially reducing computational overhead. Through extensive evaluations on the large-scale nuPlan benchmark, LAP achieves state-of-the-art closed-loop performance among learning-based planning methods, while demonstrating an inference speed-up of at most 10 times over previous SOTA approaches. Code will be released at: https://github.com/jhz1192/Latent-Planner.
♻ ☆ AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention
Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in embodied AI tasks. However, existing VLA models, often built upon Vision-Language Models (VLMs), typically process dense visual inputs independently at each timestep. This approach implicitly models the task as a Markov Decision Process (MDP). However, this history-agnostic design is suboptimal for effective visual token processing in dynamic sequential decision-making, as it fails to leverage the context of history. To address this limitation, we reformulate the problem from a Partially Observable Markov Decision Process (POMDP) perspective and propose a novel framework named AVA-VLA. Inspired by the POMDP that the action generation should be conditioned on the belief state. AVA-VLA introduces Active Visual Attention (AVA) to dynamically modulate visual processing. It achieves this by leveraging the recurrent state, which is a neural approximation of the agent's belief state derived from the previous decision step. Specifically, the AVA module uses the recurrent state to compute the soft weights to actively process task-relevant visual tokens based on its historical context. Comprehensive evaluations demonstrate that AVA-VLA achieves state-of-the-art performance across popular robotic benchmarks, including LIBERO and CALVIN. Furthermore, real-world deployments on a dual-arm robot platform validate the framework's practical applicability and robust sim-to-real transferability.
comment: 18 pages, 10 figures
♻ ☆ Multi-User Personalisation in Human-Robot Interaction: Resolving Preference Conflicts Using Gradual Argumentation
While personalisation in Human-Robot Interaction (HRI) has advanced significantly, most existing approaches focus on single-user adaptation, overlooking scenarios involving multiple stakeholders with potentially conflicting preferences. To address this, we propose the Multi-User Preferences Quantitative Bipolar Argumentation Framework (MUP-QBAF), a novel multi-user personalisation framework based on Quantitative Bipolar Argumentation Frameworks (QBAFs) that explicitly models and resolves multi-user preference conflicts. Unlike prior work in Argumentation Frameworks, which typically assumes static inputs, our approach is tailored to robotics: it incorporates both users' arguments and the robot's dynamic observations of the environment, allowing the system to adapt over time and respond to changing contexts. Preferences, both positive and negative, are represented as arguments whose strength is recalculated iteratively based on new information. The framework's properties and capabilities are presented and validated through a realistic case study, where an assistive robot mediates between the conflicting preferences of a caregiver and a care recipient during a frailty assessment task. This evaluation further includes a sensitivity analysis of argument base scores, demonstrating how preference outcomes can be shaped by user input and contextual observations. By offering a transparent, structured, and context-sensitive approach to resolving competing user preferences, this work advances the field of multi-user HRI. It provides a principled alternative to data-driven methods, enabling robots to navigate conflicts in real-world environments.
comment: Preprint submitted to a journal
♻ ☆ Agentic UAVs: LLM-Driven Autonomy with Integrated Tool-Calling and Cognitive Reasoning
Unmanned Aerial Vehicles (UAVs) are increasingly used in defense, surveillance, and disaster response, yet most systems still operate at SAE Level 2 to 3 autonomy. Their dependence on rule-based control and narrow AI limits adaptability in dynamic and uncertain missions. Current UAV architectures lack context-aware reasoning, autonomous decision-making, and integration with external systems. Importantly, none make use of Large Language Model (LLM) agents with tool-calling for real-time knowledge access. This paper introduces the Agentic UAVs framework, a five-layer architecture consisting of Perception, Reasoning, Action, Integration, and Learning. The framework enhances UAV autonomy through LLM-driven reasoning, database querying, and interaction with third-party systems. A prototype built with ROS 2 and Gazebo combines YOLOv11 for object detection with GPT-4 for reasoning and a locally deployed Gemma 3 model. In simulated search-and-rescue scenarios, agentic UAVs achieved higher detection confidence (0.79 compared to 0.72), improved person detection rates (91% compared to 75%), and a major increase in correct action recommendations (92% compared to 4.5%). These results show that modest computational overhead can enable significantly higher levels of autonomy and system-level integration.
comment: 17 pages, 2 figure
♻ ☆ Large Language Models for Robotics: A Survey
The human ability to learn, generalize, and control complex manipulation tasks through multi-modality feedback suggests a unique capability, which we refer to as dexterity intelligence. Understanding and assessing this intelligence is a complex task. Amidst the swift progress and extensive proliferation of large language models (LLMs), their applications in the field of robotics have garnered increasing attention. LLMs possess the ability to process and generate natural language, facilitating efficient interaction and collaboration with robots. Researchers and engineers in the field of robotics have recognized the immense potential of LLMs in enhancing robot intelligence, human-robot interaction, and autonomy. Therefore, this comprehensive review aims to summarize the applications of LLMs in robotics, delving into their impact and contributions to key areas such as robot control, perception, decision-making, and planning. This survey first provides an overview of the background and development of LLMs for robotics, followed by a discussion of their benefits and recent advancements in LLM-based robotic models. It then explores various techniques, employed in perception, decision-making, control, and interaction, as well as cross-module coordination in practical tasks. Finally, we review current applications of LLMs in robotics and outline potential challenges they may face in the near future. Embodied intelligence represents the future of intelligent systems, and LLM-based robotics is one of the most promising yet challenging paths toward achieving it.
comment: Preprint. 9 figures, 7 tables
♻ ☆ Learning-based 3D Reconstruction in Autonomous Driving: A Comprehensive Survey
Learning-based 3D reconstruction has emerged as a transformative technique in autonomous driving, enabling precise modeling of environments through advanced neural representations. It has inspired pioneering solutions for vital tasks in autonomous driving, such as dense mapping and closed-loop simulation, as well as comprehensive scene feature for driving scene understanding and reasoning. Given the rapid growth in related research, this survey provides a comprehensive review of both technical evolutions and practical applications in autonomous driving. We begin with an introduction to the preliminaries of learning-based 3D reconstruction to provide a solid technical background foundation, then progress to a rigorous, multi-dimensional examination of cutting-edge methodologies, systematically organized according to the distinctive technical requirements and fundamental challenges of autonomous driving. Through analyzing and summarizing development trends and cutting-edge research, we identify existing technical challenges, along with insufficient disclosure of on-board validation and safety verification details in the current literature, and ultimately suggest potential directions to guide future studies.
comment: Published in IEEE Trans. on Intelligent Transportation Systems
♻ ☆ ST-Booster: An Iterative SpatioTemporal Perception Booster for Vision-and-Language Navigation in Continuous Environments
Vision-and-Language Navigation in Continuous Environments (VLN-CE) requires agents to navigate unknown, continuous spaces based on natural language instructions. Compared to discrete settings, VLN-CE poses two core perception challenges. First, the absence of predefined observation points leads to heterogeneous visual memories and weakened global spatial correlations. Second, cumulative reconstruction errors in three-dimensional scenes introduce structural noise, impairing local feature perception. To address these challenges, this paper proposes ST-Booster, an iterative spatiotemporal booster that enhances navigation performance through multi-granularity perception and instruction-aware reasoning. ST-Booster consists of three key modules -- Hierarchical SpatioTemporal Encoding (HSTE), Multi-Granularity Aligned Fusion (MGAF), and ValueGuided Waypoint Generation (VGWG). HSTE encodes long-term global memory using topological graphs and captures shortterm local details via grid maps. MGAF aligns these dualmap representations with instructions through geometry-aware knowledge fusion. The resulting representations are iteratively refined through pretraining tasks. During reasoning, VGWG generates Guided Attention Heatmaps (GAHs) to explicitly model environment-instruction relevance and optimize waypoint selection. Extensive comparative experiments and performance analyses are conducted, demonstrating that ST-Booster outperforms existing state-of-the-art methods, particularly in complex, disturbance-prone environments.
comment: 11 pages, 7 figures
♻ ☆ ViTaMIn-B: A Reliable and Efficient Visuo-Tactile Bimanual Manipulation Interface
Handheld devices have opened up unprecedented opportunities to collect large-scale, high-quality demonstrations efficiently. However, existing systems often lack robust tactile sensing or reliable pose tracking to handle complex interaction scenarios, especially for bimanual and contact-rich tasks. In this work, we propose ViTaMIn-B, a more capable and efficient handheld data collection system for such tasks. We first design DuoTact, a novel compliant visuo-tactile sensor built with a flexible frame to withstand large contact forces during manipulation while capturing high-resolution contact geometry. To enhance the cross-sensor generalizability, we propose reconstructing the sensor's global deformation as a 3D point cloud and using it as the policy input. We further develop a robust, unified 6-DoF bimanual pose acquisition process using Meta Quest controllers, which eliminates the trajectory drift issue in common SLAM-based methods. Comprehensive user studies confirm the efficiency and high usability of ViTaMIn-B among novice and expert operators. Furthermore, experiments on four bimanual manipulation tasks demonstrate its superior task performance relative to existing systems. Project page: https://chuanyune.github.io/ViTaMIn-B_page/
comment: Project page: https://chuanyune.github.io/ViTaMIn-B_page/
♻ ☆ SPARK: Sim-ready Part-level Articulated Reconstruction with VLM Knowledge SP
Articulated 3D objects are critical for embodied AI, robotics, and interactive scene understanding, yet creating simulation-ready assets remains labor-intensive and requires expert modeling of part hierarchies and motion structures. We introduce SPARK, a framework for reconstructing physically consistent, kinematic part-level articulated objects from a single RGB image. Given an input image, we first leverage VLMs to extract coarse URDF parameters and generate part-level reference images. We then integrate the part-image guidance and the inferred structure graph into a generative diffusion transformer to synthesize consistent part and complete shapes of articulated objects. To further refine the URDF parameters, we incorporate differentiable forward kinematics and differentiable rendering to optimize joint types, axes, and origins under VLM-generated open-state supervision. Extensive experiments show that SPARK produces high-quality, simulation-ready articulated assets across diverse categories, enabling downstream applications such as robotic manipulation and interaction modeling. Project page: https://heyumeng.com/SPARK/index.html.
comment: Project page: https://heyumeng.com/SPARK/index.html. 17 pages, 7 figures
♻ ☆ EMMA: Scaling Mobile Manipulation via Egocentric Human Data
Scaling mobile manipulation imitation learning is bottlenecked by expensive mobile robot teleoperation. We present Egocentric Mobile MAnipulation (EMMA), an end-to-end framework training mobile manipulation policies from human mobile manipulation data with static robot data, sidestepping mobile teleoperation. To accomplish this, we co-train human full-body motion data with static robot data. In our experiments across three real-world tasks, EMMA demonstrates comparable performance to baselines trained on teleoperated mobile robot data (Mobile ALOHA), achieving higher or equivalent task performance in full task success. We find that EMMA is able to generalize to new spatial configurations and scenes, and we observe positive performance scaling as we increase the hours of human data, opening new avenues for scalable robotic learning in real-world environments. Details of this project can be found at https://ego-moma.github.io/.
♻ ☆ Predicting Human Perceptions of Robot Performance During Navigation Tasks
Understanding human perceptions of robot performance is crucial for designing socially intelligent robots that can adapt to human expectations. Current approaches often rely on surveys, which can disrupt ongoing human-robot interactions. As an alternative, we explore predicting people's perceptions of robot performance using non-verbal behavioral cues and machine learning techniques. We contribute the SEAN TOGETHER Dataset consisting of observations of an interaction between a person and a mobile robot in Virtual Reality, together with perceptions of robot performance provided by users on a 5-point scale. We then analyze how well humans and supervised learning techniques can predict perceived robot performance based on different observation types (like facial expression and spatial behavior features). Our results suggest that facial expressions alone provide useful information, but in the navigation scenarios that we considered, reasoning about spatial features in context is critical for the prediction task. Also, supervised learning techniques outperformed humans' predictions in most cases. Further, when predicting robot performance as a binary classification task on unseen users' data, the F1-Score of machine learning models more than doubled that of predictions on a 5-point scale. This suggested good generalization capabilities, particularly in identifying performance directionality over exact ratings. Based on these findings, we conducted a real-world demonstration where a mobile robot uses a machine learning model to predict how a human who follows it perceives it. Finally, we discuss the implications of our results for implementing these supervised learning models in real-world navigation. Our work paves the path to automatically enhancing robot behavior based on observations of users and inferences about their perceptions of a robot.
♻ ☆ MIMIC-MJX: Neuromechanical Emulation of Animal Behavior
The primary output of the nervous system is movement and behavior. While recent advances have democratized pose tracking during complex behavior, kinematic trajectories alone provide only indirect access to the underlying control processes. Here we present MIMIC-MJX, a framework for learning biologically-plausible neural control policies from kinematics. MIMIC-MJX models the generative process of motor control by training neural controllers that learn to actuate biomechanically-realistic body models in physics simulation to reproduce real kinematic trajectories. We demonstrate that our implementation is accurate, fast, data-efficient, and generalizable to diverse animal body models. Policies trained with MIMIC-MJX can be utilized to both analyze neural control strategies and simulate behavioral experiments, illustrating its potential as an integrative modeling framework for neuroscience.
comment: Corrected LaTeX issues. Project page available at https://mimic-mjx.talmolab.org
♻ ☆ Sigma: The Key for Vision-Language-Action Models toward Telepathic Alignment
To address the gap in humanoid robot cognitive systems regarding the lack of a time-updable mediating thought space between semantics and continuous control, this study constructs and trains a VLA model named "Sigma" that runs on a single RTX 4090. It uses the open-source pi05_base model as a foundation and preprocesses svla_so101_pickplace into a training dataset. The researcher independently designed an architecture for a vision-language-action model that combines deep semantic understanding and association to achieve telepathic communication. The training process involved repeated optimizations of data preprocessing, LoRA fine-tuning, and the inference-stage adapter. The experiment employed offline closed-loop replay, comparing Sigma with the untuned pure pi05_base model under data conditions. Results showed that Sigma exhibited a stable decrease in control MSE across vector, fragment, and entire trajectory timescales, while maintaining the telepathy norm and semantic-text alignment quality unchanged. It demonstrates that mind-responsive alignment control is quantified through an architecture that combines deep understanding of semantics and association without retraining the base model, which provides reproducible experience for semantic alignment and intention-driven behavior in humanoid robots.
comment: The Sigma model has been open-sourced on Hugging Face. Weights, dataset, some scripts, and logs are all available. The link is: https://huggingface.co/Veltraxor/Sigma
♻ ☆ Learning Massively Multitask World Models for Continuous Control
General-purpose control demands agents that act across many tasks and embodiments, yet research on reinforcement learning (RL) for continuous control remains dominated by single-task or offline regimes, reinforcing a view that online RL does not scale. Inspired by the foundation model recipe (large-scale pretraining followed by light RL) we ask whether a single agent can be trained on hundreds of tasks with online interaction. To accelerate research in this direction, we introduce a new benchmark with 200 diverse tasks spanning many domains and embodiments, each with language instructions, demonstrations, and optionally image observations. We then present \emph{Newt}, a language-conditioned multitask world model that is first pretrained on demonstrations to acquire task-aware representations and action priors, and then jointly optimized with online interaction across all tasks. Experiments show that Newt yields better multitask performance and data-efficiency than a set of strong baselines, exhibits strong open-loop control, and enables rapid adaptation to unseen tasks. We release our environments, demonstrations, code for training and evaluation, as well as 200+ checkpoints.
comment: Webpage: https://www.nicklashansen.com/NewtWM
♻ ☆ Image-Based Relocalization and Alignment for Long-Term Monitoring of Dynamic Underwater Environments
Effective monitoring of underwater ecosystems is crucial for tracking environmental changes, guiding conservation efforts, and ensuring long-term ecosystem health. However, automating underwater ecosystem management with robotic platforms remains challenging due to the complexities of underwater imagery, which pose significant difficulties for traditional visual localization methods. We propose an integrated pipeline that combines Visual Place Recognition (VPR), feature matching, and image segmentation on video-derived images. This method enables robust identification of revisited areas, estimation of rigid transformations, and downstream analysis of ecosystem changes. Furthermore, we introduce the SQUIDLE+ VPR Benchmark-the first large-scale underwater VPR benchmark designed to leverage an extensive collection of unstructured data from multiple robotic platforms, spanning time intervals from days to years. The dataset encompasses diverse trajectories, arbitrary overlap and diverse seafloor types captured under varying environmental conditions, including differences in depth, lighting, and turbidity. Our code is available at: https://github.com/bev-gorry/underloc
♻ ☆ LiDARCrafter: Dynamic 4D World Modeling from LiDAR Sequences AAAI 2026
Generative world models have become essential data engines for autonomous driving, yet most existing efforts focus on videos or occupancy grids, overlooking the unique LiDAR properties. Extending LiDAR generation to dynamic 4D world modeling presents challenges in controllability, temporal coherence, and evaluation standardization. To this end, we present LiDARCrafter, a unified framework for 4D LiDAR generation and editing. Given free-form natural language inputs, we parse instructions into ego-centric scene graphs, which condition a tri-branch diffusion network to generate object structures, motion trajectories, and geometry. These structured conditions enable diverse and fine-grained scene editing. Additionally, an autoregressive module generates temporally coherent 4D LiDAR sequences with smooth transitions. To support standardized evaluation, we establish a comprehensive benchmark with diverse metrics spanning scene-, object-, and sequence-level aspects. Experiments on the nuScenes dataset using this benchmark demonstrate that LiDARCrafter achieves state-of-the-art performance in fidelity, controllability, and temporal consistency across all levels, paving the way for data augmentation and simulation. The code and benchmark are released to the community.
comment: AAAI 2026 Oral Presentation; 38 pages, 18 figures, 12 tables; Project Page at https://lidarcrafter.github.io
♻ ☆ Are you a robot? Detecting Autonomous Vehicles from Behavior Analysis
The tremendous hype around autonomous driving is eagerly calling for emerging and novel technologies to support advanced mobility use cases. As car manufactures keep developing SAE level 3+ systems to improve the safety and comfort of passengers, traffic authorities need to establish new procedures to manage the transition from human-driven to fully-autonomous vehicles while providing a feedback-loop mechanism to fine-tune envisioned autonomous systems. Thus, a way to automatically profile autonomous vehicles and differentiate those from human-driven ones is a must. In this paper, we present a fully-fledged framework that monitors active vehicles using camera images and state information in order to determine whether vehicles are autonomous, without requiring any active notification from the vehicles themselves. Essentially, it builds on the cooperation among vehicles, which share their data acquired on the road feeding a machine learning model to identify autonomous cars. We extensively tested our solution and created the NexusStreet dataset, by means of the CARLA simulator, employing an autonomous driving control agent and a steering wheel maneuvered by licensed drivers. Experiments show it is possible to discriminate the two behaviors by analyzing video clips with an accuracy of 80%, which improves up to 93% when the target state information is available. Lastly, we deliberately degraded the state to observe how the framework performs under non-ideal data collection conditions.
♻ ☆ Magnetic Tactile-Driven Soft Actuator for Intelligent Grasping and Firmness Evaluation
Soft robots are powerful tools for manipulating delicate objects, yet their adoption is hindered by two gaps: the lack of integrated tactile sensing and sensor signal distortion caused by actuator deformations. This paper addresses these challenges by introducing the SoftMag actuator: a magnetic tactile-sensorized soft actuator. Unlike systems relying on attached sensors or treating sensing and actuation separately, SoftMag unifies them through a shared architecture while confronting the mechanical parasitic effect, where deformations corrupt tactile signals. A multiphysics simulation framework models this coupling, and a neural-network-based decoupling strategy removes the parasitic component, restoring sensing fidelity. Experiments including indentation, quasi-static and step actuation, and fatigue tests validate the actuator's performance and decoupling effectiveness. Building upon this foundation, the system is extended into a two-finger SoftMag gripper, where a multi-task neural network enables real-time prediction of tri-axial contact forces and position. Furthermore, a probing-based strategy estimates object firmness during grasping. Validation on apricots shows a strong correlation (Pearson r over 0.8) between gripper-estimated firmness and reference measurements, confirming the system's capability for non-destructive quality assessment. Results demonstrate that combining integrated magnetic sensing, learning-based correction, and real-time inference enables a soft robotic platform that adapts its grasp and quantifies material properties. The framework offers an approach for advancing sensorized soft actuators toward intelligent, material-aware robotics.
comment: 25 pages, 24 figures