News & Research

arXiv:2608.00946v2 Announce Type: replace Abstract: Existing 6-DoF grasp detectors typically rank grasp candidates by detector confidence. However, our analysis on GraspNet-1Billion shows that detector confidence is often poorly aligned with grasp quality, leaving successful grasp candidates at low ranks. Motivated by this observation, we study whether learned re-ranking can improve candidate ordering while keeping detector parameters and grasp candidates unchanged. We propose GraRe, which estimates grasp quality from candidate attributes, shell-stratified local geometry, and object context. Candidate attributes condition the local geometric and object-context representations, and a Transformer fuses all three feature types. The predicted quality is combined with detector confidence to produce the final ranking. Experiments on GraspNet-1Billion with five frozen detectors show consistent improvements, with gains of up to 13.56 points in Average AP. Real-robot experiments further demonst

arXiv:2606.13279v2 Announce Type: replace Abstract: In bimanual robotic manipulation, task-relevant visual information varies with the task stage and context, while the interaction of the two arms shifts between independent and coordinated modes, making policy learning challenging. However, existing monolithic VLA policies typically process multi-view observations and distinct bimanual interaction patterns without explicitly adapting their visual conditioning or action-generation structure. To address this issue, we propose a bimanual manipulation VLA framework based on Dual-Level Structural Decomposition. The View-Selective Visual Router dynamically adjusts wrist-view contributions to emphasize relevant visual cues, while the Interaction-Aware Action Mixture-of-Experts decomposes action generation into coordinated and arm-wise pathways to adapt to varying bimanual interaction modes. We evaluate the proposed method on six simulated bimanual manipulation tasks in RoboTwin 2.0 and three

arXiv:2606.10743v2 Announce Type: replace Abstract: Human videos provide a scalable source of demonstrations for robot imitation learning. However, converting them into executable and semantically aligned robot trajectories requires determining \emph{when} task-relevant interactions occur, \emph{how} demonstrated human grasps should be retargeted across embodiments, and \emph{which} post-contact motions should be reproduced. In this paper, we present \textit{HOCALT} (\emph{H}and--\emph{O}bject \emph{C}ategory--\emph{A}gnostic \emph{L}ocalization and \emph{T}ransfer), a hand-centric retargeting framework that unifies 3D hand motion reconstruction, temporal contact localization, and cross-embodiment trajectory generation. Given synchronized stereo videos, \textit{HOCALT} identifies task-relevant contact intervals by jointly reasoning over reconstructed 3D hand articulation and category-agnostic object motion cues. These intervals serve as temporal anchors for initializing cross-embodimen

arXiv:2605.18441v2 Announce Type: replace Abstract: Formation control of wheeled mobile robots (WMRs) has been widely studied due to its broad applications. However, existing studies primarily focus on tracking predefined formations, with limited adaptability to varying environments. To address this limitation, we propose CFOO, a Conflict-Free Online Orchestration framework for continuous formation navigation of WMRs that integrates centralized formation generation with distributed formation maintenance. The upper layer monitors the environment in real time and generates adaptive formations as needed, using the proposed TCF-R2T (Trajectory-Conflict-Free Robot-to-Target assignment) algorithm to compute optimal assignments in polynomial time for timely, conflict-free formation transitions. At the lower layer, each WMR continuously employs JSTP (Joint Spatio-Temporal trajectory Planning) to maintain the given formation by jointly optimizing spatial positions and segment durations, improvi

arXiv:2605.02487v4 Announce Type: replace Abstract: We consider mobile grasping in unknown and dynamic environments, where a robot must approach and grasp a target object using a camera with a limited field of view. This setting introduces two visibility constraints. Incomplete environmental observations can create collision risks along the whole-body trajectory, while occlusions can hinder object pose refinement during grasping. Previous systems often assume known or static environments and decouple perception from motion, leaving safety unresolved around unobserved obstacles. We present a unified system that enforces both visibility constraints at runtime through three core components. A velocity-aware gaze policy directs perception toward fast-moving regions of the planned swept volume, a hierarchical subgoal policy samples grasp and intermediate configurations that maintain target visibility, and a real-time whole-body planner replans over a receding horizon as the map updates. Acr

arXiv:2603.26995v2 Announce Type: replace Abstract: Autonomous robots commonly aim to achieve a nominal objective while minimizing a cost. Without contingency preparation, this can leave them vulnerable to mission failure. This is formalized as a trajectory optimization problem over the nominal cost with a contingency constraint: from each checked state along the nominal plan, an admissible contingency maneuver must exist that reaches a designated recovery target within a prescribed contingency horizon. Existing methods either optimize contingency trajectories alongside the nominal plan or evaluate this constraint through computationally expensive nested sampling-based searches. Instead, we introduce SCRAMPPI, which represents this requirement as a reach-avoid problem and leverages Hamilton--Jacobi (HJ) reachability analysis to evaluate contingency feasibility. By either computing a reach-avoid value function offline or updating it only as the environment is revealed, and integrating i

arXiv:2603.26687v2 Announce Type: replace Abstract: Hybrid aerial--ground robots can use thrust to cross obstacles that impede wheel-driven motion, but deciding how much thrust to apply during contact remains challenging. We present an energy-aware reinforcement learning framework that jointly commands wheels, tilt servos and propellers through a single continuous policy, without prescribing locomotion modes. Hardware-calibrated power models penalise estimated electrical energy, while a terrain curriculum and a terminal reward for upright, settled arrivals support learning of thrust-assisted climbing. In simulation, continuous thrust allocation improves single-step clearance over fixed-thrust and mode-switching baselines as steps become taller. At the wheel-radius step height, it draws approximately 27 percent less mean power than the best fixed allocation. An energy-weight ablation shows the accompanying trade-off between efficiency and reliability. On a physical DoubleBee prototype,

arXiv:2603.18532v3 Announce Type: replace Abstract: The strong performance of large vision-language models (VLMs) trained with reinforcement learning (RL) has motivated similar approaches for fine-tuning vision-language-action (VLA) models in robotics. Many recent works fine-tune VLAs directly in the real world to avoid addressing the sim-to-real gap. While real-world RL circumvents sim-to-real issues, it inherently limits the generality of the resulting VLA, as scaling scene and object diversity in the physical world is prohibitively difficult. This leads to the paradoxical outcome of transforming a broadly pretrained model into an overfitted, scene-specific policy. Training in simulation can instead provide access to diverse scenes, but designing those scenes is also costly. In this work, we show that VLAs can be RL fine-tuned across broad scene and object distributions and with reduced labor by leveraging 3D world generative models. Using these models together with a language-driven

arXiv:2603.15265v2 Announce Type: replace Abstract: Developing a unified policy for multi-task robotic manipulation remains challenging due to policy degradation from task interference and negative transfer. In this work, we propose Mixture-of-Experts-Enhanced Action Chunking Transformer (MoE-ACT), a parameter-efficient multi-task visuomotor framework tailored for bimanual manipulation. MoE-ACT incorporates sparse MoE layers into the ACT encoder, dynamically routing image tokens to selected experts based on task context, visual observations, and proprioceptive states. Furthermore, the framework incorporates task-conditioned Feature-wise Linear Modulation (FiLM) in the action decoder alongside multi-scale cross-attention, ensuring precise task grounding and capturing fine-grained spatial cues. Extensive evaluations on the RoboTwin 2.0 benchmark across 16 challenging bimanual tasks demonstrate that MoE-ACT achieves an average success rate of 62.0%, outperforming standard ACT by 17.4 perc

arXiv:2603.14598v2 Announce Type: replace Abstract: Microgravity rendezvous and close proximity operations (RPO) is a growing area of interest for applications spanning in-space assembly and manufacturing (ISAM), orbital debris remediation, and small body exploration. Developing autonomy for these operations requires integrating dynamics simulation with task definition, control and learning algorithms, robustness testing, and evaluation. We present \texttt{SmallSatSim}, an open-source toolkit for developing autonomy algorithms for robots operating in microgravity environments. Built on MuJoCo, \texttt{SmallSatSim} provides a common experiment and task abstraction for spacecraft models, planners, controllers and policies, actuator effects, disturbances, and evaluation, allowing model-based and learning-based approaches to operate on the same problem definitions. The framework combines conventional MuJoCo execution for model-based control with vectorized JAX/MJX execution for massively p

arXiv:2603.08485v3 Announce Type: replace Abstract: Human videos could reduce the reliance on expensive robot demonstrations, but extracting task-relevant features remains challenging. We introduce 3PoinTr, a method that provides robot policies with a global 3D point-track plan predicted from human videos. Given an initial point cloud, 3PoinTr predicts dense 3D point tracks spanning the full task and compresses them into a small set of tokens that remain fixed throughout execution. At each policy query, the policy combines these plan tokens with the current observation to predict robot actions. This formulation enables the policy to extract useful motion cues from human videos. We find this formulation transfers exceptionally well, even when human demonstration motions are infeasible for robots. In controlled simulation and real-world evaluations, 3PoinTr achieves an average success rate 25.0 percentage points higher than the strongest behavior-cloning and video-pretraining baselines o

arXiv:2603.08269v2 Announce Type: replace Abstract: In-context imitation learning allows robots to acquire skills from demonstrations, yet one-shot trajectory generation remains fragile under environmental variation. We propose SAIL, a framework that reframes robot imitation as an iterative refinement problem capable of scaling with test-time compute. SAIL utilizes Monte Carlo Tree Search, where each node is a complete trajectory and edges correspond to trajectory refinements. The process is guided by three core components: an automated archive of successful trajectories for contextually relevant retrieval, a vision language model-based scoring mechanism for trajectory evaluation, and a step-level feedback that provides trajectory-aligned scores for iterative refinement. Experiments across six diverse manipulation tasks in simulation and real-world validation clearly demonstrate that increasing test-time compute consistently improves success rates, achieving up to 95% on complex tasks.

arXiv:2603.07775v2 Announce Type: replace Abstract: Robotic systems deployed beyond controlled environments encounter abrupt and unobserved dynamics shifts that induce substantial transient performance loss, even when the nominal closed loop remains locally stabilizing. We formulate rapid inference-time recovery for physical AI as a constrained residual learning problem, in which an online correction added to a frozen reinforcement learning policy is regulated to minimize recovery time while remaining inside the robustness margin of the nominal controller. We specifically focus upon the regulation of residual's corrective authority relative to an already competent frozen policy. The learned residual produces bounded action-space corrections online, and a Stability Alignment Gate regulates this corrective authority through magnitude constraints, directional coherence, performance-conditioned regulation, and adaptive gain modulation. The method requires no environment resets, policy retr

arXiv:2603.06538v2 Announce Type: replace Abstract: Bimanual manipulation requires both temporal task structure - which actions precede or overlap others - and concrete timing - when each action starts and how long it takes. Symbolic relations enable high-level reasoning about task structure and alternative execution sequences, while concrete timing parameters are essential for coordinating two hands at the execution level. Existing approaches address these two levels in isolation, leaving a gap between task planning and movement synchronization. This work presents an approach for learning symbolic and subsymbolic temporal task constraints from human demonstrations and deriving executable, temporally parametrized plans for bimanual manipulation. Specifically, our contributions are (i) a 3-D representation of timings with methods based on multivariate Gaussian Mixture Models to represent temporal relationships between actions on a subsymbolic level, (ii) a method based on the Davis-Putn

arXiv:2603.03768v2 Announce Type: replace Abstract: Full-stack human-robot collaboration (HRC) can become brittle when replacing a planner, partner model, coordination policy, or controller changes the physical meaning of cross-layer signals. We introduce C2C, an object-centric cognition-to-control architecture that preserves these meanings through physical contracts. Rather than standardizing individual modules, C2C standardizes the physical semantics exchanged between them: task intent is represented by a geometrically verified payload path, partner information by a source-independent physical state, learned coordination by bounded 11-D task-space commands, and robot-specific feasibility remains inside whole-body control (WBC). The reference system combines a vision-language model with deterministic geometric verification and multi-agent reinforcement learning (MARL) for partner-aware coordination. Across nine transport scenarios, adaptive MARL variants achieve 77.1-82.1% mean succes