arXiv subjects

Robotics

Explore arXiv robotics papers and cs.RO metadata. Search for robot learning, motion planning and control, then check experiments in the source manuscript.

Follow this topic with RSS

336 records · Page 5Linked to original sources

arXiv2026

Drift-Based Policy Optimization: Native One-Step Policy Learning for Online Robot Control

Diffusion policies effectively model multimodal action distributions for robotic manipulation, but their iterative denoising requires tens to hundreds of network function evaluations (NFEs) for each control prediction, limiting their applicability to high-frequency closed-loop control and online reinforcement learning (RL). We present a two-stage framework for native one-step generative policies that transfers iterative refinement from inference to training. First, Drift-Based Policy (DBP) uses a fixed-point drifting objective to internalize corrective dynamics into the model parameters, producing multimodal action chunks with a single network evaluation by design. Second, Drift-Based Policy Optimization (DBPO) augments the pretrained backbone with a stochastic interface that provides exact conditional rollout likelihoods for PPO-style on-policy updates while preserving 1-NFE deployment. On the 12-task Diffusion Policy suite, DBP improves the average success rate from 0.79 to 0.83 while reducing inference from 100 NFEs to 1. Across 37 point-cloud manipulation tasks, DBP achieves an average success rate of 88.4%, surpassing the leading 1-NFE baseline OMP at 82.3%. DBPO further improves pretrained one-step policies through stable online fine-tuning on RoboMimic and D4RL. On a physical dual-arm UR5 platform, DBP achieves 123/150 successes (82%) with an average end-to-end latency of 9.5 ms, compared with MP1's 89/150 successes (59%) under the same setup. Code is available at https://github.com/YuxuanGao0822/DBPO.

cs.RO↗

arXiv2026

Toward Interaction Dynamics: A Predictive Framework for Safe Physical Human Robot Interaction

Physical human-robot interaction requires yielding transiently to contact yet recovering the commanded reference under sustained load. Finite-stiffness impedance control retains a static deflection there, while predictive alternatives typically optimize a nonlinear robot or impedance model online. Operational-space cancellation instead exposes a translational error double integrator with a fixed transition matrix and a configuration-scheduled input map, making interaction a predictive quantity rather than a property re-derived per configuration. We build on it a compact offset-free interaction-error MPC for torque-controlled manipulators: a force-domain random-walk state estimates persistent interaction and model error, and a 30-variable convex QP maps the correction through the current task inertia while constraining the applied joint torque. Conditional results establish impedance equivalence of the unconstrained passive feedback, offset-free regulation at feasible frozen configurations, and quadratic stabilizability of the scheduled backbone. In a 1kHz MuJoCo simulation of a 7-DOF Franka FR3, the estimator cuts steady-state error under a repeated 15N step from 2.77mm to 0.042mm when added to the otherwise identical 100Hz MPC. A stiffness-and-damping-calibrated impedance baseline attains 2.59mm but briefly saturates and needs 3.3x the peak positive joint power. Adding ideal measured-force cancellation to that baseline gives 1.39mm, so constant-load rejection is not unique to MPC; the sensorless controller still reaches 0.042mm in the moving task, a 65x reduction without force sensing and without the baseline's saturation or power cost. Demonstrated in simulation under a shared actuator budget, the contribution is an efficient operational-space realization complementing rather than replacing broader interaction-control architectures.

cs.RO↗

arXiv2026

AGRICAM: A Track-Mounted Crop Pollination Monitoring Robot

Insect pollination is critical for global food production, yet monitoring pollinators at commercial farm scale remains a challenge. Recent advances in computer vision and deep learning have enabled detailed analysis of pollinator behaviour, but monitoring must trade-off detail against spatial coverage and human or technological resources. This paper presents the Automated Guided Robot for Insect and Crop Activity Monitoring (AGRICAM), a purpose-built robotic system designed to meet the requirements of large-scale pollination monitoring in protected cropping systems. AGRICAM operates autonomously on low-cost, easily installed track for movement along crop rows, without disrupting farm operations or insect behaviour. The platform integrates two RGB cameras, microclimate sensors, GPS and RFID modules, motion sensors, and 4G cellular network connectivity for data transmission. A web interface enables remote device configuration and scheduling. The system autonomously captures video and image data of insects' locations and local environmental conditions. These are transferred to the cloud and analysed using computer vision models to quantify pollinator visitation and spatio-temporal activity variation. We deployed the system on a commercial blueberry farm to demonstrate and test its capability. It successfully mapped insect pollination patterns across 80 m long industrial polytunnels over 30 hours. This data enabled spatial analyses of insect activity we used to confirm a uniform pollinator distribution within polytunnels, as desired by the farm management team. The data also highlighted variation of insect activity associated with time of day and microclimate. AGRICAM therefore has been shown to be a scalable, automated crop pollination monitor that can support data-driven decisions to enhance pollination management, thereby improving crop productivity and food security.

cs.CV↗

arXiv2026

Not All Agreement Counts as Corroboration: Provenance-Conserving Multi-View Fusion for Typed Action Admission in Human-Robot Collaboration

For embodied systems, predictive agreement alone does not determine whether evidence warrants action; evidential origin matters. Repeated inference over one observation can multiply agreement without adding evidence, while source-local values do not reveal whether outputs have separately countable origins. PACT treats evidence countability as a relational variable for provenance-conserving fusion and typed action admission. A supplied provenance partition defines countable units. PACT retains coordinatewise support shared within each unit, accumulates only across units, and maps unmet release conditions to hold, confirm, or fallback. Under the stated assumptions, source-local values cannot identify countability; the coordinatewise meet is the greatest budget satisfying singleton fidelity and insertion non-amplification, with coarsening monotonicity and fixed-partition stability. Across 31,200 evaluations in 48 scene clusters, PACT attains a common-support normalized risk-coverage area (ncsAURC) of 0.0861. Excluding the constructed adversarial-consensus arm, provenance-partition aggregation reduces ncsAURC by 0.0557 relative to singleton aggregation, while the corroboration contrast vanishes. On complete-source records, native scores favor PACT, but a common posterior-peak score narrows its difference from nested Dirichlet and favors product fusion. Reassigning provenance over unchanged predictions moves evidence budgets as predicted. In offline human-robot collaboration, eightfold within-camera duplication leaves 720 typed responses per checkpoint unchanged; camera-grouped PACT admits 47 of 57 Qwen3-VL-32B reference-consistent candidates with no observed reference-inconsistent admission in 60 episodes. PACT separates computational from evidential multiplicity: agreement constitutes corroboration only when provenance permits separate accumulation.

cs.RO↗

arXiv2026

TriSAR: Task Coordination and Collision Avoidance for Aerial Robot Teams in Disaster Response

Multi-Unmanned Aerial Vehicle (UAV) disaster-response systems require coordinated task assignment and local trajectory control, yet the individual and combined contributions of these coordination layers to mission efficiency and operational safety remain insufficiently characterised under controlled experimental conditions. TriSAR is evaluated as a five-UAV coordination system operating in a physics-based Gazebo simulation of an earthquake-damaged urban environment. A 2 x 2 factorial design compares two task-allocation strategies (Genetic Algorithm and greedy fitness-based allocation) with reactive collision avoidance enabled or disabled. Each of the four configurations was evaluated over 30 stochastic episodes in a common scenario of five UAVs and eight targets. Under greedy allocation, enabling repulsion eliminated recorded collision-threshold violations, confirmed by a Mann-Whitney test (U = 885, p = 4.03 x 10^-12, rank-biserial r = 0.97). Under GA allocation, the same protective effect was confirmed (U = 675, p = 1.26 x 10^-5, rank-biserial r = 0.50). For mission-efficiency metrics, GA-based allocation showed no statistically detectable advantage over greedy allocation when repulsion was enabled, but a significant advantage in steps, path length, and energy when repulsion was disabled (Welch's t-tests, |g| between 0.92 and 1.76). These results show that reactive repulsion provides a substantial, allocation-dependent safety benefit, while the additional computational complexity of GA-based task allocation yields a detectable mission-efficiency benefit only when repulsion is disabled.

cs.RO↗

arXiv2026

Contact-Constrained Lower-Limb Joint-Offset Calibration for Humanoid Robots

Accurate joint encoder offsets are essential for kinematic consistency in humanoid lower limbs, yet existing calibration methods typically require external motion-capture systems or fiducial targets. We present a self-contained calibration framework exploiting only onboard joint encoders and a pelvis-mounted IMU during static double-support contact. The inter-foot transform from forward kinematics must stay constant when both feet are fixed; minimizing its posture-dependent dispersion yields a nonlinear least-squares problem over the 12-dimensional offset vector. A Hessian eigenstructure analysis shows that parallel pitch axes induce a rotational coupling. Orientation residuals then observe only the pitch-offset sum, while translation and posture diversity set the remaining numerical observability. For the A3 pitch-to-roll-to-yaw ordering, hip-roll and hip-yaw excitation reduce hip-pitch coupling. A standing-posture knee prior then anchors the remaining weak pitch-chain decomposition. Simulation and real-machine injection tests show consistent recovery, and on held-out recordings calibration reduces foot-height RMS residuals from 4.26 to 2.20 mm on A3 and from 8.03 to 1.43 mm on A2. An independent LiDAR-inertial reference checks the pitch-coupled channel. Removing an injected pitch offset moves the leg-odometry vertical drift back toward the LiDAR trajectory. A few static double-support stances thus provide contact-consistent corrections for well-excited directions. Individual offsets in the weak pitch chain remain prior-dependent.

cs.RO↗

arXiv2026

Toward Trustworthy Robot-Assisted Sliding Palpation for Shallow Vessel Localisation with a Calibrated Digital Twin

Reliable localisation of shallow subsurface vessels is important for safe robot-assisted venous access and vessel-aware manipulation, but collecting diverse tactile data on physical hardware is costly, time-consuming, and can degrade soft vision-based tactile sensors. We present a robot-assisted sliding-palpation framework in which a calibrated digital twin generates labelled tactile sequences, reducing reliance on real-world data. The twin models sensor-vessel contact, is calibrated against real palpation trajectories using Bayesian-optimisation-based domain adaptation, and is randomised over sliding direction and contact conditions. A spatio-temporal graph neural network trained on simulated marker trajectories performs per-node vessel classification and produces a human-verifiable top-view localisation map through 2D-to-3D-to-2D geometric projection. We evaluate three datasets: Sim, Silicone, and Meat, the latter a raw-meat phantom with vessel models at nominal depths of 0 to 30 mm, using four train-to-test configurations: Sim to Sim, Sim to Silicone, Sim to Meat, and Meat to Silicone. The calibrated twin achieves a simulated-to-real marker-alignment mean absolute error of 0.50 mm at deepest contact across four canonical interactions. After reprojection onto a 1 mm top-view grid, predicted vessel pixels lie on average 1.05 to 5.49 mm from the nearest true vessel pixel across the four models, with 1.05 to 1.31 mm for all except Sim to Meat. The larger error for Sim to Meat reflects the greater domain shift and current limit of simulation transfer. These results demonstrate progress toward trustworthy tactile palpation through calibrated simulation, interpretable localisation, and transparent cross-domain evaluation. Code, model weights, and data are publicly available on GitHub and Zenodo.

cs.RO↗

arXiv2026

PhyGile: Physics-Prefix Guided Motion Generation for Agile General Humanoid Motion Tracking

Humanoid robots are expected to execute agile and expressive whole-body motions in real-world settings. Existing text-to-motion generation models are predominantly trained on captured human motion datasets, whose priors assume human biomechanics, actuation, mass distribution, and contact strategies. When such motions are directly retargeted to humanoid robots, the resulting trajectories may satisfy geometric constraints (e.g., joint limits and pose continuity) and appear kinematically reasonable. However, they frequently violate the physical feasibility required for real-world execution. To address these issues, we present PhyGile, a unified framework that closes the loop between robot-native motion generation and General Motion Tracking (GMT). PhyGile performs physics-prefix-guided robot-native motion generation at inference time, directly generating robot-native motions in a 262-dimensional skeletal space with physics-guided prefixes, thereby eliminating inference-time retargeting artifacts and reducing generation-execution discrepancies. Before physics-prefix adaptation, we train the GMT controller with a curriculum-based mixture-of-experts scheme, followed by post-training on unlabeled motion data to improve robustness over large-scale robot motions. During physics-prefix adaptation, the GMT controller is further fine-tuned with generated objectives under physics-derived prefixes, enabling agile and stable execution of complex motions on real robots. Extensive offline and real-robot experiments demonstrate that PhyGile expands the frontier of text-driven humanoid control, enabling stable tracking of agile, highly difficult whole-body motions that go well beyond walking and low-dynamic motions typically achieved by prior methods.

cs.RO↗

arXiv2026

BinWalker: Development and Field Evaluation of a Quadruped Manipulator Platform for Sustainable Litter Collection

Litter pollution represents a growing environmental problem affecting natural and urban ecosystems worldwide. Waste discarded in public spaces often accumulates in areas that are difficult to access, such as uneven terrains, coastal environments, parks, and roadside vegetation. Over time, these materials degrade and release harmful substances, including toxic chemicals and microplastics, which can contaminate soil and water and pose serious threats to wildlife and human health. Despite increasing awareness of the problem, litter collection is still largely performed manually by human operators, making large-scale cleanup operations labor-intensive, time-consuming, and costly. Robotic solutions have the potential to support and partially automate environmental cleanup tasks. In this work, we present a quadruped robotic system designed for autonomous litter collection in challenging outdoor scenarios. The robot combines the mobility advantages of legged locomotion with a manipulation system consisting of a robotic arm and an onboard litter container. This configuration enables the robot to detect, grasp, and store litter items while navigating through uneven terrains. The proposed system aims to demonstrate the feasibility of integrating perception, locomotion, and manipulation on a legged robotic platform for environmental cleanup tasks. Experimental evaluations conducted in outdoor scenarios highlight the effectiveness of the approach and its potential for assisting large-scale litter removal operations in environments that are difficult to reach with traditional robotic platforms. The code associated with this work can be found at: https://github.com/iit-DLSLab/trash-collection-isaaclab.

cs.RO↗

arXiv2026

GhostTac: Manipulating Tactile Sensors without Physical Contact

Tactile sensors are integral components of modern robotic systems, enabling robots to perceive and interact with the physical environment through tactile feedback. Despite their importance, the physical-layer security of tactile sensors has received little attention in prior work. In this paper, we present GhostTac, to the best of our knowledge, the first contactless attack that manipulates tactile sensing via electromagnetic interference (EMI). We identify that EMI exploits the nonlinear rectification and limited bandwidth amplification effects, allowing carefully crafted EMI signals to be converted into a persistent DC offset that bypasses on-board filtering and induces stable measurement deviations. Building on this mechanism, GhostTac enables fine-grained and controllable manipulation of sensor outputs by reshaping the spatial distribution and manipulating the magnitude at the targeted location. Such interference can induce unintended and harmful robot behaviors, such as causing a domestic robot to exert excessive force, resulting in physical damage or human injury. We evaluate GhostTac on 10 sensor modules and 2 dexterous hands, covering 15 tactile sensors of different types, and demonstrate consistent attack effectiveness across all tested devices. We further present three case studies on tactile grasping, slip detection, and material classification to illustrate practical impacts in real robotic tasks. We envision that our findings shed light on a new physical attack vector against tactile sensing in robotic systems.

cs.CR↗

arXiv2026

AnyWorld: Factorized Egocentric World Models for Cross-Embodiment Generalization

Collecting contact-rich robot experiences at scale remains a major bottleneck for generalizable manipulation. Beyond data quantity, robot learning also requires diverse experiences across embodiments, viewpoints, and scenes. Human egocentric videos provide abundant physical interactions, but each video captures only a narrow slice of experience under a single body, camera trajectory, and environment. We propose AnyWorld, a cross-embodiment world modeling framework that expands a single human interaction into diverse robot-native rollouts without paired human-robot demonstrations. Our model factorizes an interaction into action, camera, and embodiment: action controls capture the motion structure, camera controls specify viewpoint evolution, and the target embodiment context defines the acting body and its interaction geometry. This formulation enables independent recomposition of embodiment, viewpoint, and scene factors, allowing a single model to generate many robot-domain experiences while preserving the underlying dynamics and object interactions. We train the model with large-scale human interaction pretraining followed by mixed-embodiment fine-tuning. Experiments show that our model supports controllable recomposition across embodiments, viewpoints, and scenes, and we further demonstrate that the generated data can improve manipulation performance on the RoboCasa GR1 tabletop benchmark and a real IRON humanoid robot. Beyond aggregate gains, we test whether unpaired human experience can be recomposed into robot-native video-action pairs that target a policy gap. Controlled IRON interventions correct a spurious completion prior and establish language-grounded spatial target selection; an action-only counterfactual intervention fails to learn the latter reliably, showing that both action calibration and visual recomposition are necessary.

cs.RO↗

arXiv2026

Parallel Reference-Centric Continuous-Time Relative Localization with Augmented Clamped Non-Uniform B-Splines

Accurate relative localization is critical for multi-robot cooperation. In robot groups, measurements from different robots arrive asynchronously and with clock time-offsets. Although Continuous-Time (CT) formulations have proved effective for handling asynchronous measurements in single-robot SLAM and calibration, extending CT methods to multi-robot settings faces great challenges in achieving high-accuracy, low-latency, and high-frequency performance. In particular, existing CT methods suffer from the inherent query-time delay of unclamped B-splines and high optimization latency. This paper proposes CT-RIO, a novel Continuous-Time Relative-Inertial Odometry framework. We adopt Clamped Non-Uniform B-splines (C-NUBS) to represent states, eliminating the query-time delay. We further augment C-NUBS with closed-form extension and shrinkage operations that preserve the spline shape, making it suitable for online estimation and enabling flexible knot management. This flexibility leads to the concept of a knot-keyknot strategy, which supports spline extension at high frequency while retaining sparse keyknots for adaptive relative motion modeling. We then formulate a reference-centric sliding-window relative localization problem that operates purely on relative kinematics and inter-robot constraints. To enable low-latency and high-frequency estimation, we decompose the tightly coupled optimization into robot-wise subproblems and solve them in parallel using asynchronous block coordinate descent. Extensive experiments show that CT-RIO converges from time-offsets as large as 264 ms to sub-millisecond within 3 s, and achieves RMSEs of 0.046 m and 1.8 degree. It consistently outperforms evaluated published methods, with improvements of up to 60% under high-speed motion.

cs.RO↗

arXiv2026

RoboGesture: Real-Time Semantic-aligned Co-Speech Gestures Generation for Humanoid Interaction

Enabling humanoid robots to respond to human speech with synchronized and semantically meaningful gestures is fundamental to natural human-robot interaction. However, this task faces three critical barriers: the scarcity of semantically rich datasets, the "modality eclipse" where models ignore audio cues in favor of kinematic inertia, and the sim-to-real gap regarding physical safety. We propose RoboGesture, a robot-centric framework that co-designs data, modeling, and control to power a complete interactive human-humanoid system in which the robot listens, responds, and gestures in real time. We first establish the RoboGesture dataset featuring over 300 gesture categories and develop an automated pipeline to synthesize large-scale collision-free, robot-specific audio-motion pairs. Our architecture features a Hierarchical Semantic-Acoustic Aligner that extracts multi-granular prosodic and semantic cues directly from raw audio tokens. These cues drive a Streaming Conditional Motion Generator based on a diffusion transformer with conditional flow matching. To ensure high responsiveness, we introduce Anti-Inertia CFG Masking, which prevents the model from collapsing into repetitive historical patterns by compelling it to proactively mine control signals from the audio modality. Finally, an MPC-based safety filter ensures real-time, collision-free execution on physical hardware. Experiments on a Unitree G1 humanoid demonstrate that RoboGesture generates safer, more rhythmic, and more semantically appropriate responses compared to state-of-the-art baselines.

cs.RO↗

arXiv2026

TRACE: Trajectory-Routed Causal Memory for Delayed-Evidence Visuomotor Imitation

Robots under autonomous operation may require decisions based on evidence that is no longer visible. We study delayed-evidence tasks, where an early cue disappears before a later decision point, so visually similar observations can require different actions. In these settings, the current observation is not a sufficient state for control. We introduce TRAjectory-routed Causal Evidence (TRACE), a memory framework for visuomotor imitation policies. TRACE stores task-relevant visual and robot-state evidence, such as object identity, target choice, or route-dependent state, in a fixed-size latent memory that remains bounded over long episodes. Instead of indexing memory by raw time or manually provided task labels, TRACE uses path signatures: compact, order-sensitive features of the executed robot-state trajectory. These signatures do not store the visual cue itself; rather, they provide trajectory-conditioned keys for writing and retrieving the evidence stored when the cue was visible. When the robot later reaches an ambiguous observation, the policy conditions on TRACE memory to recover the missing context and choose the correct branch. TRACE attaches through lightweight adapters to policies, without changing the policy backbone, action head, or imitation objective. Across real-world long-horizon manipulation tasks with visually ambiguous branch points, TRACE improves branch selection and task success over alternative baselines, including short-history and recurrent memory. Project page: https://jeong-zju.github.io/trace

cs.RO↗

arXiv2026

Hydra: Marker-Free RGB-D Hand-Eye Calibration

This work presents an RGB-D imaging-based approach to marker-free hand-eye calibration using a novel implementation of the iterative closest point (ICP) algorithm with a robust point-to-plane (PTP) objective formulated on a Lie algebra. Its applicability is demonstrated through comprehensive experiments using three well known serial manipulators and two RGB-D cameras. With only three randomly chosen robot configurations, our approach achieves approximately 90% successful calibrations, demonstrating 2-3x higher convergence rates to the global optimum compared to both marker-based and marker-free baselines. We also report 2 orders of magnitude faster convergence time (0.8 +/- 0.4 s) for 9 robot configurations over other marker-free methods. Our method exhibits significantly improved accuracy (5 mm in task space) over classical approaches (7 mm in task space) whilst being marker-free. The benchmarking dataset and code are open sourced under Apache 2.0 License, and a ROS 2 integration with robot abstraction is provided to facilitate deployment.

cs.RO↗

arXiv2026

Scalable Rao-Blackwellized Online Planning for High-Dimensional POMDPs

Online planning under uncertainty remains a fundamental challenge for robotic systems operating in partially observable environments with high-dimensional state spaces. While sampling-based POMDP solvers enable approximate decision-making in large or continuous domains, their performance degrades as belief dimensionality increases due to the high variance inherent in Monte Carlo-based estimation. In this work, we extend the Rao-Blackwellized online POMDP (RB-POMDP) framework to improve its generalizability in high-dimensional settings through hybrid continuous-discrete belief representations. By analytically propagating uncertainty associated with marginalized state components during tree-based planning, the proposed approach reduces sampling-induced variance in value estimation. We demonstrate the effectiveness of this framework in a robotic search-and-rescue task by integrating it with FastSLAM 2.0. Experimental results show that the proposed planner achieves higher cumulative rewards using significantly fewer particles and planning simulations than purely sampling-based methods under equivalent computational budgets. These results suggest that structured high-dimensional robotic problems admitting tractable sufficient statistics can be effectively leveraged within the RB-POMDP framework for computationally feasible online decision-making.

cs.RO↗

arXiv2026

Zeva: In-Context Causal Learning for Generalizable Embodied Manipulation

Generalizable embodied manipulation remains difficult to achieve through pretraining alone, due to unseen physical conditions in the real world. We argue that robots need to learn from their own physical interactions on the fly during real-world deployment and use this knowledge to inform subsequent actions. We present Zeva, the first framework that enables in-context learning from a robot's own physical interaction experience while keeping the policy model frozen. Zeva employs a Causal Interaction Extractor to encode an executed action and its induced state change into a causal interaction signal, which is stored in a dual-timescale causal memory. For subsequent actions, relevant causal interaction signals are retrieved from memory and injected into the frozen policy model as context. Experiments in simulation and real-world manipulation demonstrate that Zeva achieves the best performance among the compared frontier VLAs and WAMs and, more importantly, enables self-evolution during deployment without gradient updates. Its success rate continues to improve as the robot accumulates interaction experience. Furthermore, the acquired interaction experience can generalize across tasks.

cs.RO↗

arXiv2026

SmoothRL: Online Reinforcement Learning During Asynchronous Execution

Deploying robot policies in the physical world requires satisfying two fundamental desiderata: reliability and smooth real-time execution. However, deploying state-of-the-art generalist models presents challenges on both fronts. Achieving the precision and robustness required for real-world deployment necessitates sample-efficient online reinforcement learning (RL) to adapt pretrained models. Meanwhile, the increasing scale of robot foundation models has led to higher inference latency. To satisfy real-time constraints under high latency, modern systems adopt asynchronous inference with action chunking, overlapping policy computation with chunk execution to hide latency and enable smooth control. Despite their complementary roles, integrating asynchronous execution with gradient-based online RL remains underexplored. We present SmoothRL, an online RL framework that fine-tunes a pretrained policy within an asynchronous inference loop. SmoothRL follows a value-gradient paradigm, directly updating policy parameters using gradients of the action-value function with respect to policy actions. To enable correct optimization under asynchronous execution, SmoothRL explicitly models the asynchronous inference process during training. Specifically, each generated action chunk is partitioned by frame index into three regions: a committed region, consisting of actions committed by the previous inference cycle; an execution region, containing newly generated actions executed by the robot; and a discarded region, containing actions superseded by the next inference cycle. Gradients are propagated only through the execution region, ensuring policy optimization aligns with the trajectory distribution induced by asynchronous execution. We evaluate SmoothRL on real-world robotic tasks requiring high precision, as well as highly dynamic tasks that necessitate asynchronous execution.

cs.RO↗