新闻与研究

arXiv:2604.00055v2 Announce Type: replace Abstract: Existing robotic foundation policies are trained primarily via large-scale imitation learning. While such models demonstrate strong capabilities, they often struggle with long-horizon tasks due to distribution shift and error accumulation. While reinforcement learning (RL) can finetune these models, it cannot work well across diverse tasks without manual reward engineering. We propose VLLR, a dense reward framework combining (1) an extrinsic reward from Large Language Models (LLMs) and Vision-Language Models (VLMs) for task progress recognition, and (2) an intrinsic reward based on policy self-certainty. VLLR uses LLMs to decompose tasks into verifiable subtasks and then VLMs to estimate progress to initialize the value function for a brief warm-up phase, avoiding prohibitive inference cost during full training; and self-certainty provides per-step intrinsic guidance throughout PPO finetuning. Ablation studies reveal complementary ben

arXiv:2603.05670v2 Announce Type: replace Abstract: When humans learn new manipulation skills, they are able to generalize these skills to new contexts and environments. In particular, when learning, humans can easily separate task-relevant aspects of the environment (e.g., object location) and distractors (e.g., the table color). Ideally, robot policies should be able to learn similar, generalizable representations, but instead they often fail under small environment shifts such as changes in lighting, object instance, or initial configuration. In this paper, we propose a self-supervised method that learns a mask which, when multiplied by the observed image features, attempts to transform these features to retain only those which are relevant to the task. Our method --- which we call TransMASK --- can be combined with a variety of imitation learning frameworks (such as diffusion policies) without any additional labels or alterations to the loss function. By introducing a learned mask

arXiv:2602.16863v3 Announce Type: replace Abstract: The ability to manipulate tools significantly expands the set of tasks a robot can perform. Yet, tool manipulation represents a challenging class of dexterity, requiring grasping thin objects, in-hand object rotations, and forceful interactions. Since collecting teleoperation data for these behaviors is challenging, sim-to-real reinforcement learning (RL) is a promising alternative. However, prior approaches typically require substantial engineering effort to model objects and tune reward functions for each task. In this work, we propose SimToolReal, taking a step towards generalizing sim-to-real RL policies for tool manipulation. Instead of focusing on a single object and task, we procedurally generate a large variety of tool-like object primitives in simulation and train a single RL policy with the universal goal of manipulating each object to random goal poses. This approach enables SimToolReal to perform general dexterous tool man

arXiv:2602.10013v2 Announce Type: replace Abstract: Successfully manipulating many everyday objects, such as potato chips, requires precise force regulation. Failure to modulate force can lead to task failure or irreversible damage to the objects. Humans can precisely achieve this by adapting force from tactile feedback, even within a short period of physical contact. We aim to give robots this capability. However, commercial grippers exhibit high cost or high minimum force, making them unsuitable for studying force-controlled policy learning with everyday force-sensitive objects. We introduce TF-Gripper, a low-cost (~$150) force-controlled parallel-jaw gripper that integrates tactile sensing as feedback. It has an effective force range of 0.45-45 N and is compatible with different robot arms. Additionally, we designed a teleoperation device paired with TF-Gripper to record human-applied grasping forces. While we can train standard low-frequency policies with the collected force data,

arXiv:2602.08784v2 Announce Type: replace Abstract: Robust and accurate perception of dynamic objects and map elements is crucial for autonomous vehicles performing safe navigation in complex traffic scenarios. While vision-only methods have become the de facto standard due to their technical advances, they can benefit from effective and cost-efficient fusion with radar measurements. In this work, we advance fusion methods by repurposing Gaussian Splatting as an efficient universal view transformer that bridges the view disparity gap, mapping both image pixels and radar points into a common Bird's-Eye View (BEV) representation. Our main contribution is GaussianCaR, an end-to-end network for BEV segmentation that, unlike prior BEV fusion methods, leverages Gaussian Splatting to map raw sensor information into latent features for efficient camera-radar fusion. Our architecture combines multi-scale fusion with a transformer decoder to efficiently extract BEV features. Experimental results

arXiv:2510.11094v2 Announce Type: replace Abstract: Knee rehabilitation plays a critical role in restoring patients' mobility and functional independence. Traditional rigid rehabilitation exoskeletons are often bulky and cumbersome to wear, whereas soft pneumatic exoskeletons offer lightweight, wearable, and intrinsically compliant solutions that are better suited for human--robot interaction. However, achieving precise motion control for soft exoskeletons remains challenging due to the difficulty of accurately modeling pneumatic actuators and the patient-specific human--robot coupled dynamics during rehabilitation training. To address these challenges, this paper proposes a Koopman-based Model Predictive Control (KMPC) framework for soft knee rehabilitation exoskeletons. The nonlinear human--robot coupled system is represented through a lifted linear Koopman model, enabling predictive control with explicit handling of constraints. In addition to actuation commands used to control valv

arXiv:2509.26513v2 Announce Type: replace Abstract: Generating large and diverse obstacle datasets to learn motion planning in environments with dynamic obstacles is challenging due to the vast space of possible obstacle trajectories. Inspired by hallucination-based data synthesis approaches, we propose Learning from Hallucinating Critical Points (LfH-CP), a self-supervised framework for creating rich dynamic obstacle datasets based on existing optimal motion plans without requiring expensive expert demonstrations or trial-and-error exploration. LfH-CP factorizes hallucination into two stages: first identifying when and where obstacles must appear in order to result in a near-optimal motion plan, i.e., the critical points, and then procedurally generating diverse trajectories that pass through these points while avoiding collisions. This factorization avoids generative failures such as mode collapse and ensures coverage of diverse dynamic behaviors. We further introduce a diversity met

arXiv:2509.04094v3 Announce Type: replace Abstract: Object reconstruction and inspection tasks play a crucial role in various robotics applications. Identifying paths that reveal the most unknown areas of the object is paramount in this context, as it directly affects reconstruction efficiency. Current methods often use sampling based path planning techniques, evaluating views along the path to enhance reconstruction performance. However, these methods are computationally expensive as they require evaluating several candidate views on the path. To this end, we propose a computationally efficient solution that relies on calculating a focus point in the most informative region and having the robot maintain this point in the camera field of view along the path. In this way, object reconstruction related information is incorporated into the whole body control of a mobile manipulator employing a visibility constraint without the need for an additional path planner. We conducted comprehensiv

arXiv:2503.14753v3 Announce Type: replace Abstract: Computed tomography (CT)-guided needle biopsies are critical for diagnosing a range of conditions, including lung cancer, but present challenges such as limited in-bore space, prolonged procedure times, and radiation exposure. Robotic assistance offers a promising solution by improving needle trajectory accuracy, reducing radiation exposure, and enabling real-time adjustments. In our previous work, we introduced a robotic platform designed for accurate needle insertion within the confined CT bore. However, its performance in clinical settings is restricted by limited dexterity and a constrained workspace. In this study, we present an 11-degree-of-freedom (DOF) robotic system that integrates a 6-DOF robotic base with an improved 5-DOF cable-driven end-effector, yielding a significantly expanded workspace and enhanced dexterity. To leverage the hyper-redundant degrees of freedom, we introduce a weighted inverse kinematics controller, al

arXiv:2610.08618v1 Announce Type: cross Abstract: A robot that offloads its control loop to the network carries the receiver with it, so link quality is decided by where it goes. Simulating this requires both a physics engine and a site-specific propagation model at once; to our knowledge no simulator natively unifies both, with existing couplings of the two limited to offline analyses. We demonstrate a real-time co-simulation framework coupling NVIDIA Isaac Sim and NVIDIA Sionna over ROS 2 that closes the perception-action-communication (PAC) loop between them. Sionna ray-traces the base-station-to-robot channel over the exact geometry Isaac Sim simulates on, rather than modeling it stochastically, and feeds channel states back into the control loop in real time. Ray-traced on GPU, the coverage map is refreshed in ~16ms (~60 Hz), fast enough for real-time control. To showcase the framework's utility, we implement a wireless-aware navigation application in an OpenStreetMap(OSM)-derived

arXiv:2610.07969v1 Announce Type: cross Abstract: Scaling robotic foundation models requires diverse training data and reliable evaluation environments. Simulation offers a scalable solution, yet existing generation pipelines remain constrained by predefined assets and skills, a disconnect between scene generation and task generation, and limited support for complex embodiments and physics. We introduce EmbodiedSmith, a framework for scalable embodied data generation through recursive self-improvement (RSI). EmbodiedSmith unifies asset, scene, and task generation in a pipeline that supports autonomous creation and language-driven customization. Its core is an agentic refinement loop: scene generation anticipates downstream task requirements, while task generation guides targeted scene edits, allowing scenes and tasks to iteratively improve one another. This joint refinement improves task generation success, including for long-horizon tasks. The framework further supports mobile manipul

arXiv:2610.07910v1 Announce Type: cross Abstract: Deep Reinforcement Learning policies can produce nonsmooth action oscillations that hinder deployment on physical robots. Existing architectural and penalty-based approaches seek spatial smoothness by directly reducing sensitivity to changes in state inputs, but their broad constraints can degrade task performance as stronger smoothing is pursued. Temporal regularization instead constrains action differences along observed transitions, but has been considered unable to provide the spatial smoothness needed under observation noise. We revisit this assumption by proving that the temporal penalty bounds the expected action differences between current states sharing a next state, revealing a spatial effect that empirically extends to spatial smoothness. Building on this finding, we propose Conditioning for Action using only Temporal Smoothness (CATS), which combines a temporal penalty with linear ramp-up. We highlight temporal regularizatio

arXiv:2610.07540v1 Announce Type: cross Abstract: Robotic systems often exhibit unstable modes, along which small perturbations and disturbances can cause unbounded growth unless corrected through feedback. Controlling such systems from high-dimensional visual observations requires representations that preserve these modes. Joint-embedding predictive architectures (JEPAs) provide a natural framework for learning such representations and their dynamics from visual data. However, we demonstrate that next step prediction combined with anti-collapse regularization does not guarantee that controllable unstable modes are preserved: the training loss can be minimized while these modes are collapsed, making stabilization from the learned representation impossible. To address this, we augment world-model training with an action reconstruction objective (i.e., an inverse dynamics loss) that encourages control-aware representations, namely, visual representations that preserve crucial features fo

arXiv:2610.07477v1 Announce Type: cross Abstract: Large language models (LLMs) let users direct heterogeneous multi-robot systems (MRS) through natural language, but make task interpretation, robot assignment, and coordination difficult to inspect and change. Based on a formative study with 12 non-expert users, we developed MRPilot, a mixed reality system organized around four stages of supervision and intervention. MRPilot represents robot-team plans and execution states as structured commitments shared across synchronized situated and overview views. Across four stages, it helps users resolve ambiguous references (Forming), review plans before execution (Reviewing), monitor distributed execution (Following), and make robot-level or team-level changes when problems arise (Repairing). In a within-subjects study with 20 participants in a virtual reality-simulated home, MRPilot reduced workload, increased situational awareness, transparency, trust, and perceived control compared with a c

arXiv:2610.07192v1 Announce Type: cross Abstract: Vision-Language Navigation (VLN) enables robots to navigate through environments using natural language instructions, making human-robot interaction intuitive. Traditional VLN models often rely on navigation graphs, 360-degree views, and perfect localization which pose significant challenges when adapting these models to real-world settings. This work addresses these limitations by performing a simulation-to-real domain shift of a VLN approach that operates in continuous environments without requiring navigation graphs or panoramic views. The proposed system integrates vision-language models that align visual inputs and linguistic instructions within a shared embedding space, facilitating natural language-driven navigation. We employ a Cross-Modal Attention (CMA) based architecture trained on an existing dataset in a simulated environment and fine-tune it using real-world data collected from a custom-built Ackermann-steered robot equipp