EmbodiSteer: Steering Embodiment-Agnostic Visuomotor Policies with Joint-Space Guidance for Zero-Shot Cross-Embodiment Deployment

• URL: http://arxiv.org/abs/2606.12965v1
• Date: Thu, 11 Jun 2026 06:49:39 GMT
• Title: EmbodiSteer: Steering Embodiment-Agnostic Visuomotor Policies with Joint-Space Guidance for Zero-Shot Cross-Embodiment Deployment
• Abstract summary: EmbodiSteer is a training-free framework that steers embodiment-agnostic visuomotor policies toward zero-shot, embodiment-aware deployment.<n>It reduces collision rate by 46.1% and improves task success rate by 28.5% across 9 simulated robots.
• Score: 7.14164117393883
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Scalable robot imitation learning relies on large-scale heterogeneous data from diverse robots or body-free data, making Cartesian end-effector actions a key interface for embodiment-agnostic policy learning. However, end-effector-only abstraction leaves Cartesian policies unaware of the deployed robot body, making them brittle under robot-specific constraints such as whole-body collision avoidance. To overcome this limitation, we present EmbodiSteer, a training-free framework that steers embodiment-agnostic visuomotor policies toward zero-shot, embodiment-aware deployment. EmbodiSteer keeps policy learning in Cartesian space while efficiently lifting inference-time diffusion sampling into the target robot's joint space via forward kinematics and Jacobian-based updates. With whole-body collision-aware guidance over joint trajectories after each denoising step, the arm can be steered away from collisions while preserving learned end-effector behavior. Compared with Cartesian-only execution, EmbodiSteer reduces collision rate by 46.1% and improves task success rate by 28.5% across 9 simulated robots, and further achieves 90.0% collision rate reduction and 36.7% success rate increase on two physical robots in highly constrained scenarios. Our project page is at https://frankwang67.github.io/EmbodiSteer-Page.

Related papers

• Non-Prehensile Throwing: A Reinforcement Learning Perspective [4.519185019401122]
  Non-prehensile (grasp-free) throwing is better suited for large, heavy, and/or deformable objects.<n>Existing approaches rely on model-based optimization with simplified contact models.<n>We propose a reinforcement learning approach that directly optimize joint-space trajectories.<n>Our method throws diverse objects including heavy (790 g) and large (20x20x28 cm) items to targets up to 350 cm distance or 180 cm elevation, achieving a 97% real-world success rate.
  arXiv  Detail & Related papers  (2026-09-01T06:00:33Z)
• Hydra-0: Action Flow for Generalist World Modeling and Control [64.88849745875753]
  Hydra-0 is a world model conditioned on action flow, which represents robot actions as pixel motion.<n>Our best configuration achieves 90.4% lower robot-motion error and 60.2% lower object-motion error.<n>A trained action head maps the resulting latent features to executable actions without requiring task-specific expert robot demonstrations.
  arXiv  Detail & Related papers  (2026-08-18T17:59:30Z)
• DemoBridge: A Simulation-in-the-Loop Toolkit for Single-View Human Demonstration Retargeting [49.70827726228219]
  We present DemoBridge, a toolkit that turns a single-view RGB stereo recording of a human hand demonstration into an executable, physics-validated robot-arm trajectory.<n>At the core of DemoBridge is a single collision-aware planner.<n>A physics simulator runs in the loop. It validates each phase as it is produced and backtracks on failure, so a demonstration that cannot be reproduced as given is re-planned rather than discarded.
  arXiv  Detail & Related papers  (2026-07-10T15:31:04Z)
• cuRoboV2: Dynamics-Aware Motion Generation with Depth-Fused Distance Fields for High-DoF Robots [12.825461775215366]
  cuRoboV2 is a motion generation stack that scales from single-arm manipulators to full humanoids.<n>We present three key innovations: (1) B-spline trajectory optimization that enforces smoothness and torque limits; (2) a GPU-native TS/ESDF perception pipeline that generates dense signed distance fields covering the full workspace; and (3) differentiable inverse dynamics map-reduce self-native that achieves up to 61x speedup while also extending to high-DoF humanoids.
  arXiv  Detail & Related papers  (2026-03-05T18:58:04Z)
• Distributed Virtual Model Control for Scalable Human-Robot Collaboration in Shared Workspace [7.44764987825646]
  We present a decentralized, agent agnostic, and safety-aware control framework for human-robot collaboration based on Virtual Model Control (VMC)<n>We demonstrate safe collaboration with up to two robots and two humans, and in simulation up to four robots, maintaining inter-agent separation at around 20 cm.<n>Results show that the method shapes robot behavior intuitively by adjusting control parameters and achieves deadlock-free operation across team sizes in all tested scenarios.
  arXiv  Detail & Related papers  (2026-02-19T14:45:10Z)
• OmniRetarget: Interaction-Preserving Data Generation for Humanoid Whole-Body Loco-Manipulation and Scene Interaction [76.44108003274955]
  A dominant paradigm for teaching humanoid robots complex skills is to retarget human motions as kinematic references to train reinforcement learning policies.<n>We introduce OmniRetarget, an interaction-preserving data generation engine based on an interaction mesh.<n>By minimizing the Laplacian deformation between the human and robot meshes, OmniRetarget generates kinematically feasible trajectories.
  arXiv  Detail & Related papers  (2025-09-30T17:59:02Z)
• FAST: Efficient Action Tokenization for Vision-Language-Action Models [98.15494168962563]
  We propose a new compression-based tokenization scheme for robot actions, based on the discrete cosine transform.<n>Based on FAST, we release FAST+, a universal robot action tokenizer, trained on 1M real robot action trajectories.
  arXiv  Detail & Related papers  (2025-01-16T18:57:04Z)
• The One RING: a Robotic Indoor Navigation Generalist [58.30694487843546]
  RING (Robotic Indoor Navigation Generalist) is an embodiment-agnostic policy that turns any mobile robot into an effective indoor semantic navigator.<n>Trained entirely in simulation, RING leverages large-scale randomization over robot embodiments to enable robust generalization to many real-world platforms.
  arXiv  Detail & Related papers  (2024-12-18T23:15:41Z)
• ARMOR: Egocentric Perception for Humanoid Robot Collision Avoidance and Motion Planning [10.207814069339735]
  ARMOR is a novel egocentric perception system for humanoid robots.<n>Our distributed perception approach enhances the robot's spatial awareness.<n>We show that our ARMOR perception is superior against a setup with multiple dense head-mounted, and externally mounted depth cameras.
  arXiv  Detail & Related papers  (2024-11-30T08:39:23Z)
• Scalable Multi-Robot Informative Path Planning for Target Mapping via Deep Reinforcement Learning [11.134855513221359]
  We propose a deep reinforcement learning approach for the Multi-Robot Informative Path Planning problem.<n>Here, each robot aims to maximize discovered targets, avoid unknown static obstacles, and prevent inter-robot collisions.<n>Our approach outperforms state-of-the-art approaches by at least 26.2% in terms of the number of discovered targets.
  arXiv  Detail & Related papers  (2024-09-25T14:27:37Z)
• Towards Disturbance-Free Visual Mobile Manipulation [11.738161077441104]
  We develop a new disturbance-avoidance methodology at the heart of which is the auxiliary task of disturbance prediction. Our experiments on ManipulaTHOR show that, on testing scenes with novel objects, our method improves the success rate from 61.7% to 85.6%.
  arXiv  Detail & Related papers  (2021-12-17T22:33:23Z)
• Online Body Schema Adaptation through Cost-Sensitive Active Learning [63.84207660737483]
  The work was implemented in a simulation environment, using the 7DoF arm of the iCub robot simulator. A cost-sensitive active learning approach is used to select optimal joint configurations. The results show cost-sensitive active learning has similar accuracy to the standard active learning approach, while reducing in about half the executed movement.
  arXiv  Detail & Related papers  (2021-01-26T16:01:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.