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.

翻译：可扩展的机器人模仿学习依赖于来自不同机器人的大规模异构数据或无实体数据，这使得笛卡尔末端执行器动作成为具身无关策略学习的关键接口。然而，仅基于末端执行器的抽象使笛卡尔策略无法感知部署机器人的本体，导致其在全身避碰等机器人特定约束下表现脆弱。为克服这一局限，我们提出EmbodiSteer——一种无需训练的框架，可将具身无关的视觉运动策略导向零样本、具身感知的部署。EmbodiSteer保持策略学习在笛卡尔空间进行，同时通过正向运动学和基于雅可比矩阵的更新，高效地将推理时的扩散采样提升至目标机器人的关节空间。在每个去噪步骤后，通过针对关节轨迹的全身碰撞感知引导，可在保持已习得的末端执行器行为的同时，使机械臂避开碰撞。与纯笛卡尔执行相比，EmbodiSteer在9个仿真机器人上将碰撞率降低46.1%，任务成功率提升28.5%；在高约束场景下的两个实体机器人上，碰撞率进一步降低90.0%，成功率提升36.7%。项目页面：https://frankwang67.github.io/EmbodiSteer-Page.