Cross-Embodiment Robot Manipulation via a Unified Hand Action Space

2026-07-03

AI Summary

This work proposes a unified hand action space (UHAS) to overcome the limited transferability of existing robotic manipulation policies across diverse hand morphologies. By modeling hand actions as geometric deformations of a canonical sphere and mapping them to individual hand joint spaces via cascaded inverse kinematics (CIK), UHAS establishes the first general-purpose action representation compatible with multiple dexterous hands—including the Allegro Hand, LEAP Hand, Shadow Hand, and the MANO human hand model. The framework enables zero-shot policy transfer and rapid fine-tuning across hand configurations. Experimental results demonstrate that the proposed approach achieves efficient and robust cross-morphology dexterous manipulation in both simulation and real-world settings, significantly enhancing policy generalization and deployment efficiency.

Abstract

Robot manipulation policies are typically tied to specific robotic hand embodiments, limiting the transfer of learned behaviors across platforms with different kinematic structures. In this work, we propose the Unified Hand Action Space (UHAS), a sphere-based unified action representation for cross-embodiment dexterous manipulation. UHAS represents robotic hand actions as geometric deformations of a canonical sphere and uses a Cascade Inverse Kinematics (CIK) algorithm to map the shared representation to embodiment-specific joint configurations. Using reinforcement learning, we train dexterous manipulation policies directly in the proposed action space for in-hand cube reorientation tasks. We evaluate our method in both simulation and real-world experiments across multiple robotic hands, including the Allegro Hand, LEAP Hand, Shadow Hand, and MANO Human Hand. Experimental results demonstrate effective dexterous manipulation, zero-shot transfer to unseen hands, rapid finetuning across embodiments, and successful real-world deployment. Our experiments show that the proposed UHAS representation enables stable dexterous control and cross-embodiment policy transfer across robotic hands.

Problem

• cross-embodiment
• robot manipulation
• dexterous manipulation
• policy transfer
• hand embodiment

Innovation

• Unified Hand Action Space
• Cross-embodiment
• Dexterous Manipulation
• Cascade Inverse Kinematics
• Zero-shot Transfer