UniCross: Unified Cross-Skill Dexterous Manipulation Synthesis

2026-07-30

AI Summary

Existing dexterous manipulation approaches typically model skills such as grasping, repositioning, in-hand rotation, and translation in isolation, hindering their seamless integration for long-horizon, complex tasks. This work proposes the first unified framework that integrates these four skill categories into a shared state-action space with a common goal representation. By leveraging a unified objective function, policy distillation, and cross-morphology generalization, the framework enables end-to-end learning of a single policy capable of mastering all target skills. The approach reveals intrinsic consistency among diverse manipulation behaviors, yielding a policy that achieves strong performance across all skills, generalizes to unseen objects, exhibits robustness to disturbances, supports seamless chaining of skills, and effectively transfers to different robotic hand morphologies.

Abstract

Many dexterous manipulation tasks require the object to remain securely held throughout the interaction. From the perspective of hand-object relational motion, such manipulation comprises four canonical skills: grasping, relocation, in-hand rotation, and in-hand translation. Human hands flexibly compose these skills to accomplish complex tasks. Existing approaches, however, model these skills separately with skill-specific action constraints, objectives, or even dedicated hand morphologies, which breaks the compatibility and continuity required for long-horizon composition. In this work, we present a unified framework that models all four skills in a single formulation that shares the same state and action spaces and a common objective structure. This formulation enables straightforward distillation of a single cross-skill policy that performs strongly on every skill, generalizes to unseen objects, stays robust to disturbances, and chains skills seamlessly into long-horizon manipulation. The framework also transfers effectively across different hand morphologies. Overall, our results suggest that different dexterous manipulation skills can be viewed as instantiations of a shared task formulation, revealing the intrinsic consistency across different behaviors.

Problem

• dexterous manipulation
• cross-skill composition
• unified framework
• hand-object interaction
• skill generalization

Innovation

• unified framework
• dexterous manipulation
• cross-skill policy
• skill composition
• hand-object interaction