WristMimic: Full-Body Humanoid Control with Wrist-Guided Manipulation

2026-07-07

AI Summary

This work addresses the challenge of transferring human demonstration to physical simulation, where simultaneously reproducing body motion, object interaction, and contact forces is difficult, and hand trajectories often lack force information—leading to over-constrained finger tracking. To overcome this, the authors propose a wrist-guided whole-body control framework that treats the wrist as a natural boundary between non-contact and contact behaviors: the body and wrist are driven by kinematic targets, while fingers autonomously learn grasping strategies through object tracking and contact outcomes. The approach incorporates wrist-specific reset constraints and a reward prioritization mechanism, enabling cross-hand manipulation transfer without requiring finger pose supervision. Experiments demonstrate that the method achieves performance on par with or superior to fully supervised finger-tracking approaches across diverse anthropomorphic hands, supporting efficient, hand-agnostic policy transfer.

Abstract

Retargeting human object interaction demonstrations to physics based simulation requires reproducing not only body motion but also the object motion and contacts that make manipulation succeed. However, position only hand trajectories do not specify the contact forces needed to manipulate objects, and directly tracking them can overconstrain contact rich finger behavior. We introduce WristMimic, a wrist guided whole body control framework that explicitly separates contact free body motion from contact rich hand manipulation. The contact free body and wrist are guided by kinematic pose targets, whereas the fingers are not directly supervised by human hand pose. Instead, they learn grasping and manipulation behaviors from object tracking and contact outcomes. Our key insight is that the wrist is the natural gate between these two regimes. It is largely free from contact and can be tracked kinematically, yet it determines the global hand configuration and places the fingers within reachable grasp affordances. To ensure reliable wrist placement during interaction, we introduce wrist specific reset constraints and reward prioritization. Experiments show that WristMimic matches or surpasses methods using full finger pose supervision while enabling finger agnostic retargeting across diverse hand embodiments.

Problem

• motion retargeting
• humanoid control
• object manipulation
• contact modeling
• wrist-guided control

Innovation

• wrist-guided control
• contact-rich manipulation
• finger-agnostic retargeting
• whole-body humanoid control
• object interaction learning