Learn Weightlessness: Imitate Non-Self-Stabilizing Motions on Humanoid Robot

TL;DR

This paper introduces a biologically-inspired method called Weightlessness Mechanism (WM) that enables humanoid robots to adaptively interact with environments during non-self-stabilizing motions, improving stability and generalization.

Contribution

It proposes a novel weightlessness-state auto-labeling strategy and WM method for dynamic joint relaxation, enhancing environmental interaction in humanoid control without task-specific tuning.

Findings

WM achieves strong generalization across diverse environments

The approach maintains motion stability during complex interactions

Effective in simulation and real robot experiments

Abstract

The integration of imitation and reinforcement learning has enabled remarkable advances in humanoid whole-body control, facilitating diverse human-like behaviors. However, research on environment-dependent motions remains limited. Existing methods typically enforce rigid trajectory tracking while neglecting physical interactions with the environment. We observe that humans naturally exploit a "weightless" state during non-self-stabilizing (NSS) motions--selectively relaxing specific joints to allow passive body--environment contact, thereby stabilizing the body and completing the motion. Inspired by this biological mechanism, we design a weightlessness-state auto-labeling strategy for dataset annotation; and we propose the Weightlessness Mechanism (WM), a method that dynamically determines which joints to relax and to what level, together enabling effective environmental interaction while executing target motions. We evaluate our approach on 3 representative NSS tasks: sitting on chairs of varying heights, lying down on beds with different inclinations, and leaning against walls via shoulder or elbow. Extensive experiments in simulation and on the Unitree G1 robot demonstrate that our WM method, trained on single-action demonstrations without any task-specific tuning, achieves strong generalization across diverse environmental configurations while maintaining motion stability. Our work bridges the gap between precise trajectory tracking and adaptive environmental interaction, offering a biologically-inspired solution for contact-rich humanoid control.