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

2026-04-23

AI Summary

This work addresses the limitations of existing imitation learning approaches, which overly emphasize trajectory tracking and struggle to enable humanoid robots to naturally interact with their environment during non-self-stable motions. Inspired by humans’ ability to exploit “weightless” states to execute complex actions, the authors propose a dynamic joint relaxation mechanism (WM) that automatically identifies weightless phases and selectively relaxes joint constraints to facilitate adaptive physical interaction. By integrating imitation learning with reinforcement learning, the method requires only a single demonstration for training. Experiments on the Unitree G1 robot demonstrate that the system can stably perform diverse non-self-stable tasks—such as sitting, lying, and leaning—across varied environments without task-specific fine-tuning, highlighting its strong generalization capability.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

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.

Problem

• weightlessness
• non-self-stabilizing motions
• humanoid robot
• environmental interaction
• imitation learning

Innovation

• weightlessness mechanism
• non-self-stabilizing motions
• environmental interaction
• humanoid whole-body control
• imitation learning