HuB: Learning Extreme Humanoid Balance

2025-05-12

AI Summary

This work addresses balance control failure in humanoid robots under extreme postures—such as single-leg standing and high kicks—caused by reference motion errors, morphological mismatch, sensor noise, and unmodeled dynamics. We propose a unified framework integrating online reference motion correction, balance-aware policy learning, and robust simulation-to-reality transfer. Methodologically, we combine kinematic optimization, contact force-aware modeling, adversarial disturbance injection, and domain-randomized reinforcement learning. Evaluated on the Unitree G1 platform, our approach achieves stable execution of highly challenging motions—including “Swallow Balance” and “Bruce Lee’s Kick”—and maintains robust upright posture under strong external disturbances (e.g., soccer-style kicks). Experiments demonstrate substantial improvements over existing baselines, marking the first systematic breakthrough in overcoming three fundamental bottlenecks: morphological mismatch, error accumulation, and the sim-to-real gap.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Adversarial Learning & Robustness
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

The human body demonstrates exceptional motor capabilities-such as standing steadily on one foot or performing a high kick with the leg raised over 1.5 meters-both requiring precise balance control. While recent research on humanoid control has leveraged reinforcement learning to track human motions for skill acquisition, applying this paradigm to balance-intensive tasks remains challenging. In this work, we identify three key obstacles: instability from reference motion errors, learning difficulties due to morphological mismatch, and the sim-to-real gap caused by sensor noise and unmodeled dynamics. To address these challenges, we propose HuB (Humanoid Balance), a unified framework that integrates reference motion refinement, balance-aware policy learning, and sim-to-real robustness training, with each component targeting a specific challenge. We validate our approach on the Unitree G1 humanoid robot across challenging quasi-static balance tasks, including extreme single-legged poses such as Swallow Balance and Bruce Lee's Kick. Our policy remains stable even under strong physical disturbances-such as a forceful soccer strike-while baseline methods consistently fail to complete these tasks. Project website: https://hub-robot.github.io

Problem

• Overcoming instability from human motion tracking errors
• Addressing learning difficulties due to morphological mismatch
• Bridging sim-to-real gap from sensor noise and dynamics

Innovation

• Reference motion refinement for stability
• Balance-aware policy learning adaptation
• Sim-to-real robustness training framework