KungfuBot: Physics-Based Humanoid Whole-Body Control for Learning Highly-Dynamic Skills

TL;DR

This paper introduces a physics-based control framework enabling humanoid robots to learn and perform highly-dynamic skills like Kungfu and dancing through adaptive motion processing and imitation, surpassing existing methods in accuracy.

Contribution

The paper presents a novel control framework with adaptive curriculum and bi-level optimization for learning highly-dynamic humanoid motions, which was not addressed by prior work.

Findings

Achieves lower tracking errors than existing methods.

Successfully deploys on Unitree G1 robot with stable, expressive behaviors.

Demonstrates capability to imitate complex dynamic motions like Kungfu.

Abstract

Humanoid robots are promising to acquire various skills by imitating human behaviors. However, existing algorithms are only capable of tracking smooth, low-speed human motions, even with delicate reward and curriculum design. This paper presents a physics-based humanoid control framework, aiming to master highly-dynamic human behaviors such as Kungfu and dancing through multi-steps motion processing and adaptive motion tracking. For motion processing, we design a pipeline to extract, filter out, correct, and retarget motions, while ensuring compliance with physical constraints to the maximum extent. For motion imitation, we formulate a bi-level optimization problem to dynamically adjust the tracking accuracy tolerance based on the current tracking error, creating an adaptive curriculum mechanism. We further construct an asymmetric actor-critic framework for policy training. In experiments, we train whole-body control policies to imitate a set of highly-dynamic motions. Our method achieves significantly lower tracking errors than existing approaches and is successfully deployed on the Unitree G1 robot, demonstrating stable and expressive behaviors. The project page is https://kungfu-bot.github.io.