让机器人学会功夫和舞蹈，通过物理驱动的动态动作控制。
KungfuBot: Physics-Based Humanoid Whole-Body Control for Learning Highly-Dynamic Skills
- 分步处理动作数据并满足物理约束，提升动作真实性。
- 自适应调整追踪精度，显著降低跟踪误差。
- 适用于高动态动作学习，适合机器人运动控制研究者。
类人机器人有望通过模仿人类行为掌握多种技能。然而，现有算法仅能追踪平滑、低速的人体动作，即使采用精细的奖励设计和课程训练也难以突破。本文提出一种基于物理的类人机器人全身控制框架，旨在通过多阶段动作处理与自适应动作追踪，掌握如功夫、舞蹈等高动态人类行为。在动作处理方面，设计了提取、滤除、修正与重定向动作的流程，最大限度满足物理约束；在动作模仿方面，构建双层优化问题，根据当前追踪误差动态调整追踪容差，实现自适应课程机制。此外，采用非对称的演员-评论家框架进行策略训练。实验中，我们训练了全身控制策略以模仿一系列高动态动作，方法在多个任务上显著优于现有方法，并成功部署于Unitree G1机器人，表现出稳定且富有表现力的行为。项目主页：https://kungfubot.github.io。
原文摘要 · Abstract (English)
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://kungfubot.github.io.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。