让仿人机器人学会模仿人类多样动作，还能复用学习成果。
MuGen: Multi-Skill Generative Locomotion Controller for Humanoid Robots
- 用生成式编码捕捉人类运动模式，支持多技能动作学习。
- 训练出可部署的策略模型，能精准复现未见过的人类动作。
- 适合需要灵活运动能力的机器人研究与应用开发。
本文提出MuGen，一种数据驱动的框架，用于在仿人机器人上学习和部署多技能运动能力。该框架使机器人能在示例动作序列引导下，实现类人的富有表现力的运动。通过基于模型的强化学习训练向量量化自编码器（VQ-VAEs），从数小时异构人类表演数据中提取人类运动的关键模式，获得高效的生成式运动表征。采用教师-学生学习框架，并设计新型策略蒸馏方法，使可部署的学生策略能够学习这一高效隐空间表示。该策略不仅可追踪并模仿未见过的人类动作，还支持将学习到的隐空间复用于其他任务。我们在多种运动类型上验证了框架的有效性，实现了高精度执行。
原文摘要 · Abstract (English)
This paper presents MuGen, a data-driven framework for learning and deploying multi-skill locomotion on humanoid robots. MuGen enables a robot to perform expressive motions like humans under the guidance of example motion sequences. To achieve this, we employ vector-quantized autoencoders (VQ-VAEs) trained with model-based reinforcement learning, resulting in a generative representation of locomotion that captures key patterns of human motion from hours of heterogeneous human performance data. We employ a teacher-student learning framework and develop a new policy distillation strategy to enable a deployable student policy learning this efficient latent representation. This policy allows the robot to track and mimic unseen human motions and further enables the robot to reuse the learned latent space for other tasks. We demonstrate the effectiveness of our framework through a diverse set of motions and accurate execution.
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。