让机器人模仿人类动作时自动修正不合理的姿势，提升实际可行性。
Bi-Level Motion Imitation for Humanoid Robots
- 分两层优化：一边学机器人动作，一边调整原始动作数据
- 生成的参考动作更符合机器人物理限制，仿真中表现更好
- 适合研究人形机器人运动控制与动作迁移的开发者
从人体动作捕捉（MoCap）数据中学习模仿行为为训练人形机器人提供了有前景的路径。然而，由于形态差异，如关节自由度和力矩限制不同，直接复现人类动作对机器人可能不切实际。若在训练数据中包含物理上不可行的动作，将损害机器人策略性能。为此，我们提出一种基于双层优化的模仿学习框架，交替优化机器人策略与目标MoCap数据。首先，采用新型自洽自编码器构建生成式潜在动力学模型，学习稀疏且结构化的运动表征，同时保留数据集中期望的动作模式。该模型用于生成参考动作，其潜在表示则规范双层动作模仿过程。使用真实人形机器人模型的仿真结果表明，本方法通过调整参考动作使其更符合物理约束，显著提升了机器人策略性能。
原文摘要 · Abstract (English)
Imitation learning from human motion capture (MoCap) data provides a promising way to train humanoid robots. However, due to differences in morphology, such as varying degrees of joint freedom and force limits, exact replication of human behaviors may not be feasible for humanoid robots. Consequently, incorporating physically infeasible MoCap data in training datasets can adversely affect the performance of the robot policy. To address this issue, we propose a bi-level optimization-based imitation learning framework that alternates between optimizing both the robot policy and the target MoCap data. Specifically, we first develop a generative latent dynamics model using a novel self-consistent auto-encoder, which learns sparse and structured motion representations while capturing desired motion patterns in the dataset. The dynamics model is then utilized to generate reference motions while the latent representation regularizes the bi-level motion imitation process. Simulations conducted with a realistic model of a humanoid robot demonstrate that our method enhances the robot policy by modifying reference motions to be physically consistent.
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。