通过上千种机器人形态训练，实现跨形态零样本迁移。
Towards Embodiment Scaling Laws in Robot Locomotion
- 用程序生成1000种不同结构的机器人形态进行训练。
- 形态数量越多，新形态上的泛化能力越强，超越数据量提升。
- 在仿真和真实机器人上均实现零样本迁移，适合可重构机器人研究。
跨形态泛化是构建通用具身智能体的核心，但其关键因素仍不明确。本文以机器人运动为场景，探究形态规模定律——即增加训练形态数量可提升对未见形态的泛化能力。我们程序化生成约1000种具有拓扑、几何和关节级运动学差异的机器人形态，并在随机子集上训练策略。实验显示正向规模效应支持该假设，且形态扩展带来的泛化能力远超固定形态下的数据量增加。最佳策略在完整数据集上训练后，可零样本迁移到仿真与真实世界的新形态，包括Unitree Go2和H1。这些成果推动了通用具身智能的发展，对可配置机器人的自适应控制、形态协同设计等具有重要意义。
原文摘要 · Abstract (English)
Cross-embodiment generalization underpins the vision of building generalist embodied agents for any robot, yet its enabling factors remain poorly understood. We investigate embodiment scaling laws, the hypothesis that increasing the number of training embodiments improves generalization to unseen ones, using robot locomotion as a test bed. We procedurally generate ~1,000 embodiments with topological, geometric, and joint-level kinematic variations, and train policies on random subsets. We observe positive scaling trends supporting the hypothesis, and find that embodiment scaling enables substantially broader generalization than data scaling on fixed embodiments. Our best policy, trained on the full dataset, transfers zero-shot to novel embodiments in simulation and the real world, including the Unitree Go2 and H1. These results represent a step toward general embodied intelligence, with relevance to adaptive control for configurable robots, morphology co-design, and beyond.
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。