arXiv cs.AI· Botao He, Yiming Dou, Yiannis Aloimonos, Jitendra Malik, Akash Sharma, Ruoshi Liu·· 6 小时前AI 评分42
PlayPolicy：从人类监督式玩耍中学出机器人灵巧操作
PlayPolicy: Dexterity from Supervised Play
AI 导读
arXiv 论文提出 PlayPolicy 框架，机器人通过行为克隆从几分钟人类玩耍数据学习 play policy，自主采集数小时交互数据训练 world model，并用其引导 MPC 推理。
正文 · 原文
Abstract:How can robots autonomously collect the contact-rich experience needed to learn dexterous manipulation? We present PlayPolicy: Dexterity from supervised play (DSP), a framework in which a robot learns a play policy from minutes of human play through behavior cloning, then deploys it autonomously to collect hours of interaction data to train a world model. During inference, the same play policy then guides model-predictive control (MPC) by proposing candidate action sequences for evaluation under the learned world model. To accelerate inference, we also introduce a method for distilling a closed-loop, goal-conditioned policy through self-play within the learned world model. We evaluate this framework on underactuated pushing and any-to-any in-hand pose reorientation with a multifingered hand, in simulation and real world. In real world experiment, our method outperforms hindsight behavior cloning by 69.6% in success rate and achieves comparable success to reinforcement learning with 490 times less interaction data. These results demonstrate that supervised play can autonomously collect interaction data, guide model-based planning, and enable policy learning without any task-specific demonstrations or sim-to-real transfer. Project page: this https URL
| Comments: | Project page: this https URL |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.22340 [cs.RO] |
| (or arXiv:2610.22340v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22340 arXiv-issued DOI via DataCite |
Submission history
From: Botao He [view email]
[v1]
Thu, 15 Oct 2026 17:55:50 UTC (33,259 KB)
[v2]
Mon, 26 Oct 2026 07:47:05 UTC (33,259 KB)
来源：arXiv cs.AI · arxiv.org