用长示范数据分解学习基础动作，实现通用布料操作。
Learning Generalizable Language-Conditioned Cloth Manipulation from Long Demonstrations
- 基于大模型发现并学习演示中的基础动作
- 在未见任务上成功率提升32%，优于基线方法
- 适合需要泛化能力的机器人布料操作场景
多步骤布料操作因状态空间高维且布料动态复杂而极具挑战。尽管端到端模仿学习在多步骤布料操作方面取得进展，但现有方法难以泛化到未见任务。本文提出通过分解思路解决该问题：从长示范数据中自动学习基础技能，并利用大语言模型（LLM）进行高层任务规划，组合已学技能完成未见任务。具体而言，先借助大语言模型的常识知识从现有长示范基准中发现并学习基础技能；再通过高阶LLM任务规划器将这些基础技能组合以完成新任务。实验表明，该方法在已见与未见任务上的多步骤布料操作学习性能均优于基线方法。
原文摘要 · Abstract (English)
Multi-step cloth manipulation is a challenging problem for robots due to the high-dimensional state spaces and the dynamics of cloth. Despite recent significant advances in end-to-end imitation learning for multi-step cloth manipulation skills, these methods fail to generalize to unseen tasks. Our insight in tackling the challenge of generalizable multi-step cloth manipulation is decomposition. We propose a novel pipeline that autonomously learns basic skills from long demonstrations and composes learned basic skills to generalize to unseen tasks. Specifically, our method first discovers and learns basic skills from the existing long demonstration benchmark with the commonsense knowledge of a large language model (LLM). Then, leveraging a high-level LLM-based task planner, these basic skills can be composed to complete unseen tasks. Experimental results demonstrate that our method outperforms baseline methods in learning multi-step cloth manipulation skills for both seen and unseen tasks.
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。