让机器人通过玩耍学会组合技能，边做边改错。
Symskill: Symbol and Skill Co-Invention for Data-Efficient and Reactive Long-Horizon Manipulation
- 从无标注演示中同时学符号和动作技能
- 12步任务成功率85%，无需额外数据
- 适合需要灵活应变的现实机器人操作
动态环境中多步操作仍具挑战。模仿学习（IL）反应快但缺乏组合泛化能力，因整体策略无法在场景变化时决定复用哪个技能。经典任务与运动规划（TAMP）虽具组合性，但规划延迟高，难以实时故障恢复。我们提出SymSkill，一种统一框架，从无标签、未分割的演示中联合学习谓词、操作符和技能，兼顾组合泛化与实时恢复。离线阶段，SymSkill直接从演示中学习符号抽象和目标导向技能；在线阶段，给定已学谓词的合取，它使用符号规划器组合并重排技能以达成符号目标，并在运动和符号层面实时处理失败。结合柔顺控制器，支持在人和环境扰动下安全执行。在RoboCasa仿真中，对12个单步任务成功率达85%，并可组合成多步计划而无需额外数据。在真实Franka机械臂上，仅需5分钟玩耍数据即可完成12步任务。代码与补充分析见https://symskill.github.io/。
原文摘要 · Abstract (English)
Multi-step manipulation in dynamic environments remains challenging. Imitation learning (IL) is reactive but lacks compositional generalization, since monolithic policies do not decide which skill to reuse when scenes change. Classical task-and-motion planning (TAMP) offers compositionality, but its high planning latency prevents real-time failure recovery. We introduce SymSkill, a unified framework that jointly learns predicates, operators, and skills from unlabeled, unsegmented demonstrations, combining compositional generalization with real-time recovery. Offline, SymSkill learns symbolic abstractions and goal-oriented skills directly from demonstrations. Online, given a conjunction of learned predicates, it uses a symbolic planner to compose and reorder skills to achieve symbolic goals while recovering from failures at both the motion and symbolic levels in real time. Coupled with a compliant controller, SymSkill supports safe execution under human and environmental disturbances. In RoboCasa simulation, SymSkill executes 12 single-step tasks with 85% success and composes them into multi-step plans without additional data. On a real Franka robot, it learns from 5 minutes of play data and performs 12-step tasks from goal specifications. Code and additional analysis are available at https://symskill.github.io/ .
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。