从示范中学习可解释的任务规范，让机器人自动理解复杂操作流程。
Learning Task Specifications from Demonstrations as Probabilistic Automata
- 通过演示数据构建概率确定性有限自动机，捕捉任务结构与专家偏好
- 能识别子目标及其时间依赖关系，支持长时序任务建模
- 结果可被领域专家理解调整，适合需要灵活适应的机器人任务
为机器人系统指定任务通常需要编程技能、深厚领域知识和大量时间投入。虽然示范学习提供了一种有前景的替代方案，但现有方法在处理长时程任务时仍面临挑战。为此，本文提出一种计算高效的算法，直接从示范中学习概率确定性有限自动机（PDFA），以捕获任务结构与专家偏好。该方法能够推断出子目标及其时间依赖关系，生成可解释的任务规范，使领域专家易于理解与调整。我们在物体操作任务中进行了实验验证，结果表明该方法能使机械臂有效复现多样化的专家策略，并在条件变化时保持适应性。
原文摘要 · Abstract (English)
Specifying tasks for robotic systems traditionally requires coding expertise, deep domain knowledge, and significant time investment. While learning from demonstration offers a promising alternative, existing methods often struggle with tasks of longer horizons. To address this limitation, we introduce a computationally efficient approach for learning probabilistic deterministic finite automata (PDFA) that capture task structures and expert preferences directly from demonstrations. Our approach infers sub-goals and their temporal dependencies, producing an interpretable task specification that domain experts can easily understand and adjust. We validate our method through experiments involving object manipulation tasks, showcasing how our method enables a robot arm to effectively replicate diverse expert strategies while adapting to changing conditions.
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。