実演からタスク仕様を確率的オートマトンとして学習する手法
Learning Task Specifications from Demonstrations as Probabilistic Automata
ロボットの実演からサブゴールとその時間的依存関係を確率的決定性有限オートマトン(PDFA)として学習し、解釈可能なタスク仕様を効率的に獲得する手法を提案した。物体操作タスクで多様な専門家戦略の再現と環境変化への適応を実証した。
著者: Mattijs Baert, Sam Leroux, Pieter Simoens
分類: cs.RO
原文アブストラクト
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.
