2P1-D03 等身大ヒューマノイドの状態表現に基づく失敗検知復帰ルール記述可能な行動動作システム

概要

論文の詳細を見る
It is difficult to archive task sequences in daily life environment for home assistant robots without any faults. In this paper, an integrated robot system toward fault tolerant in task level is presented. In this system, we utilize rule-based, knowledge-based descriptions for error detection and error recovery. Goal States of operators of task planner is used in these descriptions in order to increse reusability of error detection and error recovery rules. The key features of this system are: 1) modular and layered system configuration. 2) error recovery rule and detection rule using state descriptions in task-level. 3) definition of error recovery as list operations to action sequences generated by a task planner.
一般社団法人日本機械学会の論文