マルチロボットシステムにおける知識：ダイナミクス・計算・通信の相互作用
Knowledge in multi-robot systems: an interplay of dynamics, computation and communication
連続・離散ハイブリッド系を時相認識論理でモデル化し、マルチロボットの探査・集合タスクが解けるための認識的条件を導出する枠組みを提案した。
著者: Giorgio Cignarale, Stephan Felber, Eric Goubault, Bernardo Hummes Flores, Hugo Rincon Galeana
分類: cs.LO, cs.DC, cs.RO
原文アブストラクト
In this paper, we provide a framework integrating distributed multi-robot systems and temporal epistemic logic. We show that continuous-discrete hybrid systems are compatible with logical models of knowledge already used in distributed computing, and demonstrate its usefulness by deriving sufficient epistemic conditions for exploration and gathering robot tasks to be solvable. We provide a separation of the physical and computational aspects of a robotic system, allowing us to decouple the problems related to each and directly use methods from control theory and distributed computing, fields that are traditionally distant in the literature. Finally, we demonstrate a novel approach for reasoning about the knowledge in multi-robot systems through a principled method of converting a switched hybrid dynamical system into a temporal-epistemic logic model, passing through an abstract state machine representation. This creates space for methods and results to be exchanged across the fields of control theory, distributed computing and temporal-epistemic logic, while reasoning about multi-robot systems.
