전역 정보가 제한된 상황에서의 프론티어 기반 분산 탐사 기법

Abstract

We proposes a novel exploration method for unknown areas using swarm robot systems under constrained global information. Swarm robot systems are highly applicable in scenarios such as disaster rescue, extraterrestrial exploration, and polar region exploration, where human access is challenging. In real world applications, limited computational and communicational power lead to constraint of global information of mission space, which makes distributed and cooperative mission planning for exploration challenging. The proposed approach organizes the swarm robot system into a hierarchy of a central unit (e.g. GCS), group leaders, and group members, and frontier-based explo- ration method using multi-group is utilized. By introducing the concepts of global and group frontiers, decentralized and cooperative exploration is possible under limited accessibility to global infor- mation, while minimizing communication between robots. The proposed method was evaluated through numerical simulations and compared with the result of ideal condition, demonstrating effectiveness of the proposed method without significant performance degradation.