Multi-robot patrol: A distributed algorithm based on expected idleness

摘要

Multi-robot with advantages of spatial distribution and fault tolerance is competent for patrol missions and has the potential to be used in security and surveillance applications. This article focuses on the frequency-based patrol designed to guarantee the frequent access to key positions in the environme nt. A distributed algorithm based on expected idleness is proposed, aiming to promote the efficiency of cooperation, which remains to be fault tolerant and scalable. The expected idleness is estimated with information shared between robots and utilized to avoid conflicts in the decision process. Comparisons with state-of-the-art algorithms have been conducted in a realistic simulator, Stage; moreover, the fault tolerance and scalability have also been tested. Experiments on real robots have further verified the applicability of the proposed algorithm.