Scalable, fault-tolerant and distributed multi-robot patrol in real world environments

Abstract

Despite the focus that multi-robot patrolling has received recently, there is a manifest lack of practical real-world implementations of such systems. Beyond that, the existing ones have been mainly focused on centralized policies to coordinate the team of agents. The present work addresses realistic patrol in indoor environments with teams of arbitrary number of autonomous robots performing a distributed, scalable and fault-tolerant strategy for multi-robot coordination in patrolling missions. Agents decide their actions locally and adapt to the system's needs using distributed communication. The work is validated through experiments in a large indoor real-world environment with a team of autonomous mobile robots.