Interconnecting robots using IoT for multi-agent patrolling

Abstract In this paper, we address the decentralized multi-agent patrolling problem for large-scale environments using an Internet of Things (IoT) framework. Our approach leverages a network of stationary and mobile IoT devices to support autonomous patrolling agents in achieving uniform area coverage. We design a lightweight, secure communication protocol enabling one-to-one message exchange with minimal resource usage, allowing agents to share only essential information for coordination and decision-making. We introduce a novel, local, reactive, and decentralized strategy, MRPP-IoT (Multi-agent Reactive Patrolling Protocol for IoT), that operates concurrently across distributed IoT devices. The algorithm is implemented and evaluated in a realistic simulation environment using SUMO and ROS. Extensive experiments demonstrate the algorithm’s effectiveness in achieving uniform patrolling performance and resilience to IoT station failures. Comparative analysis with existing benchmark algorithms shows that MRPP-IoT not only ensures uniform patrolling but also maintains robust performance under conditions of partial IoT infrastructure failure, highlighting its practical applicability for scalable, resilient patrolling in IoT-enabled environments.