Decentralized Leader-Following Control of a Heterogeneous Swarm Robotic System Subject to Network-Induced Imperfections

Abstract

This paper studies the decentralized leader-following control of a heterogeneous swarm robotic system composed of a group of quadrotors as aerial agents, and a group of two-wheeled mobile robots as ground agents. The interaction mechanism between the agents is based on the method of artificial potential fields. While the agents in the aerial and ground sub-swarms communicate with each other through local and direct vision, the information exchange between the aerial and ground sub-swarms is established through a wireless network. The control system of ground and aerial agents is designed using the sliding mode control technique. The controller proposed for quadrotors is finite-time, integral-terminal, and valid for both small and large angles of attitude. The effect of the communication imperfections of the wireless network on the leader-following behavior of the heterogeneous robotic swarm is investigated. Subsequently, an innovative component, called the trajectory predictor, is designed to compensate for the undesirable impacts caused by the communication imperfections of the wireless network. Finally, the performance of the presented collective control strategy is evaluated via several numerical simulations.