Leader–Follower Consensus of Multivehicle Wirelessly Networked Uncertain Systems Subject to Nonlinear Dynamics and Actuator Fault

Abstract

This paper addresses the leader–follower consensus problem of multivehicle wirelessly networked uncertain systems with nonlinear dynamics and actuator fault and proposes a class of distributed discontinuous communication protocols based only on the relative states among neighboring vehicles. By introducing a novel fault model for multivehicle wirelessly networked uncertain systems, fault tolerant consensus can be achieved with different fault modes of the actuators. It is proved in the sense of Lyapunov that, if the conditions of dwell time and the intermittent communication rate are satisfied, the leader–follower consensus can be achieved for closed-loop multivehicle wirelessly networked uncertain systems with nonlinear dynamics and actuator fault under the topology that frequently but not always contains a spanning tree rooted at the leader. Furthermore, the results are extended to the collision avoidance and formulation control problems. Four examples are presented to demonstrate the effectiveness of the proposed approaches. Note to Practitioners —This paper presents a general mechanism of leader–follower consensus for the discontinuous communication of multivehicle wirelessly networked uncertain systems with nonlinear dynamics and actuator fault. The proposed approach guarantees that the consensus can be achieved for the vehicles with nonlinear dynamics and unhealthy actuators. Specifically, the actuator behaviors include three types: miss, outage, and loss of effectiveness. More importantly, different from the existing works, the problems of leader–follower consensus, collision avoidance, and formulation control of multiple vehicles are solved by the proposed protocols, which further demonstrate the ability to extend the service life of the systems with frequent actuators faults. In addition, the proposed intermittent communication strategy also addresses the limited power consumption and the information interaction capability of each sensor.