Fault Tolerant Co-operative Formation Control of Multiple Unmanned Aerial Vehicles

Abstract

In this research, we have studied leader-following formation control problem for a group of networked Unmanned aerial vehicles in the presence of the actuator faults and external disturbances under fixed network topology, where the exosystem dynamics are not accessible by all agents. The cooperative control problem for the linear multi-agent system has been generalized to the leader following formation control of multiple Unmanned aerial vehicles. The network of multi-UAV systems comprises of a leader UAV and multiple follower UAVs, among them only a group of follower UAVs has access to exogenous signals. Therefore, information exchange among the UAVs is restricted and cannot be implemented by decentralized approach and also cannot be solved by centralized control. This research discusses the formulation of a distributed observer for the formation control. The considered exosystem in our problem is used to generate both the tracking signal and the disturbance for rejection, with dynamics not similar to those of the UAVs. The UAVs are required to track the exosystem generated trajectory and maintain their formation geometry by keeping a relative distance to the leader UAV through a dynamic distributive output control law. The Faulttolerant Co-operative control (FTCC) scheme is exploited for accommodation of the actuator faults. In fault free cases, distributive feedback control is implemented on the group of UAVs for the formation control. When one or more UAVs of the team encounters a fault, the FTCC scheme is activated and two cases are considered:1) if follower UAV encounters a severe fault and faulty UAV is not capable of completing its assigned mission while maintaining the desired formation geometry. In this scenario, the faulty UAV get separated from the formation and the mission is assigned to the remaining healthy UAVs. 2) if faulty UAV has capability to continue its mission with the degraded performance, then remaining follower UAVs reconfigure their controllers while taking into account the severity of fault in the faulty UAV. Within the proposed scheme, the two-stage Kalman filter based fault detection and diagnosis unit (FDD) is utilized for detection and diagnosis of actuator faults. Simulation results are presented to show effectiveness of the proposed fault tolerant control scheme.