Distributed consensus of linear multi-agent systems with fault tolerant control protocols

In this paper, the problem of fault tolerant consensus control of linear multi-agent systems is investigated using two control protocols, namely, the fixed-gain control protocol and the adaptive-gain control protocol. In the first part, a fixed-gain control protocol is proposed which guarantees the consensus under all possible actuator faults. Also a two-step constructive algorithm of the fixed-gain control protocol is summarized. In the second part, the failure rate-based adaptive control protocol is developed to compensate for the failure effects on multi-agent consensus. Instead of the fixed gain, the adaptive gain in the control protocol is time varying with respective to the online estimation of failure rates, which follows the projection operator adaptive law. Moreover, a criterion is derived for reaching consensus under the adaptive-gain control protocol when the multi-agent systems are subjected to actuator faults. Several procedures to construct the adaptive-gain control protocol are given. Simulation examples are finally included to illustrate the theoretical results.