Cooperative fault-tolerant tracking control of heterogeneous hybrid-order mechanical systems with actuator and amplifier faults

Abstract

This paper studies the problem of leader–follower fault-tolerant tracking control for a class of heterogeneous multi-agent mechanical systems (MAMS) with actuator and amplifier faults. The MAMS are supposed to have hybrid agents with heterogeneous first- and second-order nonlinear dynamics. The leader generates the reference of position and velocity, and only part of the agents have direct connections to the leader. All the following agents can receive the messages or information only from the connecting neighbors according to the topology of directed communication graph with fixed structure. Also, the fault-tolerant tracking problem is investigated for hybrid-order MAMS. In addition to the faults of the actuator effectiveness, the time-varying nonlinear amplification faults of the servo-amplifier are considered in particular. The ultimate synchronization of all agents with hybrid dynamics is realized under the influences of these two types of faults; a novel distributed cooperative fault-tolerant tracking controller is developed. The communication topology graph needs to contain both position and velocity spanning trees to guarantee the transmission of states information. Neural networks and some adaptive mechanisms are adopted in the proposed controller to estimate unknown nonlinear dynamics and faults. Lyapunov theory is used to prove the ultimate boundedness of the synchronization tracking errors. The correctness and effectiveness of the proposed cooperative fault-tolerant tracking controller are verified by a simulation example. The results show that the simulated hybrid-order MAMS are stable and the unknown faults of actuator and servo-amplifier can be effectively handled by the developed distributed controller.