Distributed Adaptive Fault-Tolerant Containment Controller Design For Multi-Agent Systems(MASs)

In recent years,the distributed control problem of multi-agent systems(MASs)has become a hot research topic in control theory,mainly due to the wide application of MASs in many fields,such as the formation control of unmanned aerial vehicle,swarming control,and distributed sensor network.For the cooperative control problem of multi-agent systems(MASs),it usually assumes that the system only has a leader agent even no leader agent.However,in some practical applications there are multiple leaders.Therefore,it is of great theoretical significance and application value to investigate the problem of containment control for the system containing multiple leaders.In this thesis,the containment control problem of multi-agent system with actuator faults is studied.Different fault-tolerance control algorithms are developed via state feedback and output feedback.Some simulation results are given for an aircraft model parameter of F18 developed by NASA,which further illustrate the effectiveness and rationality.The main work is as follows:In Chapter 3,the agent models in MASs subject to actuator failures and a class of nonlinear function are considered.At the same time,parameter uncertainties and external disturbance coexist in the considered MASs.Moreover,the smallest eigenvalue of Laplacian matrix representing system global information is also unknown.For the considered system,a completely distributed adaptive fault-tolerant containment control scheme is proposed and the containment control problem of system is also effectively solved.Chapter 4 investigates a class of multiple leaders MASs with unknown system state and bias faults of actuator.Because the bias faults of actuator are time-varying,system global information and system state are unknown,a reasonable adaptive observer is firstly designed.Based on the designed observer,the distributed adaptive fault tolerant controller is finally developed.The simulation results show that the adaptive observer can estimate the system states and unknown parameters quickly and accurately,and the designed controller can effectively solve the problem of fault-tolerance containment control with bias faults.In Chapter 5,the agent models in MASs subject to actuator failures,actuator bias faults and unknown system state are considered.The fault model studied in this chapter is more general than in Chapter 4.The actuator failures of system considered in this chapter are time-varying and system state is unknown,which brings a great challenge for the controller design.Thus,an adaptive fault observer is firstly designed.At the same time,in order to avoid the high frequency oscillation caused by the observer,a low pass filter is introduced.Based on the adaptive fault observer and the low pass filter,the failure parameters can be quickly and accurately estimated.By using the estimated information,an adaptive fault-tolerant containment controller is developed,which can make sure that all estimation errors eventually convergence to zero and all followers converge to a convex region formed by the leaders.Some conclusions are presented at the end of this thesis and the further research direction is discussed.