Research On H_∞ Fault-tolerant Consistency Control Of Multi-agent Systems Under Several Communication Topologies

With the rapid development of modern industry and Internet technology,multi-agent systems(MAS)has been widely used in the field of control engineering,in which consensus control is the basic problem in the research process of MAS.Due to the characteristics of information interaction,communication fusion and free cooperation between agents,MAS has uncertain topology,packet loss of channel link data and communication barrier during operation,resulting in different forms of communication topology in the system.At the same time,MAS is more prone to failure than single system during operation.Once a fault occurs,it will affect the stability and reliability of the system.In view of this,this thesis studies the fault-tolerant consensus control problem with actuator faults and external bounded energy disturbances for MAS under different communication topologies.The main contents of this thesis is as follows:1)Based on linear leaderless MAS with uncertain or packet loss communication topologies,the problem of robust H_∞fault-tolerant consensus control for systems with bounded energy disturbances and varying degrees of failure of arbitrary agent actuators is studied.Firstly,by introducing a weight matrix to linearly transform the system model,the fault-tolerant consensus control problem of uncertain or packet loss communication topology MAS is transformed into a robust asymptotic stability problem.Then,fault-tolerant controllers for uncertain and packet loss communication topologies MAS is designed respectively.Finally,the Lyapunov stability theory and linear matrix inequality(LMI)method are used to solve the control protocol parameters that make the MAS state asymptotically uniform.It is deduced that the proposed method satisfies the MAS stability and has certain disturbance suppression capabilities.2)Based on nonlinear leader-adjoint MAS in a semi-Markov jump communication topology,the problem of robust H_∞fault-tolerant consensus control for systems with external bounded disturbances and actuator failures is studied.Firstly,in order to improve the stability and convergence speed of MAS,a consensus control protocol based on the current and past state information of the attendant is designed,which ultimately converts the consensus problem of the leader-attendant MAS into the stability problem of the error system.Secondly,by using Lyapunov stability theory and LMI method,consensus the upper and lower bounds of transfer rate,gain matrices satisfying different communication topologies are derived.Finally,a sufficient condition is derived for the semi-Markov jump communication topology nonlinear leader-adjoint MAS to satisfy robust H_∞mean square consensus fault-tolerant control in the presence of actuator failures.3)Based on the linear leader-adjoint MAS under the average dwell time switching communication topology,the problem of fault diagnosis and robust H_∞fault-tolerant consensus control for system actuators under sudden,time-varying,and external bounded disturbances is studied.Firstly,considering the presence of external energy disturbances and actuator faults in each of the accompanying actuators,a class of fault observers for estimating actuator fault information in the accompanying actuators is proposed.Secondly,based on the estimated fault information,a fault-tolerant compensation controller is designed for the system under the average dwell time switching communication topology.Finally,it is inferred that MAS can maintain consistency in the state information of all agents while meeting the expected H_∞performance indicators.