Reasearch On Adaptive Control For Multi-agent System With Actuator Failure

Multi-agent system is a distributed independent system composed of multiple agents.It can solve the problems that can not be solved by a single agent.The application of multi-agent system in practical engineering is more and more extensive,and the probability of failure in the system also increases.When the actuator fails,it will affect the stability of the system.In multi-agent system,the design of each agent controller depends on the state of itself and its neighbors,the consensus control of multi-agent system is always one of the research hotspots.In this paper,several types of adaptive DSC strategies are proposed for several types of multi-agent systems with actuator failures.Firstly,for a class of nonlinear multi-agent systems with input quantization,unmodeled dynamics and actuator failure,a measurable dynamic signal is introduced to eliminate the influence of unmodeled dynamics on the system.Using Young's inequality and the properties of Gaussian function to handle the coupling effect of multi-agent neighbor nodes on subsystems in the first step of design.Representing the hysteresis quantizer as an input linear function with,bounded coefficients and bounded disturbances,and using the DSC method,an adaptive network DSC scheme is proposed,which simplifies the design of the controller and ensures that all signals of the closed-loop system are semi globally uniformly ultimately bounded,and all followers can achieve the desired consensus.Finally,a simulation example is used to verify the feasibility of the control scheme designed in this paperSecondly,the consensus tracking problem for a class of nonlinear multi-agent systems(MASs)with output constraints,unmodeled dynamics,nonsymmetric input dead-zones and actuator failures under directed graphs is discussed,and proposes an adaptive cooperative neural DSC strategy.Using the properties of invertible nonlinear mapping and Gaussian function to deal with output constraints and non-strict feedback terms separately.A measurable dynamic signal produced by an auxiliary first-order system is used to eliminate the influence of unmodeled dynamics on the system.Two input models of input dead-zone and actuator failure are linearized,each follower control signal is constructed via DSC.All the signals of the closed-loop system are proved to be cooperative semi-globally uniformly ultimately bounded,and all the followers can accomplish a desired consensus results.Finally,two simulation examples are used to verify the feasibility of the control scheme designed in this paperThirdly,the consensus tracking problem for a class of output feedback nonlinear multi-agent systems with unmodeled dynamics and actuator failures under directed graphs is discussed and proposes an adaptive cooperative neural network DSC strategy.A measurable dynamic signal produced by an auxiliary first-order system is used to eliminate the influence of unmodeled dynamics on the system.An observer is designed by introducing a k-filter to observe the unknown state of the system.An smooth function is introduced to compensate the influence of bounded stuck fault on the system.Combining the dynamic surface method to design the controller,so that all the signals of the closed-loop system are proved to be cooperative semi-globally uniformly ultimately bounded,and all the followers can accomplish a desired consensus results.Finally,two simulation examples are used to verify the feasibility of the control scheme designed in this paper.