Research On Adaptive Fault-tolerant Containment Control For Nonlinear Multi-agent Systems

As the application scope of multi-agent systems continues to expand,containment control as a sub-field of multi-agent system control has gradually become a research focus.However,the problem of sensor and actuator faults in multi-agent systems is not well considered.Therefore,this thesis analyses the problem of fault-tolerant control and containment control in multiagent systems,so that the system can remain stable and can achieve containment control in the event of failure.In addition,for the prescribed performance,finite-time stabilization and input saturation and other constraints,this thesis has carried out corresponding analysis.Moreover,for stochastic multi-agent systems with less consideration,the corresponding control method is studied in this thesis.To sum up the above,several types of fault-tolerant containment control methods are proposed for nonlinear multi-agent systems with different kinds of constraints.The main contents are as follows:(1)For stochastic multi-agent systems with sensor faults,stochastic multi-agent system dynamics model and sensor faults model are established,and neural networks are used to compensate for the performance degradation caused by sensor faults.In order to solve the problem that the output signal of the controller is unsteerable due to the input saturation of the multiagent system,an auxiliary control signal is introduced.A smooth function is used to simulate the non-derivable function of the input saturation.Therefore,a controller that can efficiently handle input saturation for multi-agent systems is designed and the invalid system input problem can be avoided.Finally,using the Lyapunov stability theorem,it is proved that all signals in the system are bounded,and proved that the followers are in the convex hull formed by the leaders.The simulation results are given to show the effectiveness of the proposed faulttolerant containment control scheme.(2)For multi-agent systems with sensor faults,an adaptive neural network controller is designed to solve the inaccurate measurement problem caused by sensor faults.In addition,according to the prescribed performance theory,based on the containment error transformed,the controller is designed,so that the error is within the specified upper and lower bounds.In order to solve the problem of complexity explosion caused by using backstepping method,dynamic surface technology is introduced,which greatly reduces the calculation time.Through the Lyapunov stability theory,it is proved that all signals in the closed-loop system are semiglobally uniformly ultimately bounded,and containm error is within the specified range.The simulation results are given to show the effectiveness of the proposed fault-tolerant containment control scheme.(3)For multi-agent systems with actuator faults,the problem of unknown control direction in actuator failure is solved by using Nussbaum gain technique.In addition,in order to ensure that the system is stable within a limited time,a finite-time controller is designed to avoid the problem of too long stabilization time.Finally,through the Lyapunov stability theory,it is proved that all signals in the system are semi-globally uniform ultimate bounded.The simulation results are given to show the effectiveness of the proposed fault-tolerant containment control scheme.