Research On Adaptive Control For Nonlinear Multi-agent Systems

Nowadays,the multi-agent systems are widely used in many fields such as UAV formation,intelligent terminal management,network resource allocation problems,and so on.The problem of consensus control is the basic problem of multi-agent systems.The purpose is that complete the desired control scheme for the output or state of the multi-agent system by the exchanging information among agents.The main research contents of the paper are summarized as follows:Firstly,the nonlinear multi-agent systems with actuator failures are mainly studied.The problem of event-triggered fault-tolerant consensus control is studied under the without the leader.Based on the model reference method and the approximation properties of the neural network,an effective and reliable controller is designed.Through the control input,the states of all agents accomplish the same objective.In addition,an event-triggered adaptive mechanism is designed,which reduces the burden of communication.And,the adaptive adjustment of the control input is completed.It is ensured that the system can complete the required control target.Finally,based on Lyapunov theorem,it is proved that all error signals are bounded in this paper,and the control method is effective.Secondly,the multi-agent systems with external interference are mainly studied.It is studied how to realize the formation control of multi-agents with external disturbances.The required formation tracking control scheme is completed by considering the safety distance among agents under the leader-following framework.And,based on the approximation effect of the neural network,the unknown model dynamics and unknown external interference have been effectively resolved when the control scheme is implemented.Finally,a simulation example is used to prove the effectiveness of the control scheme.Lastly,the leader-following consensus control problem of the multi-agent system is studied,and the influence of the time delay and external interference are considered,and the communication faults among the agents are considered.By using Lyapunov theorem,it is proved that the error can generate the required operating performance of the system is completed in the control scheme.The distributed event-triggered control method is used to reduce the communication burden in each subsystem.Finally,the simulation results of MATLAB prove the feasibility of changing the control strategy.