Research On The Consensus Of Nonlinear Multi-agent System Based On Event-triggered Control

The coordinated control of multi-agent systems has been widely used in distributed sensor networks,coordinated control of unmanned aerial vehicles,attitude adjustment of satellite clusters,and congestion control of communication networks,and has attracted the attention of researchers from many fields.Consensus is the foundation of coordinated control research for multi-agent systems,and it is also the most concerned topic.In practical applications,due to the limited communication bandwidth of multi-agent systems,as well as the limited energy and computing capabilities of agents,and the ability of event triggered consensus protocols to greatly reduce the number of communication between agents and the number of updates to agent controllers,event triggered strategy has become a hot issue in consensus research in recent years.This paper mainly studies the consensus problem of nonlinear multi-agent systems based on event triggering.Firstly,we study the consensus based on event triggering in directed network topology,and further discuss the leadership tracking control problem of dynamic event triggering.Then,the bipartite consensus under the cooperative competition network system is studied.Finally,the consensus based on sampling data and event triggering is studied.The specific work of the paper is as follows:1.A multi-agent system with nonlinear dynamics in a directed network topology is studied,and an event triggered consensus control protocol is proposed.Firstly,a centralized control strategy is designed,and it is proved that the designed protocol can solve the consensus problem.Then,a distributed control strategy is designed,and some new properties of the Laplacian matrix are obtained using matrix theory.Finally,sufficient conditions for the consensus of the system solution are obtained using Lyapunov stability theory.2.The problem of leader tracking consensus in nonlinear multi-agent systems based on dynamic event triggering is studied.By introducing a dynamic variable into the event trigger function,a dynamic event trigger strategy is proposed.The designed event trigger conditions can be dynamically adjusted according to the operating state of the system,thereby further reducing the frequency of controller updates.Using Lyapunov stability theory,it is proved that multi-agent systems can achieve leader tracking consensus under dynamic event triggering strategies.3.This paper studies multi-agent systems with nonlinear dynamics under cooperative competitive networks,and designs a bipartite consensus protocol based on event triggering.In some practical multi-agent systems,there are both cooperation and competition among agents,and the corresponding consensus problem becomes a bipartite consensus problem.The research on bipartite consensus problem also has very important application value.The centralized and distributed bipartite consensus protocols are designed respectively.The sufficient conditions for the system to solve bipartite consensus are obtained using matrix theory and Lyapunov stability theory,and it is proved that the system does not exhibit Zeno phenomenon.4.The consensus problem of nonlinear multi-agent systems based on sampling information and event triggering is studied.An event triggered consensus protocol is designed for multi-agent systems based on sampling information.This protocol fully utilizes the advantages of sampling information and event triggered control,and can further reduce the number of communication between agents in the system and the number of updates to the agent controller.In stability analysis,by introducing artificial delays,the stability of the system is transformed into the stability of time-delay systems.Sufficient conditions for the consensus of the system solution are obtained using linear matrix inequalities.