Research On Adaptive Consensus Of Event-Triggered Multi-Agent Systems Under Directed Graph

As today’s society demands more and more modern science and technology,the rapid development of advanced science such as biotechnology,computer and artificial intelligence has been promoted.and scholars in various fields have gradually begun to carry out relevant research on the collaborative control of multi-agent systems.This paper considers the restricted network bandwidth and resources in the actual system,which combines adaptive control and event-triggered mechanism to settle a matter of multi-agent systems consensus under directed graph,and the improved event-triggered mechanism can further optimize resource allocation and improve system reliability and endurance.The content is arranged as follows:Firstly,based on double static event-triggered mechanism,the adaptive consensus problem of multi-agent systems under directed graph is studied.To save the communication cost between agents and reduce the updating times of controller,the corresponding static event-triggered mechanism is constructed at the position of agent actuator and controller respectively,and an adaptive control protocol is designed based on this mechanism.Combining matrix theory,graph theory and Lyapunov stability theory,it is proved that multiagent systems under directed graph can realize adaptive consensus under double eventtriggered mechanism,and the communication between agents and controller update does not exist Zeno behavior.Finally,the possibilities of the proposed algorithm and the superiority of the proposed double static event-triggered mechanism in saving system resources are verified by a simulation example.Secondly,considering that the system state is not easy to obtain,the adaptive consensus problem of multi-agent systems based on double dynamic event-triggered under directed graph is further studied.The real-time state of agents is reconstructed by designing state observer.To further reduce the trigger frequency to save the communication cost between agents and reduce the frequency of controller update,the corresponding dynamic eventtriggered mechanism is constructed at the observer position and the controller position respectively,and an adaptive control protocol is designed based on this mechanism.Based on matrix theory,graph theory,Lasalle invariance principle and Lyapunov stability theory,it is proved that the observe-based multi-agent systems under directed graph can achieve adaptive consensus under double dynamic event-triggered mechanism,and there is no Zeno behavior in inter-agent communication and controller update.Finally,the possibilities of the proposed algorithm and the superiority of the proposed double dynamic event-triggered mechanism in saving system resources are verified by a simulation example.