Research On The Consensus Problem Of Stochastic Nonlinear Multi-Agent Systems Based On Event-Triggered Control

In recent years,multi-agent system has been applied more and more widely.Compared with a single agent,multi-agent system has a higher control efficiency by relying on the information exchange between agents.Consensus problem is the main research direction of multi-agent system,which has a wide range of applications and has attracted extensive attention of scholars.In practical applications,in order to complete tasks,the information of agents needs to be shared in time.The ideal state is that agents can communicate continuously,which is unrealistic.Based on this,event-triggered control strategy is introduced into multi-agent system.In addition,agents are subject to stochastic perturbations.Therefore,the research of event-triggered control for stochastic multi-agent system has great practical significance.Based on the existing knowledge of stability theory and algebraic graph theory,the mean-square consensus problem of leader-following stochastic nonlinear multi-agent system is studied in this paper.By designing a distributed event-triggered control strategy,the controller can be sampled and updated instantaneously in discrete time,thus it can reduce unnecessary energy consumption.In addition,the consensus criterion of the system is established under the directed networks.By constructing an appropriate Lyapunov-Krasovskii function,and using stochastic analysis method and related inequality techniques,the sufficient conditions for the system to achieve leader-following input-to-state actual exponential mean-square consensus are obtained under the event-triggered control strategy.In addition,the lower bound of the interference time is given,which proves that the Zeno-behavior does not occur.Finally,numerical simulations based on MATLAB are carried out to verify the validity and feasibility of the theoretical analyses.