Consensus Control Of Nonlinear Multiagent Systems With Dynamic Event-Triggered Strategies

In the last decade,due to the broad applications in the fields of UAV formation,satellite attitude alignment,and multi-robot systems,the coordinated control of multi-agent systems has attracted much attention.Consensus is one of the basic issues of coordinated control of multi-agent systems.The main purpose is to formulate some appropriate control protocols based on local information to ensure that all agents reach the same state.In the theoretical research of multi-agent systems,it is usually assumed that each agent has sufficient energy and excellent computing power,and real-time communications between agents.However,in practical applications,the computing and communication capabilities of a single agent depend on its embedded digital microprocessor;the energy comes from the embedded battery.Therefore,under the above mentioned limited abilities,it is difficult for the agent to meet the ideal hypothesis.Event-trigger strategies provide suitable solutions for such problems.Under event-trigger strategies,each agent only communicates and updates its controller when the event is triggered,making full use of the limited resources of multi-agent systems.This thesis focuses on nonlinear multi-agent systems and designs appropriate event trigger strategies,in order to study the consensus problem under intermittent communication of muti-agent systems.The main contents of the thesis study are as follows:We propose a dynamic event triggering strategy based on edge information,considering the limited network bandwidth in practical applications.Compared with the traditional static threshold,the time-varying threshold guarantees fewer triggers.Compared with the neighborbased event trigger strategy,the edge event trigger strategy is a more distributed information exchange method.Each time an event is triggered,interacting information is only asynchronously transmitted between pairs of adjacent agents.In order to avoid acquiring global state information,the designed consensus control protocol includes the weight updating law for each edge.Then,based on the Lyapunov stability theory,sufficient conditions for the consensus in a multi-agent system are obtained.Furthermore,we have proven that the controlled system does not exist Zeno behavior.We propose a new dynamic event trigger strategy under the framework of the multilayer network,due to the multi-layer characteristics of the real network.The trigger function is established according to the relationship between the topological structure,internal coupling patterns,and interlayer/intralayer coupling strengths.Under the proposed eventtriggered strategy,an adaptive consensus control protocol is designed for a two-layer nonlinear multi-agent system with different dynamics and different topological structures on each layer.The intralayer coupling strength is determined by an adaptive law,which reduces the conservativeness of the system.Based on Lyapunov stability theory,we have derived sufficient conditions for the two-layer nonlinear multi-agent system to achieve intralayer synchronization.Also,we have proven that the controlled system does not exist Zeno behavior.Finally,two example simulations are used to verify the effectiveness of the proposed theoretical results.