The Design And Analysis Of Dynamic Consensus Control Of High-Order Multi-Agent System

Compared with the traditional control system,multi-agent system has drawn more and more attention due to its stronger functionality,better scalability and stronger robustness.And widely used in automatic driving,cooperative control of unmanned air vehicles,traffic control.The study of the consensus problem is the basis problems of many subject.In the theoretical research of the problem of multi-agent consensus,in order to make the implementation of consensus more universal,more and more researchers simulate the problems in the real scene and add a variety of obstacles,at the same time the design and analysis of the algorithm of the multi-agent system are also become more and more complicatedThis paper mainly to solve theconsensusproblemof high-order multi-agent system.In this artical,we assume that each agent's dynamics equations is same and the state information of each agent can not be obtained completely,the internal state information can be reconstructed by using the output information and input information of each agent.Based on the reduced-order observer-based protocol,a consensus algorithm is designed.An event-triggered control strategy is added to achieve the consensus analysis of high-order multi-agent systems under fixed time-delay and non-uniform time-varying delay.Finally,the simulation of the multi-agent system is provided to illustrate the validity of theoretical results.The main contents of this paper are as follows:First,the consensus of continuous linear high-order multi-agent system is solved.Based on the related output of neighbor,a distributed reduced-order observer-based protocol is proposed.In order to construct this reduced-order protocol,a multi-step algorithm is proposed.By using integral inequality technique and matrix theory,a sufficient condition for the multi-agent systems' consensusare obtained.Using theoretical analysis and numerical simulation,proved that the system can achieve stability.Secondly,the stability of high-order multi-agent system based on reduced-order observer-based consensus is studied.On the basis of first step.We assumed that the system is at the disturbanceof fixed output delay and time-varying output delay.A novel control strategy which based on the reduced-oeder observer and the relative outputs of neighboring agents is provided.Under this control strategy,a sufficient condition for the multi-agent systems' consensus is obtained.The theoretical analysis and numerical simulation also prove that the stability of the system is achieved based on the observer-based reduced-order and fixed output delay and time-varying input delay.And obtained the estimated maximum output delay.Finally,in order to strengthen the system's practicality and reliability.The fixed output delayand maximum possible output delay of the system based on reduced-order observer-based and time-triggered mechanism is analyzed.Under this control strategy,a sufficient condition for the multi-agent systems' consensus is obtained.Using theoretical analysis and numerical simulation,it is proved that the system can achieve the stability under any non-uniform output delay not exceeding the maximum delay and the Zeno behavior of triggering time sequences is excluded.