Consensus Strategies Analysis Of Leader-following Multi-agent Systems

Cooperative control of multi-agent systems is an important issue in the field of automatic control and artificial intelligence,which has attracted more and more attention from scholars due to its wide application in robots formation,sensor networks,intelligent transportation and other fields.As one of the basic problems in the cooperative control of multi-agent systems,consensus aims to make the states of all agents reach the same index under the control protocol,and the control protocol is a kind of information update rule designed based on the local information of agents and the communication network between agents.In this thesis,the consensus problem of leader-following multi-agent systems is studied based on event-triggered control,observed control and impulsive control combined with finite time and fixed time stability theory.The specific studies of this thesis are as follows:The first part studies the observer-based event-triggered leader-following consensus problem for linear multi-agent systems.Firstly,a class of output based event-triggered control protocol is presented for the multi-agent systems,which solves the consensus problem of true state and estimation state of agents.Secondly,a class of distributed adaptive event-triggered control strategy is proposed.Then,Lyapunov stability theory and linear matrix inequalities are employed to guarantee the leader-following consensus for multi-agent systems.Moreover,it is proved that there is no Zeno behavior.Finally,to illustrate the effectiveness of control strategies,a numerical analysis is given.The second part proposes a theoretical framework to study the finite/fixed time leaderfollowing consensus problems for a class of Markovian jumping stochastic multi-agent systems.Firstly,state feedback switching nonlinear control protocols with impulsive effects are presented to achieve the leader-following consensus of the addressed systems.Then,by employing stochastic analysis theory and Lyapunov stability theory,two types of sufficient criterion are derived to guaranteed finite time and fixed time leader-following stochastic multi-agent systems consensus.Finally,the validity of the control strategies of the theoretical results are verified by numerical simulation.In the third part,the finite time consensus problem of linear leader-following multiagent systems with Markovian jumping based on event-triggered control strategy is discussed.Firstly,the distributed event-triggered control protocol is proposed based on the multi-agent system.Secondly,the finite time consensus problem of leader-following multi-agent is studied by using Lyapunov stability theory.Moreover,it is proved that there is no Zeno behavior.Finally,the numerical analysis is provided to demonstrate the effectiveness of control strategy.