Consensus Of Multi-agent Systems With Random Time-delays And Topologies

Multi-agent systems can solve the complex and difficult problems that beyond a single agent capability,as a consequence it has great potential for industrial application.In recent years,the coordinated control of multi-agent systems has attracted extensive attention of scholars.Consensus of multi-agent systems with random delays and topologies is a difficult problem in research,and the results about the consensus problems at present are rare.However,such systems have very important practical significance,due to the random delays or topologies changes caused by internal failures or external attacks.In addition,it is more difficult to study that time-delay govern by Markov chain than random time-delay,consequently it is very important to consider the consensus problem of the multi-agent systems with Markov delays and random topology.This paper deals with the problems of consensus for multi-agent systems with time-delay and switching topologies,which is governed by Markov chain in consensus of multi-agent.Its main work is divided into the following aspects.(1)The problem of consensus for multi-agent systems with switching topologies and stochastic delays was investigated.The switching topologies and time-delays are governed by Markov chains.By using a model transformation approach and Lyapunov function approach,sufficient conditions are established such that consensus multi-agent systems are achieved.A numerical example is provided to verify the effectiveness of it.(2)The problem of consensus for second-order multi-agent systems with switching topologies and time-delays was investigated.The switching topologies and communication time-delays are described by Markov chains.By using a model transformation approach,sufficient conditions are established under which consensus is guaranteed.(3)For the consensus multi-agent systems with Markov time-delays.In addition,the state of the communication is unmeasurable and communication loss exists in the communication process.Sufficient conditions for consensus are established and numerical examples illustrate the effectiveness of the proposed design procedure.(4)The problem of event-triggered consensus for multi-agent systems with stochastic delays was discussed.To save the cost of computation and transmission,a trigger function is designed based on state error.Further,the problem about following consensus of original system is transformed to the problem of mean square stability via model transformation.The sufficient conditions of consensus for multi-agent systems are given by linear matrix inequalities.