Consensus Problem In Multi-agent Systems

Multi-agent systems has being a new type of control system in recent years,with the rapid development of computer and network communication technology.It is widely used in engineering,biological and social economic field as a frontier subject in the field of control system.Consensus problem of multi-agent systems,is not only to reveal the inherent law of objective existence in the nature of biological cluster phenomenon,but also to understand complex systems and human social life service through understanding the inherent law.Based on the existing research work,by using the knowledge of algebra graph theory,matrix theory,control theory and stability theory,this thesis has done the following main works:(1)Consensus problem of first order systems with weighted factors based on PD control Firstly,a basic first order multi-agent systems with time delay is constructed.Secondly,PD control is added to the system,and the effect of weighting factors on the system is fully considered,and the consistency algorithm based on PD control is proposed.Finally,Matlab simulation and the model of various weighting factors system are used to verify the correctness of the theoretical results.The simulation results were compared to the system that can achieve consistency.Weighting factors will affect the convergence speed of the system.(2)Consensus problem of two-order systems with different input delay.Firstly,a basic two order multi-agent systems with time delay is constructed.Secondly,considering the input delay of each agent in the system,a new algorithm for the consistency of the two-order systems is presented.Finally,Matlab simulation is used to simulate the system which satisfies the algorithm and does not satisfy the algorithm.The simulation results can be obtained when the system meets the algorithm conditions.(3)Consensus problem of two-order systems with outdated information and two layer neighbors.Firstly,a two-order multi-agent system with outdated information is constructed.Secondly,to fully use the two tier structure of the system,and the two layer neighbors into the two order multi-agent systems model,the two layer neighbor consistency algorithm with outdated information is proposed.Finally,Matlab simulation is used to analysis and compare of the simulation experiment can be obtained by adding two layers of the neighbor system that the convergence rate of the system is greatly accelerated.