Research On Group Consensus Of Heterogeneous Multi-agent Systems

In recent years,with the combination and rapid development of computer science,signal processing technology,biology,artificial intelligence and other disciplinary,the technology of coordinated control group movement has also made great progress,and has become one of the hot issues for researchers in various fields.Inspired by the complexity of collaborative tasks requiring the cooperation of agents,the group consensus of multi-agent systems research has attracted highly attention by numerous domestic and foreign researchers.Considering the application of multi-agent problems in practical engineering need to have different properties with each other to complete complex tasks.This dissertation mainly focuses on further studies the group consensus of heterogeneous multi-agent systems(HMAS)based on the related knowledge of graph theory,pinning control,adaptive control and the foundation of studying and summarizing the related research results.The research focus of this paper is summarized as follows:1.Aiming at the current group consensus of multi-agent systems research,how to divide the sub-group is not clear.In particular,the complex multi-agent network should be divided into several sub-groups.By referring to the community structure characteristics of complex networks,the community partitioning algorithm is introduced into the group consensus control strategy of heterogeneous multi-agent systems composed of linear first-and second-order agents.Considering the fixed and switching topological networks,the sufficient conditions for the realization of group consensus are strictly proved by using graph theory,matrix theory and Lyapunov theory.2.For the heterogeneous multi-agent system composed of linear first-order,second-order and nonlinear Euler-Lagrange(EL)structural agents,in order to realize that the final state of agent in each subgroup can tend to the expected states,aiming at undirected fixed topology,a distributed control protocol based on pinning scheme proposed;the feasibility of the controllers is proved based on grapy theory,Lyapunov theory and Barbalat's Lemma.3.Aiming at the heterogeneous multi-agent systems composed of linear first-and second-order agents,distributed adaptive control is applied to the coupling weights between neighboring individuals,and node-based and edge-based distributed adaptive so that the whole system does not need to obtain the global information of the network topology.Then,the pinning control is introduced and the adaptive control is applied to the feedback gain of the pinning nodes,a distributed control based on node and edge-based distributed group consensus control protocols are proposed,so that each sub-group reaches the desired state.Finally,compared with the simulation results without the adaptive control,the proposed control algorithm can accelerate the convergence speed of the group consensus.