An Improved Two-time-scale Consensus Algorithm For Multi-agent Systems With Binary-valued And Directed Communication

Agent is a kind of computing entity that can compute independently,per-ceive the environment and communicate with each other.Multi-agent system is formed by communication,cooperation and competition among multiple agents.This system model can provide analysis methods for various cluster problems.Consensus control is one of the most fundamental problems in the distributed control of multi-agent systems,which aims to make the state of agents in the sys-tem agree through protocol algorithm.Different topologies and communication environments,such as directed communication,communication failure,connec-tion interruption,information packet loss and so on,will have a great impact on the convergence of the system.Set-valued communication is widely used in industry,biomedicine and other in-dustries,in which nodes can only know whether the information sent by neighbors belongs to a certain range.Due to digital communication,set-valued communica-tion appears more and more in multi-agent systems.In order to reduce the influ-ence of binary-valued communication,researchers use effective set-valued recogni-tion method to design consensus algorithm for multi-agent system in undirected communication topology.However,how to reduce the impact of binary-valued communication in multi-agent systems under different communication topologies,especially the impact on consensus value,has not been studied.Based on the above considerations,this paper studies the distributed consensus problem of multi-agent systems with directed topology and binary-valued com-munication,and proposes an improved two-time-scale algorithm with alternating control and estimation,and discusses how to reduce the influence of binary-valued communication on the consensus value of multi-agent systems around the selec-tion of algorithm parameters.The main contents of this paper are as follows1.We consider the distributed consensus problem of multi-agent systems un-der directed topology and binary-valued communication,and propose an improved two-time-scale consensus algorithm.Compared with the origi-nal two-time-scale consensus algorithm,we relax the selection conditions of holding time and state bound,so as to give a wide selection range of algorithm parameters when reducing the influence of binary-valued com-munication,so that each holding time period can get a.more accurate es-timation.In the analysis of convergence,we use the matrix decomposition and inequality theory to analyze the convergence of the algorithm and give the convergence limit.Finally,the process of parameter selection is given through simulation experiments,and the convergence of the theorem and the effect of parameter selection are verified respectively.2.We consider the distributed consensus problem of multi-agent systems un-der cyclic switching topology and set-valued communication.Combined with the topology cycle switching mode,we adjust the holding time of the improved two-time-scale consensus algorithm,so that the set-valued identi-fication part can meet the requirements of estimation accuracy,and reduce the influence of the binary-valued communication on the consensus value.At the same time,the convergence of the algorithm is verified by simulation,and the necessity of modifying parameters is verified by comparison.This paper provides a solid theoretical support for the consensus control method of multi-agent system in directed topology and set-valued communication,and provides ideas and preliminary theoretical results for the research of switching topology,and provides an effective solution for the problem of parameter selection in practical application.