Consensus Of Multi-Agent Systems In Noise Environment And Its Application In Formation Flight

In recent years,with the development of computer technology and network communication,coordinated control of multi-agent systems has been widely used in industry,military,aerospace and other fields,and has become a research hotspot in many disciplines.As the most basic problem in coordinated control of multi-agent systems,the consensus problem has attracted increasing attention of researchers.Considering the actual influencing factors,the communication between agents is inevitably affected by noise,delay,nonlinear items and other complex environmental factors.The stochastic consensus problem of discrete-time nonlinear multi-agent systems with multiplicative noises and its application in UAV formation control algorithm is studied in this dissertation.The main contents are as follows:(1)The consensus problem of discrete-time nonlinear first-order multi-agent systems with and without a leader is studied.For first-order multi-agent systems,firstly,we design leader-follower and leaderless control protocols based on position measurement information of adjacent agents.Secondly,by applying stochastic analysis tools and algebraic graph theory,we transform the consensus problems into the Lyapunov stability problems of the first-order multi-agent system.Then,the Lipschitz condition is used to deal with the multiplicative noises and nonlinear term of the system.In the end,the inequality conditions for mean square and almost sure consensus are obtained.(2)A consensus problem of discrete-time nonlinear second-order multi-agent systems with multiplicative noises is studied.Firstly,for a leader-follower second-order multi-agent system in a directed communication topology,the control input protocol is designed based on position and velocity measurement information of adjacent agents.The same applies to a leaderless second-order multi-agent system in an undirected communication topology.Secondly,by applying Kronecker product,Schur lemma and stability theory,the scalar inequality conditions for mean square and almost sure consensus are obtained,and the method of selecting appropriate control gains is given.(3)Based on the above research on the consensus theory of the multi-agent systems,firstly,the corresponding UAV formation control protocol is designed,which enables the neighboring UAV to obtain the relative formation command,relative position and speed information with the neighboring UAV through limited information interaction.Secondly,by using MATLAB,numerical simulations are given to illustrate the effectiveness of the control protocols and the validity of the obtained consensus conditions.Then,the Rflysim simulation software and UAVs experimental platform based on indoor optical positioning are used to achieve autonomous formation flight of a number of quadrotor UAVs and verify the validity of the formation algorithm.