Research On Consensus Of Saturated Constrained Nonlinear Multi-agent Systems

With the rapid development of artificial intelligence and information technology in recent years,Multi-Agent Systems,as a new subject,have been applied in the formation control of aircraft,distributed sensor network,traffic systems and other important fields.Because real systems are mostly nonlinear systems,and there are saturation constraints in the systems.Therefore,this paper focuses on the research on the consensus for saturation constraints nonlinear Multi-Agent systems.Euler-Lagrange is a typical nonlinear model which can describe the robot arm,the space satellite and the robot model.So it has the wide application prospect.In this paper,the Multi-Agent systems with Euler-Lagrange dynamic characteristics are selected as the research object.And in order to ensure that the system can achieve the control objectives in a given time,this paper studies the finite-time consensus in the consensus problems.Firstly,this paper studies finite-time consensus of Multi Euler-Lagrange systems under ideal conditions.According to whether there is a leader in systems,they can be divided into leaderless and leader-follower systems.And the corresponding finite-time consensus protocols are also proposed.Using Matrosov theorem,LaSalle theorem and the theory of finite-time stability,a sufficient condition is obtained that the system can achieve stability in a finite-time and the state of the agent remains consistent.Secondly,this paper studies the finite-time consensus of Multi Euler-Lagrange systems under external disturbance.By using the fast terminal sliding mode theory,a finite-time consensus protocol is proposed in the leaderless case.Using the finite-time Lyapunov stability theory,a sufficient conditions for the system to achieve finite-time stability are obtained,and the upper bound of the convergence time is given.Finally,this paper studies the finite-time consensus of Multi Euler-Lagrange systems under saturation constraint.Aiming at the saturation limitation in the Euler-Lagrange model,the saturation function is used to deal with the control saturation phenomenon,and finite-time consensus protocols under the saturation constraint are proposed.By using Matrosov theorem,LaSalle theorem and finite-time stability theory,a sufficient condition is obtained that under the action of bounded control input,the systems can be stable in finite-time and the state of the agent remains consistent.To sum up,this paper has made some breakthroughs in the following aspects: finite-time consensus protocols are proposed in under ideal conditions.And the control objectives and protocol parameters are optimized.In the leaderless case,the systems can operate in an aggregation controlled manner.In the leader-follower case,coupling coefficient is used to reduce control variables,which is more convenient to adjust and control.The proposed finite-time consensus protocol under external interference adopts fast terminal sliding mode theory,which greatly improves the convergence speed of the systems and makes the systems have stronger robustness and anti-interference ability.The finite-time consensus protocols in the saturation constrained case ensure that the systems can achieve finite-time consensus and the control input is limited to a bounded range to avoid the occurrence of over saturation.This paper further enriches the existing theoretical achievements and also has higher practical application value.