Consensus Coordination Control Of Nonlinear Multi-agent Systems

In recent years,the coordination control of multi-agent systems has gradually become a research hotpot in various fields.By designing a valid control protocol for each agent in the systems,the state information of each one eventually converges to a certain common value,which is called the consensus problem.As the most essential coordination control problem,consensus problem is also widely concerned by scholars at home and abroad.It not only reveals many natural phenomena in reality,but also has a widespread range of applications in many areas,such as robots,unmanned aerial vehicle,sensor networks,just to name a few.In the synthesis and analysis of control systems,nonlinearity is always an inevitable problem.Due to its own or external influence,many practical systems have certain nonlinear characteristics,and many nonlinear terms with small effects are often ignored.With the rapid development of system identification and modeling technology,research objects on multi-agent systems has gradually changed from the simpler first-order,second-order and higher-order linear multi-agent systems in the early stage to the more general nonlinear multi-agent systems.Compared with linear multi-agent systems,the existence of nonlinear characteristics makes the nonlinear multi-agent systems more complicated and the research on the consensus problem is more challenging.The main contents of this paper are as follows:(1)The consensus for second-order heterogeneous multi-agent system is investigated.The heterogeneous systems consist of linear and nonlinear second-order agents.Consensus control protocols to achieve the consensus are presented for linear and nonlinear agents.Based on Lyapunov stability theory,the consensus of the systems is discussed with and without a static leader,and sufficient conditions are established.Hence,both leaderless and leader-following consensus problems are solved under the undirected and connected communication topology.(2)The leader-following consensus issue is addressed for a class of nonlinear multi-agent systems,whose unforced dynamics is linear.For each follower in the systems,the distributed nonlinear control protocol is constructed.By variable substitution,the leader-following consensus problem is converted to the asymptotic stability problem of the error systems.According to Lyapunov stability theory,sufficient conditions are obtained.Hence,the leader-following consensus is achieved when the communication network among followers is an undirected and connected graph.Based on the above analysis,a new control protocol is given.It is proved that this new protocol can also make the multi-agent systems reach leader-following consensus.(3)Based on the previous chapter,the leader-following consensus problem is considered for a class of complex nonlinear multi-agent systems.An effective nonlinear consensus control protocol is given for each follower.Similarly,the leader-following consensus problem is transformed into the asymptotic stability problem of the error systems by variable substitution.Applying Lyapunov stability theory,we establish sufficient conditions which ensure the consensus of the nonlinear multi-agent systems.Hence,the leader-following consensus problem is addressed under the undirected and connected graph.Then,a new control protocol is presented,and it is proved that the new protocol can also make the systems achieve leader-following consensus.On this basis,the input saturation of the nonlinear multi-agent systems is further considered,and the problem of the leader-following consensus is solved under input saturation.