Group Consensus Of Several Classes For Multi-agent Systems Based On Observer

In recent decades,many experts and scholars at home and abroad have studied the cooperative control of multi-agent systems.Its application is mainly embodied in: consensus problems,synchronization,congestion,and formation problems.Consensus has always been a hot spot in the control field.Given the wider applicability of group consensus,study group consensus has practical significance.This paper investigates several classes of multi-agent systems based on observer.The main results are as follows:(1)The group consensus problem of observer-based for multi-agent systems is considered.According to the immeasurable speed of the leader in the control system,put all of the intelligent agent into two groups firstly,each group corresponds to a leader,and presents a distributed control protocol,at the same time for each group is divided into by the designed an observer,and then by establishing the Lyapunov function,the sufficient conditions for the stability of the control system are obtained.The results are verified by numerical simulations.(2)The lag group consensus of the observer-based for heterogeneous multi-agent systems is considered.Firstly,a distributed control protocol is given based on the relative output information of the neighbor agent,and all the agents in the control system are required to achieve group consensus according to that control protocol.Since the leader's speed is unmeasurable,an observer is designed for each group to observe the speed information of the corresponding leader in each group.Later,sufficient conditions for the group consensus are obtained by graph theory and Lyapunov function.The results show that the proposed model can not only make the position error,velocity error approach zero,but also realize the group consensus problem.(3)The group consensus problem of observer-based multi-agent systems in competition-cooperative networks is considered.In real life,there exists a wide range of competition-cooperation models among biological populations,namely,cooperation between agents of the same group and competition between agents of different groups.To achieve group consensus,based on the state output information of the adjacent agents,a bipartite consensus control protocol is designed to achieve group consistency among different groups.Given the inaccessible state information of the agent in the system,an observer is designed to observe the state information of agent in real time.Sufficient condition for the system consensus are obtained by designing the Lyapunov function.