Distributed Cooperative Control Of Several Classes Of Disturbed Multi-Agent Systems With Directed Network

Multi-agent system has plenty of application scenarios,such as multi-robot systems in the industry,battlefield,space,underwater,disaster areas,urban transportation network,smart grid,Internet,to name just a few.However,during the application,each agent cannot avoid the influence of disturbances.For example,robots can be affected by friction torque,UAVs can be disturbed by wind,and DC motors could be influenced by the load change.The existence of disturbances may not only affect the normal operation but also hinder the realization of cooperative objectives.For a single system,the anti-disturbance control methods have been rapidly developed,while there are new challenges when considering the cooperative anti-disturbance control problem in multi-agent systems due to the existing couplings.This paper mainly focuses on the topic of cooperative control for disturbed multi-agent systems and has accomplished the following works:1)addresses the finite-time bipartite consensus problem for multi-agent systems(MASs)on a directed signed network.Some properties for signed digraphs are first investigated and two nonlinear control protocols are then designed for first-and second-order MASs,respectively,where agents may be influenced by bounded disturbances.Particularly,for MASs with second-order dynamics,a new estimation technique is developed to estimate the settling time.2)considers the finite-time output consensus problem for a class of second-order multi-agent systems(MASs),where the mismatched disturbance exists in the dynamics of each agent,and the communication topology is directed.First of all,a basic backstepping control protocol is proposed to solve the finite-time consensus problem without mismatched disturbance.Then,a finite-time dis-turbance observer is designed to estimate the mismatched disturbance,based on which,two adaptive finite-time consensus protocols are proposed to solve the finite-time output consensus and track-ing consensus problems without using any global information with respect to the communication topology.3)investigates the distributed consensus control problem for a class of disturbed second-order multi-agent systems(MASs)with directed networks.Two cases are investigated.For the tracking problem,the leader is assumed to have a bounded input.Each follower is affected by disturbances generated by exosystems.A novel distributed adaptive observer is first proposed to cooperatively estimate the real state of the leader even the upper bound of its input is unknown to all the followers.With the distributed observer,the fully distributed tracking problem can be further solved.For the leaderless consensus problem,each agent is assumed to be affected by local disturbances generated by exosystems.Each agent here cannot obtain the exact exosystem model.Different from existing works,a local composite disturbance observer is proposed,with which the exact states of the local disturbances can still be estimated.Accordingly,the fully distributed leaderless consensus problem can be solved4)considers the fully distributed tracking consensus problem for general nonlinear multi-agent systems with a leader whose control input is nonzero and bounded.First,a new class of distributed state observer for the leader is proposed without the knowledge of the upper bound of the leader's input.Then,the situations that followers are affected by disturbances with unknown upper bound or disturbances generated by exosystems are investigated.Specifically,two distributed control protocols based on the distributed state observer,neural networks,and adaptive laws are proposed.