Consensus Of Classes Of Nonlinear Multi-agent Network Systems Via Intermittent Control

Consensus problem of multi-agent systems has attracted great attention of many researchers in various disciplines such as mathematics,physical science,control and network science during the past decades.This is mainly due to the fact that not only it can elucidate many natural phenomena observed,but also has extensive applications in many different fields,including swarming,flocking,rendezvous,formation,consensus to name a few.Consensus is one of the significant and active topics of research in multi-agent cooperative control,which can drive the group of agents to agree upon a desired trajectory by sharing local information with its neighbors as time goes on.This paper analyses the collective phenomenon and coordination control methods for complex multi-agent networks to explore some more general dynamic behaviors,such as partial component consensus and lag consensus.In addition,since the intermittent information flows exist among the agents,then we design several intermittent actuators to satisfy the discontinuous communication property.Moreover,this paper uses a pinning policy in which only a few agents need to pin under the aperiodically intermittent signal assumption.Some sufficient conditions are derived to ensure that the network system will reach consensus.This paper main focus on several nonlinear multi-agent network systems via intermittent control.1.Partial component consensus of leader-following multi-agent systems via intermittent pinning control.Partial component consensus means that some components of all state variables in a multi-agent system tend to be convergence as time tends to be infinite.It is a dynamics behavior that is weaker than identical consensus.In chapter 3,partial component consensus of leader-following multi-agent systems via intermittent pinning control is investigated for the first time,and the intermittent signal can be aperiodic.With the help of permutation matrix method,the corresponding error system is reduced to a new error system.Then,partial component consensus in the multi-agent system is converted into the stability of the new error system with respect to partial variables.Based on matrix theory,graph theory and stability theory of partial variables,some sufficient conditions to guarantee exponential partial component consensus are derived.Finally,numerical simulations are given to clarify correctness of the proposed control methods.2.Second-order lag consensus tracking of multi-agent systems via aperiodically intermittent control.Lag consensus is a coordination behavior to make the followers' states track to that of the leader with a time delay.In chapter 4,we investigate the second-order lag consensus for nonlinear multi-agent systems with aperiodically intermittent pinning control.The proposed consensus scheme possesses three merits: 1)efficient communications can be achieved by aperiodic and intermittent signal transmission.2)a simple external pinning policy needs to pin just a few agents.3)distinctive adaptive control is based only on local intermittent communications and time-varying state.We give several sufficient criteria for the exponential convergence to lag consensus via Lyapunov theory.Moreover,we consider the network systems in the presence of random noise perturbation.Chua's circuit simulations are shown to clarify the effectiveness of the obtained results.3.Aperiodically intermittent pinning control for lag consensus of multi-agent systems with delays has been investigated in chapter 5.This chapter studies the leader-following exponential lag consensus of nonlinear coupled second-order multi-agent systems under the directed network topology,where it is not necessary that the network structure is strongly connected or contains a directed spanning tree.Unlike the single control method,we exploit an aperiodically intermittent pinning control(combined intermittent control,adaptive control,pinning control),in which not only can solve the aperiodic and discontinuous signal transmissions of network communication,but also just need to pin a small fraction of agents in large-scaler network.We also address the nonlinear coupled networked system with delayed dynamics,and derive some novel criteria by virtue of matrix theory and Lyapunov function method.Numerical simulations clarify that the effectiveness of the aperiodically intermittent adaptive pinning policy for nonlinear coupled systems.