Research On Consensus Control Of Multi-Agent Systems Under Network Attack

In recent years,due to the widespread application of multi-agent systems(MASs)in satellite communication,unmanned aerial vehicle formation,smart power grid and many other fields,the problem on consensus control of the multi-agent systems has gradually become a research focus.With the development of sensing,computing and network communication,more and more multi-agent systems use network for long-distance data transmission and information interaction.Considering that the network attack caused by the openness and uncertainty of the network will seriously damage the stability of the system,and in the physical environment,the multi-agent system is often affected by the noise environment,which brings great challenges to the consistency control of the multi-agent systems.The specific research content is summarized as follows:Firstly,the mean-square bounded consensus is studied for a nonlinear multi-agent system subjected to randomly occurring deception attack,process and measurement noises in this paper.Considering the measurement tampered by the attacker,an estimator is presented to obtain relative accurate state estimation,where the gain is acquired by a recursive algorithm.On this basis,a type of centralized controller is designed combined with cloud computing system.Moreover,from perspective of the defender,a detector is proposed at the side of agent to detect whether the current actuator input is attacked.Using linear matrix inequality,sufficient conditions are given for achieving mean-square bounded consensus and an upper boundary is derived.The validity of the proposed method is illustrated via two simulation examples.Secondly,for a class of linear multi-agent systems with unknown-but-bounded process and measurement noises,the consensus control problem of multi-agent systems under mixed attacks is studied in combination with cloud computing systems.Based on the set-membership estimation theory,the set-membership estimator and attack detector are designed to determine whether the current estimation state is available.According to detection results of the attack detector and information of the neighbor agents,a one-step predictive controller is constructed.Furthermore,by using linear matrix inequality and mathematical induction,sufficient conditions for ensuring quasi-consensus of the multi-agent systems are solved.The effectiveness of the proposed theory is verified by simulation.Finally,on the basis of the previous chapter,the consistency control problem of nonlinear multi-agent systems with unknown-but-bounded process noise and measurement noise under mixed attack is further studied.Using the idea of set-membership estimation,the current state is estimated and attack detector is constructed to judge whether the current measured value is tampered by attackers.According to the detection result of attack detector and the information of neighbor agents,a one-step predictive controller is constructed.By using linear matrix inequality and mathematical induction,the sufficient conditions for achieving quasi-consensus and realizing stability of multi-agent systems are solved.The effectiveness of the algorithm is verified by numerical simulation.