Coordinated Optimal Control Of Dynamic Multi-agent Systems Based On Event-triggered Mechanism

With the rapid development of distributed artificial intelligence technology,agent in multi-agent systems can communicate and coordinate with each other,and can complete large-scale,complex tasks that cannot be performed by a single agent.In the past two decades,coordinated control of multi-agent systems in unmanned aircraft,coordinated movement of intelligent robots,formation control of distributed sensor networks,posture queue control of satellite groups,communication Congestion control in the network,networked computer-aided teaching and medical treatment,etc.all aspects of the social field have a wide range of applications,but also attracted the attention of a large number of researchers.The control and management of a large-scale multi-agent systems requires a large amount of computing resources and channel bandwidth,which makes managing the systems in a centralized control manner expensive or even impractical.In order to solve this problem and improve the reliability and scalability of multi-agent systems control,distributed control can be carried out through the shared communication network using local information exchange between neighbor's agents.However,the communication load of large-scale dynamic multi-agent systems continues to increase during the process of information transmission,which makes distributed event-driven mechanisms a more effective control method.The problem of collaborative consistency control of multi-agent systems based on event-driven mechanism has caused great concern in theory and practice,because it not only enables all agents to achieve consensus,but also significantly reduces the number of agent transmissions has saved the agent's limited computing power and its own battery energy.This article summarizes the current research status of multiagent systems with completeness.Based on modern control theory and matrix theory,combining the related theories of multi-agent systems,this paper focuses on the synergy of multi-agent systems based on event-driven mechanism.The main work of this paper is as follows:1.For second-order multi-agent systems with a dynamic leader,the problem of leader-following consensus is investigated herein.To reduce wastage of hardware resources and improve the communication efficiency of multi-agent systems,a distributed event-triggered control algorithm is proposed,which depends only on the agent states of each agent and its neighbors.The closed-loop model of second-order multi-agent systems is established with the distribute event-triggered mechanism.Through the matrix and modern control theories,the convergence condition of multi-agent systems' cooperative motion based on the distributed event-triggered mechanism is obtained.Through theoretical analysis and calculation,the absence of Zeno behavior is confirmed in this control protocol.Finally,the effectiveness of the algorithm is verified through simulation experiments.2.For the multi-agent systems with external disturbances,this paper studies the consensus control problems of Leader-Following.A distributed disturbances observer is proposed to estimate the disturbances in the systems,and an active control scheme based on the disturbances observer is designed.The cooperative motion algorithm is analyzed by using modern control theory and matrix theory.The consistency convergence condition of the multi-agent systems is obtained by means of the distributed eventtriggered strategy,and the lower bound of the event-triggered time interval is given.Finally,the effectiveness of the proposed control algorithm is verified by computer simulation.3.Consider the following consistency control problem in the case that only some followers can receive the position state of the leader in the multi-agent systems.In order to reduce the communication cost and improve the control efficiency,a distributed leader observer based on dynamic event-driven mechanism is proposed.At the same time,considering the influence of external disturbance in the practical application process,the distributed disturbance observer is designed to suppress the disturbance in the case of external disturbance in the channel,and it is proved that the consistent distributed control protocol proposed in this paper can achieve the leader following consistency in the sense of probability.4.The distributed optimal control problem for second-order multi-agent systems with external disturbances under undirected graphs is considered in this paper.To effectively reduce the waste of communication resources and the energy consumption of the agent,the controller is based on the dynamic event-triggered mechanism.In addition,the distributed disturbances observer is designed to estimate the exogenous disturbances in the input channel of the systems.Next,the external disturbances are suppressed by feedback.Finally,using the Lyapunov theory and the matrix theory,the sufficient conditions for the second-order multi-agent systems to achieve consensus are obtained.It is also evidenced that the distributed dynamic event-triggered mechanism designed does not have the Zeno behavior.Finally,the effectiveness of the algorithm is proven by a numerical simulation example.