Distributed Average Tracking Of Multi-Agent Systems Based On Event-Triggered

In recent years,the distributed average-tracking(DAT)problem has become a research hot-spot in the direction of multi-agent systems(MASs),and has been widely used in many fields such as distributed map detection,dynamic area formation tracking control and wireless sensor networks.The DAT problem involves a group of agents with different time-varying reference signals.Using the local communication of multi-agent to design a suitable control protocol to make each agent track the average value of time-varying reference signals.Considering the limitation of actual communication resources,the event-triggered control strategy is applied to the DAT algorithm design,which means the agent only communicates with each other and updates the controller when the triggering function is greater than the preset threshold.At the same time,in order to reach average tracking convergence faster,fixed time control is applied to the design of DAT algorithm.The main contents of this paper are as follows:(1)For the distributed average tracking problem of first-order MASs,a distributed robust average tracking algorithm is designed under the event-triggered control.First of all,the proposed algorithm is robust to initialization errors and can deal with network interruption.Then,by constructing a triggering function with dynamic parameters,the event-triggered control is realized,and there is no Zeno behavior.Finally,a sufficient condition for the average of time-varying reference signals on MASs asymptotic tracking is obtained by using Lyapunov stability method.(2)In order to solve the DAT problem of first-order MASs with input delay,an event-triggered fixed-time distributed algorithm is proposed under the premise of considering the convergence rate and reducing the update frequency of the controller.First,in order to overcome the input delay,a new state variable is constructed by integral and applied to the algorithm.Then,based on the new state variable,a triggering function is designed to reduce the update frequency of the controller and ensure that there is no Zeno behavior.Finally,the bounded fixed-time convergence of MASs under fixed undirected topology is analyzed by Lyapunov stability method and inequality method.(3)For the DAT problem of second-order MASs,an event-triggered fixed-time distributed algorithm is proposed.First,by using the signum function,a fixed-time DAT estimator driven by the event-triggered method is designed.The proposed estimator can estimate the average value of time-varying reference signals,and does not need continuous neighbor information under the event-triggered control strategy,and reduces resource consumption.Then,a fixed-time controller is proposed based on the sliding mode control method.Finally,the upper bound of convergence time can be obtained by Lyapunov stability method and inequality method.