Iterative Learning Control For Multi-agent Systems With Impulse

The consistent tracking problem of impulsive multi-agent systems is an important research direction of multi-agent systems.It has a wide range of applications in the fields of information communication,control science and formation control.This paper studies the problem of consistent tracking of the leader in a limited time interval in a pulsed multi-agent system with a virtual leader.For the fixed batch step size,the D-type and PD-type learning laws with initial iterative state are used to realize the consistent tracking problem of the multi-agent system under impulse input.Under the fixed directed topology graph,for the impulsive multi-agent system,it proves that when the parameters and learning laws of the system meet the given conditions,and all followers can obtain the information of the virtual leader through the given topological relationship,as the number of iterations increases,all agents can track the leader consistently.And the validity of the theoretical results is verified by numerical simulation.In view of the change of batch step size,a domain alignment operator is introduced to correct the tracking error and update the control function.On this basis,the convergence of the impulsive multi-agent system is analyzed by using the D?D-type learning law and the I?D-type learning law.Furthermore,the local average operator is used to optimize the control function,so that the control function can make full use of the iteration error to update the learning law,and through strict mathematical theory,a sufficient condition for the multi-agent system to track the leader consistently is given.And the validity of the proposed learning law is verified by numerical simulation.