With the rapid development of artificial intelligence and autonomous systems,the need for fault-tolerant control mode under network attacks is becoming more and more urgent.At present,the network attack methods and means of multi-agent system are increasingly complex and covert,and the traditional fault-tolerant control mode is difficult to deal with effectively.Therefore,this paper designs a distributed control mode based on state error feedback,and uses the Riccati equation solving method in the optimal control theory to minimize the performance index of the system,so that it can not only improve the security and reliability of the multi-agent system,but also provide reference for other related fields of research and application.