| In real life,the communication relationship between agents is not only cooperative,but also competitive.Therefore,it is of the great significance to study bipartite consensus of the multi-agent systems.At the same time,the frequent communication between agents will cause unnecessary waste of communication resources,and also cause a huge burden on the system’s computing capacity.Therefore,in order to solve the above problems,this thesis combines the prediction mechanism with the event-triggered control strategy for the linear multi-agent system,further reduces the system resource consumption and computing burden,and achieves the bipartite consensus of the multi-agent system.The main research contents of this paper are as follows:On the one hand,the problem of event-triggered leader-following bipartite is studied for continuous-time multi-agent systems based on prediction mechanism under directed topology graph.The state difference between the agent and its neighbors is estimated,and then the control input of each agent is estimated.The estimated value of the next trigger state is obtained from the last trigger one of the agent.A predictor based event-triggered control strategy is proposed,and sufficient conditions are given to ensure that the multi-agent systems can achieve the leader-following bipartite consensus,and then reduce the number of triggering,and consumption of system resources.In addition,in order to avoid the unnecessary waste of system resources caused by continuous monitoring of agent’s own state,a distributed self-triggering bipartite consensus control strategy is proposed.Finally,the correctness and effectiveness of the algorithm are verified by numerical simulation.On the other hand,compared with continuous-time systems,discrete-time systems are better at digital correction,improving computer efficiency and resisting external factors.Therefore,the problem of event-triggered bipartite consensus is studied discrete-time multi-agent system based on prediction mechanism under directed graph.In the case of discrete sampling,the control input of the agent is estimated,and then the next event-triggered state is predicted in the predictor.A new event-triggered control strategy is proposed based on the predictor.By defining inconsistent vectors,the problem of bipartite consensus is transformed into a stability problem,and the sufficient conditions of achieving bipartite consensus are given.Finally,the correctness and effectiveness of the algorithm are verified by numerical simulation. |
