Consensus For Multi-Agent Systems Based On Predictive Control Scheme

In recent years,the application of multi-agent system has been widely used,which has attracted a large number of researchers to conduct in-depth research.Analysis and research for multi-agent system from the perspective of control theory is becoming a hot research topic at home and abroad.The main work is that through the design of control law to achieve the corresponding control performance requirements.The research contents mainly include the collaborative control,decentralized control and formation control of the multi-agent system,and the collaborative control problems are widely concerned by researchers in different fields.Consistency is the prerequisite and foundation for the cooperation of the multi-agent system,which plays an extremely important role in the further study of the collaborative control of multi-agent systems.Model predictive control is widely used in recent years for the multi-agent system coordination mechanism and the effective control of research,mainly because it can accelerate the convergence speed,solve multi-agent system with multi-variable constraints and expanding the scope of effective sampling period and other advantages.This paper focuses on the consensus of multi-agent system based on predictive control,including first-order model and second-order model.For the undirected topology first order multi-agent systems,the study is based on an iterative predictive control method with state predictor to accelerate the convergence speed of the system.And then prove it in theory.Then,we studies the use of centralized and decentralized event-driven functions respectively.We know that the predictive methods can increase the frequency of event triggering in multi-agent systems.In order to reduce the impact of the prediction method on the frequency of event triggering,the target is to optimize the event-driven function.By constructing the Lyapunov function,we give the sufficient conditions for ensuring the average consensus or the ball consensus of the system.Simulation examples verify the rationality of the control protocol and the effectiveness of the drive function.