Research On Formation Control Method Of Milti-uav Based On Game Theory

With the continuous development of modern science and technology,more and more multi-agent systems have appeared in our lives.As an important application field of multi-agent systems,multi-unmanned aerial vehicle(UAVs)systems have been applied more and more widely.As the tasks they can perform become more and more complex and changeable,it is of great significance to study the formation control algorithm.The existing formation control algorithm still has some shortcomings in dealing with the interference environment and time-varying tasks,and it is difficult to deal with the complex environment.Based on the theory of differential game,this paper studies the control algorithm of UAV formation under the interference environment and time-varying tasks.The main research contents and results are as follows:First,the control problem of multi-uav formation is transformed into a linear quadratic Nash differential game for solving.By proving the stability of open-loop Nash control strategy,an open-loop Nash equilibrium control algorithm based on differential game is obtained.Then,a terminal state estimation method is introduced to solve the problem that the distributed formation system cannot obtain all the current UAV state information during flight.Simulation results show that the Nash equilibrium strategy obtained by this method is basically consistent with the classical open-loop Nash equilibrium strategy,and the required formation can be formed.Secondly,a Receding Horizon Nash control algorithm is designed on the basis of terminal state estimation and combining the idea of model predictive control,so as to provide a real-time state feedback for formation control and solve the unexpected situations such as external interference that multiple UAV formation may encounter during flight.In this algorithm,the current state of the UAV is feedback to the formation system.When the current state of the UAV is deviated by unexpected circumstances such as external interference,the current state can be used as the initial state to recalculate the open-loop Nash strategy in the remaining period,so as to solve the interference problem.Simulation results show that the algorithm can still form the required formation when the state of a UAV deviates.Finally,this paper analyzes the reason why direct formation switching in time-varying formation control will lead to discontinuity,through with the property of the sigmoid function smooth can guide design a continuity of time-varying adjacency matrix and time-varying describe matrix formation,combined with the open-loop Nash control strategy based on terminal state estimation,a time-varying formation control algorithm is proposed.Simulation results show that the proposed time-varying formation control algorithm can ensure the continuity of formation switching when formation formation changes and formation members change,so that the formation can execute flight mission stably and has good scalability.