Research On Multi Vehicle Cooperative Control Based On Fuzzy Game

With the development of the automatic driving technology and the support of the communication technology,the information interaction of the vehicles became more secure and effective.In this context,there were great progresses in the development of platoon control technology.Platoon control technology has a wide application prospect in many fields of national economy,and has a significant advantage in improving traffic flow,improving security and reducing energy consumption.Therefore,a large number of enterprises at home and abroad have carried out platoon control technology research.Aiming at the problem of vehicle collision avoidance in the platoon,this paper studies the feasibility of game theory in solving this problem.The main contents of this paper are as follows:Firstly,the idea of hierarchical control is used,in which the upper controller uses the method of model prediction control,takes the requirements of safety,comfort and driving efficiency as the optimization performance index of the system,and determines the expected acceleration of the vehicle.And the lower controller adopts the method of inverse dynamics model,through the analysis of the longitudinal force of the vehicle,the expected throttle opening and braking pressure can be obtained under the expected acceleration.But the inverse dynamic model also has disadvantage of requiring high modeling accuracy.To solve this problem,PID control method can be used.The advantages of both sides can be effectively utilized,and the expected acceleration can be followed more accurately.In the joint simulation environment,this method can effectively meet the needs of controller design.Then,the collision conflict problem of the rear two vehicles may happen in the process of following the leader vehicle independently,and game theory is an effective mathematical tool to solve such conflicts.The inequality of the rear two vehicles can be solved in the Stackelberg game.Generally,the leader car makes the acceleration and deceleration strategy choice first,and then the follower car can make the most favorable choice for the maximum of its own interests based on the strategy selection of the leader vehicle.For the treatment of the weight coefficient in the income function,the method of variable weight coefficient is adopted.Combined with fuzzy control,the emphasis of the income function is different under different working conditions.Therefore,the variable weight coefficient can be used to deal with the needs of various complex situations.Finally,three methods are used to compare the steady follow-up and start-up conditions through the combined simulation of CARSIM and SIMULINK.The simulation results show that the performance of Stackelberg game is the best.Compared with the other two methods,the vehicle in the queue can be kept stable faster and the whole platoon length is the shortest,and it also performs better in the vehicle spacing.