Adaptive Controller Design Based On Data-driven And The Application In The Connected Cruise Control

With the advent of the 5G era,the acceleration of data transmission rates has made V2 X communication of great practical significance in the area of connected cruise control(CCC)system which is composed of human-driven vehicles and autonomous vehicles.At present,the design of CCC system is mainly based on traditional model-based methods,that is,the system matrix composed of parameters related to driver’s driving behaviors and vehicle system dynamics need to be known,however,these parameters cannot be accurately obtained in many cases.Therefore,the focus of this research is to realize the adaptive cooperative control of the autonomous vehicle only using the online state data obtained from V2 X communication.Firstly,considering the influence of vehicle longitudinal dynamic inertia delay,a data-driven adaptive dynamic programming control strategy is designed for the connected cruise control system on a single straight lane,which realizes the adaptive cooperative control of the autonomous vehicle.At the same time,the effectiveness of the algorithm is verified by the simulation test.The simulation results show that under the control strategy,only using the speed and position information of the front vehicles obtained by V2 V communication,the optimal control obtained can ensure the autonomous vehicle travel to a stable and expected state,and the convergence speed is fast.What’s more,it also shows that the inertia delay of vehicle longitudinal dynamics cannot be ignored in the design of connected cruise control system,which has a great impact on the control effect of autonomous vehicles.Secondly,considering the lateral movement of the vehicle,the steering wheel control model of autonomous vehicle is constructed based on the lateral error and heading error,and the vehicle lateral control strategy based on data-driven is designed by using the road information obtained from V2 I communication and the state information of the vehicle.At the same time,based on the joint simulation of Simulink and Car Sim,the driving state of autonomous vehicle on different curvature roads are analyzed,and compared with the traditional model-based control method.Finally,three complex scenes are constructed,that is the random speed change of the lead vehicle,the merging of two platooning vehicles on a single lane and the combination of curves and straight roads.The simulation results based on Simulink and Car Sim show that the designed control strategy based on data-driven has good adaptability and can respond to the change of state in the platooning and the change of road environment,at the same time,it can achieve the desired speed and maintain the desired headway on the straight road,and can ensure that the lateral error tends to zero when driving on the curve road.