Research On Adaptive Cruise Control With Steering Control System

Due to the serious traffic congestion and the rapid development of science and technology,autonomous vehicles are recognized as a priority for development of the automobile industry in future.There are many key technologies involved in autonomous vehicles.At present,the main research focus in motion control are the lateral control and longitudinal control of autonomous vehicles.For the longitudinal control of automatic driving vehicle,an adaptive cruise system with two layers of controller was designed.The upper controller was divided into adaptive cruise control system and cruise control system.The main purpose of the upper controller was to get the desired longitudinal acceleration under the current working conditions and a simple switching logic to ensure the safety and stability of the vehicle was designed.Adaptive cruise control system used model predictive control(MPC)to design a safety,comfort and stability control strategy and the cruise control system used proportional integral differential(PID)control algorithm to achieve the desired speed.The lower level controller developed the throttle and brake switching logic according to the vehicle state,and used PID algorithm to control throttle and brake respectively,so as to ensure that the vehicle realized the following longitudinal acceleration of the upper controller.For lateral control,this paper adopted two ways to realize the steering control of the autopilot vehicle.Firstly,the MPC algorithm and Preview-Follower Theory(PFT)were combined to track the desired trajectory.By combining the MPC and PFT algorithm(MPCPFT,Model Predictive Control with Preview-Follower Theory),the PFT algorithm was used to increase the length of the effective reference path under the premise of not increasing the amount of computation,which improved the accuracy and stability of the trajectory following control algorithm.In the meantime,a MPC steering controller based on lane departure error was also proposed,the vehicle got the deviation of the current track from the sensor.Through the fitting and interpolating,a limited number of preview points were processed to obtain enough preview points in the predicted time,and the controller achieved steering control by the purpose of reducing the lane departure error.In the process of control,the process of coordinate transformation was reduced,so the difficulty of the controller was reduced.Based on the longitudinal control and lateral control of the autonomous vehicle,the control architecture considering the longitudinal and lateral dynamic coupling was designed.First,I decoupled the longitudinal and lateral controllers,but the longitudinal vehicle speed will affect the lateral stability of the vehicle,so the changing vertical speed caused by the cruise system is the one of inputs to the steering controller,which improves the control stability,safety and effectiveness of the whole controller.Finally,the adaptive cruise control system with steering control was realized.A reasonable simulation conditions were set up for the longitudinal controller,lateral controller and adaptive cruise controller with steering system.The appropriate sensors were set up in CarSim and the controllers mentioned above was built in MATLAB/Simulink to verify the reliability,validity and security of the algorithm proposed in this paper.