Research On Longitudinal Control Strategy Of Vehicle Autonomous Driving

In recent years,with the rapid development of the Internet of things and artificial intelligence,and gradually to the automotive industry penetration,making the vehicle self-driving technology gradually mature,and gradually to the market.The application of the automatic driving technology,the realization of the vehicle to replace the driver to the operation of the car,self-driving cars to forecasting the potential dangerous surroundings and take the corresponding security measures,reduce the caused by pilot error to the operation of the vehicle traffic accidents,can significantly improve the safety of vehicles,and comfort of their cars.Firstly,this paper analyzes the working condition of the vehicle equipped with millimeter-wave radar in longitudinal driving,and determines the three functional objectives that the controller should achieve: cruise control,distance keeping and anti-collision.According to the three objectives of the controller,three operating modes of the vehicle are defined: cruise control,distance keeping and anti-collision.In order to improve the safety of the vehicle in the process of driving,this paper adopts a variable vehicle distance safety distance model which takes into account the influence of the main vehicle speed and the relative speed of two vehicles on the safety distance of the workshop,greatly improving the accuracy of the safety distance model.Secondly,this article in CarSim2016 software has set up a vehicle dynamics model for the car,because this article focuses on the longitudinal motion of the car,so only to body model,the aerodynamics model of CarSim,wheel and tire model and the braking system model parameter is set,the other had a little effects on vehicle longitudinal control system model,such as steering system model and the suspension system model is using the default Settings.In order to simulate the autonomous driving environment of the car,a 3D road model of the car driving is designed in the ADAS environment model of CarSim,and the motion parameters of surrounding vehicles are set.In addition,based on the millimeter-wave radar sensor installed in front of the vehicle,a longitudinal effective target vehicle discrimination method is proposed.Then,this paper adopts the method of hierarchical control,in Matlab/Simulink to build the upper controller of the three operating modes of the car,based on the PID control theory to design the cruise operating mode of the upper controller;Based on the optimal control theory,a distance preserving upper controller is designed.The upper controller of anti-collision operation mode is designed based on the three-stage braking control strategy of braking hazard coefficient(?).Based on the ideal of the upper controller output acceleration,the article in the Matlab/Simulink has set up a lower level controller for the vehicle longitudinal control,including: PID controller model,drive/brake switch logic model,drive control model,the brake control model and actuator model,and the lower level controller parameter calibration,in order to make sure the upper controller and the simulation model of the longitudinal running smoothly.Finally,in combination with the practical condition of the car and IVISTA-ACC.TP-A0 and IVISTA-AEB.TP-A0 ADAS technology working condition of test scenarios to determine the simulation conditions of different working modes include: working condition of acceleration/deceleration cruise vehicle,target vehicle follow condition,target vehicle driving speed following condition,target vehicle static collision conditions,target vehicle driving low-speed collision condition and the vehicle deceleration driving collision conditions.The simulation results show that the controller or control strategy designed in this paper can achieve the functional objectives well,and the simulation results curve can reflect the dynamic response characteristics of the vehicle well.