Study On Steering Avoidance Control Of Intelligent Vehicle Based On Fuzzy MPC And APF

Automobile intelligent technology effectively improves vehicle safety and driving convenience.Active collision avoidance technology is one of the important technologies,which can control the vehicle to avoid obstacles,ensure the vehicle to drive along the established path,and greatly reduce the possibility of traffic accidents caused by poor operation of drivers.Considering that under the condition of high speed and low road adhesion coefficient,collision avoidance by steering control is more efficient than collision avoidance by braking control.Aiming at the technology of collision avoidance by steering control,this paper focuses on the path tracking and path planning algorithm of intelligent vehicles.Firstly,according to the requirements of path tracking of intelligent vehicles,the MPC path tracking controller is designed based on the model predictive control(MPC)algorithm;according to the requirements of "real-time control" of intelligent vehicle automatic driving,the path tracking problem under the framework of MPC algorithm is transformed into a standard quadratic programming problem to improve the real-time calculation.The path tracking control algorithm based on MPC is built in the Simulink/CarSim co-simulation platform,and the performance of MPC path tracking controller is verified by simulation at different vehicle speeds.In view of the fact that MPC path tracking controller with constant weight coefficient is difficult to meet the variable requirements of path tracking under different working conditions of vehicles,this paper studies the influence of the weight coefficients in the MPC objective function on tracking performance,designs an adaptive adjustment algorithm of weight coefficient based on lateral deviation and road curvature,and builds a fuzzy MPC path tracking controller with adaptive weight coefficient.The simulation results show that the path tracking control system with adaptive weight coefficient has smaller lateral deviation and better stability.Secondly,considering the requirements of intelligent vehicle to avoid collision when driving in structured road environment,based on artificial potential field(APF)and MPC algorithm,the research of path planning to avoid collisions is carried out,analyze the impact of road curvature and obstacle speed on intelligent vehicle's collision avoidance by steering control,design obstacle and road potential field function,build vehicle driving environment model,and design collision avoidance path planning controller based on MPC algorithm.Finally,a double-layer collision avoidance control system with collision avoidance path planning and path tracking is built in the Simulink/CarSim co-simulation simulation platform.The simulation of the collision avoidance control system is carried out under multiple combination conditions of speed and driving environment.And the results show that the control system to avoid collisions can plan the path satisfying human driving habit under both static and dynamic environment,and the path can be tracked by the vehicle easily.Finally,in order to further verify the real-time performance and effectiveness of the designed collision avoidance control system in the actual controller,the hardwarein-the-loop simulation is carried out based on the NI real-time simulator,D2 P rapid prototype development platform and host computer.Vehicle dynamics model is compiled and deployed to the NI real-time simulator through the NI VeriStand,based on the D2 P rapid prototype development platform,download the path tracking algorithm with adaptive weight coefficient to the controller,and carried out hardwarein-the-loop simulation experiments under different vehicle speeds,which proved that the path tracking control algorithm with adaptive weight coefficient can meet the realtime performance and robustness requirements when running in the real electronic control unit.