Lane Keeping of Intelligent Vehicle under Crosswind Based on Visual Navigation

In this work, a framework of the co-simulation of intelligent vehicle based on visual navigation is presented, which is used to analyze lane keeping of vehicle under crosswind. Firstly, we define a virtual road environment in CarSim. Next, the method of lane extraction based on MATLAB is also presented. A rectangular region of interest in road image is set to speed up image processing, and lane boundaries are detected by combining with the edge extraction and the Hough transform. Subsequently, vehicle model is built in CarSim, and three types of crosswind model are built in MATLAB, which are used to co-simulation lane keeping of vehicle under crosswind. Simulation results show that the proposed lane detection method has good performance and the proposed controller can guarantee the stability of the vehicle under crosswind.