Research On Lane Keeping Control System Based On Monocular Vision

In recent years,with the increasing number of vehicles and t he construction of intelligent transportation,automatic driving system has become a popular research direction in the vehicle industry.Many schools and research institutions around the world are working on autonomous driving functions to reduce unnecessa ry accidents due to driver negligence and improve driving efficiency.As one of the main functions of autonomous driving,lane keeping assist has been paid attention to by many research institutions,so research on lane keeping has very important practical value.The main contents of this article are as follows:Road information is recognized by a method based on adaptive threshold.In this method,the camera perspective distortion is eliminated by inverse perspective calibration experiment,and the transformation from image coordinate system to world coordinate system is realized;the contrast method based on the lateral gradient information of lane lines can effectively eliminate the influence of shadow;the improved RANSAC and least squares are used to obtain the best matching model of lane lines.The method has good illumination invariance and fast calculation ability,which provides the basis for the follow-up.Taking advantage of the stability of linear quadratic regulator and the advantages of multi-objective control,a lateral controller based on incremental front wheel rotation is proposed.Based on the linear quadratic regulator,a two-degree-o f-freedom vehicle dynamics model is built,which takes the slope and lateral distance as input and the incre ment of front wheel angle as control variable;the system is controlled by continuously adding a moment of front wheel angle as the output of the system.The controller has good stability because of control ing the front wheel rotation angle.The performance of the controller is verified by the joint simulation of Car Sim and Simulink.According to the national recommendation document on automatic driving test in early 19 th year,the road environment is designed,and s curve and closed road which meet the test standard are used as the simulation test environment.Then,the controller is tested under three operating conditions.The simulation results show that the vehicle has good ce ntering performance and running stability in the simulation process.