Lane Detection and Trajectory Tracking Control of Autonomous Vehicle Based on Model Predictive Control

• Published in: International journal of automotive technology Vol. 21; no. 2; pp. 285 - 295
• Main Authors: Hu, Jianjun, Xiong, Songsong, Zha, Junlin, Fu, Chunyun
• Format: Journal Article
• Language: English
• Published: Seoul The Korean Society of Automotive Engineers 01.04.2020; Springer Nature B.V; 한국자동차공학회
• Subjects: Automotive Engineering; Autonomous vehicles; Control stability; Control systems design; Controllers; Edge detection; Engineering; Fuzzy control; Lane changing; Lane keeping; Object recognition; Predictive control; Robust control; Simulation; Steering; Tracking control; Trajectory control; Trajectory optimization; Yaw; 자동차공학; Model predictive control; Lane detection; Trajectory tracking; Autonomous vehicle
• ISSN: 1229-9138; 1976-3832
• DOI: 10.1007/s12239-020-0027-6

A novel approach of combining lane detection and model predictive control (MPC) is presented to maintain the accuracy and stability of trajectory tracking control for autonomous vehicles. For the approach, the autonomous vehicles equipped with a camera act as the research object; image recognition technology such as dynamic region of interest (ROI) extraction, edge detection and Hough straight line detection are applied to extract the lane line. Next, to track the extracted lane line, the MPC based on a three-degree-of-freedom vehicle dynamics model is constructed; and the front wheel steering angle is corrected by the fuzzy controller based on the yaw angle and the yaw rate. In order to verify the effectiveness of the designed controller, a simulation model is established based on Matlab/Simulink-Carsim. And the simulation is performed under double lane change road. The simulation results show that the root mean square error of the optimized tracking trajectory and the expected trajectory is reduced by 19.35 %, which indicates that the designed controller shows good robustness and adaptability.