Laser‐based SLAM automatic parallel parking path planning and tracking for passenger vehicle

Abstract

Automatic parking system (APS) plays an important role in alleviating drivers' burden and improving vehicle safety. This study proposes a laser-based simultaneous localisation and mapping (SLAM) automatic parallel parking and tracking control scheme, which include environment perception and reconstruction, parking path planning, and path following. First, the proposed approach selects short-range lidar to enlarge its perceptron on parking environment and senses available parking space to build neighbourhood mapping with the help of simultaneous localisation. The test evaluation indicates that laser-SLAM has achieved good performance for recognising free spaces. Then, scene-dependent B-spline parallel parking path planning algorithm is proposed to optimise the parking path curve and makes the path meet vehicle kinematics model constrains well, in which vehicle actuator's dynamic performance and parking path smoothness are taken as key attributes. The experiment results indicate that the designed parallel parking path is better smooth. At last, a path following control method based on model predictive control (MPC) algorithm is proposed. The authors built an intelligent vehicle APS platform, and the results indicate that the designed MPC algorithms has excellent robustness and can minimise the tracking error. All the proposed scheme is quantitatively evaluated using different parking scene simulation with satisfactory performance.