Path planning and tracking for autonomous vehicle based on model predictive control with variable horizon parameters

Abstract Lane-changing (LC) path planning and tracking are the main technologies used to avoid obstacles in autonomous vehicles. To increase safety in path planning and improve tracking performance, an path planning and tracking scheme based on the safe LC area algorithm and the optimized variable horizon parameter model predictive control (MPC) is proposed. First, a judgment method for the safe LC area is presented based on the comprehensive synthesis variable time-headway model, and a quintic polynomial LC path is planned. Second, under different driving speeds and curvature of the path, the weight-adaptive particle swarm optimization algorithm is used to optimize the horizon combination (HC), and the fitting relationship between driving speed, curvature and HC is obtained to calculate the optimized HC in real time, which improves the accuracy of path tracking and reduces the calculation time. Finally, the results show that compared to the other MPC algorithms, the average sum of lateral errors of the proposed adaptive horizon parameter-MPC algorithm is reduced by 21.69%, and the economy and comfort are improved by 13.27% and 13.39%, respectively.