MPC-based Path Tracking Control with Forward Compensation for Autonomous Driving
Jiangfeng Nan, Bingxu Shang, Weiwen Deng, Bingtao Ren, Yang Liu
Abstract
Path tracking is a fundamental and important part of the automated vehicles, which ensures the vehicle to drive along the desired path accurately. Although great progress has been made in path tracking control in recent years, it still faces serious challenges in adaptability and robustness under different driving conditions. In this paper, an improved path tracking control method is designed with combining the feedforward compensation and model predictive control (MPC). The proposed method uses a pure pursuit algorithm to calculate a feedforward compensation based on expected geometric path, as a steady-state input for feedback tracking control. Then the MPC algorithm is utilized to calculate the optimal control input increment in the feedback controller in order to reduce tracking error and deal with the system nonlinear constraints. Finally, the simulation results show that the proposed method has good path tracking performance, even operating on the nonlinear region of tire dynamics, and good robustness on roads with different friction coefficient.