MPC-Based Path Tracking Control for Autonomous Vehicles on Varying Friction Roads

The focus on autonomous vehicles (AVs) has been steadily increasing lately, owing to the development of new technologies. While path tracking control is a significant module of autonomous driving, some control methods fail to consider friction variations on actual roads. By integrated consideration of road friction variations on the tracking performance, this paper proposes a new model predictive control (MPC) method taking into account desired curvature and designed adaptive speed. Moreover, an adaptive weight strategy is proposed in this new MPC framework based on variations in road friction and vehicle speed. Co-simulations considering varying road friction conditions were conducted for autonomous vehicles by CarSim and MATLAB/Simulink to verify the effectiveness of the proposed scheme. Findings indicate that the suggested MPC can effectively achieve safe and comfortable path tracking under different friction coefficient road conditions, both at low and high initial speeds.