A Path Tracking Controller for Heavy-Duty Buses Based on Robust Model Predictive Control With a Novel Enhanced Kinematic Model

In this paper, a novel Enhanced Kinematic (E-Kinematic) model for Heavy-Duty Buses (HDBs) is suggested and employed to construct a robust Model Predictive Control (MPC) system for efficient path tracking. The E-Kinematic model is derived from actual vehicle yaw motion data, with its efficacy validated at both low and high speeds through frequency domain analysis. This model is then utilized in the development of a robust MPC system for the improved path tracking performance. To enhance the path tracking performance of HDBs, a new reference trajectory is introduced within the delay frame to account for the vehicle's inherent lateral motion delay. This approach prompts the vehicle to adhere to the reference trajectory based on its anticipated future state, effectively considering lateral motion delay of the vehicle. Furthermore, the output constraints of the MPC controller are designed to take into account the geometric characteristics of the vehicle, such as sidewalls and overhangs, to prevent potential lane boundary intrusion. Techniques of constraint tightening and smoothing are applied to ensure robust control in the presence of model uncertainty. The performance of the proposed system is comprehensively validated through simulations and real vehicle experiments over various curvatures.