Vehicle Lateral Control for Highway Automation

TLDR: In this article, the authors proposed a frequency-shaped linear quadratic (FSLQ) control theory for lateral feedback and feed-forward adaptive control for highway automation, which allows to track the center of a lane with small error and maintain good ride quality under different vehicle speeds, loads, wind gust disturbances, and road conditions.

Abstract: The objectives of lateral control for highway automation are to let vehicles track the center of a lane with small error and to maintain good ride quality under different vehicle speeds, loads, wind gust disturbances, and road conditions. In this paper, the lateral feedback and feedforward controllers are designed to satisfy these objectives by utilizing the frequency-shaped linear quadratic (FSLQ) control theory. This design method allows that the ride quality be included in the performance index explicitly, and the high-frequency robustness characteristic be improved by properly choosing the weighting factors. It is shown that a controller with fixed gains does not perform satisfactorily under all conditions. Therefore, an estimator for comering stiffness of the tires is proposed to enhance performance. Simulation results show that this adaptive control approach works satisfactorily under a variety of conditions including intermittent measurement of lateral tracking error.