Research on the Intelligent Vehicle Trajectory Tracking Control Based on Optimal Preview Distance

Accurate and responsive trajectory tracking is a critical challenge in intelligent vehicle control system. To improve the adaptability and real-time performance of intelligent vehicle trajectory tracking controllers, we propose a genetic algorithm adaptive preview (GAAP) scheme that offline optimizes the preview distance based on vehicle speed and reference path curvature. The goal is to obtain the optimal preview distance that balances tracking accuracy, stability, and real-time performance. By establishing a relationship between optimal preview distance, speed, and curvature, we enhance real-time performance through online table checking during trajectory tracking. Our trajectory tracking error model takes into account not only position errors but also heading errors. A feedback–feedforward trajectory tracking controller is then designed to achieve rapid responses without compromising robustness. Simulation tests conducted under straight circular arc condition and double lane change condition using CarSim/Simulink validate the effectiveness of our proposed scheme. Experimental results indicate that our proposed GAAP scheme improves real-time performance by approximately 86%, with a maximum response adjustment time of only 0.2 s, demonstrating significant advantages over existing schemes.