Speed and Path Planning Optimization of Autonomous Vehicle to Minimize Lap Time in Racing Track

This study proposes a trajectory optimization method for racing vehicles, aiming to maximize speed and path planning performance by estimating tire friction coefficients. This method addresses the challenges inherent in high-speed racing environments, where tire friction and vehicle dynamics are critical for achieving optimal lap times. A friction circle model was used to determine the friction coefficient, which is subsequently applied to calculate the maximum feasible speed and acceleration constraints for different track segments. The optimization problem was formulated to minimize both path curvature and lap time, balancing these factors to determine the optimal trajectory. Experimental validation was conducted using a real vehicle on the AMG Speedway short track in Yongin, South Korea, where the proposed method demonstrated significant improvements in lap times across different tracks compared to the minimum curvature method. The proposed algorithm effectively minimized lap times and demonstrated applicability, with an average computation time of 30 s on a standard computer. This research provides valuable insights into autonomous vehicle path planning, particularly in racing contexts, and provides a robust framework for further advancements in autonomous driving technologies.