An analytical eco-driving trajectory planning method with field test validation at intersections

Abstract

Existing eco-driving strategies at intersections often rely on computationally intensive optimization or predictive models, with limited validation through field tests and challenges in robustness. This study proposes an analytical method for efficient eco-driving trajectory planning at signalized intersections. The method uses a cubic function approach to generate smooth trajectories for three distinct cases based on different spatial and temporal scenarios. Theoretical analysis demonstrates the smoothness and control feasibility of the proposed method. Numerical simulations and field experiments compare its performance with stop-and-go and optimization-based methods. Results show that the proposed method achieves over 90% of the energy-saving performance of optimization-based methods, with computation times reduced by more than 95% due to its analytical nature. Field tests reveal that real-world disturbances may reduce effectiveness by approximately 3%. Our code will be opened upon the acceptance of this paper.