Enhancing Vehicle Safety in High-Traffic Environments: An Optimal Segmented Trajectory Planning Approach

Abstract

This paper proposes a novel approach for segmented trajectory planning of autonomous vehicles in high-traffic environments, which utilizes optimization techniques with state constraints to avoid collisions with obstacles. The approach focuses on generating an optimal trajectory for lane changes while maximizing ride comfort and maintaining safe distances from obstacles. To achieve this goal, this paper presents an trajectory optimization approach based on Pontryagin's Minimum Principle, well-suited for solving optimal control problems. The proposed approach incorporates segmented and optimized trajectory planning using a jump condition to prevent collisions with state constraints on the trajectory. This approach provides a geometric method for avoiding collisions while considering the movement of obstacles, ensuring the vehicle navigates safely and smoothly through its environment. Simulation results demonstrate the effectiveness of the proposed approach regarding the enhanced driving safety and ride quality.