A Dynamic Emergency Lane Clearing Method for Urban Roads Considering Vehicle Turning Information in a Connected Vehicle Environment

Abstract

In complex urban traffic environments, the timely passage of emergency vehicles (EVs) is a significant challenge, often hindered by traffic congestion and insufficient coordination among road users. Enhancing emergency response capabilities is crucial for safeguarding lives and property. With advances in automated driving technologies, connected vehicles (CVs) are expected to cooperate to reduce congestion and ensure prioritized EV passage on urban roads. This study proposes a real-time dynamic emergency lane clearing method for CV environments. A bi-level optimization model is developed for segmented road sections. The lower-level model employs an improved A* algorithm that incorporates vehicle turning information to generate optimal lane-changing trajectories for normal vehicles in the shortest possible time, aiming to minimize their lane-changing costs. The upper-level model is formulated as a mixed-integer nonlinear programming (MINP) problem to determine the optimal trigger points for segmented clearance. By triggering lane clearance in a stepwise manner, the model aims to minimize EV interference and maintain its desired speed. Numerical experiments show that the proposed method significantly reduces deceleration delay by 50% and the number of affected vehicles by 45% compared to traditional strategies. Sensitivity analyses further demonstrate its adaptability to varying road saturation levels and segment lengths, highlighting its potential for real-world deployment in CV-enabled urban environments.