Cooperative CAVs optimal trajectory planning for collision avoidance and merging in the weaving section

Abstract

Weaving sections may cause massive congestion and accident problems. Connected and automated vehicles (CAVs) are acknowledged to improve traffic safety and efficiency through effective communication and control. To this end, this study proposes a centralized cooperative vehicle trajectory planning framework for SAE Level 4 or 5 automation. Specifically, focusing on the complex movements at weaving sections, the longitudinal optimal trajectory control is proposed to avoid collisions. This improves traffic efficiency and reduces fuel consumption and driver discomfort. A sideswipe collision prediction algorithm takes into account the geometric features of vehicles and predicts the time of the collision. The merging sequences model with safety constraints is developed to avoid the collision between the on-ramp and off-ramp vehicles. The effectiveness of the proposed model is validated through simulations, where the proposed method is compared with the baseline to demonstrate its potential in improving safety and reducing the fuel consumption and travel time.