Blind zone accidents caused by queuing on adjacent lanes are a common type of intersection accidents, where conflicts between right-turning vehicles and pedestrians frequently occur, often with serious consequences. To improve the safety of intersections, this paper proposed a novel methodology to avoid blind zone collisions between right-turning vehicles and pedestrians based on the vehicle infrastructure cooperative technology. The methodology first detects the dynamic blind zones when right-turning vehicles are affected by queuing on adjacent lanes, and then identifies potential conflicts between right-turning vehicles and pedestrians in blind zones based on the GT indicator, using roadside LiDAR to obtain the real-time high-resolution pedestrian trajectory data, and provides a collision warning mechanism and speed advisory strategies of stop/pass for right-turning vehicles. The proposed algorithm was tested using the microscopic traffic simulation software VISSIM and VISSIM COM interface, results indicated that the algorithm could accurately warn conflicts without missing or false warnings, and achieve better performance even in the case of low CV penetration.
A V2X-based Approach for Avoiding Potential Blind-zone Collisions between Right-turning Vehicles and Pedestrians at Intersections*
20.09.2020
610303 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Transportation Research Record | 2021
|