Effect Assessment of a Motorcycle Collision Warning System at Intersections

Abstract

With the advancement of information and communication technologies, specific intelligent transportation systems can now deliver timely warnings to road users before potential traffic conflicts, enhancing drivers’ risk awareness and ultimately contributing to road safety improvements. This study used five distinct intersections as testing grounds to develop a motorcycle collision warning system (MCWS). The system was designed by integrating motorcycle onboard units (OBUs), roadside units, and warning signs. Furthermore, artificial intelligence-based image recognition was employed to analyze the post-encroachment time (PET) associated with right-angle conflicts. The findings indicated that the PET between motorcycles generally decreased, while the PET between motorcycles and cars increased. However, when the intersection was located on an incline, changes in conflict patterns were minimal. Moreover, at four intersections, vehicle speeds were notably lower when warning signs were present than when they were absent. Future research should explore various aspects to further enhance the applicability of MCWS, particularly in reducing traffic conflicts at unsignalized intersections.