A Study on Density Map Based Crash Analysis

摘要

Travel path of vehicles contains valuable information on real-time traffic environment and situation. Recently, due to the technology advances, collecting travel paths of vehicles became easier. In this paper, we tried to extract the crash related information from the vehicle trajectories. We assumed that the distribution of vehicle trajectories in opposite directions will have relation with the accident and performed an exploratory study using bicycle GPS data. The bicycle GPS data were collected from a segregated bicycle track including accident black spots. Collected data were divided into two groups according to the heading, and utilized to the construction of the density maps. Based on the density map, an intersection map which indicates the intersection ratio was built. The intersection ratio was compared with the crash risk and it showed minor relation.