Title :
Road Danger Estimation for Winter Road Management
Author :
Moiseets, Pavel ; Tanaka, Yuichi
Author_Institution :
Grad. Sch. of Inf. Sci. & Technol., Hokkaido Univ., Sapporo, Japan
Abstract :
With over a billion vehicles on the road of the world traffic and traffic related issues are becoming extremely important topics of research. The identification of dangerous sites can help with managing the traffic and introduction of new policies, but sometimes lack of detailed data can preclude the use of sophisticated methods. In this paper we use probe car taxi data and retrospective accident data for evaluation of road danger in the city of Sapporo in winter time. We discuss the relationship between weather and traffic accidents and the difficulty of predicting the location of new accidents from available retrospective data alone. Then we show a correlation between different types of traffic, obtained by clustering probe car data, and the accident rates. We propose a method for estimating the danger levels of road segments in a broad area of the city based on the traffic data. Finally we give an evaluation of the proposed method and discuss improvements that can be made in the future.
Keywords :
data handling; road accidents; road safety; road traffic; road vehicles; Sapporo; probe car taxi data; retrospective accident data; road danger estimation; road traffic accident prediction; road vehicles; winter road management; Accidents; Cities and towns; Probes; Roads; Snow; Vectors; clustering; probe-cars; traffic accidents; winter road management;
Conference_Titel :
Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC), 2014 International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4799-6235-8
DOI :
10.1109/CyberC.2014.61