Division of beijing road based on the driving pattern

Abstract

Congestion is the main feature of traffic in megacities, which has led to severe problems for energy and the environment. At the same time, moving vehicles as the traffic participants can real-time feedback fundamental traffic parameter information such as the interval velocity, average energy consumption and so on. This paper does research on the division of Beijing road which is the basis to calculate the link energy consumption of Beijing. The paper firstly defined the driving pattern and put forward a clustering model of Beijing road link based on the vehicle-specific power (VSP) distribution. Then, by associating the clustering result and other link attributes, a link-based driving pattern classifier is established. At the end of the paper, some experiments are taken to verify that this method made a great classification accuracy of 86 %. © 2013 Springer-Verlag.