Elaboration of Traffic Simulation System for Japan Nation-wide Network

Abstract

Providing proper advice for the low skilled driver is one effective way to improve drivers'skill. But there are mainly two problems. One is the classifier of high and low skilled drivers. The other is to make clear the difference between two groups. In this paper, we proposed a driving skill analysis method to utilize of data collected by telematics system. A feature extraction technology based on machine learning algorithm is introduced to detect the characteristic difference between high and low skilled drivers and build a classifier. Results of experiments operated on driving simulator show the feasibility of the proposed method.