A Proposal of Real-time Safety Driving Advice System Using Automated Evaluation Method of Driver's Skill

Abstract

In this paper, we propose a real-time safety driving advice system based on automated evaluation method of driver's skill. Our system can detect potentially dangerous situations (e.g. a driver enters a blind intersection with excessive speed without scanning around) and can provide safety advice for a driver to prevent traffic accidents in real time. Through experiments with 20 subjects, we confirmed that we can improve drivers' behavior by providing real-time safety driving advice.