An investigation of factors contributing to major crash types in Japan based on naturalistic driving data

Abstract

Traffic accident statistics suggest that the human errors contributing to major crash types in Japan are predominantly failures in safety confirmation and hazard recognition that result in delayed response. A naturalistic driving data acquisition system was developed to investigate the human factors that contribute to such accidents. A preliminary analysis was performed to evaluate the efficiency of the collected naturalistic data. An analysis of vehicle-to-motorcycle conflict data demonstrated that types of recognition failure differ by types of traffic situation encountered. This result suggests that naturalistic driving data can provide valuable information for investigating the factors that contribute to the risk of human error.