Development of Driver Identification Method using Driving Behavior at Intersections

Abstract

In recent years, the technology development related to automobiles has been remarkable. In particular, ADAS (Advanced Driver Assistance System) is a technology that improves the safety and convenience of drivers. In order to improve the performance of ADAS, it is important to identify the driver. By identifying the driver, appropriate support and information can be given to the driver. In particular, the technology of identifying a driver using the driving behavior has many advantages over identifying the driver using biometric authentication.In this study, we considered that the driving operation at intersections is complicated　and individual differences are likely to appear. Therefore, we constructed a driver identification method using the driving behavior of intersections, which has not yet been focused on in previous studies. It was shown that the driver can be identified with high accuracy by applying LSTM (Long Short-Term Memories) to the driving behavior at the intersection.