Study on dynamic estimation method for detecting distracted driving using eye-movement and driving data during driving

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00247
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Tokyo University of Science
• Principal Investigator: Nishiyama Hiroyuki 東京理科大学, 理工学部経営工学科, 教授 (80328567)
• Co-Investigator(Kenkyū-buntansha): 大和田 勇人 東京理科大学, 理工学部経営工学科, 教授 (30203954)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 機械学習 / 帰納論理プログラミング / 表情情報 / 眼球運動 / 自動車運転 / 散漫運転検知 / 居眠運転検知

Outline of Final Research Achievements

In this research, we have studied the estimation method to detect distracted driving during driving of the car and developed a dynamic detection system using this estimation method. First, we developed software that dynamically collects gaze information and facial expression information using eye movement measurement device and 3D camera sensor. Next, from the obtained information, we generated a learning rule to detect distracted situations including dozing status using a logical machine learning system. In addition, we developed dangerous location identification software on map information and developed a software and a graphical user interface on the tablet PC to confirm whether distraction in the dangerous position occurred by dynamically checking the map information and the distraction situation during driving.

Free Research Field

分散人工知能