Feasibility study of drowsy driving prediction based on eye opening time

Paper #:
  • 2017-01-1398

Published:
  • 2017-03-28
Abstract:
Because drowsy driving causes serious traffic accidents, the prevention technologies are highly required. In this study, we propose a drowsy driving prediction method based on eye opening time. The challenge of this method is to predict driver’s strong drowsiness before they feel sleepy. We pay attention to eye opening movement during driving, because overlooking hazards is one of the causes of traffic accident and is closely related to recognition and drowsiness. Hence, we attempt to predict driver’s drowsiness from eye opening time. At first, we form hypotheses of drowsiness and eye opening time based on the results of previous studies. We assume that the standard deviation of eye opening time (SDEOP) indicates the driver’s drowsiness and consider two types of transition, that is, the increase and decrease of SDEOP. In order to confirm our hypotheses, we investigate a relation between drowsiness and SDEOP. As a result, the two types of transition were observed via a preliminary experiment in our test course (the number of drivers : 7, speed : 80km/h, run time : 1 hour ). Then, we develop a drowsy driving prediction method founded on the aforementioned hypotheses. The proposed method has upper and lower thresholds, and predicts the drowsiness when SDEOP goes across one of the thresholds. The thresholds are determined by an adaptation session to address the individual difference of SDEOP. Finally, we confirmed the possibility that our method can predict strong drowsiness 5 to 25 minutes before its occurrence by experiments in the test course (the number of drivers : 10, speed : 80km/h, run time : 1 hour).
Access
Now
SAE MOBILUS Subscriber? You may already have access.
Buy
Attention: This item is not yet published. Pre-Order to be notified, via email, when it becomes available.
Select
Price
List
Download
$22.00
Mail
$22.00
Members save up to 36% off list price.
Share
HTML for Linking to Page
Page URL

Related Items

Technical Paper / Journal Article
2004-11-16
Article
2016-02-02
Article
2016-02-02
Training / Education
2017-04-03