Home // International Journal On Advances in Life Sciences, volume 10, numbers 1 and 2, 2018 // View article

Modeling of Drivers Distraction State based on Body Information Analysis

Authors:
Kazuhito Sato
Masafumi Sawataishi
Hirokazu Madokoro
Momoyo Ito
Sakura Kadowaki

Keywords: body information; driver behavior; human engineering; near miss; non-regulated intersection

Abstract:
For this study, we defined a "concentration state" as that when a driver performs only driving tasks and a "distraction state" as that when a driver performs a driving task and a mental arithmetic task simultaneously. Based on results of these driving tests, we elucidate the characteristics of safety confirmation behaviors by near-misses according to differences between two driving conditions when approaching an intersection. Specifically, using Bayesian Networks (BNs) to express the relation between safety confirmation behaviors graphically for driving scene and driver’s internal state, we analyze correlations between characteristic body information (i.e., eye-gaze / face orientation) and operation information (i.e., steering wheel, accelerator, and brake) when switching from a "concentration state" to a "distraction state" based on viewpoints of driving style and driving workload sensitivity. Using evaluation experiments, by constructing internal state estimation models of subjects for whom driving style and driving workload sensitivity are mutually opposite, we try to analyze body information and the operation information by which the influence of a distracted state appears easily through the specification of unique behavior patterns associated with each driving scene.

Pages: 42 to 53

Copyright: Copyright (c) to authors, 2018. Used with permission.

Publication date: June 30, 2018

Published in: journal

ISSN: 1942-2660