Effects of a Preventive Warning Light System for Near-Miss Incidents

Effects of a Preventive Warning Light System for Near-Miss Incidents

Akira Yoshizawa (Denso IT Laboratory, Inc., Shibuya-ku, Japan) and Hirotoshi Iwasaki (Denso IT Laboratory, Inc., Shibuya-ku, Japan)
DOI: 10.4018/IJSSCI.2018010105
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

This article describes how the number of fatal traffic accidents has been decreasing in Japan because of recent safety technologies of vehicles, such as stiff cabins, antilock braking systems, and seat belts. Automated vehicles and advanced driver assistance systems can advance the trend. However, many traffic accidents occur on narrow streets in residential sections, where it is difficult for even advanced vehicles to drive safely. In this research, this paper utilizes a near-miss incident database to analyze driver gazing. The result showed that preventive warning systems are useful for avoiding traffic accidents.
Article Preview

Introduction

Even though the number of transport fatalities in Japan is steadily decreasing, we still have many traffic accidents. For those promoting safety systems, such as antilock braking systems and airbags, drivers and passengers are becoming relatively safer. However, pedestrians, cyclists, and motorcyclists still face dangerous situations. Statistics showed that 83% of such accidents occurred on straight roads, and 35% of them were caused by aimless driving (ITARDA, 2012).

Driver-monitoring systems have been marketed to prevent such situations. For example, the Driver Status Monitor (Denso, 2014) developed by Denso detects drivers’ head angles. If the driver looks away for a certain time, it alerts the driver with beeping sounds to get the driver’s eyes back on the road. Of course, the system can be effective, but some dangerous situations still remain. For example, the driver may gaze off the road without turning his/her head. We thus need to develop a new system with an eye-tracking device.

There is research analyzing driver gazing. Most of this research has been conducted with driving simulators or with real vehicles but in a very simplified scenario of a planned procedure in advance in a test field (Fletcher & Zelinsky, 2009; Harada, Iwasaki, Mori, Yoshizawa, & Mizoguchi, 2013; Harada, Yoshizawa, Mori & Iwasaki, 2014; Harada, Iwasaki, Mori, Yoshizawa & Mizoguchi, 2014; Harada, Mori, Yoshizawa & Iwasaki, 2015; Kircher, Ahlstrom & Kircher, 2009; Mizoguchi, Ohwada, Nishiyama, Yoshizawa & Iwasaki, 2015; Nishiyama, Yoshizawa, Iwasaki & Mizoguchi, 2015; Yamashiro, Deguchi, Takahashi, Ide, Murase, Higuchi & Naito, 2010; Yonetani, Kawashima, Hirayama & Matsuyama, 2012; Yoshizawa & Iwasaki, 2014; Yoshizawa & Iwasaki, 2015; Yoshizawa, Nishiyama, Iwasaki & Mizoguchi, 2016; Yoshizawa, Nishiyama, Iwasaki & Mizoguchi, 2017).

Starting with a simple scenario is a reasonable to try to understand how traffic accidents occur. However, there is no guaranty of reaching the goal of explaining sufficiently all traffic accidents by summing up the knowledge we acquire from this research. Real road scenes are far more complicated than experiment scenes. There are many objects, such as vehicles, bikes, pedestrians, traffic signals, shops, buildings, mountains, and blind corners, and they change dynamically. Sometimes they change interactively with changes made by the driver. Weather and sunshine change as well, changing the appearance of all the objects. The real world is so complicated that the exact same situation never recurs.

There is also research to create driver models. Itti et al. proposed a saliency map, a fundamental model of eye fixation (Itti, Koch & Niebur, 1998). It shows that different appearances of objects lead to different gazing behavior. The saliency map is a bottom-up model of human visual attention. There is also a top-down aspect (Oliva, Torralba, Castelhano & Henderson, 2003). The top-down mechanism drives change of gazing direction based on the meaning of each object, prediction, driving custom, or searching action, in which the driver looks back and forth when he/she wants to turn left/right for example, independently of the saliency map, considering total risk. Attention is a core feature of human intelligence (Wang, Patel & Patel, 2012). And it is considered that the feature controls the top-down and the bottom-up process. It works so flexible that

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2019): Forthcoming, Available for Pre-Order
Volume 10: 4 Issues (2018): 1 Released, 3 Forthcoming
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing