An Analysis of Elderly Drivers', Cyclists' and Pedestrians' Behaviors Using Wearable Sensors

Abstract

Rapid aging of Japan's population has caused serious traffic problems in the country. In 2015, over half of the fatalities in traffic accidents were elderlies. However, only a few studies have investigated the elderly drivers', cyclists' and pedestrians' behaviors in real traffic conditions. We have developed a system to evaluate elderly persons' driving/riding/walking skills from the viewpoint of preventing accidents, using small wireless wearable sensors that directly measure their behaviors. In this paper, we study characteristics of risky behaviors in the elderly. Through an experiment involving 749 elderly drivers, we found that elderly drivers' tendency to neglect proper scanning to confirm safety contributed to their hazardous driving performance. We also found that elderly persons with no driving license behaved in a more hazardous manner while walking or riding bicycle compared to those with license.