Elderly driver retraining using automatic evaluation system of safe driving skill

In Japan, although the rapid aging of the population has caused serious traffic problems, only a few studies have investigated the behaviour of elderly drivers in real traffic conditions. The authors have been developing a system to automatically evaluate safe-driving skill through small wireless wearable sensors that directly measure the driver´s behaviour. The authors aim is to promote safe driving by providing a personalised training program according to the individual´s own shortcomings in driving behaviour. By employing the sensors together with GPS and driving instructors´ knowledge, our system can automatically identify shortcomings in driving skill with an accuracy of over 80%. In February 2010, the Kyoto Prefecture Public Safety Commission, in Japan, certified our system as the first and only support tool for its `mandatory retraining course for elderly drivers´ that all elderly drivers, aged over 70 years, are required to take when renewing their driver´s license. In this study, the authors discuss the effectiveness of our system and investigate elderly drivers´ behaviour through a large-scale demonstration experiment, involving 749 elderly drivers, in the mandatory driver-retraining course on public roads. The authors results reveal that although elderly drivers are able to maintain a safe vehicle speed, their tendency to not scan around their vehicle to ensure safety makes their driving risky.