Kazuhiro Kozuka

Effect Analysis of Intelligent Speed Adaptation (ISA) in Advisory Mode on Community Roads

Abstract ISA is considered as an effective measure to reduce traffic accidents. To make the possibility analysis, an experiment is conducted by using a driving simulator. The outputs from the simulator are analyzed combined with the ages and road conditions. The experiment shows that the ISA can improve recognition to speed limitation and drivers’ behavior. Furthermore, drivers react significantly to voice assisted ISA; audible advisories by the ISA such as reduction of speed are acknowledged by drivers concluding positive effects on traffic safety.

Detection of the gaze direction of a driver using the time-varying image processing

Abstract To realize driving support systems friendly to drivers, it is important to detect the driver states (consciousness degradation, inattention etc.). This paper describes the development of a technique to detect the gaze direction of a driver during driving by the time-varying image processing. In this method, the facial direction and the gaze direction can be detected separately. The verification experiments proved that the developed method is applicable for the detection of inattention.

Effects of Intelligent Speed Adaptation on Elderly Drivers’ Driving Behaviors and Mental Workloads

This study investigates the potential impacts of implementing mandatory and advisory intelligent speed adaptation (ISA) for elderly drivers driving on neighborhood streets. Using a driving simulator, driving behaviors such as vehicle speed and position of the gas pedal were analyzed. In addition, the mental workloads that could be represented by pupil dilation were measured using an eye camera device. Because elderly drivers tended to drive below the speed limit, implementation of ISA had little effect on reducing the speed driven at on neighborhood streets by elderly drivers. However, the implementation of ISA did serve to significantly reduce the mental workload for elderly drivers.

Probability of driver's state detection based on systolic blood pressure

We attempted to detect a typical drivers surprised state based on blood pressure using a driving simulator. The results show that the drivers systolic blood pressure after a child and a bicycle rushed out on the road is higher than that before child and bicycle rushed out, and these two data are statistically significant. This means that the systolic blood pressure became high when the bicycle rushed out, thus suggesting that the feeling of “traffic near-miss” (surprised) can be detected via blood pressure. In addition that, systolic blood pressure is more adequate index to detect drivers surprised state than LF/HF, which is generally used to detect or estimate drivers surprised state.