Drowsy driving is one of the main causes of traffic accidents. To reduce such accidents, early detection of drowsy driving is needed. In previous studies, it was shown that driver drowsiness affected driving performance, behavioral indices, and physiological indices. The purpose of this study is to investigate the feasibility of classification of the alert states of drivers, particularly the slightly drowsy state, based on hybrid sensing of vehicle-based, behavioral, and physiological indicators with consideration for the implementation of these identifications into a detection system. First, we measured the drowsiness level, driving performance, physiological signals (from electroencephalogram and electrocardiogram results), and behavioral indices of a driver using a driving simulator and driver monitoring system. Next, driver alert and drowsy states were identified by machine learning algorithms, and a dataset was constructed from the extracted indices over a period of 10 s. Finally, ensemble algorithms were used for classification. The results showed that the ensemble algorithm can obtain 82.4% classification accuracy using hybrid methods to identify the alert and slightly drowsy states, and 95.4% accuracy classifying the alert and moderately drowsy states. Additionally, the results show that the random forest algorithm can obtain 78.7% accuracy when classifying the alert vs. slightly drowsy states if physiological indicators are excluded and can obtain 89.8% accuracy when classifying the alert vs. moderately drowsy states. These results represent the feasibility of highly accurate early detection of driver drowsiness and the feasibility of implementing a driver drowsiness detection system based on hybrid sensing using non-contact sensors.

In seating comfort research, it is known that the pressure should not exceed a certain threshold from the viewpoint of tissue compression and should be widely distributed. However, its ideal distribution is not defined in past research. It is also known that the comfortable pressure distribution is not always constant and has individual differences. It is assumed that this is due to the influence of individual differences in body shape, such as skeletal shape and flesh of the seated person, and individual differences in sitting posture, but the mechanism has not been clarified by analyses including these factors. From the above, it is considered that the comfortable pressure distribution cannot be explained only by the mechanical state. In this study, we focused on the pressure sensitivity of thighs and buttocks and performed an analysis assuming seating in an automobile seat. We determined the exponent of Steven’s power law for seat pressure by measuring local perceived pressure load that felt the same pressure feeling at the reference load point, and the sensitivity distribution of 29 participants were measured and classified them into 4 types. The comfortable pressure distribution of five participants was measured using the experimental seat and converted into a perceived pressure distribution using the sensitivity distribution. The results show measured pressure distribution is not the same as perceived. Analysis of the perceived pressure distribution suggests that the comfortable perceived pressure distribution is a uniform distribution that falls within a certain range for the minimum pressure.

Indoor temperature affects the arousal level and thermal comfort of drivers. In this study, we aim to prevent the drowsy state and to improve the thermal comfort of drivers in driving for a long time using changes in ambient temperature in cabin. Targeting the scene of driving for 90 minutes, the hypothesis that repeated thermal stimulation to the driver can prevent the decrease of arousal level and thermal comfort of the driver was tested. We investigated the drowsiness level, thermal comfort based on subjective evaluation, and physiological parameters of the driver in the condition of repeated thermal stimulation for 90 min with driving in a simulator. As a result, the drowsiness level with a scale from 0 (alert) to 4 (extremely drowsy) was kept below 1 (slightly drowsy), and the comfort sensation with a scale from -3 (very uncomfortable) to 0 (comfortable) was maintained above -1 (slightly uncomfortable). These results suggest that repeated thermal stimulation is valid to prevent the decrease of arousal level and thermal comfort of the driver.

A digital human body model as a virtual occupant surrogate for the riding comfort simulation is developed for both 1D lumped network (Modelica) and 3D mesh based (Finite Element) solutions. Since the composition of 1D and 3D versions of the human body model has a similar multibody system architecture, the kinematic responses from both solutions are almost equivalent. The models are therefore complementary, since the economic 1D models can serve effectively in design exploration and optimization, while their sophisticated 3D counterparts can serve in final design validation. The detailed modeling process and validation results against standard seat vibration excitation test are introduced in this paper, preparing the models for use in seat design.

This study focused on lower leg swelling as a typical physical load in prolonged sitting postures such as driving. We obtained prediction equations for lower leg swelling (ratio of the inverse of the measured impedance to the initial impedance and lower leg swelling assessment value, BI [%]) from thigh pressure distribution, participants’ physical characteristics, and sitting time. A total of 22 participants (11 males and 11 females) were recruited. The impedance in the lower leg and thigh pressure distribution were measured over 90 min in a sitting posture at three tilt angles (8°, 0°, and −8° from the horizontal plane). Multiple regression analysis was performed to construct prediction equations for lower leg swelling in the males, the females, and all the participants. Bioelectrical impedance was selected as the dependent variable, with height, body fat percentage, thigh pressure distribution, and sitting time as the independent variables. The validity of all constructed prediction equations for the males, the females, and all the participants was confirmed by an adjusted R². These findings can be used to develop a device to prevent lower leg swelling (the main problem resulting from a prolonged sitting posture) and can be applied to automobile seats, aircraft seats, and office chairs.