We have developed a vehicle simulator consisting of a three-dimensional CAVE display and a steering system which considers a steering reaction force. The CAVE provides stereoscopic vision in real time according to the movement of an eye position of a driver. Experiments for test subjects are carried out to validate the vehicle simulator and analyze the driver's behavior. Also from the analyzed data the feasibility of contribution to the driver's safety utilizing the vehicle simulator are discussed.

View full abstract

The steering reaction force is the most important information during car driving in the meaning that the situation between the tire and the road is feed back to the driver through the steering wheel. The result of developing a driving simulator that uses the motor that generates the steer reaction force is reported in the paper. The steering reaction force is generated by controlling the current of the motor according to the reaction force torque calculated using to the bicycle model. The study result of the lane keeping control utilizing the developed driving simulator is also reported.

View full abstract

This paper shows that it is possible to improve the control performances of the Stewart platform motion using a dynamic model and a new control method namely IDCS. By means of numerical simulations we compare the IDCS performances with traditional control methods. First, we obtain the inverse and forward dynamic models of the Stewart platform motion simulator in terms of the Newton-Euler method. Second, we perform numerical simulation for attitude control of the platform by means of PD control or DDCS control. Finally, we verify that the tracking performances and the system responsiveness of IDCS control are superior to traditional PD control.

View full abstract

The main objective of this study is to investigate the factors which affect the roadside signboard cognition during driving. As the experimental subject to this research, 10 kinds of signboards on suburban roadside shops are studied. Regarding the study objective, the visible distance of signboards in Driving Simulator were divided into three groups; detectable distance, confirmable distance and legible distance. These three distances, solid angle, and horizontal angle of boards were used as analysis data to examine how the factors of color size, and, width of boards easy the signboard cognition. The results of this study are as follows: Placement of color in high conspicuity over wide surface is effective at detecting a signboard. Placement of wide board closer to the road is effective at confirming a signboard. Placement of big letters or letters with logotype is effective at distinguishment a signboard.

View full abstract

We evaluated the probability of stopping inside a railway crossing due to lead vehicle's sudden stopping when "stop" before entering the railway crossing is deregulated. Road environment of driving along the railway crossing was reproduced in a driving simulator. 30 drivers participating this experiment drove at low speeds or at high speeds while following a lead vehicle. The headway distances to the leading vehicle and drivers' reaction time to the brake lamp of leading vehicles were measured in the driving simulator experiment. Additionally, the deceleration from the application of brake pedal to the vehicle stop was collected from driver behavior data in an actual road environment. The probability of stopping inside the railway crossing was estimated based on the measured data. The results suggest that the probability of entering the railway crossing differs according to the driving speed range: Slower driving speed leads to higher probability of stopping inside the railway crossing.

View full abstract

This paper proposes a method to characterize hurried driving from the viewpoint of attention allocation. When overtaking another car, drivers need to manage the collision risk by allocating their attention appropriately, thus increasing their mental workload. However, drivers' hurrying mental status likely arises in such situations, even though it is risky. Therefore, this study investigates the relationship between the risk of collision with surrounding vehicles and attention allocation when overtaking other vehicles. We will show that hurried driving can be characterized by outliers from the modeled distribution of normal driving in terms of the relationship between collision risk with the lead vehicle and continuous gaze time to the rear vehicle. Furthermore, a potential method to detect hurried driving is proposed by applying the results.

View full abstract

Eco-driving is a driving mode through driver behaviors mainly about accelerator works, such as smooth acceleration at start, preservation of steady speed, and active uses of engine brake. In this study, in virtue of a universal driving simulator (DS), driving experiments by subjects were conducted in the modes of normal driving, eco-driving, and eco-driving with eco-indicator. In the experiments, surface electromyography (sEMG) of leg muscles of driver was measured to clarify muscle fatigues in the different driving modes, and effect of mental workload was also analyzed from a subjectivity evaluation of NASA-TLX. The studies validated that the fuel economies can be improved by the eco-driving mode; however, the eco-driving requires more muscular activities than the normal driving. But the experimental results suggested that the muscular activities can be decelerated by using the co-indicator.

View full abstract