No abstract

Advanced light distribution control technology of the headlamp has developed and it is becoming possible to present information to the driver by graphics on the road surface as well as forward irradiation by ADB. On the other hand, the effect of presenting risk information by the function of the advanced headlamp on driver's behavior is not clear. In this study, we set various irradiation patterns of headlamp on the driving simulator and conducted experiments on collision avoidance behaviors at night. As a result, it was confirmed that the brake reaction time of drivers presented risk information by the headlamp to the road surface was shorter than drivers without presenting information. Also, we found that drivers perform brake operation more strongly by presenting rectangle-line based on the stopping distance algorithm on the road surface. Presentation of rectangle-line indicating the braking distance seems to reduce the frame line from the front of drivers, it was confirmed that this presentation method was easily recognized as a warning by drivers. Furthermore, the probability of collision against the obstruct was determined by data simulation based on the driving operation data obtained in the experiment, and it was found that the probability of collision decreases due to the risk information presentation.

In this research, we investigated the effect of allowance time on driver behavior when the Request-To-Intervene (RTI) at nighttime driving. We also analyzed driver behavior at RTI using the concepts of Situation Awareness (SA) and the behavioral analysis description method. The experiment was performed using a driving simulator, and a RTI occurs when the subjects was driving using level 3 of automated driving at night. After that, the subjects operate the steering wheel to avoid obstacles. As a result of the experiment, the lighting area of the driving beam was limited at night, so the obstacles moved out of the lighting area when Time-To-Collision (TTC) exceeded 7 seconds. Therefore, the driver was unable to responds to RTI driving behavior because no obstacles were detected. On the other hand, it was found that if TTC is shorter than 5 seconds, there is a high possibility of inducing a sudden operation the driver, and there is not enough time to confirm driving safety, so we think this driving condition is not suitable for RTI timing. Therefore, we think 7 seconds is enough for RTI timing at night. If this timing is secured, safety operation to avoid the collision to the obstacle is possible.

No abstract