U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

 
REPORT
This report is an archived publication and may contain dated technical, contact, and link information
Back to Publication List        
Publication Number:  FHWA-HRT-17-016    Date:  April 2017
Publication Number: FHWA-HRT-17-016
Date: April 2017

 

Leveraging the Second Strategic Highway Research Program Naturalistic Driving Study: Examining Driver Behavior When Entering Rural High-Speed Intersections

STUDY LIMITATIONS

This project succeeded in modeling driver behavior when approaching to rural high-speed intersections but was limited in two key ways. The number of observations was sufficient for basic analyses but prevented the estimation of more complex models. More than 5.4 million trips were collected as part of the whole NDS, but only 411 of those (less than 0.01 percent) were relevant to this project. Only by loosening the restrictions on intersection selection could this number be increased, but doing so could introduce differences by site, making the ability to make inferences about driver behavior even more difficult. There were several additional intersections identified in the RID that satisfied all criteria for this project, but the minor routes lacked link IDs, thereby making it impossible to request associated crossings. It was possible to expand coverage of the RID to include more rural intersections as defined here, but the authors are aware of no such plans.

The greatest limitation to this study lies in the reduction of eyeglances. Reduction of traffic conditions was relatively straightforward, but eyeglances were plagued by a number of issues. Participants wore sunglasses in 17.3 percent of crossings, so glance targets had to be inferred based on head movements alone. VTTI staff made notes regarding the eyeglance reduction process as well. For 12.2 percent of crossings, notes such as “dark/grainy video,” “bad sun glare,” and “face covered by sun visor” were made. Even under ideal circumstances, however, the reduction of eyeglances based on face video was much less accurate than existing eye-tracking systems common in highway driving simulators and field research vehicles. This report relies on the assumption that head movements correlate to glances to the intersecting roadway, but that is impossible to confirm.

This study was also limited in its ability to assess the effect of daylight on stopping and scanning behaviors. The time of day for each crossing was requested and successfully extracted. However, these times were given in 2-h bins. Such wide bins would have resulted in an unacceptable level of uncertainty regarding the daylight status of many crossings. Future attempts to assess this effect could be addressed by including daylight status in the video reduction protocol.

 

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000
Turner-Fairbank Highway Research Center | 6300 Georgetown Pike | McLean, VA | 22101