Driver Behavior at Highway-Rail Grade Crossings Using NDS and Driving Simulators
Researchers at Michigan Technological University used Second Strategic Highway Research Program (SHRP2) naturalistic driving study (NDS) data and a driving simulator to perform a quantitative evaluation of driver behavior at highway-rail grade crossings (HRGCs). They developed a 3-point scale to generate driver behavior scores for over 9,000 NDS traversals. They used the mean scores to perform statistical comparisons of driver behavior at HRGCs with different traffic control devices (TCDs), with and without an accident history, and in various environmental conditions. They also simulated two HRGCs in a driving simulator and compared the driver behavior scores between naturalistic and simulated environments.
This study revealed that most drivers did not visually scan for trains and did not prepare to stop, regardless the type of warning device present at the crossing or the environmental conditions at the time of traversal. The results were fairly consistent in both NDS and simulated approaches. The NDS data analysis showed little statistical difference in driving behavior between any of the TCDs analyzed. The only exceptions were the significantly higher mean scores at passive HRGCs equipped with stop signs and the higher mean scores for day versus night traversals.