Guidelines for safer pedestrian crossings: understanding the factors that positively influence driver yielding to pedestrians at unsignalized intersections.
Principal Investigator(s):Raphael Stern, Assistant Professor, Civil, Environmental and Geo-Engineering
When a pedestrian tries to cross an unsignalized intersection that a driver is approaching, there are two primary potential outcomes: the driver can yield or the pedestrian can yield. Many factors influence the outcome of the interaction between a driver and a pedestrian at an unsignalized intersection. This project is focused on understanding what factors influence the driver yielding rate at unsignalized intersections, and using that knowledge to create guidelines for designing intersections that encourage drivers to yield to pedestrians.
One factor thought to influence driver yielding is the speed at which the driver is approaching the intersection. Another factor that may influence yielding rate is the sight distance, which influences how easily a driver is able to perceive a pedestrian approaching an intersection. Other factors that may influence yielding rate include land use context and lighting (e.g., morning, midday, evening), among others. This study is collecting data on how these factors influence yielding rate my collecting field data using pole-mounted video cameras in conjunction with radar units that can identify the speed that the driver is approaching the intersection. The video data will be manually labeled to identify the outcome of each driver-pedestrian interaction. The data will then be analyzed using statistical analysis (e.g., logistic regression) to identify the likelihood of a driver yielding to a pedestrian as a function of the variables considered. This data will also be used to construct a table that tells the empirical probability of a driver yielding for each combination of speed and sight-distance observed.