Pedestrian-Vehicle Interaction in a CAV Environment – Explanatory Metrics

Principal Investigator(s):

Jon D. Fricker, Professor of Civil Engineering – Purdue University

Project Abstract:
The motivation for this study is to measure the interaction between pedestrians and motorists, so that the variety of expected interactions between pedestrians and autonomous vehicles can be documented. This knowledge is intended to better inform the programming of driverless vehicles and to provide guidance to officials contemplating alternative crosswalk designs. To seek data that would assist the experiments described above, videos of pedestrians crossing one-way streets at specially-marked crosswalks are being recorded to create an archive that can be studied to detect and document the variety of behaviors by pedestrians and motorists at semi-controlled sites. Here, “semicontrolled” means sites marked with “State Law Yield to Pedestrian Within Crosswalk” signs. [MUTCD Figure 2B-2. Unsignalized Pedestrian Crosswalk Signs] The timing of this study is quite good, because the one-way streets being videoed will soon become twoway streets, as part of a redesign of streets at Purdue’s campus. This will permit the study of the same pedestrian population at the same location under modified circumstances. Based on a preliminary viewing of the videos made to date, a list of metrics is being developed. The current categories are: General environment, Pedestrian behavior, Motorist behavior, and PedestrianMotorist interaction.

Institution(s): Purdue University

Award Year: 2017

Research Thrust(s): Human Factors, Modeling & Implementation

Project Form(s):