Urban transportation is experiencing rapid changes with the introduction of new micro-mobility options. These include shared e-scooters, sit-down scooters, e-bikes, and a range of other options. Within this mix of new modes, pedestrians face substantial and growing risks on American streets where road designs cater to drivers and automobiles dominate numerically. Over 6,000 pedestrians died on U.S. streets in 2018, an 14% increase over 2015 and a 27% increase over 2014. Human behavior, road design, environmental conditions, and technology each influence the prevalence of crashes and improving traffic safety for all users likely requires improvements along several of these dimensions. The rarity of pedestrian fatalities and those injuries associated with e-scooters makes disentangling the causal factors difficult. Hence, this project focuses on “near misses” between vehicles and pedestrians, e-scooters, e-bikes, and bicycles. The objectives of this project are to (a) use new methods to gather better data on what determines pedestrian and micro-mobility risk, (b) create tools that deliver more integrated solutions in collaboration with industry micro-mobility partner Bird, and (c) test the tools in the service of the needs of the real communities of New Brunswick and Highland Park, NJ. These tools will explicitly integrate both the social and technology solutions to improve safety. Project tasks include: (a) create an enhanced near miss detection capability using multiple visual sensors and advanced computer vision techniques; (b) perform behavioral experiments using both traditional tools (signage, temporary road reconfigurations) and smart-city tools (sensor-equipped and networked mobile actors and intersections); (c) conduct technological experiments integrated into a prototype mobile-phone-based app for pedestrians, e-scooters, e-bikes, and drivers; and (d) convene a community deliberation process that informs development of a local smart transportation plan.

The intellectual merit of this proposed project resides, first, in the development of a test bed equipped to evaluate social, technological, and integrated risk-reduction strategies for vulnerable road users. One component is the refinement of computer vision algorithms to much more accurately detect pedestrians, e-scooters, e-bikes, vehicles; measure trajectories (direction, velocity); measure near misses (deceleration, proximity, avoidance behavior); and distinguish key user attributes (clothing brightness, gender, race, size). A related contribution is to use digital models of the built environment to improve the performance of the computer vision algorithms and allow spatially explicit tracking of different entities. The second major contribution is to acknowledge the sequencing and layering of social and technological strategies as part of an integrated risk reduction portfolio. Vulnerable elderly, children, and under-represented minority pedestrians and cyclists will benefit because they are currently at disproportionate risk. The project will involve undergraduate and graduate students in the research activities through multidisciplinary capstone design classes, the summer Rutgers' Aresty Undergraduate Research Program, and the summer RISE (Research in Science and Engineering) program which introduces outstanding minority students to graduate-level research through summer jobs with research groups.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.