Characterizing and visualizing physical world accessibility at scale using crowdsourcing, computer vision, and machine learning

Abstract

Poorly maintained sidewalks and street intersections pose considerable accessibility challenges for people with mobility-impairments [13,14]. According to the most recent U.S. Census (2010), roughly 30.6 million adults have physical disabilities that affect their ambulatory activities [22]. Of these, nearly half report using an assistive aid such as a wheelchair (3.6 million) or a cane, crutches, or walker (11.6 million) [22]. Despite comprehensive civil rights legislation for Americans with Disabilities ( e.g. , [25,26]), many city streets, sidewalks, and businesses in the U.S. remain inaccessible. The problem is not just that street-level accessibility fundamentally affects where and how people travel in cities, but also that there are few, if any, mechanisms to determine accessible areas of a city a priori. Indeed, in a recent report, the National Council on Disability noted that they could not find comprehensive information on the "degree to which sidewalks are accessible" across the US [15]. This lack of information can have a significant negative impact on the independence and mobility of citizens [13,16] For example, in our own initial formative interviews with wheelchair users, we uncovered a prevailing view about navigating to new areas of a city: " I usually don't go where I don't know [about accessible routes] " (Interviewee 3, congenital polyneuropathy). Our overarching research vision is to transform the way in which street-level accessibility information is collected and used to support new types of assistive map-based tools.