University of Maryland project looks to crowdsource an accessibility map of DC’s sidewalks

Other programmers at the recent Walk Hack Night presented data visualizations that explore walkability and walking connections in the Washington, D.C., region.

Across the United States, 30.6 million Americans older than 15 live with mobility impairments, a significant portion of whom require an assistive aid like a walker or wheelchair. Despite progress since the Americans with Disabilities Act of 1990, it is still difficult for many of them to navigate cities.

Manaswi Saha presenting Project Sidewalk's progress.

Manaswi Saha presenting Project Sidewalk’s progress.

To identify the conditions creating these barriers, a team at the University of Maryland is crowdsourcing a map of sidewalk impediments within the District of Columbia. Presenting at the latest Transportation Techies’s Walk Hack Night, Manaswi Saha showed attendees the mapping progress of the project, now called Project Sidewalk, since it was last demoed for the group in 2015.

Inspired by Walk Score, which rates the walkability of neighborhoods nationally, Project Sidewalk seeks to provide a better understanding of the District’s walking accessibility, especially for people who depend on assistive aids. Instead of creating ratings based on proximity to amenities, Project Sidewalk catalogs the quality and accessibility of sidewalks and curb ramps.

The web tool, deployed publicly last fall, currently depends on public participation to evaluate D.C. sidewalks. Users mark problem spots with one of four tags: missing curb ramp, obstacles like fire hydrants, surface problems like crumbling pavement, and overall missing sidewalks (as shown above). Using Google Maps’ Street View, the tool drops participants into random locations throughout the city and prompts them to audit up to 1000 feet of that neighborhood. So far, the 475-person Project Sidewalk community has covered 463 miles of D.C. roads with 64,000 labels. This data is will be available for specific routing instructions, and also informs a WalkScore-esque neighborhood rating system.

Saha hopes the project will eventually cover 100 percent of the District, but its dependence on crowdsourced information is time-consuming, labor-intensive, and makes it difficult to expand to other locations. Ultimately, Saha and her team plan to automate the process, teaching the program to identify problem spots on its own, and to communicate the information to relevant government entities tasked with fixing them. In addition, they aim to build accessibility-aware navigation, providing users directions based on their dependence on assistive aids.

walkshed capture

These tools were made for walking

The other presenters at Walk Hack Night followed similar paths, demonstrating tools to better understand the distribution and quality of pedestrian infrastructure.

  • Aaron Ogle shared the history of Walkshed.js (above), a visualization of what people can reach on foot from any given point. When it first began years ago, Ogle approached the project by thinking of walkability as a surface, rather than a network – similar to how water in a given area returns to the ocean. From this lens, a walkshed map can better address how a person can reach a certain point or amenity from any location. Using a color gradient that accounts for obstacles like rivers or highways, walkshed.js creates a strong visual sense of where walking is most convenient.
  • Similarly, Transportation Techies organizer Michael Schade has been developing an Arlington Walkshed tool to help employers in the county educate their workers about the myriad commute options within the radius of a short walk. Schade’s tool highlights areas that users can reach within a chosen timeframe from 10 to 60 minutes, using walking and other modes.
  • University of Virginia’s Andrew Mondschien describes Tysons Corner – a historically car-centric area in the midst of an urban transformation – as having only a “nascent” level of walkability. Mondschein and students have been mapping the current quality of walking around Tysons with wearable sensors, collecting a range of ambient information, from air quality and noise levels to types of land use. Over time, Mondschein hopes to use these datasets to measure walkability for people at a certain time within a complex area, and to understand how this changes over time. This data, he said, should be “as community-based as decision-making is,” leading residents to be more invested in their environment and providing the information to ensure planning is making their city more livable.
  • District Ninja’s Matt Triner and Amir Farhangi shared District of Pedestrians, a 2015 analysis of the safety issues facing pedestrians, built from D.C.’s Vision Zero website. Drivers failing to stop for pedestrians was the number one infraction, and the VZ data showed that neighborhoods close to each other tended to share similar problems, creating pockets of geographically-linked safety issues.
  • Finally, many companies are working to understand indoor walking habits (mostly to encourage visitors to buy stuff). Chris Fricke of Geometri explained how Bluetooth sensors placed in lighting fixtures track customer movements within buildings. Using this data, programmers can determine the position of people within three feet, analyzing how they move around stores and where they spend the most time.

Images, from top: A screen capture of the Project Sidewalk crowdsourcing tool. Manaswi Saha presenting at Walk Hack Night (MV Jantzen, Flickr, Creative Commons). A capture of Walkshed.js in Philadelphia.

Share this item

Be the First to Comment