Kotaro Hara: Improving Accessibility Navigation

By Marcus Fedarko
Student Staff Writer

When most people get directions, they assume the directions will be accurate and navigable. In the age of Google Maps and other geographical technologies, it's usually an easy assumption to make. For some users, however, conventional shortest-path directions may not be enough.

Graduate student Kotaro Hara's work, done in the Human-Computer Interaction Lab and advised by Assistant Professor Jon Froehlich, focuses on mobility technology. Certain sidewalks in cities are difficult to access for individuals in wheelchairs or with other mobility impairments, due to a lack of curb ramps or other accessibility features. Furthermore, data about these deficiencies is not readily available to individuals with mobility impairments before visiting a certain area. Hara's project facilitates the crowdsourced accumulation of this data: using Google Street View, individuals can easily label curb ramps at an intersection, and the accumulation of this data can be applied in a multitude of ways to improve navigation for people with mobility impairments.

“We want to show the most accessible path,” notes Hara. While Google Maps and other technologies can indicate the shortest path between two points, they won't account for accessibility—so a lack of curb ramps could make certain routes untraversable for the mobility-impaired.

The project is currently focusing on Washington, D.C. as a proof of concept, with a software framework designed to be scaleable to other cities and, hopefully, worldwide.

When asked where he sees accessiblity technology going in the future, Kotaro smiles—“It's a very broad area,” he says. When I specify accessibility technology for mobility impairments, he gives the same answer: it's still incredibly broad.

He hopes to integrate data into mapping technologies in the future, but notes that there's still a lot of possibilities available with the project. “It's a great combination of application and technology,” he says. He's currently working on training algorithms to detect curb ramps using this crowdsourced data. There's a lot of room to make a difference, and to change things for the better.

Hara admits that accessibility was not a planned PhD topic. “Jon [Froehlich] introduced me to this project,” he says. “I thought … it's a great project, a great cause, and then I got into it [laughs], and now I'm working on it years later.”

Interest Areas: 

The Department welcomes comments, suggestions and corrections.  Send email to editor [at] cs.umd.edu.