A Change in Direction

We are Team AMATA, a student group from Carnegie Mellon University, MHCI program. We partnered with SWPPA and pathVu to advocate for older adults and people with disabilities, and create a pedestrian navigation experience that is safe and accessible for ALL.

“We can carry you up the stairs!” — well, don’t call that accessible!

From our prototype testing, we discovered that education is a huge barrier for able-bodied people to crowdsource sidewalk conditions, as they generally have little understanding of the traveling experiences of a person using a wheelchair. Users also lack the motivation to participate in crowdsourcing activities, especially if they can’t get practical benefits or value from it.

Therefore, we turned directly to wheelchair users to hear their stories related to accessibility and techniques evaluating route conditions and obstacles. We hypothesized that their experiences traveling on sidewalks are greatly affected by path conditions, and they will be motivated to actively use an app that provides them with relevant accessibility information.

We conducted 4 semi-structured interviews in total, with 1 manual wheelchair user and 3 power wheelchair users. We the following to be common across these users:

• Many wheelchair users have similar ideas of what is uncomfortable and/or difficult to navigate.
• People will choose an efficient route over a comfortable one, as long as the efficient route is significantly more efficient and still possible for them to navigate.
• People care about the accessibility of their destination, and some will pre-plan to ensure it’s accessible.
• People want to make their own judgments about whether something (a route, an entrance, etc.) is accessible for them, so they want objective data about it.
• People are motivated to help others, especially other people with disabilities.
• People find it difficult to trust accessibility information online; What people say it’s accessible, it may not be accessible for me.

Models, models, models!

After the interviews, the team felt that the best way to consolidate what we learned was to make a journey map of how our participants traveled! This model helped us really see where people are struggling on their journeys and where we could find opportunities. We found four main parts to the journey and mapped thoughts, experiences, and feelings along all parts of the trip to see where struggles existed. Everyone we talked with had to plan their trip, travel some distance on a sidewalk, and finally arrive at the destination but some people had an optional second step of driving or taking transit as part of their journey. The most individual differences we found were actually at the planning stage while most people experienced similar phenomena during the ending parts of the journey.

(Alt text: A model showing the four stages of a trip from planning a route to arriving at the destination with notes, thought bubbles, and emotional highs and lows along each section of the journey.)

Our team hypothesized that the last bit of the journey would be the hardest part of the journey and according to who we talked to, that was definitely the case! We mapped the emotional journey along all four stages and found the most frustration, anger, and upset when traveling on sidewalks and trying to find the entrance of a destination.

(Alt text: A close up photo of the above model showing only the stages for traveling on the sidewalk and arriving at the destination with highlighting on the very frustrated emotions in these stages.)

Overall, mapping the findings from our interviews was helpful for the team to really understand where frustrations are most common. We’ll be revisiting and editing these journey maps as we learn more to help ensure we’re working on the problems with the most impact for our users.

Meet Abi, Moe, and Olson

In order to ground our findings in our users and to help us design a solution that keeps their needs at the center, we created some personas. We started by identifying what we currently believe are the facets most likely to impact our design and how people use it. We ended up with 6 categories and created continuums for each:

• Tech literacy — how comfortable and able are they exploring and using current technology, such as smartphone apps
• Motivation — to take action to help themselves and/or others
• Age
• Vision ability
• Motor ability
• Education — how knowledgeable about accessibility issues are they

We mapped all of our interview participants and other key stakeholders on these spectrums to see where there are commonalities and help us build personas that are based in real data.

Once we had done this more analytical work, we narrowed down to the ones that we felt would be users of our design and built them out into traditional personas, with backstories, personalities, and goals. Meet Abi, Moe, and Olson:

Abi!

Moe!

Olson!

Though this was a bit of a longer process, it helped ensure we were basing our personas in data and that they would be useful for the remainder of the design process. Then infusing the more human elements at the end brought our focus back to the people we’re designing for.

Now that we have learned a lot and consolidated those findings into models and personas, we realized we were at a point where we could reframe our project.

The Reframe Game

What is the “Last Mile”?

The “Last Mile” refers to the part of the journey that is not supported by navigation systems at the moment. These current systems (like Google Maps) generally work well to navigate the user to the rough vicinity of the destination, but don’t help the user actually complete the journey — finding an accessible parking spot, an accessible route to the building, an accessible entrance.

From our primary research, we have learned that “Last Mile” data is desired by almost all users, and access to this data adds value to the journey — since being able to access the destination is crucial for a successful journey.

Narrow focus to people with mobility impairments

From our research, we learned that people with mobility impairments are more affected by sidewalk data than people with visual impairments. People with visual impairments rely more on their mental models of familiar places and therefore have much less value to gain from navigation systems that offer re-routing. Therefore, we have narrowed our scope to focus on people with mobility impairments.

Investigate crowdsourcing to influence the “Last Mile”

Simply put, crowdsourcing affects the design of everything — all parts of the journey AND all user groups are affected by the data provided by users. Initial interviews show crowdsourcing could be optimized to encourage more data contribution, allowing the service to be more accurate and usable.

Expand to include “Last Mile” info

If people aren’t able to access their destination, the journey loses an immense amount of value. Currently, there aren’t good enough data/solutions for the “Last Mile” — so we see a large opportunity here to improve the end-to-end navigation experience.

Additionally, when people with mobility impairments travel on sidewalks, it’s short routes; or if they travel long distance, they travel with car/bus → less opportunity to re-route and a higher % of the journey is the “Last Mile”.

Explore how we can allow users and reporters of the data to be the same population

We found that people who don’t have mobility impairments have a very hard time understanding what obstacles and conditions cause difficulties for people with mobility impairments. We believe empowering wheelchair users to report their own obstacles will have more of an impact for other users facing similar issues.