Uber for streetlights


The challenge was to empower people to feel safe and independent while walking alone at day or night.

Our solution is in the implementation of drones with streetlights built-in, video streaming, and wayfinding lighting features to be a companion to a person walking to any destination.


Interaction Designer, UX Researcher, 3D Model Prototyper


Model making, FIGMA, Adobe Illustrator, Photoshop, Indesign, and After Effects


Emily Cho (UX/UI Designer)
Peyton Westman (UX Researcher + Writer)


First-quarter graduate course in the MS HCDE program at University of Washington

Fall 2019 (10 weeks)


How might we improve perceived and objective safety to empower people to walk alone?

On our first night of brainstorming, my group and I felt we were being watched in an empty study hall. Someone was hiding around a dark corner, dressed in a sweatshirt and baggy clothes with their face concealed. Walking away from the library, we found ourselves caught up in a rant about what a ridiculous problem this was to have—we’re three independent women in the year 2019, yet we can’t even get back to our cars on our own… how is that still a problem?

As we realized what a common issue this was, and one without an apparent solution, we set out to find a way to help people feel safer when walking alone.


Our process began with three primary methods of research:


We studied many personal safety apps/services and compared features between apps of variable popularity/reviews. Our analysis included Life360, Safe24, Noonlight, ADT Go, bSafe, SirenGPS, Kinetic Global, My Safetipin, and Kitestring.

We learned that most of these focused on just-in-time interventions, whereas none of the apps focused on preventative or proactive solutions.
Check out our comparison of different products HERE.


We wanted to hear directly from the target population. Utilizing this method ensured getting the necessary information, while providing the latitude to explore interesting insights.

Utilizing affinity mapping, we gathered data on users’ personal experiences with walking alone, any experience with safety apps, and their opinions on the concept of a personal safety app. We described features to them to test out what was meaningful, and to understand some pain points on their day-to-day walks.


Literature reviews yielded strong results that could be generalized to the larger population, which interviews could not, as convenience sampling was used to recruit participants.

We utilized info from literature reviews to pursue various target audiences, including women, minorities, and LGBTQ+. Our user interviews confirmed that the experiences of walking were less relevant to who was walking and more about the human experience of fear and being alone.

to persona or not to persona—

Our personas embody the three primary types of people we came across during our user research. Though we weren’t able to isolate them further than age, which was mostly based on availability of participants, we were able to interview people with various backgrounds to see if we could find anomalies that differentiated the experience of being fearful while walking alone from women to men, various races, or various sexual orientations/gender identifications.

User interviews would continue to inform our work as we worked to narrow down our target audience through the ideation and prototyping phases. By the time we finished our mid-fidelity deliverables, we realized our users had so much in common that the best way to isolate our target audience was to focus on everyone because, from our user research, we realized everyone is capable of fear, no matter their background, race, sexual orientation, etc. Ultimately, we found our personas to not be highly relevant to our research. However, the practice of developing them helped us understand our target audience that much better, especially since our research concluded in not having a specific target audience.

What does the user need?

Despite the wide array of personal safety apps on the market, the usage of such apps appears to be fairly low. We investigated this by asking questions in our interviews that related to existing features and potential new features. Based on our user research, we identified the following design priorities.

  • Users felt a large contributor of an environment feeling unsafe was a lack of peripheral awareness
  • It would help to have something that was easy to access discreetly and immediately
  • Location tracking and wayfinding information about new neighborhoods would help the feeling of uncertainty/fear

Narrowing down the key interaction elements—

We started with paper prototyping so that we could focus on the structure and more rudimentary aspects of the app in our initial usability testing. This helped us dial in our user flow and focus on the higher-level interactions so that we weren’t distracted by smaller details.

Writing out the structure of the app in the form of its information architecture helped inform crucial decisions of what flowed well/what didn’t, how the hierarchy informs the most important aspects of the app, and to make sure there were no gaps in the interactions we had focused on.

What does the walking experience look and feel like?—

I created a drone prototype out of chipboard to perform some usability tests with our class. This method utilizes Wizard of Oz prototyping by using custom inserts built to house a “streetlight” component, phone for video recording, and a laser for wayfinding indication. The drone design is based on a branding concept of a “firefly”.

Usability testing—

In our usability testing, we received a lot of great feedback about what to consider moving forward with our mid-fidelity prototype. As we thought through these ideas, we found that many users had the same questions, so I find this information best addressed in the format of a “frequently asked questions” list.

Since this app and drone solution is sort of a novel idea, we figured we should add a heavy emphasis on education of how the app works. There are many “if we had more time” cases, since we only had 10 weeks from coming up with the idea to the final product, so those are also explained in the FAQ link below.


visual design—

I began my ideation by isolating different features of the firefly. I knew that the drone emphasized the lighting and body shape, so I started with that and developed numerous sketches to explore the firefly’s body with lines. I brought these into Adobe Illustrator to refine them and to test out more iterations of a firefly design. In the end, I landed on a design that emphasized a playfulness with the intention to feel safe and inviting.

design guidelines

These design guidelines were partially adapted from Apple’s iOS human interface guidelines, with a focus on primary colors and easy-to-read text so as to portray a sense of familiarity/friendliness and safety/security. Emily Cho led the efforts on creating the visual design palette and UX/UI design of the mobile application seen below.

final design—

Introducing, Firefly, a drone and mobile app solution to help people feel, and be, safer walking alone.


Lighting from above to provide peripheral awareness of environment
Light projection of directional arrows to provide wayfinding
Video recording/live streaming for your trusted contacts

know what's ahead

When users open Firefly for the first time, they will be greeted with onboarding screens, which give a high level synopsis of how Firefly works. It is important to educate users on how Firefly works, as the solution is particularly novel. Users can navigate through the screens by swiping left or by tapping “Next” and then “Done” when they have reached the final screen. If users wish to skip these screens, they can do so by tapping “Skip.”

connect your safety network

Users will have the option to add trusted contacts, to create a safety network. Users can add trusted contacts manually or directly from their contacts. When users start a walk, the trusted contacts, who the user has chosen to share their location with, will get a notification to view the users’ walk. Users can decided who they’d like to share their location with by tapping “Edit.” The profile icons will then begin to shake, similar to how iPhone app do when being rearranged.  Users can by pressing, holding, and dragging.

browse the map, find a place to go

When a user is ready to go on a walk, they enter their desired destination. The origin point will prepopulate with user’s current location. If the user would like to select a different origin point, they can clear the field by tapping the x icon and type the the desired origin point.

Using an algorithm that considers recent crime data, how trafficked an area is, and the state of the built environment, Firefly will find the safest route to get the users from point A to point B. The route is highlighted in red.

confirm the most important details

On the payment screen, users will see information about their intended walk including the time and length of the walk, the estimated cost, and how far away the drone is. Users can place their order by tapping the “Order Drone” button. Afterwards users will be redirected to a order confirmation screen.

awareness of your surroundings and your route

When users are actively walking they will have two view options: the live video stream or a map view with directions. Users can toggle between the two views by tapping on the video box in the upper left screen and the back arrow. Trusted contacts, who the user is sharing their location with, will also have this view when users are actively walking.


when in doubt, do more research!

There were lots of ideas that my group members and I went back and forth about, so it really helped us realize the benefit of working directly with users. When you have a question or are uncertain in advocating for a certain feature/change, it often means you need more user research, or you need to return to the research you have and re-analyze your notes.

consider ethical consequences

Consider the different use cases—from the most positive, ideal situations to the most evil, corrupt situations, consider the best and worst intentions to design a more effective solution. There were many privacy and ethical concerns about our proposed solution, so this helped us prepare to answer concerns of users.

think about edge cases

If I had more time, I’d love to address some of the edge cases, like “what if someone wanted to use this in a football stadium?” or “what if the weather is extreme?”. I think the edge cases are very useful in developing a more effective solution.

check out another project

Click on the link above to check out a design challenge where I explored how students report issues to their facilities team on campus.