2. Problem Definition
  3. Solution
  5. Generative Research
  6. Affinity Clustering
  7. storyboarding

F.O.M.A. the Artbot

User Research | Experience Design | Project Management
An adorable art-loving robot that attracts citizens to interact with public art.
Team: Wendy Chang, Eileen Lee, and Matt Franklin
September 2019 — December 2019


  • User Researcher
  • Project Manager
  • Strategist
  • Client Presentation


In partnership with Metro 21 and the Heinz College, Arts Management at Carnegie Mellon University


  • Observe & Intercept
  • Affinity Clustering
  • Think Aloud Protocol
  • Storyboarding
  • Speed Dating
  • Experience Prototyping

Problem Definition

How might we leverage the power of social influence to increase people’s engagement with public art?
Our team was tasked with researching the current state of how Pittsburgh citizens are interacting with public art. The ultimate goal was to improve the livability of Pittsburgh by increasing engagement with public art, and thus to help justify the cost of public art.
After preliminary research on the domain, both primary and secondary, our team focused on definition of the problem statement and research goals. We conducted Walk the Wall and Abstraction Laddering activities to share our individual insights and to challenge current understanding of the space. One research goal rose to the top: How can the power of social influence subconsciously persuade passers-by to take a second look? This led to our team name, FOMA: Fear of Missing Art, and our problem statement.


A miniature robot who frequents public art and, through its companion web application, offers an assortment of services related to the nearby artwork.
Our design, a small mobile robot named F.O.M.A, draws citizens towards public art with eye-catching movements. It then prompts them, through text and voice, to visit its companion web-application. Among the offerings on the web-application are photography services to form new memories, and background history on the piece of art to learn about its significance.
Learning about the art’s history
FOMA will tell users about the piece of artwork and its history, and encourage them to find more details on the web app.
Getting their photo taken with the art
FOMA will gladly take a photo of users in front of the public art. This feature is especially appealing tor friends and families, who can then save or share the images through the web app.
Locating other public art nearby
FOMA will give instructions, via Google Maps, about other public art sites that can be found nearby.
Browsing the robot’s daily photo gallery
FOMA will take photos of and around the public art throughout the day. Users can see the world through FOMA’s eyes.

My Contribution

In addition to my research and design contributions, I took on a Project Manager role.
Our team members all had a strong hand in the research and design throughout the project, but at the outset we designated specific roles. As the team’s Project Manager, I led communication at our team meetings, ensured that tasks were delegated appropriately, and was accountable for deliverables.

Generative Research

Individually, we explored the domain of public art through primary and secondary research.
Collectively, we analyzed over 20 academic articles on public art. We also utilized informal sources in our initial research reports, including our own experiences. Perhaps most importantly, we physically visited many pieces of public art in Pittsburgh, observing and noting citizen behavior in such environments.
Together, we observed and intercepted public art passers-by.
We chose Dippy the Dinosaur as a public art environment to investigate, observing 30+ citizens look at or photograph Dippy and stopping 5 for a follow-up interview. After observing their behavior, we used a semi-structured interview format to elicit their thoughts on Dippy and public art in general, especially on the topic of sharing and discussing public art.

Affinity Clustering

We then held an interpretation session and created an Affinity Map, revealing common themes among our diverse interviewees.
> Insight 1
People inherently associate public art with its physical location and form opinions on the art based on its surroundings.
> Insight 2
Psychological barriers
People often show initial reluctance when discussing public art, but, after enough prompting, can readily recall personal and emotional stories related to public art.


When we felt ready to ideate potential solutions, we conducted a “Crazy Eights” exercise.
After reviewing and synthesizing our research to date, we created a list of user needs and values.
With these needs and values in mind, we each individually came up with 8 ideas in 8 minutes. We encouraged ourselves to think outside of the box, and even push the boundaries of user comfort. These 32 ideas became the seeds for our storyboards.
We conducted speed-dating sessions with 12 storyboards and gained valuable reactions, some of which were unexpected, from 4 Pittsburgh locals.
One surprise was that research participants were largely unfazed by concepts that involved physically altering public art, such as painting or writing messages.

After speed-dating and interpretation of the sessions, we voted on which solution to further explore. The winning storyboard was the ‘public art robot’ approach, but close seconds were a social-media photoshoot at the art site, accessing augmented reality information about the art, and locating art nearby. In the end, we focused on the robot as the experiential component, but aimed to merge all four of these ideas into one solution through the robot and its accompanying website.
> Insight 1
Social proof
For many of the storyboard scenarios, even some of the riskier ones, our participants said that they would be much more likely to participate if others were already doing so.
> Insight 2
Prompting users to interact with public art can often be seen as inconvenient or obtrusive.

Experience Prototyping

We prototyped a functional robot, complete with voice lines, dance moves, and a companion website.
We first constructed a low-fidelity robot using a small cardboard box and wheels and simulated the website with paper screens on a cell phone. One team member voiced the robot, while the other team members moved the robot and switched out the paper screens. From the low-fidelity version, we learned that the robot would need more perceptual affordance and the web link needed to be more visible.
We moved forward into a more refined prototype, using a remote-control car, a Bluetooth speaker, and a GoPro camera. We fashioned a flag for the robot advertising the website, which it could wave to grab user’s attention.
To test our robot prototype, we hid near the public art, remotely controlling the robot and feeding it a simulated voice.  We observed how people would react to encountering a public art robot.
Our plan was to stand back and let the robot engage participants, seeing how citizens would react before we approached them and asked them to participate in an interview. We gained valuable insights, both from those who stopped and agreed to talk with us, and from those who walked by and simply expressed an impression, such as a smile or a quizzical look.
The people who most appreciated the robot were groups of people walking together, especially families with small children.

Design Decisions

1. Give citizens, especially groups, a new reason to stop and visit public art
The robot is eye-catching, which draws people into the space. Once they are engaged, FOMA helps them make memories by providing an informative experience and creating a digital souvenir (taking their photo)
2. Small size, cute dog-like movements, a bow-tie, and a high-pitched robotic voice made FOMA approachable and charismatic
The personality of the robot is also a huge draw; users consistently likened the robot to a dog, the way it moved, ‘sat’, and wiggled added character and charm.
3. The web app allows multiple users to simultaneously experience FOMA
If there is a large gathering of people, while FOMA can only physically interact with one person at a time, FOMA can provide conversation and information through the website.
4. Conversational interface supports users with varying levels of technological savvy
Those without a smartphone, including children, can still interact with FOMA via a verbal conversation about the artwork.


We presented our research findings to our stakeholders at a poster presentation on December 4th, 2019.
We created a poster detailing our process and solution, and even brought our robot along for a live demonstration!


Public art is at the mercy of the weather.
One aspect that made experience testing difficult was cold weather and rain during the Pittsburgh winter.
One area which deserves special attention in future iterations is ensuring that the robot does not distract from the public art.
While drawing people into the public art’s space is beneficial on its own, we hoped the robot would also elicit more direct interaction with the artwork. We hypothesize that solidifying an immediate connection between the robot and the public art will help to achieve this.