In my Pervasive Interaction Design course, my team decided to tackle an issue that’s been on the minds of many Ann Arbor citizens for several years now: electric scooter safety. While fun and convenient, we’d all heard anecdotes and seen statistics about how dangerous they could be, and we made it our mission to come up with a way to make them a safer option for short-distance commutes.
Project Overview
-
Project name: Scoot Scoot
-
My roles: User research, product design, prototyping, user testing, marketing
-
Tools/skills utilized: User interviews, comparative analysis, interaction maps, user surveys, heuristic evaluation, preference tests, usability testing
-
The key problems to solve: How can we make electric scooters easier and safer to use by integrating some form of sensors and internet connectivity? What degree of automation will users trust?
Research Findings & The Design Process
Initial Research & Ideation
We began by looking into statistics related to scooters and conducting field observations of people riding scooters on and around campus.
-
Most riders were very focused, never taking their hands off the handlebars or trying to use their phones while riding; they often had backpacks or bags, and never wore helmets despite most rental apps encouraging helmet usage; pedestrians would sometimes get spooked by scooters suddenly seeming to come out of nowhere​
We discussed our findings and spent some time brainstorming to each come up with a different product concept. My idea was for a pair of augmented reality (AR) glasses that would sync with the scooter and the user's phone to display directions and hazard warnings in real-time.
​Next, we sent a survey out to UMSI students asking about their experiences with electric scooters, including whether they owned vs. rented them, why and how often they rode them, and what safety concerns they had.
-
We learned that most riders rent rather than own their scooters, only doing so occasionally
-
Their main worries while riding were pedestrians and rough pavement conditions​
We then conducted interviews with several of our respondents, asking them to go into more detail about their answers and showing them our initial product concepts and asking for feedback​.
-
The most popular concept involved adding a smartphone mount to the handlebars for the user's phone display route information and hazard warnings; they did not like the idea of having to purchase any external accessories
​
The First Prototype
We settled on designing a system that would provide route navigation and use a combination of onboard sensors and crowdsourced data to warn the rider about obstacles and guide them to alternate routes if necessary, similar to the Waze navigation app, all to be displayed through the user's phone mounted on the handlebars. To test our initial design, we recruited students to run user enactments using a lo-fi prototype consisting of a children’s scooter with cards on the handlebars representing different screens, an indoor course with cardboard boxes as obstacles, and five scripts incorporating different scenarios and levels of automation​.
We found that users were much more comfortable with lower levels of automation, such as giving warnings about hazards or suggested alternate routes, than they were with higher levels, such as automatic braking or route changes; warnings about route congestion and low battery were generally ignored​. Users also expressed worries about their phones falling out of the handlebar mounts and about having to look down at the screen while moving.
-
We attributed the latter largely to the fact that we'd used a children's scooter, which had much lower handlebars and put the screen at more awkward angle; if we had had more time, we would have run additional tests using a larger adult-sized scooter
Refining The Design
Based on the findings from our user enactments, we further refined our design by changing the handlebar smartphone mount to a built-in screen, adding haptic feedback to reduce the need to look down, and removing unpopular features involving higher levels or automation.
Our finalized design (digital mockup pictured below) involved a handlebar-mounted screen that would display navigation instructions, battery status, and warnings about obstacles and congestion, as well as onboard sensors, vibration motors in the handlebars for haptic alerts, and buttons to dismiss warnings on the screen.
We built our second prototype around an adult-sized push scooter, featuring a smartphone standing in for a built-in screen that would show screen mockups created in Figma, as well as an improvised vibration motor controlled by an Arduino.
Finally, to wrap up the course we conducted a live demo using our final prototype, and then mounted it on a real electric scooter and shot an infomercial-style video showcasing our main product features, which I edited.
Project Reflection
Outcome
Based on our on-campus observations and survey, we decided to focus on navigation, because many people said that they often used scooters when in a hurry and/or in unfamiliar areas, and road safety, because obstacles and bad road conditions were people’s main worries, as our main problems to tackle. We ditched the idea for connected smart glasses after none of our interviewees indicated that they would want to purchase any scooter-related accessories.
​
Our user enactments revealed that most users were uncomfortable with higher levels of automation, such as the navigation system automatically changing the route to avoid a crowded sidewalk or the scooter automatically braking upon detecting an obstacle, so we opted to limit the automation to warnings that users could choose to heed or ignore. We also chose to incorporate haptic alerts via vibrations in the handlebars (an earlier idea we’d initially decided against) after our users expressed concerns about having to look down at the screen to see or check for warnings. Finally, after another round of user testing we changed the handlebar smartphone mount to a built-in screen that integrates with the user’s smartphone through the rental app, due to users’ concerns about their phones falling out of the mounts while riding.
Challenges
-
In terms of conceptual design, the biggest obstacle we faced was that our interviewees weren’t particularly enthusiastic about any of our initial ideas
-
​They had the fewest issues with the idea of mounting a screen on the handlebars, but our test users raised new concerns as we kept working on it, so we had to keep modifying the design to work around them by lowering the level of automation, adding haptic feedback to reduce the need to look down at the screen, and integrating the screen into the scooter rather than using a phone mount
-
If we hadn't been under such a time crunch for the class, I think we should have spent more time rethinking the entire concept before going ahead with building a prototype
-
-
In terms of practical design, the platform of an electric scooter proved quite challenging to prototype for
-
All we had access to for our initial prototype was a children’s push scooter that had an unorthodox method of steering and was quite awkward for adults to operate, which no doubt did nothing to improve the test users’ impressions of our design (and which may have contributed to their worries about looking down at the screen, which was much lower than it would have been on a full-sized scooter and required the user to look almost straight down to see)
-
We were able to get an adult-sized push scooter to build our demo prototype around, which was definitely easier to work with, but with a much slower speed than electric scooters, we were still unable to accurately recreate the sensation of riding one, as using a real electric scooter for testing would have been highly impractical, not to mention much more unsafe
-