Paul Goodwin - BikeSense

BikeSense imagines a future in which ubiquitous computing fosters awareness and empathy between drivers and cyclists. Our team prototyped an IoT road sign that alerts drivers to the presences of upcoming cyclists, while simultaneously gathering data about cycling patterns throughout the city. This data can be used to improve urban densification and optimize bike lanes.

Roles

Researcher. I conducted several interviews with drivers and cyclists to better understand their relationships.

Physical Prototyper. I lead the physical prototyping exercises and helped run our wizard-of-oz tests with volunteer cyclists.

Video Producer. I wrote the script, planned, shot, edited and animated our bikeSense video.

Skills

User Research. Fly-on-the-wall, semi-structured interviewing, secondary research

Synthesis. Affinity diagramming, body storming, storyboarding

Prototyping. Physical prototyping, interaction design, wizard-of-oz experiments, animation

Communication. video production, presentation, design reviews

Research

How do we improve the relationship between drivers and cyclists?

We set out to use sensors to improve road safety.

Traditionally, Pittsburghers are drivers. As part of the post-war boom, significant investments were made in road and highway systems to connect and extend the downtown core. Despite modern programs that promote cycling in the city, many Pittsburgh drivers we interviewed still view cyclists as second-class traffic.

Inspired by the DOT Smart City initiative, our group explored how ubiquitous computing could provide new interventions to age-old urban challenges.

"The Smart City Challenge finalists recognize that to get smarter they need to enhance their ability to collect, process, analyze and share data. They plan to take in data from an immensearray of sources from connected DSRC infrastructure, crowdsourced data from smartphone users, and a variety of new sensors..."
- US Department of Transportation

Gathering Observations

We took to the streets to see how drivers and cyclists interact on the road.

Hills, rain, and blind spots all contributed to a tense atmosphere as we observed some of the more treacherous corners around 5th Avenue. One driver we interviewed mentioned that they only see cyclists in certain neighbourhoods, and outside of that, they are less likely to be on guard.

contextual inquiry — Blind Spot for drivers near Mellon Institute

Not all of our observations showed fault on the driver's part: we spotted a cyclist running a red light near the University of Pittsburgh.

Uncovering the problems

Drivers blame reckless cyclists; cyclists blame inattentive drivers. How can we help both types of commuters?

Our initial ideation and interviews focused on improving rental bike safety. We interviewed 6 people (3 students and 3 adults) between the ages of 21 and 39 about their bike rental experiences. We conducted several rounds of ideation to generate ideas.

As we reviewed our interview notes, we uncovered an underlying lack of empathy between cyclists and drivers: each group blamed the other for the majority of road safety issues. Drivers blame reckless cyclists; cyclists blame inattentive drivers. We focused our next round of ideation and subsequent research on the relationship between drivers and cyclists.

We used storyboards to explored how bikes, cars, cyclists, and the environment could be leveraged to increase awareness. Because we were interested in Pittsburgh's 2016 Smart City application, we focused on near-term technologies. We tested these 6 storyboards with 4 cyclists and used their feedback to guide our prototyping.

Bike Lottery - would you participate in a program that rewards (and punishes) good driving behavior for bikes and cars?

Biker Gang - have you ever wanted to rent a bike but were afraid of doing so because of the local traffic?

Safety Spotlight - have you ever wished you had a guardian angel that always protected you from the dangers of the road?

Automatic Breaking - do you forget to slow down at intersection sometimes? Do you find it useful if your car / bike can automatically slow down at intersections, especially where there is another car / bike?

Pacing Dot - what if you had a reliable external system that ensured your safety through optimal speed and route?

Traffic Visualization - would you ever choose a different transportation method if traffic looked heavy? Do you make an effort to bike more safely in heavy traffic?

Physical Prototyping

Creating a smart road sign

Pittsburgh already uses signs to help raise driver awareness in high pedestrian traffic areas. We decided to extend this existing approach to include cyclists.

We envisioned bikeSense as more than a static sign. It is environmentally responsive: changing its form and messaging to reflect the actual presence of cyclists. We imagined several states for the sign and focused our design on clearly communicating the current situation to drivers in under a second.

We tested the signs in a wizard-of-oz study with a driver and a cyclist. As the cyclist passed by our prototype, we switched the signs and raised the red flag at the top. Then, after a delay, the driver passed us and could interpret from the prototype how close a cyclist was. Using this method, we determined that the sign is most effective when it passes through the four states (represented above) and returns to a base state over 80 seconds. We also changed from our initial rectangular flag to a triangular flag so the driver could more easily interpret the distance to the cyclist.

BikeSense would use simple, readily available parts to detect bikes. A camera and computer vision algorithm provide reliable cyclist counts in an area. Since the sign is self-contained and portable, it can be moved to problem intersections to intervene and gather data. The data could be collected periodically or delivered live through the cell networks.

Results

"Pittsburgh received nearly $11 million to deploy smart traffic signal technology – proven to reduce delays at street lights by up to forty percent – along major travel corridors."
- US Department of Transportation

Although bikeSense was not one of the projects selected as part of Pittsburgh's initiative, the city did receive a $11 million grant to pursue other smart traffic sign projects.

Exit interviews with our wizard-of-oz participants indicated that cyclists felt safer knowing that drivers were being alerted to their presence. In turn, drivers appreciated knowing specifically to be on the lookout for cyclists, rather than seeing the generic "share the road" reminders.

My Contribution

This project required many different design considerations: exploring bikes, cars, and signs as IoT platforms; testing the benefits and drawbacks of ubiquitous computing; creating interactions that are understandable in less than a second; and situating a project to deliver real value to a larger initative.

As a researcher, I pushed for a wizard-of-oz study to help our team step outside of our comfort zone, and gather meaningful data about our physical prototype. As a designer, I honed my skills as a visual storyteller by leading our video production and communicating bikeSense's value to pursuade a broad audience. And as a leader, I motivated members of the team to persist until we had clarified the nature of the problem at hand, rather than settling for an existing solution.