Designing for Pedestrians at Stop Light

As we all know that technology has empowered human relationships and communication. While we are very used to screens, smart devices, connected wearbles, wifi-enabled objects, Internet of Things (IoT) seems to be the picture of near future. However, how might we enhance social relationships without being intrusive and manipulating? In this IoT class, we were challenged to create an IoT set that can enhance social relationships in an ambient way.

Time Span: February 2019
Team Project: 1 person designed the interface and made lighting animations, 1 person built the circuit and enabled cloud data sharing, I conducted research and programmed the game algorithm


Learning from the field

In class, we had a lot of literature reviews, product discoveries and discussions of "internet of worries". Some of my key takeaways are that, first, efficiency may not always be the goal when designing for human relationships, such as in a home environment. Second, as we making more devices connected, they should not draw unnecessary attentions and require too much efforts from the users. When IoT objects are everywhere in our day-to-day life, they should integrate into the ecosystem seamlessly.

Finding the moment

So how do we choose a daily problem to focus on? We had a brainstorming session in which we imagine how random daily objects can be useful when connecting to the internet. We also each wrote down 5 daily life observations for three days and discussed them.

We found that crossing the street is an interesting daily scenario. It's too short to be even realized. But people are still too impatient to wait for the street light.

While we are so used to design for fast, how might we design for slow?

75 Secs

average wait time at red light

2.5 Million

intersection accidents annually

+ 50%

people killed in red-light running were pedestrians

Talking to people

From my observation in the CMU campus, jaywalking is a daily common scene. Sometimes people can get impatient easily and push the push-to-walk button a lot to fulfill their desire of urging the light changing time. While they mash the button, they are spreading some sort of negative emotions. The less patience and tolerance people give to each other, the more inappropriate behaviors and safety risks can occur.

Ideation & Iteration

We are interested in transforming what is normally a boring waste of time into a fun playful experience. Our goal is to create an augmented object to allow pedestrians to interact with other pedestrians on the other side of the street.

we decided to create a pair of gamified objects with which pedestrians can challenge each other to see who can win the game. The game needs two sides of the pedestrians to participate. Once both people press the start button, the neopixel will light up to indicate that the game has started. Within a limited time (before they can go across the street), they will press the game button as fast as they can and the first person reaches the threshold number will win the game. We believe that adding this simple tug-of-war type game will make waiting to cross the street more entertaining.

User Feedback

We explained the concept to our potential users. I found that some of them are reluctant to use the shaker because they worry that they would be unsanitary. So we decided to change the game input to be buttons which requires less hand exposures from people and can potentially integrate with the push-to-walk buttons in the future.


(Created by teammate)


Building the circuit

List of Parts:

2 Breadboards, 2 Particle Argons, 2 USB Micro B Cables, 2 Neopixels, 4 Push Buttons,  20 Jumper Wires

Programming the game

We then drafted the state flow of the game and built the code without adding complicated light effects. In this step, we built and tested the code piece by piece so that it's easier for us to debug it. According to the National Association of City Transportation Officials, an average wait time at red light is 60 - 90 seconds. Assuming someone takes 1 second to press a button on average, we decided to make 50 times as a threshold that won't take too long or too short.

testing the concept

Creating the light animations

After testing the devices and making sure they are working properly, we tweaked the lighting effects in the ready state, the game state, the winner's side and the loser's side. We were careful to choose the colors so that they are not similar or contradict with the traffic light colors.

Making the interface

Low-fi Prototype

High-fi Prototype

Interactive Prototype


We tested it with 30 pedestrians, I found that 90% of them actually played it at the stop light, 20% of them interacted with other pedestrians after the game and 45% potential negative behaviors were prevented from occuring.

Check out my code and other IoT projects on my GitHub