Safe Square

Design Interactions between autonomous Cars and Pedestrians

Date: 2018

Collaborator: Xiaoyan Song; Jiaqi Wang

Instructor: Wayne Chung

Date: 2018

TYPE: Experience prototyping, Production Design, Interaction Design, Mobility - project at the CMU School of Design - Tools for UX design, Sponsored by BridgeStone

design prompt

We were tasked to explore how we could design to support people’s mobility in next 5-10 years by Bridgestone. The mobility of people and goods is changing as the application of various technologies advance. What does mobility look like for people in 5-10 years? How will people live, work, and commute at that time? These questions are the starting points for our journey of exploration.

What does mobility look like in the future? -


With rapid development of AI and Big Data, autonomous driving empowered by these technologies will transform the way people transport. It is expected that, by 2030, there will be some 20.8 million autonomous vehicles in operation in the U.S.


These several years have already seen the boom of electric vehicles. With increasing support from governments and maturity of charging technologies, this will continue and transform the car industry.


In near futures, vehicles will use a number of different communication technologies to communicate with the driver, other cars on the road, roadside infrastructure, and the “Cloud”. This will not only improve vehicle safety, but also to improve vehicle efficiency and commute times.


Sharing economy isn’t new to people living in 2010s, many of whom are using Ubers day to day for commuting purposes. It is predicted that by 2030, only 20% of Americans will own a car, enjoying the seamless travel services by shared cars.

Today and in the future


- Traffic elements provide explicit guidelines

- Some decisions, such as “who will go first”, are made by implicit negotiations between road users

social cue.jpg

In the future...

Shared spaces will become universal: Traffic elements will be removed; Users of the space will have to interact to pass through.

Autonomous vehicles will be prevalent: Pedestrians cannot rely on non- verbal cues to interact with cars.


Shared Space Definition “Shared space is an urban design and traffic engineering concept that integrates pedestrian, vehicles and other road users through the removal of traditional street elements such as signs, traffic lights, pedestrian barriers, road markings and kerbs”. The goal of shared space is to improve the road safety and vibrancy of roads and junctions, particularly ones with high levels of pedestrian traffic, by encouraging negotiation of priority in shared areas between different road users. Such shared spaces have already been implemented in Australia, Netherland, the UK and a few other countries, and this design approach is gaining its popularity in the urban design (CNN).

The Problem we are solving:

In a future world where autonomous cars are prevalent, the long-established interaction patterns will no longer be applicable between pedestrians and human drivers. In this case, how might we support the communication between pedestrians and autonomous cars in the shared spaces?


Exploratory Research

To solve the problem for the future pedestrians, we first want to explore how pedestrians today are currently interacting with cars.

The essential questions we probed into at this stage are: 1. What do people do when crossing the street? 2. How do people currently interact with drivers?


A Poster Collecting Real-World Pedestrians’ Opinions

We put a poster at a bus stop near CMU, where the context will serve as a cue for people to recall their experience of interaction with drivers. We had two questions for participants:

1. How do you communicate with drivers when you are a pedestrian?

2. What do you usually do when you cross the street?

The underlying rationale for designing these questions was that we want to know the context of the topic that we were dealing with and see how we could design for users based on their current walking habit.

survey result.png

We received 30 responses in total.

We found out from the poster survey that most of the people pay attention to the road when crossing the street

However, there are still some people who would like to do other things.

Some quotes:

Based on the results we got from the survey, we created the affinity diagram:

affinity diagraming.jpg


Personas: According the findings, we created our personas with different behavior patterns when crossing the street. Here are some quotes that these three types of people gave.

“I usually make eye contact and make sure they are seeing me.”

“Listening to the music. Sometimes also look at the phone.”

“I run as fast as possible regardless of the light.”


Journey Map:

We also made journey map for these three personas in which we delineated their experience of crossing the road and interacting with drivers.

1'_Journey map.png



Generative Research

From there, we started our second round of research, aiming at exploring people’s perception on possible solutions of the problem. We used a participatory design approach, attempting to actively involved users in our process to help ensure the result meets their needs.

We tried to understand: 1. What do people want to know about a coming autonomous car? 2. How do people want to get receive information?

poster survey 2.JPG

A Poster Generating Insightful Solutions

We put a poster at the University Center in CMU, where various people would pass by and take a look at the prompts that we gave them.

We had the context set up as following: It’s 2030 right now. Autonomous cars are prevailing on the road. You are crossing the street. A car is coming towards you.

Based on the given context, participants will be asked to generate insights according to the next two questions:

1. What do you want that car to tell you about itself?

2. Please indicate WHERE you think the message would be delivered (with the color coding dots) and tell us how (with the sticky notes).

2_codesign session.jpeg

Co-Design Session - Futuristic Card Game

We were inspired by Stuart Candy’s lecture and decided to hold a co-design session where participants would have the opportunity to propose their solutions. We conducted the co-design session with 9 participants.

We designed three sets of cards which consisted of roles, locations, and intentions. Participants would first be asked to draw one card from the two roles that we had for them, either an autonomous car or a pedestrian. Then they would draw one card for the location and one for intention, both of which would depend on the role that they chose. After that, participants would be given a piece of paper and wrote down in text or drawing following the prompt that we gave them.

Through the co-design session, together with the poster survey, we collected a wide range of ideas from participants, which served as our starting point of our design.

co-design session

Some quotes from the Co-Design Session:

“How does it know I want to cross the street? “

“Will it slow down or stop for me?”

“Can it see me? Does it know if I am there?”

Co-Design Session Results and Insights: through the co-design session, together with the poster survey, we collected a wide range of ideas from participants, which served as our starting point of our design. We created the affinity diagram shown below

2 affinity a.jpeg

From the affinity diagram, we could tell three large categories of information that people wanted to know about an autonomous car: 1. Technological details for safety 2. Information related to pedestrian crossing 3. Information about its current stage and human control.

We realized that the first category tapped into a fundamental issue – people did not trust that autonomous cars are safe enough to run on the road. We deemed that people’s acceptance of a new technology always takes time, and their answers were based on their current understanding of the technology. We think that in a future where autonomous cars are prevalent, the issue will be not be an issue anymore. Both the second and third categories are situational – they are about the information people want to know at that particular situation when they want to cross the road. However, the third category is, in a way, associated to the lack of trust of pedestrians towards the autonomous car, which requires time to be built up. On the other hand, we found the second category was most informative to our design – how do we want to help build timely and effective communication between pedestrians and autonomous cars? And this question guided our subsequent design process.

Insights: People want to be informed about the intention of cars, and be assured that autonomous cars are able to understand their intention.

How might we facilitate two-way communication between pedestrians and autonomous cars in the shared spaces, while bringing a sense of safety, and minimizing effort for pedestrians?


How do we design for the three personas who own different personalities but share the same space?

1'_conclusion_who do we deisgn for.png



Taking together our research findings and Bridgestone business focus, we developed Safe², Safe² surface will be used in the shared spaces where traffic signs and lights are removed.

- It is paved on the surface of the shared space which will have direct contact with both pedestrians and autonomous cars which enter this zone.

- Also, the sensors embedded in the Safe² surface can detect the movement of both cars and pedestrians and send signals to both parties of how they should do in the shared space. 

- The surface consists of connected units of smaller rubber tiles, with sensors embedded underneath.

- The details of the product will be shown in the following pages.

design 1.png


Paved on the surface of shared spaces

design 2.png


Consists of grid system in which units of rubber tiles connect with each other.

design 3.png


Sensors embedded beneath can sense the movement of the road users and send out signals.

module 2.png


The single unit of our product is a square-shaped rubber tile with sensors embedded which can be connected with other ones.

Scenario 1 : Pedestrian Crossing

scenario 2 : Emergent Situation


Safety: Safe² provides the users with a safer environment where they can be rest assured about the safety of the autonomous cars and their movement/intention.

Efficiency: Through the embedded sensors, the whole system becomes intelligent and can maximize the efficiency of the shared spaces through its data collection and calculation.

Accessibility: Since Safe² will create the physical change which every pedestrian can feel, this will increase the accessibility of this product, making it accessible for people with disabilities.