Revolution

In a world ever-accelerating towards an autonomous future, user acceptance of automation technology is crucial to realizing its forward trajectory. In the automotive industry, automation is of particular concern due to the potential safety and efficiency benefits it may provide. While the engineering and manufacture of core technologies to support this function are both large factors in the automation's efficacy, little attention is being paid to educating consumers on these technologies as they become available. Inflammatory news articles cite single autonomous crashes as a tragedy, while every day, 90 people die in fatal crashes from non-autonomous vehicles. Users are primed by media and personal experiences to eschew automation in their personal lives and would rather let someone else take the risk with new technologies.

To better reach those users, we've made Revolution: a product that prepares users for the upcoming revolution in Autonomous Vehicles (AVs). Revolution is a showroom simulator that acclimates users to the controls, terminology, and overall experience of operating an AV. Users are guided by a retail associate to customize their driving experience, selecting options that change the behavior and driving style of the vehicle. They customize the interface to their taste, and are then guided through the main interface of the vehicle. After the introduction, they start the simulation. The user is given a test track to learn how to control the AV and test its handling. After the user feels comfortable, the retail associate has the option of multiple different scenarios they can load up to show the user how the AV reacts to certain edge cases, such as an oncoming vehicle, construction, fauna in the road, EMS interaction, or all in one big experience. After the user is done, they're given a card that holds their user preferences to be scanned by their new vehicle to bypass the setup process.

Research

This product is backed by weeks of secondary and primary research, including scholarly studies and user tests. 

We wanted to see the current state of consumer AV technology and generate AV-user empathy, so we scheduled a test drive in a Tesla Model Y. I drove while the other team members helped document.

To further our primary research, we went to the middle of Forsyth Park in Savannah to get real answers from real people. We gathered around 50 responses, and found more users were excited for AVs than we expected.

I called DriveSafety about their simulators and they provided me with two papers, one on avoiding motion sickness and one on optimal positioning of the three displays for the simulator. I used this for the final prototype.

Team member Nandika organized a Card Sort exercise with multiple users in our demographic and we used that data to inform our final user interface information architecture. I performed a few of these exercises, while team member Arjun took notes.

Finally, team member Arjun, team member Alejandro, and I put together a simulator using three displays, a driving game controller, foam core, a laptop, and a Unity simulation I put together using free resources and some custom code. I was in charge of the simulator code, so that's what I'll cover for the most part.

Simulation

We knew in order to make a viable user test, we needed to effectively simulate the driving environment. In our case, that meant a triple-monitor display, a racing wheel and pedal controller, and a Unity simulation driving the entire thing. While many of the projects on this site involve experienced coders, none of us even opened Unity before this project. I had to learn how to code in C#, at least well enough to reference the car and the path object. 

In the images above, you can see the X + Z path above the path, the entire route the AV takes, and a closeup of a turn. This path was made my me using the PathCreator add-on for Unity in the demo scene from EasyRoads3D.

This was a particular struggle: finding a way to make the car object find, go to, then follow the path I made, and on only the X + Z axes. I edited the PathFollower code and made it happen. I have three states, one for finding, going to, and following the path once it's there. You can also see it handles the joystick button press of the "X" button on an Xbox controller, which is the type we were using. I had to do deep research into Quaternions and Eulers in order to make the code I did, but in the end I got it to work.

Once that was done, we were able to test our solution on users. We went to the Gulfstream Center for Design and tested our solution on as many people as we could. I acted as the Retail Associate, while the rest of the team took notes. In the following images, you can see Alejandro and Arjun helping me set it up, me explaining the simulator to a test user, and various shots of the other users in their user tests.

To make things even more realistic, we also made a UI to go along with the prototype. This UI brings the user through setup, navigation, and an edge-case (construction). Team members Angel, Nandika, and Arjun worked primarily on these parts of the experience. This UI was put onto an iPad to the users' right side, where an In-Vehicle Infotainment center would be. While the user was able to interact with it during the initial setup section of testing, we controlled it using Figma Mirror as they drove so it would update with the user through their journey.

I also updated the Unity file to better visualize what a more realistic experience would look like. I used the Unity Windridge City package as a backdrop, and various other free 3D assets to better visualize three city environments users might be concerned about an AV handling: an ambulance, a construction zone, and an animal on the road.

Ultimately, retail employees will determine the success of Revolution, as they're the ones who will guide the user through their experience. I used my knowledge of retail training systems to create a document that explains the Revolution experience.

That's all for now!

Revolution addresses the pre-purchase touchpoint for potential AV users who hold a number of fears: how the AV handles edge cases, what the car feels like to drive, and how to interact with it. It offers a solution for customers who are curious about the technology but don't quite feel ready to take it onto the real road, where they may be involved in an accident. It acclimates customers to driving with an AV, and prepares them for driving the real thing should they choose to do so at a later time. It offers valuable experiences that users most certainly have questions about but would not experience in a normal, real-world test drive. Most importantly, it retrains users' common knowledge with experiential knowledge, which is the gateway to their acceptance later down the line. Revolution isn't the only part of the AV Revolution, but it's a new angle we're excited to explore.