Last summer I spent 4 months at Renault-Nissan's research facility in Silicon Valley, conducting research on a new haptic HMI (Human Machine Interface) for autonomous driving. The HMI, better known as Stewart II, was implemented in a driving simulator for an exploratory research study. The goal of the study was to explore indirect control and feedback by enabling a haptic dialog between the driver and the autonomous car.
Company: Renault Innovation Silicon Valley
Collaboration: Nissan Research Center Silicon Valley
Center for Design Research - Stanford University
Special thanks to Nikhil Gowda
Sunnyvale, CA, USA
Jun - Nov 2016
The goal of the internship at Renault was to push the Stewart concept out of its concept phase and use it as a research tool for the exploration of haptics in autonomous driving. The prototype Stewart II that I designed at Eindhoven University, The Netherlands, prior to the start of the internship, was duplicated and modified for the implementation of the final research study that employed an advanced car simulator.
The final study was set up to identify the effects of such a haptic HMI like Stewart on the driving experience of autonomous cars. Participants were able to "feel" the car's intentions. They were also able to express their intentions to the car by manipulating the HMI. These novel interactions - that blend non-visual feedback and intent - yielded interesting insights related to haptics and the use of indirect control for autonomous driving. The study consisted of 16 participants that experienced 30 min in the driving simulator and ended with a 30 min interview about their experience. Ultimately, the results and insights will be shared with the public in the form of a paper.
Setting up a research study involving a car simulator requires specific methodologies that enhance the quality of the retrieved data. Luckily, experts at the Center for Design Research at Stanford and Renault were there to guide me in setting up the end study. Besides setting up a study itself, there are many technicalities that need resolving before running the study.
A simulator study of an autonomous vehicle requires a couple of things: A road or world, events that happen during the drive, a working software model that lets the car drive itself whilst allowing for control externally. Besides setting up the simulator, I also had a custom build prototype that needed to be installed in the vehicle, to send and receive data from and to the sim.
Cutting a long story short, doing this type of research is time-consuming and demands a wide variety of skills. This setup involved ample programming (C++, Java, and JS), prototyping, and academic research. I remember working together with one of the engineers on setting up a UDP connection between my prototype and the simulator where I explained I was not trained to do any of this type work. His reply was that he wasn't either but that I was in Silicon Valley now, and if I could not do it, probably nobody could. A moment I'll never forget because it holds a mental value I kept and hope to never lose, the essence of trusting my own ability to overcome any challenge.
Building a car requires many different types of expertise. However, building the car of the future takes even more. As a result, I had the opportunity to work together with experts in the field of software engineering, automotive, mechanical engineering, UX/UI design, anthropology, business and many more. Albeit that everybody has a different background, they have one thing in common - a love for cars. We found a shared passion, especially, in rethinking the way mobility works today so we can make it better in the future.
Silicon Valley is a place like no other, full of positive and smart people that are all determined to make a change. Before arriving there, I was unsure if I, a Dutch ID grad student, would be able to keep up the pace. However, after working with great people from Nissan, Renault, Stanford, and MIT I realized I was no less and got excited to work with these amazing people. What I like most is the level of trust and the many collaborations that happen between people, companies, and universities.
If I was stuck with my work it was easy to get help, and if that person couldn't help me he or she probably knew someone that could. Due to this closely connected network between people, innovation and technical iterations happen fast - really fast! Everybody is laser focused and lean, yet at times laid back. For me as a designer, this is great. Being able to work hard in a stimulating environment whilst being able to take your mind off things if needed. Silicon Valley and the rest of California have been really good to me. I felt at home there and perhaps one day... I can return to this place and add my part in this ever-changing world - for the better.