Uber self-driving system for seamless transportation of people and goods – Interview with Nastasha Tan

As Head of Design at Uber’s Advanced Technologies Group, Nastasha Tan is shaping the experience of self-driving cars.

With technology advancing and improving every second, it is inevitable to recognize the greatest changes to our city life will involve the world of mobility and most specifically: cars. It is no longer a matter of if it happens, it is a matter of how quickly it will come. The transition to self-driving cars will mark a significant change in the way we live and the way we think about them. They aren’t just a new sort of car, they are teachable machines that come with new perceptions, ownership models, and rules.

While autonomous vehicles give incredible opportunities that can potentially liberate not only those who do not want to drive, but critically, those who cannot, they also come with new and important aspects that need to be taken into account when designing its aesthetics and its overall experience. Yes, self-driving cars need to be first and foremost safe, but beyond that, there are crucial choices that still need to be made, complex choices such as how people perceive them, what do they look like inside and out, who owns it, how it drives, what is the user experience like, and even whether or not riders notice the difference between driven and driverless cars. In short, designers still need to define what the experience of not driving a car should be

The success of self-driving cars depends on how well design and creativity are harnessed to create the best experiences for future riders. The enormity and importance of this task cannot be underestimated and it needs great designers who can address all these challenges to make the visions of the future come to life. One of those designers is Nastasha Tan, Head of Design of the Advanced Technology Group at Uber. DesignWanted went straight to the source and had the opportunity to interview Nastasha to find out more about Uber’s Advanced Technology Group, its self-driving system, their approach to the self-driving experience, and how Uber sees the future of mobility.

Having a double-major in Cognitive Science and Psychology and a masters in HCI, how has that influenced your way to approach design?

Nastasha Tan: “Psychology, Cognitive Science, and Human Computer Interaction have all fundamentally shaped how I understand people, and in practice how to apply this understanding to any challenge or problem space. My belief is that good design begins with good problem solving, and fundamentally understanding people can orient us to identify the right problems to solve for people, and use design’s toolkit to solve them in compelling ways. 

Understanding human behavior — what drives people, what people want, what people need, what their obstacles are — they all inform the core of what our design solution should strive to solve for. And how we choose to solve them, whether it’s designing a product, service, or experience — or whether it is making existing things better, like improving our experience with purchasing groceries — or whether it is creating something entirely new for an emerging space, like what we’re working on at Uber Advanced Technologies Group (ATG): defining Uber’s self-driving experience, we should be striving to design things that form a bond and establish a relationship with the people we’re designing them for.”

Curious about the developments in the mobility field? Don’t miss Seymourpowell promotes Appropriate Mobility: the right mode of transport at the right time.

Uber - SD Vehicles
Uber believes the future of mobility is increasingly shared, sustainable, and automated

What are the main responsibilities of the Design department at Uber ATG?

Nastasha Tan: “Uber ATG is focused on creating Uber’s self-driving experience. Our mission is to bring safe, reliable self-driving transportation to everyone, everywhere. At Uber, we believe the future of mobility is increasingly shared, sustainable, and automated. 

Design at Uber ATG is focused on solving challenges for people across the end-to-end lifecycle of creating a self-driving experience — from the development of the self-driving technology to the experience of the self-driving car itself.  

Our team is an interdisciplinary team of designers: from Product Design to Visual Design to UX Research to Industrial Design to Design Operations — we all work together on two key areas: 

The first: 
For an industry that is emerging, like self-driving, the tools to create self-driving technology does not exist off-the-shelf as ubiquitously. For Product Designers, tools like Photoshop, Figma, and Framer are tools that are essential for our craft as design practitioners. Without them, we cannot create the UI of our digital products or iterate on them without a way to prototype. Equivalent essential tools like these need to be created from scratch for the various people who help to create our self-driving technology and experience. 

And it’s quite complex — we have to understand a range of user needs — from how our autonomy engineers develop and debug their code, to how map producers translate log data from our cars to produce meaningful data for our cars to interpret the world, to how to best support our mission specialists (the humans in the loop who help us test and evolve our software) who capture feedback about our self-driving cars that is essential for iterating on our self-driving technology. Yes, it’s complex, but it’s quite fun — in addition to having good design craft, our team is built around a core skill of technical fluency. To some extent, as a Designer at Uber ATG, you get an unofficial mini degree in self-driving development.”

Uber - SVD testing on street
Developing self-driving technology is one of the biggest technical challenges of our time and Uber is taking this challenge head-on

Nastasha Tan: “The second:
In addition, our team designs all parts of the user experience of the self-driving ride. This includes things like: what should our self-driving car look like? We ensure that the sensor kit, which includes various sensors that enable our cars to see and sense the real world, are designed in a way that doesn’t scream “I’m a science experiment” or “I’m a robot” because it’s crucial to establish norms that are human and familiar for us to build trust with our riders, and also those we share the road with (other vehicles, pedestrians, cyclists,etc.). We have a rad industrial designer, Johad, on the team who, with extreme care, balances the needs of our engineers who specify the sensors that our car needs to have, and also the needs of our future customers to ensure the sensor kit’s final appearance feels approachable and as human as possible. 

In addition, we also design the rider’s end-to-end journey with our self-driving car. From requesting a ride on our rider app (building on the experience of calling an Uber as you know of it today) to the experience inside our car to the experience of arrival and drop-off. For example, we ensure that it’s safe to start the trip by checking to ensure your seatbelts are on, etc., to visualizing the right kinds of information throughout your trip on a digital surface in the car to ensure that we’re providing essential information for you to feel informed and safe, to providing a way to always be able to reach our remote operator if you need real-time assistance during the trip. We want our self-driving experience to feel as intuitive and self-guided as possible, and if we’ve succeeded, riders will liken it to a human-driven experience, or dare I say, even more preferable.”

Uber - SDV safety testing
Uber’s vision for self-driving technology is built on a foundational commitment to safety

Mapping the experience of any Uber user might be quite complex, how do you get to find the pain points and areas of improvement within such a well established company?

Nastasha Tan: “It’s a great question. For us, the users that develop our self-driving technology that we build tools for are within the walls of ATG, but we don’t take closer access for granted — we are intentional about the relationships we want to have with them to serve their needs across the tools we build

Foundationally, it’s crucial to form a relationship with our internal users through a regular dialogue. Tactically speaking, what I mean by ‘regular dialogue’ would include three parts to ensure reciprocity: first, a conversation to learn about our user’s needs, pain points, and bright spots, second, time for us to synthesize and iterate on the design work, and third, a follow-up conversation with our users with a prototype or something tangible to show how we’ve applied our learnings to the design work. This kind of dialogue helps our users feel heard and reinforces that we’re helping to drive outcomes in the things we design to support their key jobs to be done so that they can be successful. This builds a foundation of trust with our users that helps us continue to further our understanding of their highly expert domain to continue evolving the tools we build for them. 

Our user group shifts a bit when we look at the self-driving car and the experience that surrounds it — our users here, are future riders. Given Uber’s core business in shared rides, we work with their design teams to understand current needs, pain points, and bright spots across riders as a baseline — riders have inherent needs in mobility, whether it’s a human-driven car or a self-driven car. 

We also work with a broad spectrum of prospective riders on fine-tuning the experience of a self-driven ride — from the self-driving car’s vehicle behavior (how it moves within space) to understand more about how a vehicle’s motion informs trust and safety, and how to make the end-to-end rider experience: from requesting the trip to taking the trip to concluding the trip, as desirable as it can be for future riders. Solving for real challenges our riders have in mobility is what fundamentally guides our design approach and process — it helps us assess what’s most crucial to solve for people, and informs how our self-driving technology can help, instead of it being the other way around.”

Uber - in setting
Uber self-driving vehicles have the potential to make our roads safer and transportation more affordable for everyone

With all the changes that have been happening in the world, and having changed our way of interaction as humans, how has it informed your perspective on the future of mobility?

Nastasha Tan: “My personal perspective is that it has helped widen our empathy for a greater spectrum of mobility challenges, in addition to giving us a stronger call-to-action to find more ways people services and automated services (like self-driving) need to work together to solve for them. 

Our global pandemic is a very unexpected circumstance where our entire world has had to refrain from normal human-to-human interaction, so it has broadened the use cases that our self-driving cars can solve for, especially for unexpected needs that emerged with this pandemic. Like, the need to access essential goods like food and water safely without risking exposure, or the desire for social connection with our friends and loved ones without putting anyone at risk of exposure. 

It’s one of design’s superpowers to take unexpected challenges and imagine potential futures that could exist to broaden what is possible and enable that reality. So this pandemic has inspired more ‘what if’ brainstorms.  

What if our self-driving cars could deliver essential goods to those in need that is safe and timely? What if our self-driving cars could enable a more mobile shelter-in-place experience with our immediate loved ones that allowed us to travel and safely maintain social connections? 

This pandemic has also opened up new layers to privacy and safety needs in self-driving cars that we’ve already been building towards. We have thought a lot about privacy in the context of shared self-driving rides, very much in the way that we’ve seen airlines implement privacy dividers between passengers, etc. but for circumstances like the one we’re persevering through today, we have to widen our definition of privacy and safety that may not be solved by a physical privacy divider.”

Uber - SDV programming
Uber starts with a vehicle known for its strong safety track record then modifies the vehicle to integrate the sensors and software needed for their self-driving system

What is your perspective on the key use cases for self-driving technology?

Nastasha Tan: “We’re still in the early exploration phases, but given Uber’s rich platform of food delivery, flying taxis, freight delivery, etc. we’re excited that self-driving technology will make the transportation of people and goods more seamless. Personally, I think that self-driving technology will enable us to be with the people that we care and love a lot more (because traveling becomes even more seamless with fewer obstacles with self-driven cars) and that we won’t be confined to our immediate physical location for who we want to spend time with. 

Mobility and our desire to move from place to place is really driven by our desire to be with people that matter to us — whether it’s to see a co-worker, a friend, or a loved one. Our job as designers is to facilitate the ability to connect with these people seamlessly, so all of our use cases are centered around the principle that our products and services in self-driving cars should really be ambient and facilitate our desire to connect with the people we care about.”

Uber self-driving - white volvo
Uber is leading the industry with cutting-edge research for self-driving cars in broad areas of computer vision, machine learning, and robotics

How will the ever more interconnected infrastructure change the service that Uber delivers to its users?

Nastasha Tan: “I think it will enable transportation to be more seamless and hopefully more invisible so we can pay attention to our ‘end’ instead of the ‘means’. At the moment, mobility requires a lot of mental calculus — there is a lot of planning and strategy involved to get to where we want to get to.  

I think our infrastructural solutions will start with things that are possible to build without too much systemic change — for example, like a bike rack installed on the sidewalk, or an app we can download to access a real-time bus schedule — it will start with things that can be designed in post, and may feel very hacky in this way to start.   

But as self-driving cars become more ubiquitous and the value of self-driving cars evolve, our infrastructure will inevitably drive change at the systemic level and allow for more integrated solutions. I could imagine, less street parking and more public spaces for people, and perhaps even designated bus stops becoming a lot more integrated with the environment, despite it still giving us the affordance to know where to catch our ride. Integrated infrastructural changes like this will make our end-to-end mobility journey feel a lot more seamless. If we do our job well as designers, you shouldn’t see much of the infrastructure that makes self-driving possible or sees much of the technology of self-driving cars celebrated as much, and instead experience the value it adds to your life because you’re able to move more seamlessly around the world.”

Uber self-driving - white volvo testing
Uber wants to bring self-driving vehicles to their network around the world

What do you think about the role of design in future mobility? What should it be focused on?

Nastasha Tan: “Design has a super crucial role to play in shaping the future of mobility, especially as our world becomes more automated, shared, and sustainable. For me, Design would include: the design of the infrastructural system across cities and the services within it (as we had just discussed in your previous question), the regulatory changes that drive these desired changes, the design of the self-driven car itself, the services and products that enable a seamless journey around the self-driven car, and the tools that we need to design for experts like autonomy engineers to create a high quality self-driving technology

Designers, because we are advocates for users, we will also be crucial in informing what a good experience feels like when you’re in a self-driven car — ensuring our rider experience meets the standard of human drivers and also removing the risk that jeopardizes safety since our self-driving technology doesn’t get distracted.  

And also discovering high-value services for riders — where do riders want to go, and how can our self-driving cars help? We’ll know that we’ve succeeded if the self-driving technology itself becomes invisible and that the service it provides riders just allows riders to get to where they want to get to a lot easier. 

And we must not forget, Designers will also inform how self-driving cars can share our public roads with other human-driven cars, pedestrians, cyclists on the road — yes we design for our future riders, but we have to design our self-driven cars in such a way that they coexist in harmony with those we share the road with.”

If you want to know more about the future of automotive design, don’t miss With elegance, purity, and innovation at heart, Pininfarina keeps creating the vehicles of tomorrow.

Uber self-driving - overview
Uber is focused on the safe development of safe self-driving technology, starting their trials in a number of cities in the US and Canada
No related posts
Send this to a friend