Designing robots

I want to design robots.

I’ve been doing design work at software companies for the past five years, and I like it a lot. Beyond designing web apps, I make a lot of other things: musical instruments, prints, comics, purses, clothes… I love the process of figuring out what new tools I can use to create the things I want to see in the world. My very favorite feeling is stretching my brain to learn some wild new thing, and I want to honor that feeling by pursuing a persistent preoccupation of mine: robotics. Robots have reached an exciting level of capability — they’re showing up in our homes, on our sidewalks, and among our crops. People are starting to have surprise encounters with them… are robots ready for that?

Here’s an excerpt from a Wired article from 2017 about a San Francisco city ordinance restricting robot use (emphasis mine):

The SF Board of Supervisors voted on Tuesday, December 5 to severely restrict the machines, which roll on sidewalks and autonomously dodge obstacles like dogs and buskers. Now startups will have to get permits to run their robots under strict guidelines in particular zones, typically industrial areas with low foot traffic. And even then, they may only do so for research purposes, not making actual deliveries.
…

Just this last year has seen an explosion in robotics, as the machines escape the lab (thanks in part to cheaper, more powerful sensors) and begin rolling and walking in the real world. They’ve arrived quickly and with little warning.

And that’s made folks both curious and uneasy. Go to a mall and you may well find a security robot scooting around keeping an eye on things. Robot nurses roam the halls of hospitals. Autonomous drones fill the air. The question is: How are we supposed to interact with these machines? It’s a weird and fundamentally different kind of relationship than you’d form with a human, and not even experts in the field of human-robot interaction are sure how this is going to play out.

The big thing is safety. Machines are stronger than us and generally unfeeling (though that’s changing with robots that have a sense of touch), and can be very dangerous if not handled correctly. Which is what spooked Yee. San Francisco’s sidewalks are bustling with pedestrians and runners and homeless people and dogs and the occasional rat stacked on a cat stacked on a dog. How can the city make sure that roaming delivery robots and citizens get along?

What does it mean to design robots? I’m not sure yet — that’s actually one of my research questions. I have a hunch that it’s going to look a little like engineering, a little like industrial design, and a little like service design.

There’s a blossoming field called Human-Robot Interactions, and that’s what I want to explore.

Where to begin?

I like to let my projects take on a life of their own. When I start pulling on a thread of inquiry and my question unravels, I usually find ten new questions — more, if I’m lucky. So here are a few questions I’m starting with.

How can we make robots not just effective at completing tasks, but comfortable to share space with?

Humans and robots have a hard time working alongside each other. We can’t communicate well, robots don’t know much about the world, and some of them are dangerous.

How can robots best work alongside humans? What verbal or non-verbal cues do they have to be attentive to when collaborating? People learn a lot from watching each other work — how can robots learn through observation?

How can we make robots approachable?

By “approachable” I don’t just mean “non-threatening”. Because robots are new, we don’t have a universal framework for understanding them: it’s hard to tell what a robot is doing, why it’s doing it in that way, what its intentions are, or what it’s aware of. By contrast, people have a lot of subtle ways to understand or “read” each other from a distance. We can often tell if someone is busy, sleeping, relaxing, thinking, or in need of help merely by looking at them.

How might we make robots’ behavior more “legible”, even to people who have never interacted with a robot before? How can we make a robot’s goals clear?

What are the “tools” of robot design?

I currently design web apps, and while we tackle new and varied design challenges all the time there’s a rich, pervasive understanding of what makes a UI “well-designed”. We know the rules of page structure, color, contrast, and iconography. We know how to draw a user’s attention around the page. We know how to make great experiences across platforms. We know how to introduce new users to a product. All of this creates a toolkit to help us usher people efficiently and effectively through their many tasks. What should be in the toolkit of a robot designer?

What makes a robot “well-designed”?

There are well-documented heuristics for doing UI design work — do they apply to robotics?

How can we use robotics for the benefit of society?

In an age where capitalism has created astounding and ever-increasing inequality and where having a job is often the only way to secure healthcare, shelter, and food, robotics could easily become the new tool of the wealthy to accelerate profits and automate people out of society. But in an age where we need to redefine our relationship to work, our planet, and each other, robotics could free us from tedious, difficult, or dangerous work, allowing us to become the artists, naturalists, and explorers we should be.

How can we make sure that robots benefit society as a whole, not just people seeking profit? What does an equitable and just transition into a more automated future look like? How can our relationship to work change for the better as robots become more capable of doing jobs that humans perform today?

Not every robot has to be built — how might we choose the right ones?

Opening the curiosity door

I have a lot to learn on my way to becoming a robot designer! I applied to a couple of graduate programs in computer science, specifically for Human-Computer Interaction (the larger field to which Human-Robot Interaction belongs). I didn’t get in this year, but I still intend to continue my learning journey — so I’ve started sketching out a list of things I want to learn in this kind of independent study I’m embarking on. Maybe this can be Freshman year at Ramona University!

Robotics 101

• The basics of robot control. How do you move a robot around the world? How does it sense its environment?
• How to build a robot. I know you can run Robot Operating System on a RaspberryPi. Maybe that’s a place to start?

Recent HRI research

• What’s the latest in HRI research? There are some awesome labs & research teams out there. My favorite at the moment is the InterACT lab at Berkeley. I started a Notion page to compile a list of HRI papers I want to read (and notes).

Machine learning

• The basics of machine learning. A lot of the papers I read describe machine-learning-based methods for generating robot behavior.

Statistics and game theory

• Machine learning requires a good understanding of statistics, so it’s time to brush up on Bayes!
• When you start talking about giving a robot goals to optimize for, it becomes an agent that other agents (humans) have to interact with… an understanding of basic game theory ideas might be helpful here.

Electrical engineering

• I want to work on (and hopefully design and build!) real live robots. I have a basic, arduino-project-level understanding of electronic circuits, so I signed up for an Electronics 101 class at a local community college.

Mechanical engineering

• Similar to above, if I’m working on robot hardware, I’ll need to understand how to make things that can move and survive in the real world.

Principles

As a designer, I like to have principles for my projects to help remind me of the things I want to get out of my work. Here are some of the principles I’m using to approach this learning adventure.

Follow the ✨

My most important goal is to honor my love of learning. I’m still trying to figure out how my interests and skills fit into the enormous field of robotics, and I have many other hobbies besides. I want to make sure that this work always feels exciting and nourishing, so I intend to follow my instincts towards whatever sparkles on the horizon.

Leave a trail

One of my favorite things about working as a user experience designer is that, as a new-ish field, everyone seems to be at least a little self-taught. I don’t really know what’s out there regarding robot design yet, and I want to leave a trail so that others who are interested in this topic might have an easier time finding their way. That’s why I wrote this post!

Ask every question

As a new person to this field, I want to leave no stone unturned. I also want to connect with the rest of the community! So I plan to keep on throwing questions out there — maybe they’ve been answered, maybe they haven’t. Either way, I hope we can talk about it!

Want to talk about robots? Me too! You can find me on twitter @designingrobots.