Companion Robots to Assist People
BY PAKSY PLACKIS-CHENG
Maja Matarić has spent decades building robots but not for some nifty entertainment purpose. She is building companion robots for children and seniors to have a better quality of life.
impactmania first met with Matarić at USC in 2014. We caught up with Matarić, November 2016 to discuss her latest venture, still in stealth mode. She shares on how robotics will impact our lives; starting a company after being in academia; and her ultimate goal of improving the quality of people’s lives.
Tell me about your startup.
Embodied is a robotics and AI [Artificial Intelligence] company. We’re using my 14 years of research in founding the fields of socially assisted robotics and many more years than that in robotics research.
The idea of socially assisted robots would help people with special needs, by empowering them to help themselves rather than doing the work for them. As we know from all of the literature, taking autonomy and work away from people doesn’t result in longer and better lives. It’s the opposite.
How did a startup come about after decades in academia?
I’m doing this in partnership with Paolo Pirjanian who is my former post doc from seventeen years ago. Paolo went on to do a lot of amazing things and has a PhD in robotics. One of a few CEOs in robotics that actually has a strong background in robotics.
Most recently, Paolo was Chief Technology Officer (CTO) of iRobot. We got together to start this endeavor. It’s truly a meeting of the minds. One of the many things that we agree on is that we don’t want early hype and over promising.
Why socially assistive robots?
Companion robots are fundamentally about helping people. To do that right, it takes a bit of time. It’s not meant to be a luxury technology.
We are in Pasadena, and already supported by Intel, Osage, Amazon Alexa Fund, Grishin Robotics, and Hive Ventures. This is a strong group of technologists as well as strategic and institutional venture capitalists. I feel extremely fortunate, because I’m doing this at a relatively mature time in my life, compared to some other entrepreneurs and I’m fortunate to be teamed up with someone extremely experienced. I don’t have horrible stories to tell! [Laughs.]
When I was a MIT grad student the metaphor was and still is: drinking from a fire hose. It’s awesome. It’s great to know that I could be learning so much and hopefully doing some good.
I’ve waited 11 years to take the sabbatical from USC. A part of me feels like, “Boy, that was stupid to wait that long.” But there’s another part of me that feels like, “Well, if this is what I was waiting to do, then that was really worth the wait.”
When it came out, I said, “Wow, that’s fantastic. Good for her!” It is great to see a woman in tech and an academic who took all the expertise and build a product.
Maja Mataric at Wall Street Journal Digital live. With Cynthia Breazeal on a panel on robots in the home.
Are startups taking enough advantage of the research done in academia?
Startups seem to for all the difficult challenges that they have, want to reinvent things. It’s sort of, “Okay, don’t read any papers that I spent 30 years doing!” [Laughs.] That’s a little frustrating, right? But it is also the motivation for us to jump in. Because if I keep publishing, the only people who will be reading it are academics, which is great, but it’s not enough.
We recently did an interview with a millennial entrepreneur. She took her findings in laser lighting out of the lab and into the world. Do you see more students becoming entrepreneurs?
What I see definitely is much more interest in going into startups. It’s exciting, in a way, I had my students beat me to this. Brian Gerkey in particular, founder of Open Source Robotics Foundation who wrote ROS [Robot Operating System] which is the language that’s being used for all of the research and development in robotics.
I was not a great advisor, career-wise for them, because I breed academics. Then I thought, “The world is changing.” It makes me a better advisor for the students who are coming in and saying, “I don’t think I’m going to be interested in academia. I want to do all this stuff, but then I want to do a startup.”
Give me a glimpse of the future; what will the impact of robotics be in 10 years?
It’s a very exciting time because we’re finally developing products that are going into people’s actual lives. Out of the lab, out of the factory, out of any kind of assembly. Whether you’re assembling at a macro-scale like cars or at a Nano scale like DNA sequencing.
The idea is that we might be un-caging robots. That’s not my work. But obviously, companies are looking at that. There is more and more robotics in healthcare, the autonomous car, and the autonomous flying stuff.
Although the most interesting things that will touch people’s lives are robots in the homes. This is very, very early stage. Currently, people are creating machines that either have very little function or they’re platforms that are entertaining.
Ten years from now, we should be looking at machines in the home that will be doing functional things to actually improve quality of life. Right now, you cannot have a robot that’s going to make your bed, do the laundry, or clean your dishes safely.
There are niches that don’t involve as much manipulation, ‘cause manipulation is still very hard. There are niches of service that don’t even involve physically moving around. You see this interesting set of products that are emerging, like Amazon’s Alexa. Everything from phone-like applications, all the way to the things that sit in the home, will be going on to machines that will move around. In ten years, one should hope, we’ll have a nice merger of services that are physically embodied and are mobile.
Historically, these fields in academia are separate. This is another reason why it’s a great time, because the companies that are creating products are forced to merge these technologies.
That also creates pressure on the academics because it’s hard to tell what’s going to happen to research funding in the near future. If the industry is going where the resources are, then industry can also drive the agenda. That is not always healthy, right? [Laughs.] But this real world push for these challenges is going to be helpful to both industry and academia. It’s going to take both sides because companies still have limited resources.
That’s the trouble with certain small startups, you focus and do something really well, but you only have so many resources. If you choose and execute well, it’s great. In a big company, even Google and Apple, huge as their resources are, can end up going down the wrong path because you can.
What specific problem would you like to solve with your company?
Yeah, that’s what I can’t talk about yet. In my previous research, I’ve worked with elderly with Alzheimer’s and children with Type 2 Diabetes, and kids with autism. We’re going to focus on one of those groups to start with, as a wedge into the market.
It’s incredibly fundamental to be solving a problem. Because if you look at a lot of startups, they’re creating things that are not solving anyone’s problems. They’re completely technology push, right? It’s like, “We’ll give you this cool thing!”
It’s cool but we added another problem instead of taking one away.
I couldn’t care less about entertaining for amusement. There’s a lot of need out there. I’ve spent the last 14 years learning about the needs that different populations have. I would feel like I would be wasting my time if I weren’t trying to help with some of that. We are working on a machine that provides companionship, coaching, and motivation to allow and enable a person to reach their potential. That is what I have been doing all along, but to do it in a product that people can have at home. That is a whole different challenge.
What has been the most surprising learning starting the company?
There are practical learning things. The expense side is a fundamental bottleneck that I’ve had to get my head around because in academia, we don’t worry about expense at small levels.
I don’t want to build something that costs tens of thousands of dollars ‘cause that’s not even affordable for a researcher. But I don’t worry if some component is $100 or $200. I just slap it on and we go. Literally the room I just walked out of, the last phrase was, “…$40? $40 being $200 retail, there’s no way.”
There is a lot going on right now in tech. It’s all in these pretty early stages, it’s too soon to tell what is robust or affordable. It’s going to be very interesting to see which of these things survive.
The large companies will immediately absorb some of the better products, and then you can’t use it, ‘cause they won’t play. That’s a new challenge for me. It would be grand if people collaborated more but I’m too old to expect that. [Laughs.]
Some of these problems are very hard — people will have to bring technologies together in order to get to the next step. That’s like academia, right? In academia, you can say, “I do machine learning but I don’t care about robotic data.” Or you say, “I do machine vision.” If you look at only images from the web. Well, then it’ll never run on a robot. When industries or companies start to make it real — putting it on a robot — you suddenly realize how shallow some of the technologies are.
They’re just not very deep. No one’s getting to that level, because system integration is really hard. In academia, no one rewards you for that. No one cares; just write a paper.
Then in companies, you’re in tech, except, I can’t quite be as bleeding edge because it’s too expensive or is it really gonna run? And are we gonna take that risk? That’s very sobering.
Your product needs to be affordable, right? The community you are serving needs to be able to have access to your product.
Exactly. To me, that is incredibly important. I absolutely don’t want to create something that’s inaccessible due to cost. Or something that’s flakey, right? You don’t want something that’s like, “Well, it kinda works most of the time, but you have to be a techie person to make it work.” No, no, no.
You have these amazing VCs on board. Kauffman Foundation and Babson College reported that women-led ventures bring 12% more revenue than male-led companies, yet only 2.7% of venture-backed companies have a woman CEO.
You have just raised funds for the company. What have you learned?
I don’t think we were seen as a woman-led company. My CEO and partner is male. He is the one that has the industry experience. I’m in no way deluded into thinking that most of the VCs we talked to looked at us and said, “We’re going to invest because of her IP.” Especially in robotics hardware — it’s starting to be easier to get funding for hardware. Software is still the easiest thing; hardware takes longer to build, is a bigger risk, and cost more. It was their confidence in Paolo’s past performance and the fact that he’s made money for people. [Laughs.]
There are some investors that care about IP. Certainly Osage, they were aware and interested in the research. Amazon cared about the idea. I had a very unique opportunity to present in an Amazon forum. It was by-invitation; a bunch of cutting-edge stuff got presented to Bezos [Founder, chairman, and CEO of Amazon.com] directly. I got to talk about my stuff. I don’t think that it was him necessarily that said, “Yeah, let’s.” But it helped. In other cases, the access we got was in large part because of Paolo.
If it were me and a bunch of graduate students raising funds, then it would be seen as a woman-led company. But it would also be much harder.
What would you say to entrepreneurs who are starting out?
I tell them absolutely to do it. But be careful. I can’t claim some massive expertise here, but having gone through a funding cycle, you’re in a vulnerable position, ‘cause you need something from someone. It’s a tremendous negotiation. It’s mostly gonna be a room full of men and even the women in the room often are going to be doing the male thing.
Talking with VCs at some level it is like high school. Who is popular, how do you position yourself, how do you establish the dynamic in the room, and who is in control? These are incredibly important things and this is much easier for someone who’s senior because, “Yeah, I have a job, I’ve done this.”
Most people are going to be insecure. The power dynamic in the room is very strong when you’re pitching. My best advice is be passionate about what you’re doing and don’t compromise away.
Early on, people will say, “Well, yeah, this is great, but you’ll have to move to San Francisco.” Or, “We have to save this budget with that.”
Helpful advice, but it is basically bullying. We’ve all been told, “Well, you’re going to have to do x, y, and z.” It’s like any endeavor in life; you don’t wanna get into it and go downstream. You’ll have to compromise on some things, but what is fundamental? For young people, it’s hard to know where that line is and then to uphold it.
I’m sure Paolo, with this venture, has impacted your professional DNA. Who else has left an imprint on your professional DNA?
I don’t believe in the whole professional versus personal thing. It tears people in half. Playing different roles, trying to shield one from the other, is very unhealthy, especially for women.
For me, my mother is the role model for both. Here’s a woman, a full time working mother, in her 40s when she got her PhD. She moved as a widow to a new country, where nobody would respect her PhD. So, she had to work terrible, terrible jobs.
Now she is 82 and still publishing books. I’m looking at her, “My God. It can be done.” She never did business, although if she had, help us all. [Laughs.]
You’re small fries, compared to your Mom, Maja!
It’s true! People have told me when they meet my mom, “Boy, she’s scary but really funny.” When I watched her get her advanced degree, I was a 13. She was typing late into the night on a manual typewriter. She’s in English literature, typed a 900-page thesis and made three copies of it manually, until she had an inflamed clavicle bone. I remember watching and thinking, “I’m never going to do that. That’s just way too much work for too little.”
Then when I was applying for the PhD program, there wasn’t a part of me that thought, “I can’t do it.” I wasn’t the first; she had already shown that anything could be done.
What’s is next?
We’re going to raise money. We’re building prototypes by early next year. That’s gonna be so exciting because we can put it in front of folks and see what they think.
I don’t want to leave USC; I love working with students. I want to see how universities will deal with this new era in which interesting IP [Intellectual Property] come out of universities. From a parent’s and a woman’s perspective, I want to know about my quality of life. When I go back, are they going to treat me like I don’t have this company? Can I accommodate this?
How can I negotiate, not just for me, but also for the other faculty who are otherwise going to flee? There are many of us who would love to have one foot in each field.
That’s important to figure out because I believe that some of the very best IP come from students and academics because of the freedom to roam the space of ideas. In a way that you’re not free in a company, you’re just not.
Give me some words that describe your journey so far.
Definitely meandering. It is important ‘cause women especially get told this story about how you are going to be an engineer when you tinkered in your basement. Well, I didn’t tinker and I didn’t have a basement!
I did computers because my uncle said that I needed to learn a trade. Psychology is what I love; meandering is key; and it’s ultimately all about people. That mantra keeps repeating itself. Even research has everything to do with foibles of humanity! In business too, it’s not just who you know, it is the dynamics in pitching and negotiation. It is power play out the wazoo.
No wonder that is what I am focusing on. It is all about human dynamics with technology. I want to create something that helps people tick better.
My main word is push. You never get what you deserve. You may not get what you need. It is gonna take pushing.