This content was produced by Boston Globe Media in collaboration with the advertiser. The editorial department of Boston.com had no role in its production or display.
sponsored content guidelines
Russ Tedrake is the Vice President of Robotics Research at Toyota Research Institute (TRI). Tedrake manages a team devoted to producing a world-class simulation capability for TRI with a simple vision: What if we could develop real-world robots and autonomous vehicles completely in software (and trust that they will work in the real world)? Tedrake’s team also pursues fundamental research on “Enabling Technologies” for TRI Robotics, with a specific focus on manipulation and soft robotics. Tedrake is also the Toyota Professor at the Massachusetts Institute of Technology (MIT) in the Department of Electrical Engineering and Computer Science, Mechanical Engineering, and Aero/Astro, and he is a member of MIT’s Computer Science and Artificial Intelligence Lab (CSAIL). He received a B.S.E. in Computer Engineering from the University of Michigan in 1999, and a Ph.D. in Electrical Engineering and Computer Science from MIT in 2004.
Lindsay Gearheart: Could you talk a little about your background and how you found your way to Toyota Research Institute?
Dr. Russ Tedrake: I’m a professor at MIT. I’ve been here building robots at MIT for a number of years, I guess 15 years roughly now. Just before Toyota Research Institute came into existence, I was part of a big national competition called the DARPA Robotics Challenge. We had 300-pound humanoid robots, the early predecessors of the robots that you see doing backflips for Boston Dynamics now. We had to program those robots to go into a disaster response scenario and open doors and turn valves and walk across stairs. That was the first time I had to take the technology we’d been building at MIT and actually deploy it into something that was closer to the real world. It really opened my eyes to the challenges of deploying these advanced algorithms.
Right after that challenge, Gill Pratt (CEO at TRI), who was actually involved in the robotics challenge too, had this new opportunity with Toyota. So I was actually one of the very first people at the Toyota Research Institute, and we had this opportunity to somehow do something more real, more in the real world, than anything I could do at MIT, and have an impact in the real world by taking some of the technology coming out of academia and partnering with the biggest manufacturing company in the world and maybe getting robots into everybody’s home.
LG: Certainly for someone like me, who doesn’t have a robotics background, learning about how you’re making this accessible for people potentially in their homes one day is super interesting.
RT: It’s been such an exciting few years. A few years ago, we were working on unmanned aerial vehicles – they’re called drones these days. The types of things people were working on in research just a few years ago are now something you can buy at the mall to take a selfie (a quad-roader that takes a selfie, for example). I’ve watched that technology transition, and it’s been super fun to watch.
Autonomous driving is sort of the next thing that’s started to transition, where industry started pouring resources in. Of course, Toyota is working on autonomous driving too. But for me, I think the one that’s coming next, right after that, that’s just going to have a huge impact on the whole world, is robots that have hands, for instance, and can do meaningful work around the home. I think that’s the next hard problem that maybe we can transition from academia into industry.
LG: Can you talk a little more about how the robotic technology being developed at Toyota Research Institute has the potential to impact people in their everyday lives beyond transportation?
RT: If someone hears the name Toyota Research Institute, I think they won’t be surprised to hear that we’re working on autonomous driving. But I don’t think many people realize that we’re also making a massive investment in robotics. It’s a very targeted investment actually, it’s trying to help people age in place in the home. Toyota as a company is extremely aware of the silvering society. More and more of our population is living until an age where they need care, and the ratio of caregivers to people in need of care is going to get out of whack. It’s not a problem that a company would invest in to get a return in a year or two, but it’s maybe the big problem that we really have to solve. I’m incredibly excited about the opportunity to be working on this real problem of getting robots in the home that could help people age gracefully in place.
LG: How does your team ensure that the robots and autonomous vehicles you’ve developed in software will work in the real world?
RT: That’s a big question and a good question. I think one of the biggest challenges to this whole enterprise is that nowadays we’re using machine learning and artificial intelligence to program these robots, and nobody knows how to guarantee that it’s going to work, or not do something silly or crazy, in the real world. So that’s a big part of our research emphasis, is how do you have confidence in a system that was programmed with machine learning for instance.
And you’re right, we do a lot of our work in simulation. There’s just not enough hours in the day to run robots doing home tasks in the lab. You just can’t run enough experiments. Same way with my robotics competition at MIT. So we use simulation to do many more experiments in the cloud than we could ever do in reality. But it’s even more than that, we use simulation to try to understand the sensitivity of our robots to strange things happening. So we can vary lighting conditions, we can make plates that are super slippery, we can put a banana peel in the middle of the floor, things that would be very hard for us to try all of these things in reality, we can try them very quickly in simulation.
LG: I imagine it’s a lot of fun to think up those situations! So, how do you think public expectations about robots (from movies, books, or television) have influenced innovation in the industry?
RT: I think one of the biggest challenges about that is that people expect the robots to be much better than they are right now. When we talk about robots with hands that want to do something like load a dishwasher or pack groceries or fold laundry, expectations are super high because these seem like such easy problems for humans, but they’re very hard for robots. Ironically, doing a backflip is somehow easier than folding the laundry for a robot.
I think people’s expectations are very, very high. It’s interesting that culturally this varies across the world. Unfortunately, in the U.S., we grew up with robots being Terminator, and other things like this, so that also colors our view of robots. But I think robot vacuum cleaners, robot toys, even Alexa, these are the things that are getting people used to having technology like this in the home, so I hope that it’s paving a path to where we really wouldn’t be so surprised to have more dexterous, more capable robots in our lives.
LG: You’ve mentioned that in addition to your work with TRI, you also teach at MIT. What has working with these students taught you?
RT: I wouldn’t be the person that I am if it wasn’t for hanging around MIT for the last almost 20 years now. I think MIT, I think Boston more generally, is just such a wonderful culture of academics and brilliant people and constantly over-stimulating conversations that cause you to think differently. To give a lecture in front of a bunch of kids that you know are going to be thinking about every detail of everything you write on the board, that causes you to think about problems in a different way to prepare for that lecture.
Toyota was very strategic setting up the Toyota Research Institute. They’re just down the street from MIT, and there’s actually two other locations for TRI. One is right next to Stanford and one is right next to the University of Michigan. They actually fund a great deal of research at those universities and collaborate a lot with universities, and I think that was incredible good and strategic.
Whenever I teach a class, I always end completely exhausted, like this took everything out of me, but completely convinced that this is part of who I am, that it caused me to be stronger than I was before.
LG: Was there anything that we didn’t cover that you wanted to share before we wrap up?
RT: It really is amazing how much robot technology has come out of Boston. If you count the number of universities contributing but also the number of big companies that are now investing in Boston and the number of startups that have come out of Boston. Even 10 years ago, you could’ve said that most of the robots that were fielded by the Department of Defense actually were built by one of the companies in Boston, whether it was iRobot or Foster-Miller or something. There’s just so many robotics companies that are in Boston. I think people think of biotech and they think of other things but robotics is a super strength for Boston and vice versa.
Watch a recent video from Dr. Russ Tedrake and TRI to learn more about manipulation for human-assist robots here.
The HubWeek Change Maker series showcases the most innovative minds in art, science, and technology making an impact in Boston and around the world. Know a change maker you think should be interviewed for this series? Nominate them here.