Today in Tokyo, Toyota announced that it is investing US $1 billion over the next five years to establish a new R&D arm headquartered in Silicon Valley and focused on artificial intelligence and robotics. The Toyota Research Institute (TRI) plans to hire hundreds of engineers to staff a main facility in Palo Alto, Calif., near Stanford University, and a second facility located near MIT in Cambridge, Mass.
At a press conference, Toyota president Akio Toyoda showed a photo of Gill Pratt working on a Corolla some 35 years ago. “Nice hair, Gill,” he said. Image: Toyota via Ustream
Former DARPA program manager Dr. Gill Pratt, an executive technical advisor at Toyota, was named CEO of TRI, which will begin operations in January. Toyota president Akio Toyoda said in a press conference that the company pursues innovation and new technologies “to make life better for our customers and society as a whole,” adding that he wanted to “work with Gill not just because he’s an amazing researcher and engineer, but because I believe his goals and motivations are the same as ours.”
We spoke to Dr. Pratt to find out what kinds of things TRI wants to work on, how they plan to transform pure research into practical applications, and where all that money is going.
Dr. Gill Pratt, the former DARPA program manager responsible for the DARPA Robotics Challenge, will lead the new Toyota Research Institute in Silicon Valley. Photo: Toyota
FOCUS ON CARS AND ROBOTS
TRI’s initial focus is AI for cars and robots, Dr. Pratt told us. He detailed some of that strategy in September, when Toyota, the world’s largest automaker by sales, announced a research collaboration initiative with Stanford and MIT, a first step in the company’s big push into AI and robotics. With TRI, Toyota wants to significantly speed up the development of AI with applications to smarter and safer vehicles, as well as robots that can make our lives better at home, especially as we age. But although vehicles will be equipped with more powerful sensors, computers, and software, that doesn’t mean they’ll be designed to operate fully autonomously all the time, as Dr. Pratt explained:
“Our belief is that you actually don’t have to give up the joy of driving, and in fact we can make the joy of driving much safer, and much more available to all kinds of people across a very wide spectrum. But we’re not also shutting off the idea that some of the time you actually want the car to take over from you and to either drive on a highway on its own or drive in a parking lot by itself, or maybe if you’re too tired to drive home, to take the lead for you.”
In terms of robots, he said they’ll be using what they learn and hardware that they develop for indoor mobility applications, hoping to develop technology that can help people get around as well as move objects in their homes.
FROM PROTOTYPES TO PRODUCTS
TRI has a broad mandate, with lots of flexibility, and Dr. Pratt said they’ll be exploring research on numerous fronts. But he emphasized that, as part of its mission, the organization will strive to bridge the gap between fundamental research and product development:
“We think that the answer is to have projects that have a feasibility prototype or some other type of milestone as an intermediate goal in order to see what is possible to do. Then we want to see if we can translate those results into a new capability that we will test to be able to say, ‘Okay, does this work or doesn’t this work? What are the advantages and disadvantages? Should we turn this into a real component of a future car? Or in the case of the robotics field, some new type of capability in a robot that might be used indoors?’ ”
Dr. Pratt added that they’ll have a “modest fund” that they’ll be using to invest in startups: not as a way of making money, but to “help nurture technology” that their projects can benefit from. What kinds of technology is Toyota interested in the near term? He gave us one example:
“There are new kinds of imaging and proximity sensors coming out all of the time now, because [LIDAR] has transformed the field of both map making and autonomous driving quite a bit. The trouble is that those sensors tend to be very expensive, and many people right now all over the world are looking at how you can have that same capability in a much less expensive sensor. Some of that research is being done by Toyota, some of it is being done by other places and we’re going to be looking at all of it.”
Toyota recently demonstrated a Lexus test vehicle capable of driving itself on highways, aiming to offer the technology in commercial models by 2020. Photo: Toyota
GET YOUR RESUME READY: TOYOTA IS HIRING
For the new TRI facilities in Palo Alto and Cambridge, Toyota plans to hire about 200 employees. That’s a lot of people to hire, and Dr. Pratt is aware that the process might ruffle some feathers. Uber found itself in the middle of a controversy when it hired dozens of researchers away from Carnegie Mellon’s National Robotics Engineering Center. Google did something similar to establish its own autonomous driving program. But Dr. Pratt is not concerned, noting that the first step for Toyota was to support the academic programs at Stanford and MIT directly:
“We think that it’s really important to nurture the pipeline of new talent that grows, and that means trying to make sure that professors and research scientists that are at universities are not only used as a source of talent but also that we put support back into the system to try to grow more talent. We also think that when students get to the stage that they have decided that they’re ready to move from the academic world into the private sector that our job is to present them with a really good option—here’s a place where you can be creative, work with other people who are extraordinarily smart, and where you can become even more talented than you were before. And of course have a whole lot of fun.”
For our part, we’re very excited to see what happens with this. We write a lot about robotics research, and far less about robotics products that make your life better, simply because getting from the former to the latter is surprisingly difficult. We feel that this level of commitment Toyota is demonstrating towards AI and robotics is exactly the kind of big push we need to get these technologies out in the real world. With these new research facilities in California and Massachusetts, we’re hoping that we’ll be able to follow along as TRI develops its advanced cars and robots, as well as other technologies that are probably going to be a complete surprise.
Erico Guizzo is a senior editor at IEEE Spectrum. He has written stories on a wide range of science and technology topics, including Japanese androids, French computer codes, Icelandic video games, American crash-test dummies, and Canadian bacteria. His main area of interest is robotics, and he has written and edited hundreds of articles and videos featuring the latest advances in this field. He is also the cocreator of Spectrum’s critically acclaimed Robots for iPad app. For his robotics coverage, Guizzo has won four Neal Awards and has been a finalist for two National Magazine Awards. An IEEE member, he holds a bachelor’s degree in electrical engineering from the University of São Paulo, in his native Brazil, and a master’s in science writing from MIT.
Evan Ackerman is the senior writer for IEEE Spectrum’s award-winning robotics blog, Automaton. Since 2007, he has written over 6,000 articles on robotics and emerging technology, covering conferences and events on every single continent except Africa, Antarctica, Australia, and South America (although he remains optimistic). In addition to Spectrum, Evan’s work has appeared in a variety of other online publications including Gizmodo and Slate, and you may have heard him on NPR’s Science Friday or the BBC World Service if you were listening at just the right time. Evan has an undergraduate degree in Martian geology, which he almost never gets to use, and still wants to be an astronaut when he grows up. In his spare time, he enjoys scuba diving, rehabilitating injured raptors, and playing bagpipes excellently.