Four Questions For: Daniela Rus

You have said before that you believe robots will be as commonplace as smartphones. How do you envision robots to be used in people’s everyday lives?

Recent years have seen major advances in fields that are vital for producing useful robots. Computer vision researchers are developing algorithms that allow machines to “see” increasingly more like humans. Experts in manipulation and control are creating increasingly more agile robots that can do increasingly more dexterous tasks. Research in natural language processing is creating intuitive ways for machines and humans to interact.

Taken all together, this work has gotten us closer than ever to a world where robots will be able to help us with everything from space exploration and search-and-rescue operations to manufacturing and folding our laundry!

Robots will help with physical tasks in a way that is analogous to how smartphones help with computational and data tasks. Imagine a world where anyone can design and build their own robot, customized for their needs. This would could create a whole new industry of “24-hour manufacturing” – people could go to a 24-hour robot-manufacturing store to customize a robot to their own needs.

For example, say you want to retrieve a ring from a vent, or tidy the toys on your floor. From this task specification, the store will be able to create a robot quickly and at low cost. We do not have these capabilities today, but we are developing technologies that will someday enable us to create our own robots like that.

How do you create limits that safeguard humans against being harmed by robots?

We need technological solutions and policies. We should always be cognizant of new technologies’ potential for both good and bad. When airbags came out, many people were understandably concerned that they occasionally malfunctioned, causing injuries or worse. But we now accept that the many lives that airbags save far outweigh the harm they cause.

I believe that in the coming years we will go through a similar process with robots. That said, I’m glad that there has been a healthy debate to make sure that we implement clear guidelines on robotic technologies and how they are used.

My team is acutely aware of the fact that traditional factory robots are caged to make sure that we are safe around them. Our recent research has explored the potential of soft-bodied robots, which we think could be a safer alternative. (For certain tasks, they might even be more effective, since their agility lets them change direction and squeeze into tight spaces.)

How can American workers prepare themselves for careers in technology? How can people already in career paths who have already been working for 10, 20, 30 years keep from getting left behind?

Rapid progress in computing over the past 50 years has made it so that we simply cannot live without computers today. The world is undergoing a big change in the nature of work, and we need to adapt to this change and be tech-literate. Although technology is developing rapidly and the jobs are also changing, the job changes are more gradual. This means we should be concerned about the young generation and make sure that they learn the skills required in the future.

Fortunately, with online education I believe that it’s never been easier to pick up new skills even when you’re a decade or two past your college years. Companies like edX offer many courses to keep you up to speed on emerging technologies and skills. In the last couple years we’ve created several courses aimed directly at working professionals who want to learn the latest skills in big data, cybersecurity and the Internet of Things.

As more jobs require using computers in complex ways, we have to address the knowledge gap in teaching coding. I think that this means not just “learning to code,” but what you might call “coding to learn” – learning these skills not merely as an end in itself, but to look at the world in a different way.

What advancements in robotics do you expect to see in your lifetime?

Robots will be able to help us with physical tasks big and small. I think we’re only a few years away from a world where you will be able to walk into a store and order your own robot for specific tasks around the home or office.

For example, I believe that artificial intelligence will transform transportation profoundly. In the next decade we will see more customized and safer transportation with autonomous and near-autonomous technologies doing a lot of the heavy lifting with getting people and packages around, whether by train, plane or car.

One challenge to making robots commonplace is that they take a long time to design and build. Today’s fabrication processes are slow and costly, with even one small change resulting in days or even weeks of lost time to re-evaluate the designs.

My team is among many around the globe who is working to develop systems that let you design robots more quickly. In our case, we have developed an interface that lets you design a robot in just a few minutes, to be 3D-printed in a matter of hours.

Daniela Rus is the Andrew (1956) and Erna Viterbi Professor of Electrical Engineering and Computer Science and Director of the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT.  Rus’s research interests are in robotics, mobile computing, and data science. Rus is a Class of 2002 MacArthur Fellow, a fellow of ACM, AAAI and IEEE, and a member of the National Academy of Engineering. She earned her PhD in Computer Science from Cornell University. Prior to joining MIT, Rus was a professor in the Computer Science Department at Dartmouth College.


Leave a Reply