Q&A with Hod Lipson
October 13, 2017
Columbia Technology Ventures (CTV) spoke with Hod Lipson, Professor of Mechanical Engineering, about soft robotics, his quest to develop robotic self-awareness, and the future of food.
CTV: You recently applied for a patent through Columbia Technology Ventures on one of your innovations in soft robotics. Tell us about it.
HL: Yes, we applied for patents through CTV for a new material that is basically a synthetic muscle. Soft robotics is a fast-growing field, and you can make soft anything-- but there’s always a hard component like a motor, a compressor, or something else mechanical. Up until now, it’s been difficult to make a soft muscle. Our material is soft, inexpensive, and easy to make—you can 3-D print it—and when you pass a current through it, it expands 900 percent. You can do all sorts of interesting things with it—we’re still in the early stages of figuring out the best application, but the material has properties that are unique and exciting.
CTV: Some of your previous work in soft robotics involves robots that evolve and “learn” how to walk. How could these learning algorithms be used outside the lab?
HL: Most robots today are used in manufacturing and assembly, where they perform set tasks in a controlled, repetitive way. But there’s a trend in robotics toward taking robots out of contexts where every move has been programmed and every scenario is controlled. To do this, we need robots that have some sense of autonomy, and can respond to unpredictable environments and situations. In a sense, that kind of learning is robotic evolution. We have yet to master this concept, but there’s tremendous engineering and commercial potential for robots that can evolve their functionality. We’re years away, but the kinds of steps we’ve taken in our lab by evolving soft robots that learn to move are critical steps in the right direction.
CTV: Another major theme in your research is robotic self-awareness, or the notion of developing robots that can model and understand themselves. Is there a commercial market for these kinds of robots, and how far along are you in the development?
HL: There is absolutely a commercial market for robotic systems that have a sense of self-awareness. Most robotic systems today are packed with sensors that enable them to sense the external world around them and to model, reason, and attempt to make predictions. We have developed systems that can look internally and sense themselves—robots that can detect if one of their components breaks or isn’t functioning. They can feel themselves, and it’s not the same as the way we feel pain, of course, but it’s a sense that something isn’t right. This is a crude form of robotic self-awareness.
In terms of commercial potential, there are so many applications. Imagine building a bridge that could sense areas of weakness, or machinery that can model itself so that every jet engine in every commercial airplane understands what it is and when something needs maintenance? Right now, we have built robots that understand when something breaks or isn’t functioning the way it was built to function, and these robots can react and recover. This area is growing quickly, and we’re going to see it develop much more in the coming years. You could conceivably apply this to every area, every discipline.
CTV: Switching subjects—3D printing is everywhere now, but you’re adapting this technology to something very unique: cooking. Would you explain the work you’re doing with food?
HL: 3D printing has reached the point where we can print with many different materials— plastic, metal, cells. You can print almost anything. Our project is focusing on food printing with raw ingredients, like flour, sugar, eggs, water, even insect powder. Here at Columbia, we built a food printer that can handle 8 ingredients, and we’re in the process of building one that can handle 24 ingredients. The printer mixes the ingredients according to a software program we created, but it bears no resemblance to the way we traditionally think of mixing ingredients—in a bowl, with a spoon, in our home kitchen. The printer mixes the ingredients the way an inkjet printer mixes red, green, and blue ink an inkjet printer to create a new shade of color. Then we use lasers to cook the food.
It’s an entirely new frontier—there’s no cookbook or food science to guide this work—but I really believe that cooking with software is the future of food. These two major forces—software and cooking— have never come together, and once they do, we’ll never look back. Food printing has enormous commercial appeal, not only because everyone cares about food, but because this is a way to produce fresh, novel foods on the spot, with infinite variety. Food printing also creates opportunities for marrying food and health in new ways—we could create foods based on feedback from biometric devices that provide information about an individual’s specific nutrition needs on any given day, and we can meet those needs immediately.
CTV: Is this something you feel is ready for commercialization?
HL: Much of the technology is there already, but the systems aren’t there to support widespread use of food printing at this point. But that’s going to change. I do want to mention that the notion of commercializing our work can be very motivating for students. When a student comes to me with a great idea and they file on it, they’re suddenly thinking long-term, and they own the idea in some way. For many students, that gives them a new perspective and a lot of motivation. I’ve had quite a few students graduate, spin off their own company, and do great things.
Columbia Technology Ventures works closely with Columbia researches to commercialize early-stage technology innovations, connecting industry and investment partners with researchers to bring impactful, in-demand robotic technologies to market as quickly as possible. To see Columbia’s Robotics portfolio available for licensing please click here: http://innovation.columbia.edu/robotics
Please send all inquiries to: firstname.lastname@example.org