Hod Lipson | Building Self-Aware Robots
Professor of Mechanical Engineering
From chocolate and other edible items to prosthetic limbs and electronic devices, 3D printers have far surpassed their original capabilities when invented 30 years ago. But pushing those limits even farther intrigues Hod Lipson, professor of mechanical engineering.
“3D printing is a universal technology that has the potential to revolutionize our lives by enabling individuals to design and manufacture things with unprecedented freedom,” he says. “If we can leverage this technology to allow robots to design and create new things for us, we can achieve immeasurable potential.”
Lipson, who is a member of the Data Science Institute, focuses his research on evolutionary robotics, a branch of robotics that uses processes inspired from biological evolution to “breed” new robots, rather than design them manually. Using biologically inspired approaches to the design and manufacture of robotic lifeforms, Lipson hopes to be able to make self-aware robots that exhibit characteristics such as creativity and even curiosity.
“Biology is my primary source of inspiration when looking for approaches to meet these challenges. I use those ideas to bring biology insight into engineering, and new engineering insights into biology,” Lipson says.
“It’s always surprising to see new systems evolve or learn on their own,” he continues. “Seeing a robot learn to do something you didn’t program it to do is a pretty amazing experience.” He compares it to seeing your child do something on their own for the first time.
Some of Lipson’s work has reached mainstream use, like his artificial intelligence (AI) software Eureqa that “breeds equations” to fit complex data, software that is now used by thousands of people to detect equations and hidden mathematical relationships in big data. Other technologies pioneered by Lipson and his team include food printing and bioprinting, which is the printing of biological tissue and organs through the layering of living cells.
“We developed a process for robots to learn to simulate themselves, a process which I think is the beginning of self-awareness,” says Lipson.
Lipson’s research continues to focus on methods for automatic construction of robots and methods for automatic control of robots—essentially the body and brain. This work has led to new optimization algorithms that effectively help robots evolve.
“My goal is to see self-aware robots walk out of a 3D printer. I don’t know if we’ll get there in my lifetime, but we’re on the path that will eventually lead there,” he says.
Lipson joined Columbia Engineering in July, 2015. Prior to Columbia, Lipson was a professor of mechanical and aerospace engineering, and computing and information science at Cornell University. In 2013, he coauthored the award-winning book Fabricated: The New World of 3D Printing (top 25 book in China, 2013). His new book, Driverless, coauthored with Melba Kurman on the technologies and trends of autonomous cars, is due out fall 2016.
BS, Technion - Israel Institute of Technology, 1989; PhD, Technion - Israel Institute of Technology, 1998
—by Jamie Della Croce