This isn’t quite how most of us imagined the future: You walk into your local, 24-hour robot-manufacturing store — a sort of latter-day Kinko’s — and describe the kind of robot you want.
Maybe it’s a toy for the cat to chase around; maybe it’s an army of crawling robots that can be used to search for survivors of a natural disaster. Whatever the purpose, in a matter of hours, you’re handed a thin piece of paper, metal, or plastic, which then springs to life, folding along hinges into a customized machine.
That vision of cheap, self-folding robots — based on the ancient Japanese paper art of origami — is still a long way off. But a team of Harvard and Massachusetts Institute of Technology researchers published a proof-of-concept study Thursday that demonstrates such an approach can work. Their paper in the journal Science is accompanied by a fantastical video, in which a flat, thin sheet of paper and plastic that looks like a jigsaw puzzle piece suddenly stirs. Flaps fold up and creases form to reveal a crouching, bug-like creature that scuttles out of view.
“Today, it costs a lot in time and money to make a new robot,” said Daniela Rus, director of the Computer Science and Artificial Intelligence Laboratory at MIT. “The question is, can we develop the tools that will allow us to automatically and rapidly generate one robot for any task? This could impact many aspects of our lives and the economy.”
Origami might seem a peculiar inspiration for the next generation of technology, but over the last 15 years, a growing number of scientists, engineers, and mathematicians have become intrigued by the possibility of building useful things by simply folding up various materials.
Engineers began to think about using the principles of origami to build structures, and robots, which are time-consuming to design and assemble, were an obvious place to start. While other researchers have shown folding robots can be built, the paper by the Harvard and MIT team is a true feat, said Zhong You, a professor of engineering science at the University of Oxford. It is more complex and truly origami-like than the simple folds that have been done before, he said. “I think it opens doors for more complicated patterns being done.”
The work was led by Sam Felton, a graduate student at Harvard’s Wyss Institute for Biologically Inspired Engineering. He works with Harvard professor Robert Wood, best known for his efforts to build miniature “RoboBees” and other robotic insects. What interested them about folding was the possibility that the design and manufacturing of a robot could be greatly simplified and accelerated. But their initial hand-folded robots were labor-intensive. Though they didn’t need assembly with nuts and bolts, they required at least an hour of manual, skilled folding.
To automate the process, the team decided to build a robot body out of the most basic components: layers of paper, a thin piece of plastic with a circuit etched onto it, and Shrinky Dinks — children’s toys that contract when heated. The total cost was $100, most of which is accounted for by the motors and batteries (which still need to be installed by hand).
To trigger the folding, circuits embedded at different parts of the robot’s joints would heat up, causing the Shrinky Dinks layer to contract.
The potential for the self-folding robots is at this point theoretical — and can range as wide as a person’s imagination. Felton is excited by the prospect of sending into space flat sheets that could assemble into satellites, but the technology could also be used to enhance science education by giving students a fun project in which they could design and build their own foldable robots.
“Robots today are relatively difficult to fabricate, assemble, and transport. New digital fabrication techniques, such as 3-D printing and origami folding, can alleviate this bottleneck,” Hod Lipson, an associate professor of mechanical and aerospace engineering at Cornell University, wrote in an e-mail. “This work is a major step towards that goal.”
For now, Felton is still working on basic techniques — trying to build robots of different sizes. He’s examining whether other technologies and materials could be used to trigger folding, to solve limitations of the current method. For example, the paper sometimes smoked a bit when heat was applied to a joint, and the robot began to sag after a while because it was made of paper.
The study is one of two being published in Science that demonstrate the application of concepts taken from origami to areas that are far different than paper cranes.
A team from the University of Massachusetts at Amherst, Cornell University, and Western New England University in Springfield described a way to use the concepts of origami to come up with new materials, whose properties — such as how rigid they are — can be changed by tweaking the folding pattern.
That work was born when a UMass physicist, Christian Santangelo, was having dinner with Jesse Silverberg, a graduate student at Cornell. The two began passing back and forth a piece of origami that Silverberg had folded, using an intricate pattern that creates a series of mountains and valleys.
As they fiddled with the origami, the scientists noticed that if they disrupted the folding pattern by popping one of the folds down, they could drastically change how stiff the paper was. That dinnertime insight led to two years of work, as they attempted to unravel the math and physics at play, so that it might be used to build materials in the future that would have flexible properties — stiff one moment and loose the next.
“In robotics, this would be a useful thing — to be able to make structures that can basically lock or unlock at will — rigidify at will. You could imagine a floppy arm or something that then can lock into place and hold and withstand large forces,” Santangelo said. “In a certain sense, this is the first step in the path to making Transformers and robots that turn into other things.”