It's alive! Using some paper, a circuit board and the plastic used in Shrinky Dinks, a team of researchers has designed an origami-inspired crawling robot that folds itself into working order in about four minutes.
These transformer-like paper bots, described in the journal Science, could offer a new way to make lightweight, self-building structures and machines – and give 3D printing a run for its money.
"I thought it was really exciting," said Michael Dickey, a chemical engineer at North Carolina State University who was not involved in the research. "It's quite a feat of engineering."
Even though the paper bot was made out of a single sheet of composite material, the robot that it folded itself into could walk and even turn. It easily carried the circuitry and batteries required for it to work, running at 2.1 inches per second.
The mechanical crawler is made out of a piece of flexible circuit board layered with sheets of polystyrene and paper with fold lines cut into a specific design. When a current is sent through the circuitry board, the resistive circuits heat the polystyrene layer to 100 degrees Celsius (212 degrees Fahrenheit) along the fold lines. The polystyrene – the same stuff used in Shrinky Dinks toys – then contracts, pulling at the paper layer and causing the joints to fold.
The mathematics of origami folding has long impressed scientists. And folding is a technique used by living things on many levels, whether it’s the folding of proteins on the cellular scale or the folding of tissue in the brain. For engineers, it can be a way to maximize surface area or strength (or both) with very little material. Another report in Science led by Jesse Silverberg of
Scientists have looked into building by folding before. In fact, a
But for this paper, another team of researchers also anchored by Harvard scientist Robert Wood wanted to see if they could make a complex machine that could assemble itself.
"We picked a crawling robot because it's actually a very difficult problem," said study leader and Harvard roboticist Sam Felton. A stationary machine, with a heavy, boring base, would have been much easier to fold into position – and then it would stay put. But for a crawling robot, "we had to make sure we could fit all the power and the controls in a mobile structure."
This method for building machines – creating two-dimensional patterns that then fold themselves into complex three-dimensional shapes – could be cheaper than 3D printing, Felton said. The materials used to make it didn't cost much: The paper was akin to very thin poster board, and the polystyrene cost about 50 cents a sheet.
Origami-inspired assembly could make machines potentially easier to mass produce, and it's simpler to install chips and batteries on a flat surface than around or inside a complex, 3D-printed object.
Such self-folding machines could one day be used as satellites. They could be sent to space in an easily packed stack of sheets, and unfold themselves into their proper shapes once in orbit. They could land on another planet and quickly assemble into a shelter. On Earth, they could be tossed into disaster zones to enter an earthquake-ravaged buildings, or perform other work in areas unsafe for humans -- similar to these robotic termites.
There are a whole lot of steps to be taken before that can happen, Felton pointed out. For one thing, scientists need to develop powerful software that will come up with the fold pattern for a given robot or machine. For this robot, Felton had to do much of the patterning by hand.
"I tried using some programs but a lot of it came down to trial and error," he said.
The technology is in its infancy; future robots may need to use different materials. And for now, the assembly is a one-way process: Once a robot has folded, they can't unfold themselves.
But researchers are working on building tiny pouch motors that could fold and unfold at will, Felton said. And with those kinds of joints, this origami-tech could one day lead to robots that could, like the fictional Transformers robots, change into different configurations, depending on the situation.