People have been folding paper for nearly as long as the material has existed. In the sixth century, monks in Japan transformed flat sheets into shapes replete with meaning. But using the term origami to describe the art became popular only in the 1900s, when Japanese kindergartens discovered that it offered inexpensive amusement for their tiny charges, and the idea proliferated.
But folding materials into new shapes is not just for toddlers. Now a team of engineers from Harvard and MIT has made the concept behind Transformers—the multibillion-dollar franchise based on cars that turn into robots—into a reality. In a new paper published Thursday in Science, Sam Felton and his colleagues show how they created the world’s first self-assembling robot, which forms itself, Transformer-like, from a flat sheet into a four-legged creature that crawls.
Advances in origami were key to making a fold-up robot possible. Mathematicians have been studying fold properties since the late 19th century, and more recent work has described properties conferred by different types of folds. Moreover, it has been shown that any arbitrary three-dimensional shape can be created from folding a flat one, so origami shape possibilities are endless.
Like origami, the technology to design, cut, and coax a self-folded robot together mixes old with new. The material for making the robot consists of three others glued together: a sheet of paper, a layer with plastic and copper, and a layer of pre-stretched polystyrene, or Shrinky Dink, a children’s toy invented in the ’70s. Shrinky Dink changes shape when heated to a particular temperature, which allowed the researchers to determine when a robotic fold would activate. Modern manufacturing “printers” allowed researchers to quickly make the flat component shape they needed.
Designing the shape and creases took some effort. “We knew when we started that we wanted something with four legs that could crawl,” Felton told me. From there, the team puzzled out how to cut and fold their sheet material so that legs would be connected and movable, but also leave enough space for batteries and a motor. Creases also had to be placed far enough apart so that heat would only affect one set of folds at a time. With the help of computer algorithms, they arrived at a design that worked.
Felton and his colleagues think this kind of self-assembly could have many practical uses. Once a machine has been designed, it is much faster and less expensive to make than it would be via traditional nuts and bolts production. Since any three-dimensional shape can be made from a two-dimensional one, a manufacturer could input a digital design for a robot that does a particular function, and have it created in a matter of hours, without needing lots of additional material. Materials for the newly created prototype robot, for example, cost just $100.
Having a robot that can go from flat storage into three-dimensional function is also advantageous for transport or access. Solar panels, for example, are launched into space in a compact state known as a Miura fold (after a Japanese astrophysicist), and expanded once they reach their mark. With robots that are too small to be efficiently put together with nuts and bolts, folding is an attractive alternative. In emergency situations or hard-to-reach places—under a crevice or pile of rubble, let’s say—the ability to deploy a compact robot that can then rearrange itself into a functional one could be a godsend.
But it may be some time before they can fold into cars.