Microscale Robot Folds, Crawls

The legacy of microscale robotics at Cornell continues to unfold — and refold and unfold itself again.

The latest addition is a robot less than 1 millimeter in size that is printed as a 2D hexagonal “metasheet” but, with a jolt of electricity, morphs into preprogrammed 3D shapes and crawls.

The robot’s versatility is due to a novel design based on kirigami, a cousin of origami in which slices in the material enable it to fold, expand and locomote.

The project was led by Itai Cohen, professor of physics in the College of Arts and Sciences (A&S), whose lab has previously produced microrobotic systems that can actuate their limbs, pump water via artificial cilia, and walk autonomously.

The kirigami robot is the next step in that evolution, and is the result of longtime collaborations with Paul McEuen, the John A. Newman Professor of Physical Science (A&S); Hadas Kress-Gazit, the Geoffrey S.M. Hedrick Sr. Professor in Cornell Engineering; Nicholas Abbott, a Tisch University Professor in the Robert F. Smith School of Chemical and Biomolecular Engineering in Cornell Engineering; Alyssa Apsel, the IBM Professor of Engineering in Cornell Engineering; and David Muller, the Samuel B. Eckert Professor of Engineering in Cornell Engineering – all co-authors of the paper.

In a sense, the origins of the kirigami robot were inspired by “living organisms that can change their shape,” Liu said. “But when people make a robot, once it’s fabricated, it might be able to move some limbs but its overall shape is usually static. So, we’ve made a metasheet robot. The ‘meta’ stands for metamaterial, meaning that they’re composed of a lot of building blocks that work together to give the material its mechanical behaviors.”

Such metamaterials can often be designed to have properties that are difficult to achieve with natural materials, Wang said.

Here is an exclusive Tech Briefs interview, edited for length and clarity, with Cohen.

Tech Briefs: What was the biggest technical challenge you faced while developing this kirigami robot?

Cohen: This is a massive feat of engineering. The origami robot has something like 60 big panels, and in order to get them to splay, we had to create an origami linkage. And each origami linkage had three hinges that you had to actuate. And so this thing had just a massive number of components, almost 200 different hinges that we had to control in order to get the robot to function. So that was a massive piece. Just the fabrication itself was a tour de force because a good fabrication is at about 90 percent. But, for these robots, 90 percent means that about 20 of your hinges are not working. So, we had to get much, much higher and the students really pulled it through. That was a big one.

Then, the other big engineering feat was trying to figure out when it's locomoting across the surfaces, what mechanism is it using to crawl and how do we create something that's controlled? Because if you have all these different surfaces, depending on which ones are touching the ground, you end up getting motion in different directions. That makes it very hard to control. So both of those were the big challenges with this robot.

Tech Briefs: Can you explain in simple terms how it works?

Cohen: The basic idea is that we want to be able to print robots in a 2D sheet, and then we want the sheet to expand and essentially morph into a 3D shape. Now, if you've ever picked up a sheet of paper, you know that the best you can do with a sheet of paper is kind of roll it up into a cylinder, right? If you tried to wrap a sheet of paper on an orange, you get all these crinkles and defects. So, that doesn't work.

What you have to do is you have to put cuts in the paper; that's a technique called kirigami, paper cutting. And then what happens is, as you wrap the paper around some sort of 3D object, it can locally expand. So, there's the broader paper that stays the same, but the area that's kind of expanding into the third dimension is expanding. You can change the Gaussian curvature of the sheet; a sheet normally has a Gaussian curvature of zero. And by having these locally expanding regions, you can change that; that allows the sheet to adopt any 3D shape. Now you get to print something in 2D, but you get to morph it into any 3D shape that you want because you gave it these cut patterns that make it morphable.

Tech Briefs: Do you have plans for any further research, work, etc. on the horizon?

Cohen: Right now, the way that we morphed the robot was we either had all the hinges actuating at once, or we split the robot into two pieces and actuated each piece separately — maybe an inner piece and an outer piece, or the left and the right. But, in the end, eventually what we want to do is we want to put a little microchip — like the same CMOS microchip that drives all of the technology in your phone — on each of these 60 micron panels. Each panel will have its own little brain telling it what it should be doing with the hinges. The idea is that these chips will sort of have to talk to one another and act in concert to essentially be able to communicate with one another, how to actuate the hinges in a coordinated fashion.

This is sort of the way an octopus’ brains are distributed over their tentacles. It's the same kind of idea. We want to distribute the brain over the different panels in the sheet and have the sheet locomotion arise as an emergent behavior from the interactions of these electronic chips. Once you have chips controlling everything, then your material can start to do things that you could never do with a natural material. So, it's really giving it additional functionalities.

I’ll give you an example. I have a sheet and I deform it on one side. The only way this side knows that it's been changed is that there are sound waves that travel all along the sheet.

But, if I have a sensor on this side and it communicates to the other side at the speed of light, I could actually beat the speed of sound. So, that means that this side of the sheet could know that something happened on this side, on the other side before the sound waves even reach it.

Tech Briefs: Oh, wow. That’s very interesting stuff!

Cohen: Yeah, we're having fun. What's amazing is that once you start playing in this realm, you realize that 50 years of Moore's law has given you this enormous potential to completely make everything electronic at the microscale. That opens up possibilities that you could never imagine with just a passive material.

What's really made these robots and technologies possible now, was creating something that can bend in response to voltage at the microscale. So, the idea is that the chips can provide some voltage to your actuator, the bending part of the robot. And, up to now, there just haven't been any microscale actuators; even MEMS devices really only bend to the millimeter. We needed something that's going to bend on the microscale because we wanted to make microscopic sheets. That's really the technology that's been undergirding the whole enterprise.

---