For Robots In Sand, The Secret Is In The Step Sand is difficult terrain for robots. But Daniel Goldman, a physicist at Georgia Institute of Technology, and his colleagues looked to lizards and cockroaches for inspiration.

For Robots In Sand, The Secret Is In The Step

  • Download
  • <iframe src="https://www.npr.org/player/embed/101547339/101547326" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

IRA FLATOW, host:

Up next: Flora Lichtman is here with our digital video pick of the week. Hey, Flora.

FLORA LICHTMAN: Hi, Ira.

FLATOW: Welcome back.

LICHTMAN: Thank you very much. Good to be back.

FLATOW: That's great. What have you got for us this week?

LICHTMAN: This week, a new robot. The robot is called Sandbot.

FLATOW: Sandbot.

LICHTMAN: And it's very cute. And as you might have guessed from the name, it's designed to move in sand, which is not - it turns out it's not trivial to get a robot that can maneuver soft ground like sand. And these researchers, you know, started with this robot called Rex. And this robot has been around for many years, but - and it was excellent on hard ground, just great, one of the fastest robots out there.

FLATOW: Great in the parking lot.

LICHTMAN: Yes, exactly.

(Soundbite of laughter)

LICHTMAN: Yeah exactly.

LICHTMAN: So, Daniel Goldman, who's a physicist at Georgia Tech, said to himself, well you know, he's interested in more complicated terrain and said what if I put kind of a Rex sibling, a smaller version of Rex on sand?

FLATOW: Huh, huh.

LICHTMAN: And they put it on sand, and it went nowhere. I mean, it just spun holes, sort of dug itself into a hole.

FLATOW: I watched the video about that. It was just like, yeah. It was like digging a hole. He's, like, digging a hole.

LICHTMAN: You can just see it, sort of sink into the…

FLATOW: Like my car on the beach. Exactly.

LICHTMAN: Right.

FLATOW: Yeah.

LICHTMAN: So this is sort of an interesting physics problem. How do you make a robot that can deal with this terrain? And we've talked about this before. Sand is actually is really difficult to model. So they did a lot of experimentation, and what Goldman and his colleagues did was, you know, look to organisms like lizards that are really good.

FLATOW: They - run very quickly through this.

LICHTMAN: No problem with sand. And what they found after some trial and error is that if you change the way the robot walks, you can make some progress here.

FLATOW: So you have video of - at sciencefriday.com. And it's right there on our front page, on our home page is the video, this pick of the week. And you can just watch the progression of this experiment.

LICHTMAN: Right. You can see it at the beginning, sort of faltering, this poor little robot's going nowhere. And then, you know, eventually, it's like the little robot that could.

FLATOW: It looks - that one looks like a motorized cockroach.

(Soundbite of laughter)

FLATOW: You could say that about it.

LICHTMAN: No, it's true. But, you know…

FLATOW: Yeah.

LICHTMAN: …it's based off of cockroaches.

FLATOW: Yeah, I would imagine. They run - they do very well in the sand. But it's interesting. You go to the evolution of how the engineers, the scientist had to study the right way to get the limbs to move in the sand.

LICHTMAN: Right. And it takes observation. I mean, it really took looking at these organisms that do well in the sand and saying what about them - is it the toes? Is it how fast they move their limbs? Is it their gait, you know, the order?

FLATOW: Which limbs are used together, right?

LICHTMAN: At once. Right.

FLATOW: He said there were three at a time, always, in contact?

LICHTMAN: Right. So, this robot Sandbot has six legs, and it's an alternating tripod gate. That's the sort of technical jargon. But it just means that three legs go at once, and then the other three legs go at once. So they're 180 degrees out of sync. It's easier to see.

FLATOW: Yeah. And it's a small robot, right?

LICHTMAN: Yeah, it's small. A couple of kilograms.

FLATOW: They're about as big as a shoe box. That's how we - you can put it in a shoebox.

(Soundbite of laughter)

LICHTMAN: I think it would fit inside. Yup.

FLATOW: Inside. And you have, you have interviewed the scientist who worked on it.

LICHTMAN: Right, Dr. Goldman.

FLATOW: Dr. Goldman.

LICHTMAN: And I think, you know, his next steps are using this robot to better understand, you know, how organisms are adapted to deal with these environment. So he kind of goes both ways. Kind of interesting.

FLATOW: Well, we'll look forward to it. It's a Flora's video pick of the week. It's on our sciencefriday.com Web site. Just go to sciencefriday.com. You'll see the video there, and you can see all the other videos that we have in our library that Flora has made. We'll see you next week, Flora.

FLATOW: Thanks, Ira.

FLATOW: Thank you. That's about all the time we have for today.

Copyright © 2009 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.