Efficient as a Clam


RoboClam technology is based on the digging mechanisms of Atlantic razor clams. Images: MIT Gear Lab

MIT assistant professor Amos Winter knew he wanted to create an underwater robot, partnering with Bluefin Robotics, a company he once interned for. After adding MIT professor Anette Hosoi to the mix, along with her interest in animal locomotion, they set about finding the best animal to emulate when it came to digging skills.

“I just remember going through endless books,” says Winter, “trying to find the perfect one.” He needed something that could efficiently move a rigid object through underwater soil and be compact. It needed to be low power and he wanted the robots to stay stationary and not fight a current. He wanted this application to be for anchoring.

Clams were your winner because, according to Winter, they: 1) Live in every substrate of water and 2) Can burrow through seemingly everything. They can go through rock, the shipworm version goes through wood, etc.

But with so many clams in the ecosystem, which one to choose? The Atlantic razor clam stood out, a creature with a shell that opens and closes like a book and a foot, which is a tongue-like structure that sticks out from the bottom of the shell and inflates like a balloon. The Atlantic razor clam can also dig a centimeter per second.

Dan Dorsch (left) and Amos Winter testing a RoboClam prototype. Image: MIT GEAR Lab

“We compared it against manmade technology to see if it provided a benefit,” he says. “I figured out how the anchoring force goes with depth and size of an anchor and figured how much anchoring force came from a razor clam. The numbers showed it was very efficient.”

Next was identifying how to utilize the way it digs. “But how is it so efficient?” Winter says. “We realized by visualizing and digging through soil that the razor clam moved the shell a certain way in order to liquefy the soil and make a pocket of quicksand. That’s much easier to move through and requires a lot less force. If you look at differences between fluid and static soil it’s justified why they can dig as deep as they can. The crux of the problem for us was how to move the rigid part of the shell through the soil. It was the movements of the shell that reduced the drag.”

“Instead of using a foot like the clam, the RoboClam used a piston and a rod that pushed downward on a clam-like shell puppet. Also, instead of pulling from below, [the RoboClam] pushed from above in order to push the digging implement down.”

Winter says the puppet opened and closed with symmetric expansion on each side of the mechanism, expanding southwards and contracting inwards.

But the RoboClam may not be fit for anchor use alone. Underwater cable installation is another possibility “When you run transatlantic cable, you have the boat that pulls a sled and the sled digs a trough,” he says. “But when you get to 10 meters of water depth it can be too shallow for a sled and you have to manually dig in cable with divers—that takes time and is expensive. It’s better to have a system of clamps on the cable and get into the soil. That’s where the RoboClam can come in.”

Other applications could include deep sea oil recovery and even exploration on the moons around Jupiter, Winter adds. “There are so many potential opportunities when you have a low-energy means of digging in the soil,” he says. “It will be fun to see where it leads.”

Eric Butterman is an independent writer.

There are so many potential opportunities when you have a low-energy means of digging in the soil. It will be fun to see where it leads.

Prof. Amos Winter, MIT


June 2014

by Eric Butterman, ASME.org