Fire Ants Studied To Develop Better Digging Robots

Researchers at Georgia Tech are drawing from the characteristics of fire ants to fuel their research into tunneling robotics. The scientists are studying the way the ants dig, and have so far gleaned a number of principles they are hoping to apply to future robots.

The fire ants dig in a way that creates tunnel environments that minimize slips and falls, as well as making it easier to navigate without complex neural processing. The ants also rely on their antennae for both locomotion, as well as sensing the space around them while digging.

"The environment allows the ants to make missteps and not suffer for them," said Daniel Goldman, an associate professor in the School of Physics at the Georgia Institute of Technology. "These ants can teach us some remarkably effective tricks for maneuvering in subterranean environments."

To study their technique, the ants were placed in tubes full of soil and allowed to dig for 20 hours under observation. The researchers simulated a wide range of environmental conditions by altering the soil consistency and moisture.

The scientists found that the ants' tunnels remained widely consistent throughout these variables, which demonstrated some very interesting principles.

"The size of the tunnels appears to be a design principle used by the ants, something that they were controlling for," Professor Goldman stated.

Researchers also observed ants using their antennae to grab onto tube walls when they fell.

"A lot of us who have studied social insects for a long time have never seen antennae used in that way," Michael Goodisman, a professor in the Georgia Tech School of Biology said. "This is an adaptive behavior that we never would have expected."

Scientists believe the discoveries made during these experiments will be extremely valuable when designing digging robots.

"The problems that the ants face are the same kinds of problems that a digging robot working in a confined space would potentially face - the need for rapid movement, stability and safety - all with limited sensing and brain power," said Goodisman. "If we want to build machines that dig, we can build in controls like these ants have."

The study was published in the Proceedings of the National Academy of Sciences.