The robotics team at GE Research recently showed off its DARPA-funded autonomous digging robot. The bio-inspired soft robot is designed to tunnel and navigate underground obstacles similar to an earthworm.
Funded with $2.5 million from DARPA’s Underminer Program, the robot achieved a big milestone during a recent demonstration.
To test the prototype, GE researchers had the robot worm burrow under GE’s research campus in Niskayuna, New York. GE partnered with InnoVital Systems to design the earthworm. InnoVital helped create the robot’s artificial muscles that mimic earthworm muscular structure and function. Powered by fluidic artificial muscles, the worm carved out a tunnel with a 10 cm (about 4 inches) diameter.
According to the researchers, the project is a breakthrough for autonomous robotics and soft robots designed to manage tighter turning radiuses and operate in extreme environments.
Soft robots are designed without joints, which can give them more degrees of freedom than conventional robots. The robots can be more dexterous and flexible to make better turns and squeeze through small spaces or pick up and move various objects. For inspiration, GE’s team picked the earthworm, one of nature’s most proficient tunnel diggers,
Biomimicry, the design of products inspired by nature, has led to some incredible advancements in robotics and aerospace (and a couple ridiculous ones, as well). For example, two years ago, Russian researchers made a drone that looked like a snow owl to avoid detection. In June 2019, Airbus engineers created the AlbatrossOne, a prototype aircraft with bird-inspired free-flapping wingtips to fend off turbulence while reducing drag and wing weight.
GE’s work won’t just advance tunneling technologies but also advanced inspection and repair capabilities using robots. The company previously explored using snake-inspired robots for jet engine inspection and repair.
DARPA’s Underminer program is working to develop new solutions that are better than current commercial drilling capabilities, including new tunneling approaches, downhole sensing and operations concepts. Initially, the hope is to create robots that can quickly construct tactical tunnel networks for secure logistics infrastructure to pre-position supplies or resupply troops.
Teams from the Colorado School of Mines and Sandia National Laboratories were also selected for the project.
