Robot swarms could mine and build on lunar surface
Image credit: Chris Richards / University of Arizona
Engineers from the University of Arizona are mapping out a plan for mining the surface of the Moon, involving autonomous swarms of robots using innovative excavation techniques and machine learning.
Lunar mining could be moving into the realm of reality. Last September, Nasa announced an initiative to pay private companies to mine lunar regolith (rock and dirt covering the surface), in the hope of establishing international standards for business in space.
Mining on the Moon could be a more efficient way of building structures on the lunar surface, as it would not require rock, dust and dirt to be transported from Earth. It could also provide rare earth metals needed for technologies such as smartphones and medical devices, as well as titanium, precious metals and helium-3 (which is rare on Earth).
Now, a University of Arizona team is working on advanced space-mining methods, using Nasa funding.
“It’s really exciting to be at the forefront of a new field,” said Professor Moe Momayez, who is interim head of the department of mining and geological engineering. “I remember watching TV shows as a kid, like Space: 1999, which is all about bases on the Moon. Here we are in 2021 and we’re talking about colonising the Moon.”
To extract core embedded in rock on Earth, miners must drill through the rock. Momayez is a specialist in this process and has developed an electrochemical process to drill through rock five times faster than any other method. However, lunar mining presents new challenges.
“Here on Earth, we have an unlimited amount of energy to throw at breaking rocks,” he said. “On the Moon, you have to be a lot more conservative. For example, to break rocks, we use a lot of water and that's something we won't have on the Moon. So, we need new processes, new techniques. The most efficient way to break rocks on Earth is through blasting and nobody has ever set off a blast on the Moon.”
The University of Arizona researchers hope to harness autonomous robotic swarms to resolve this problem. They plan to build and train robots on Earth – using a unique neuromorphic learning architecture technique known as HEART – before sending them into space. This will not only train robots to work collaboratively on mining, excavation and building, but will also allow the robots to improve their collaboration skills over time.
Eventually, the researchers hope the robots will work in a fully autonomous swarm and be able to mine materials and construct simple structures without any instructions from Earth.
Professor Jekan Thanga, who developed the HEART system, said: “In a sense, we're like farmers. We're breeding talent out of these creatures, or a whole family of creatures, to do certain tasks. By going through this process, we help perfect these artificial creatures whose job it is to do the mining tasks.”
While the team considers humans a critical part of space exploration, they suggest that these robot swarms could free up astronauts’ time so they can focus on other mission elements: “The idea is to have the robots build, set things up and do all the dirty, boring, dangerous stuff, so the astronauts can do the more interesting stuff,” said Thanga.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.