Artificial intelligence has proven itself extremely efficient in various research, including the exploration of exoplanets. This time, a group of scientists have managed to uncover new craters on the moon using an AI algorithm. Thanks to the algorithm, scientists have successfully revealed 6,000 previously unknown craters on Earth’s natural satellite.
Their discovery comes as a part of a Natural Sciences and Engineering Research Council (NSERC) project. This project has been funded so that scientists can see the results of how tech proves itself useful while participating in space explorations. Aside from discovering new craters on the moon, scientists used AI to determine the size of the craters on the lunar surface.
Normal explorations are the result of manually using a lot of tools. To do so, scientists have to take photos of the moon’s surface and manually count them to determine the crater count. After that, they are supposed to determine the size of the craters by manually checking the size of each image.
“Tens of thousands of unidentified small craters are on the moon, and it’s unrealistic for humans to efficiently characterize them all by eye,” Ari Silburt from University of Toronto, a co-developer of the algorithm said in a statement.
In this case, scientists used the convolution neural network. With it, a lot of time and effort was saved and the search provided more accurate results. The AI algorithm was trained for this mission from crater images caught by satellites, covering two-thirds of the moon. This algorithm helped determine the 6,000 previously unknown craters on the other one-third of the lunar surface.
“It’s the first time we have an algorithm that can detect craters really well, for not only parts of the moon, but also areas of Mercury,” Ali-Dib, another co-developer of the algorithm said in the statement.
Both the moon and Mercury are celestial bodies that don’t have air, water and plate tectonics. That makes them experience little erosion that helps them keep their oldest craters, with some of them dating back 4 billion years, remain visible.
Thanks to the observations, scientists have an insight in the shape and size of the discovered craters. However, they determine the age of the impact point by the interior structure of the hole, and by counting other, smaller craters located inside the larger craters.
AI is already known for its capabilities to discover exoplanets around the universe. Now, scientists have proven AI is capable of solving harder challenges. Being able to find impact craters could help scientists understand how the universe formed, especially the violent history of how Earth, its natural satellite, the Moon, and other planets came to exist.
“There’s real potential for machines to help identify these small craters and reveal undiscovered clues about the formation of our solar system,” Silburt noted.
The scientists wrote a paper that looks at the work AI performed in finding new craters on the Moon. They published the paper under review in the journal Icarus.