AI helps to colonize Mars by searching for caves where the first humans can settle
A new artificial intelligence system has successfully detected dozens of entrances to caves on Mars using images of the planet’s surface where people could live.
This was reported by the British edition of The Independent.
Researchers from Durham University in the UK have trained the CaveFinder machine learning algorithm to recognize potential cave entrances from images of the Martian surface.
These caves, formed as a result of the collapse of ancient lava tubes, can serve not only as a shelter for future researchers, but also as potential areas for studying signs of microbial life on Mars. Researchers suggest that many such tubes may be interconnected under the planet’s surface.
The CaveFinder system was able to identify 61 entrances to such caves by analyzing images in four different regions of Mars. This opens up a new approach to finding caves on Mars, as previously scientists manually studied thousands of satellite images, which required considerable effort and time.
Even if the artificial intelligence system is not yet ready to detect caves on the entire planet, scientists believe that it can be effective in identifying potential caves in limited regions where it is already known that they may exist.
