The NASA-IBM Lunar Foundation Model is a publicly available artificial intelligence tool developed to support the study and mapping of the Moon’s surface.
The model was trained using more than 30 layers of data collected by nine instruments aboard four NASA missions, including the Lunar Reconnaissance Orbiter. It is part of IBM and NASA’s Prithvi family of open foundation models, which also covers applications in geospatial analysis and weather.
The AI tool can assist researchers in detecting potential ice deposits in the Moon’s permanently shadowed regions, mapping craters to identify safer landing sites and analysing volcanic features. Such tasks have traditionally required scientists to manually examine maps and images or use machine-learning systems operating with lower-resolution data.
According to NASA and IBM, benchmark tests showed that the model identified key lunar surface features with up to 23 per cent greater accuracy than widely used methods.
The search for lunar ice is particularly important for space agencies because water deposits could provide access to resources such as water and oxygen. These resources could support a future sustained human presence on the Moon and potentially be used to produce rocket fuel for missions to Mars.
NASA’s Artemis programme aims to return astronauts to the Moon in 2028 as part of efforts to test technologies for a sustained lunar presence and lay the groundwork for future missions to Mars.
