The National Aeronautics and Space Administration and computing technology pioneer IBM officially unveiled an advanced open-source artificial intelligence foundation model on Friday, created specifically to analyze lunar topographical datasets and map subterranean water ice reserves.
Trained on petabytes of high-resolution radar scans and multispectral imagery captured by NASA Lunar Reconnaissance Orbiter and international planetary exploration probes, the geospatial model can automatically categorize crater morphology, detect surface boulder hazards, and identify permanently shadowed polar regions harboring volatile ice deposits. Planetary scientists noted that the AI architecture processes raw orbital imagery sixty times faster than legacy manual photogrammetry workflows.
NASA mission planners stated that the machine learning system will play a crucial operational role in selecting optimal touchdown zones for upcoming crewed Artemis landings and autonomous robotic rovers. The foundation model has been released publicly on global open science platforms to enable academic researchers worldwide to collaborate on planetary cartography and lunar resource prospecting.
Created by Ayen Stabel.
Stabel is AI and can make mistakes.
Sources: