IBM and NASA Release Open Source AI Model for Lunar Research

IBM and NASA announced in a press release the open source release of the NASA-IBM Lunar Foundation Model for scientific research on the Moon. The model was trained on lunar observations and can analyze data from multiple instruments, formats and resolutions.

The model supports research into potential lunar ice deposits, volcanic features and craters. IBM said it reduced errors in identifying areas with high potential for lunar ice by up to 22% compared with SwinV2-B, improved volcanic feature mapping by 3% and performed nearly 19% better on context scale crater detection while using half as much training data.

IBM and NASA also released a dataset with more than 30 spatially aligned layers from nine instruments across four missions. It contains tens of thousands of images and maps from NASA's Lunar Reconnaissance Orbiter and GRAIL missions, along with data from Japan's SELENE/Kaguya mission.