IBM and NASA Unveil AI Model to Aid Lunar Research
IBM and NASA have collaborated to launch an open-source AI model that will aid in lunar research. The model, called the NASA-IBM Lunar Foundation Model, has been developed using data from decades of lunar observations and combines information obtained by various instruments at different resolutions.
The model is part of IBM's Prithvi family of open foundational models designed for scientific applications, including geospatial data, meteorology, and heliophysics. Its purpose is to address the challenge of analyzing large volumes of data accumulated from multiple missions in an integrated way.
Researchers have tested the model on various tasks, including identifying possible ice deposits, mapping volcanic formations, and detecting craters. According to a technical study by IBM and NASA, the model was able to reduce errors in identifying areas with potential for lunar ice by up to 22% compared to another model.