AI framework developed to assist lunar mission planning

Researchers at The Ohio State University have developed an artificial intelligence framework called LUMINA to assist in planning lunar missions [225]. The system is designed to evaluate potential landing sites, identify hazards for future crews, and scour new regions for scientifically significant geologic features [225]. Preliminary results indicate the model can make complex lunar observations and provide scientific commentary regarding candidate sites, which could assist planners in locating resources such as water ice [225].

LUMINA is built upon a previously developed lunar vision-language foundation model known as LLaVA-LE [225]. Unlike passive tools that only provide answers to specific questions, this framework is intended to function as an active assistant that can conduct multistep analyses and explain the evidence behind its recommendations [225]. Developer Gokce Inal, a Ph.D. candidate in civil, environmental and geodetic engineering, stated the system is a decision-support tool meant to work alongside mission planners rather than acting as a semiautonomous machine [225].

While currently focused on the moon, the researchers suggest the technology could be applied to other planetary bodies, such as Mars or asteroids, as well as Earth [225]. Co-author Alper Yilmaz, a professor in civil, environmental and geodetic engineering, noted that the effort aims to create a model that is generalizable to various surfaces [225]. The team intends to continue developing the system by testing it against new, high-quality lunar data, including information that may become available through NASA's Artemis program [225]. The research findings were presented at the 58th Annual Meeting of the Division for Planetary Sciences of the American Astronomical Society [225].