New AI analyzes indicate possible location of the historic Luna 9 probe on the lunar surface
The mystery surrounding the final resting place of the first spacecraft to make a soft landing on Lua appears to be close to a definitive solution thanks to the use of modern data processing technologies. Pesquisadores applied advanced artificial intelligence algorithms to high-resolution orbital images to try to identify the exact position of the Soviet probe, which landed on the natural satellite in February 1966 and revolutionized space exploration by sending the first photos of the lunar surface.
The historic mission operated for just a few days before losing contact, leaving the exact location of its remains unknown for six decades. Equipes Independent scientists now pinpoint promising coordinates in the Oceanus Procellarum region, using visual data captured by Lunar Reconnaissance Orbiter from Nasa combined with machine learning models trained to distinguish human artifacts from natural geological formations.
Discrepancies in the coordinates found
Different groups of scientists have reached different conclusions about where the module would be located, which keeps the academic debate heated and demonstrates the complexity of tracking small objects in Lua. A team led by University College London used an artificial object detection model, trained with images of landing sites from the Apollo missions, to scan a search area of approximately 25 square kilometers.
Another strand of investigation, led by Russian and Japanese experts, focused on the detailed comparison of the horizon features visible in the original 1966 panoramic photos with the current topography mapped by satellites. Essa approach resulted in a position about 25 kilometers away from that estimated by British researchers, suggesting that Soviet-era calculations about the descent trajectory and bounce of the probe’s inflatable shock absorbers may have been inaccurate at the time.
Technology applied in space search
The algorithm used in the new scan, known as YOLO-ETA, processed hundreds of images to isolate geometric objects and shadow patterns that do not correspond to ordinary rocks or craters. Automatic detection allowed researchers to filter vast areas of the satellite’s surface with an efficiency impossible for conventional human observation, identifying candidates with a high probability of being the mission’s components, such as the spherical capsule and petal-shaped panels.
Definitive visual confirmation of these hypotheses depends on new orbital observations scheduled to take place later this year. The Indian probe Chandrayaan-2 is expected to carry out strategic overflights in the candidate regions in March, using high-resolution cameras to attempt to photograph the module and validate the mathematical models applied in the search, rescuing a fundamental chapter in the history of the space race.
