- Scientists may have located the Luna 9 spacecraft using AI analysis of lunar images
- Luna 9 was the first spacecraft to achieve a soft landing on the Moon in 1966
- Researchers used the YOLO-ETA algorithm to scan NASA's Lunar Reconnaissance Orbiter data
Scientists may have spotted the long-lost Luna 9 spacecraft, an uncrewed Soviet space mission that achieved a historic soft landing on the Moon in 1966 and returned images of its surface. It operated for just three days, but provided humanity with valuable data about the lunar surface. Despite its historic significance, the exact location of Luna 9 remained a mystery for decades. The spacecraft's landing site was estimated to be in the Oceanus Procellarum region, but the uncertainty was too large to pinpoint its location.
Also read | NASA's Hubble Telescope Captures Stunning Image of 'Space Egg'
But now, the researchers in the UK and Japan used an artificial intelligence (AI) to identify several promising candidate locations for the spacecraft.
They used an advanced machine-learning algorithm called YOLO-ETA (You-Only-Look-Once-Extraterrestrial Artifact) to scan images of the lunar surface by NASA's Lunar Reconnaissance Orbiter (LRO).
The team, led by Lewis Pinault at University College London, published the results in npj Space Exploration. The researchers hope that their model's predictions could soon be tested using new observations from India's Chandrayaan-2 orbiter.
"The Lunar Reconnaissance Orbiter Camera (LROC) Narrow Angle Camera has imaged the Moon continuously since 2009, providing a unique record of both natural and anthropogenic surface features at up to 0.25 m pixel-1 resolution," the new team explains in their study. "Identifying artificial objects within the resulting vast dataset remains a challenge owing to illumination variability, complex backgrounds, and the small pixel footprints of many targets."
The algorithm detected a cluster of candidate artifacts near 7.03 Celsius N, -64.33 Celsius E, which match the expected dispersal pattern of mission components. The terrain and horizon in the images appear consistent with the flat lunar landscape seen in Luna 9's 1966 photographs.
Also read | IITian At Yale University Lists "Massive Life Upgrades" After Moving To US, Internet Reacts
"These findings identify promising locations for follow-up imaging and demonstrate that compact, edge-deployable machine-learning models can support future orbital surveys of lunar artefacts and surface assets," researtchers wrote in the study.
While the findings are promising, they are not yet definitive proof of Luna 9's location. Further imaging by LRO or future orbiters is needed to confirm the discovery. Overall, the study highlights how AI can be used efficiently in planetary exploration.
