An international team appears to be scanning the Moon with an AI system and identifying areas for landing and exploration.
Selecting future landing and exploration sites on the Moon can result in the most promising locations for construction, minerals or energy resources, but optical scanning over such a large area is laborious and often inaccurate.
See also: AI: How does it help us better understand the universe?
Siyuan Chen, Xin Gao and Shuyu Sun from KAUST (King Abdullah University of Science and Technology) in Saudi Arabia, along with colleagues from the Chinese University of Hong Kong, have now applied machine learning and artificial intelligence (AI) to automate the identification of potential landing and exploration areas on the Moon.
See also: Cancer and AI: Researchers develop tool for early diagnosis of prostate cancer
Machine learning is a very effective technique for training an AI model to search for specific features on its own. The first problem Siyuan Chen and his colleagues faced was that there was no dataset with labeled rilles that could be used to train their model.
According to KAUST, the next challenge was to develop a computational approach that could be used to simultaneously detect craters and rilles, which has never been done before.
See also: Politicians: AI measures how much time they spend on their phones during debates
The team's approach is said to have achieved an accuracy of up to 83.7%, which is higher than existing crater detection methods. Their findings have been published in Applied Energy.
Information source: theengineer.co.uk
