Photometric Search for Exomoons by using Deep Learning and Convolutional Neural Networks

Astronomers know of more than 4,000 exoplanets. These are celestial bodies that orbit a star other than our Sun. However, none of these has yet been proven to have a moon, although there are some possible candidates. Lukas Weghs wrote an intelligent programme for a high-performance computer that helps identify possible exomoons. This is done using the transit method, which is also used to detect exoplanets: as soon as a celestial body passes in front of a star from the Earth’s point of view, it minimally decreases its brightness according to a defined pattern. An existing exomoon would change this pattern a little more. The young researcher’s algorithms help search astronomical brightness measurements for traces of exomoons.

Category: COMPUTING Country: GERMANY Year: 2021


Lukas Weghs