[ad_1]
Discovering planets that no one has ever seen before is even more difficult than it seems. Space telescopes from NASA and other scientific organizations have collected an incredible amount of data that will take years and years for astronomers to sift through, and often there is nothing to find.
But humans do not have to do all the work and Anne Dattilo, a senior at the University of Texas at Austin, used artificial intelligence to study data from NASA's Kepler Space Telescope and found out not one, but two new exoplanets the process.
Researchers led by Dattilo, 22, have devised a computer algorithm that can detect weak allusions from orbiting planets to stars observed with Kepler's telescope.
Using a computer to quickly scan research instrument data from around the world such as Kepler is not new in itself, but the data that Dattilo and his colleagues used required a unique approach. The second Kepler telescope surveillance tower was a little different from its first, and the extended mission called K2 provided data that was confused by instability.
Kepler developed mechanical problems later in life and, although the data collected during K2 is still usable, traditional planet detection algorithms simply would not work because of sensor movement. This required a specially designed AI to take this motion into account and cancel it effectively, allowing the computer to detect the drops in brightness associated with a planet passing a distant star.
Dattilo and his team have created a specially designed algorithm for this purpose. When this was applied to some of Kepler's K2 data, the computer was able to detect a pair of worlds never discovered before. The planets, located about 1,300 light-years away from Earth, can now add to Kepler's incredible legacy, which has detected more than 2,500 planets during his multiple observation campaigns.
Source of image: NASA
[ad_2]
Source link