The discovery of planets that no one has ever seen is even more difficult than it sounds. NASA's cosmic telescopes and other scientific bodies have gathered an incredible amount of data that will require astronomers years and years to sift and many times there is nothing to find.
But people do not have to do everything Work and Anne Dathillo, a senior scientist at Texas University in Austin, helped for artificial intelligence to study NASA's Kepler Space Telescope data and found not one but two new exoplanets.
Dathilo developed a computer algorithm capable of detecting the weak hints of planets circling the stars spotted by the Kepler telescope.
Using a computer to quickly scan data from hunting craft such as Kepler is not new. but the data that Dathillo and her colleagues have undertaken require a unique approach. The second round of the Kepler telescope on planetary shades was slightly different from the first, and the expanded mission, known as K2, provided data that was blurred by instability. during K2 could still be used, the traditional planet detection algorithms would simply not work due to the movement of the sensor. This required an AI specifically designed to take this movement into account and effectively cancel it, allowing the computer to notice the failure of the brightness associated with a planet passing in front of a distant star.
Datilo and her team created an algorithm that did exactly that, and when it was applied to some of Kepler's K2 data, the computer manages to blow a pair of undetected worlds. Planets, which are about 1