MENU

2 New Super-Earths Discovered Using Artificial Intelligence, Deep Learning

Close

Astronomers from the University of Texas at Austin use artificial intelligence to discover two new exoplanets hidden in Kepler space telescope's data archive.

Researchers said the technique will help scientists unveil additional planets that are missed using traditional search methods.

Kepler Extended Mission

UT Austin undergraduate Anne Dattilo developed a deep learning algorithm that sifts through the existing data taken by Kepler. It aims to identify signals that were missed by previous planet-hunting methods.

AstroNet-K2 is a neural network that is capable of being trained to look for exoplanets and identify false positives with an accuracy of 98 percent. The system currently still needs human supervision to ensure the reliability of the candidate planets.

The team is joined by NASA Sagan fellow at UT Austin Andrew Vanderburg and Google engineer Christopher Shallue. They carried out the task during Kepler's extended mission called K2.

"AI will help us search the data set uniformly," Vanderburg said. "Even if every star had an Earth-sized planet around it, when we look with Kepler, we won't find all of them," Vanderburg said.

Vanderburg explained that some of Kepler's data are too noisy or that the planets are sometimes not aligned the right way, making it difficult to detect the actual planets.

To determine the total number of planets, Vanderburg said it is important to count the planets they may have missed.

The planets, named K2-293b and K2-294b, orbit their host stars which are 1300 and 1230 light-years away respectively. Both are located in the constellation of Aquarius.

Future Of Planet Hunting

Dattilo said the new algorithm can help uncover new potential planets in the K2 archive. The Kepler space telescope has observed images of approximately 300,000 stars.

She also believes that AstroNet-K2 can help Kepler's successor planet-hunting mission TESS.

"Our method is a step towards automatically identifying new exoplanets in K2 data and learning how exoplanet populations depend on their galactic birthplace," the authors concluded.

The study has been accepted for publication in an upcoming issue of The Astronomical Journal.

ⓒ 2018 TECHTIMES.com All rights reserved. Do not reproduce without permission.
Real Time Analytics