• AI discovers 50 new planets from old NASA data

    Date:27 August 2020 Author: Kyro Mitchell

    The telescopes used by researchers to find new planets have become so good that they are now able to locate thousands of potential planets both inside and outside of our solar system.

    The problem with finding this many potential planets is the fact that scientists have to confirm whether they are indeed planets or not, and having to sift through that much data is extremely time consuming.

    When a potentially new planet is located by instruments like NASA’s Transiting Exoplanet Survey Satellite (TESS) for example, researchers confirm its existence by looking for a distinctive dip in brightness, which indicates something is passing by a star. If they detect this dip in brightness, it could mean a planet has passed in front of its star, but it could also be nothing more than asteroids, dust, or a glitch.

    To get around this problem, David Armstrong and his team of researchers from the University of Warwick in the UK are now using AI for planet confirmation. The research team developed a machine learning algorithm and trained to identify potential planets using previous data on confirmed planets and false-positives from NASA’s retired Kepler mission.

    Once the AI had received enough ‘training’ and learned to accurately separate real planets from false positives, they used it to analyze old data sets that had not yet been confirmed, also from the Kepler data. To their surprise, the AI system managed to confirm the existence of an additional 50 planets from that data set.

    The fifty planets range from worlds as large as Neptune to smaller than the Earth, with orbits as long as 200 days to as little as a single day. By confirming that these fifty planets are real, astronomers can now prioritise these for further observations with dedicated telescopes.

    “In terms of planet validation, no-one has used a machine learning technique before. Machine learning has been used for ranking planetary candidates but never in a probabilistic framework, which is what you need to truly validate a planet,” Armstrong said in a Warwick release.

    “Rather than saying which candidates are more likely to be planets, we can now say what the precise statistical likelihood is. Where there is less than a 1% chance of a candidate being a false positive, it is considered a validated planet.” Added Armstrong.

    Image: Pixabay

     

     



    Latest Issue :

    October 2020