A fresh AI
algorithm has revealed over 300 unknown exoplanets in data collected by a
now-defunct exoplanet-hunting telescope.
The Kepler,
NASA's first dedicated exoplanet hunter telescope, has detected hundreds of
thousands of stars in the hunt for potentially habitable worlds outside our
solar system. The collection of potential planets it had accumulated continues
generating new discoveries even after the telescope's expiration. Astronomers
analyze the data for signs of exoplanets. But a new algorithm called ExoMiner
can now caricaturist that procedure and search the catalog faster and more
efficiently.
The telescope, which stopped working in late 2018, looked for partial decreases in the brightness of the stars that might be affected by a planet crossing in front of the star's disk as viewed from Kepler's perspective. But not all such darkening are caused by exoplanets, and astronomers had to follow elaborate procedures to differentiate false positives from the real stuff, according to a statementfrom NASA.
ExoMiner,
is a so-called neural network, a type of AI algorithm that can study and
improve its abilities when provided a sufficient amount of data. And Kepler created
plenty of data: In the less than 10 years of its service, the telescope revealed
thousands of planet candidates, nearly 3,000 of which have since been
confirmed. That is a immense majority of the overall 4,569 exoplanets currentlyknown.
For each
candidate exoplanet, astronomers poring through the Kepler data would look at
the light curve and evaluate how large a portion of the star the planet looks
to be covering. They would also explore how long it appears to take the
would-be planet to cross the star's disk. In some scenarios, the observed
brightness variations are not likely to be explained by an orbiting exoplanet.
The ExoMiner algorithm follows exactly the same procedure but more efficiently,
which allowed the astronomers to add a group of 301 previously unknown
exoplanets into the Kepler planet catalog at once.
0 Comments