A team of astronomers led by a student in Texas has discovered two planets circling the stars of more than 1200 light-years from the Earth. But this is what has revealed them and how is it that attracts the attention.
Anne Dathilo, a senior scientist at the University of Texas, Austin, found the planets using an Artificial Intelligence program to sift out a mountain of data collected from Kepler's Space Telescope. Using AI, the 22-year-old man helped introduce a new era in astronomical research. Dattillo joined the project about a year and a half ago after astronomer Andrew Vanderburg spoke during one of them. classes. Vanderburg uses data from Kepler to search for planets circling in distant stars.
"And at the very end, he said," I'm taking students if any of you want to do research on this by finding planets, "Dathillo recalls." I figured it was what I wanted to do, so I sent it by email. "
Kepler, which launched in 2009, is meant to point to a small part of the sky and measure the light of about 100,000 stars in the field of vision. one of the stars, the light coming from this star will diminish slightly as the planet passes in front of it.
Towards the end of Kepler's life, mechanical damage means that it can not measure the light so accurately that the data collected is more difficult to interpret.
Dattilo modified an artificial intelligence program called AstroNet-K2 to work on data from the last part of the Kepler mission. After the modified program discovered stars that seemed to have planets, Dathillo and her colleagues used terrestrial telescopes to confirm the findings.
"If we want to know how many planets we have in common, we need to know how many planets we have found," Vanderburg said in a statement from the University of Texas at the University of Texas. , "But we also need to know how many planets we have missed, that's where [AI] comes."
Vanderburg and Google engineer Chris Shallu "first used AI to find a planet around a star by Kepler", according to
Jesse Christiansen, a researcher at the NASA Science Institute at the California Institute of Technology, says , that she was impressed by what Datilo's team had achieved. And in a sense she is not so surprised.
"NASA makes all data publicly available," says Christiansen. "You just have to think of a new idea of what to do with data that no one has ever done before."
Michel Nampampa of the Harvard-Smithsonian Center for Astrophysics expects to see many other astronomers who use artificial intelligence techniques for analysis in the future. This is because the newer telescopes do not collect as many stars and galaxies as digital data for these celestial objects.
"We're just going to see unprecedented volumes of data and we have to invent new ones
So writing a machine-learning program as a student is a good preparation for Dathillo as she gets a diploma in astronomy
are accepted for publication in The Astronomical Journal