4 min read<\/p>\n
NASA AI Model That Found 370 Exoplanets Now Digs Into TESS Data<\/h1>\n<\/div>\n<\/div>\n<\/div>\n\n\n\n![\"A](\"https:\/\/assets.science.nasa.gov\/dynamicimage\/assets\/science\/missions\/hubble\/artist-concepts\/hubbleart_67.jpg?w=3000&h=2400&fit=clip&crop=faces%2Cfocalpoint\")
<\/a><\/figure>\nThis artist\u2019s impression shows the star TRAPPIST-1 with two planets transiting across it. ExoMiner++, a recently updated open-source software package developed by NASA, uses artificial intelligence to help find new transiting exoplanets in data collected by NASA\u2019s missions.<\/div>\nNASA, ESA, and G. Bacon (STScI)<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\nScientists have discovered over 6,000 planets that orbit stars other than our Sun, known as exoplanets. More than half of these planets were discovered thanks to data from NASA\u2019s retired Kepler<\/a> mission and NASA\u2019s current TESS<\/a> (Transiting Exoplanet Survey Satellite) mission. However, the enormous treasure trove of data from these missions still contains many yet-to-be-discovered planets. All of the data from both missions is publicly available in NASA archives, and many teams around the world have used that data to find new planets using a number of techniques.<\/p>\nIn 2021, a team from NASA\u2019s Ames Research Center in California\u2019s Silicon Valley created ExoMiner, a piece of open-source software that used artificial intelligence (AI) to validate 370 new exoplanets from Kepler data. Now, the team has created a new version of the model trained on both Kepler and TESS data, called ExoMiner++.<\/p>\n
\n\n\n![\"Artist's](\"https:\/\/assets.science.nasa.gov\/dynamicimage\/assets\/science\/cds\/general\/images\/2024\/08\/tessinspacerender.jpg?w=2560&h=1440&fit=clip&crop=faces%2Cfocalpoint\")
<\/a><\/figure>\nArtist\u2019s impression of NASA\u2019s Transiting Exoplanet Survey Satellite (TESS), which launched in 2018 and has discovered nearly 700 exoplanets so far. NASA\u2019s ExoMiner++ software is working toward identifying more planets in TESS data using artificial intelligence.<\/div>\nNASA\u2019s Goddard Space Flight Center<\/div>\n<\/figcaption><\/div>\n<\/div>\n<\/div>\nThe new algorithm, which is discussed in a recent paper<\/a> published in the Astronomical Journal, identified 7,000 targets as exoplanet candidates from TESS on an initial run. An exoplanet candidate<\/a> is a signal that is likely to be a planet but requires follow-up observations from additional telescopes to confirm.<\/p>\nExoMiner++ can be freely downloaded from GitHub<\/a>, allowing any researcher to use the tool to hunt for planets in TESS\u2019s growing public data archive.<\/p>\n\u201cOpen-source software like ExoMiner accelerates scientific discovery,\u201d said Kevin Murphy, NASA\u2019s chief science data officer<\/a> at NASA Headquarters in Washington. \u201cWhen researchers freely share the tools they\u2019ve developed, it lets others replicate the results and dig deeper into the data, which is why open data and code are important pillars of gold-standard science.\u201d<\/p>\nExoMiner++ sifts through observations of possible transits<\/a> to predict which ones are caused by exoplanets and which ones are caused by other astronomical events, such as eclipsing binary stars. \u201cWhen you have hundreds of thousands of signals, like in this case, it\u2019s the ideal place to deploy these deep learning technologies,\u201d said Miguel Martinho, a KBR employee at NASA Ames who serves as the co-investigator for ExoMiner++.<\/p>\n\n\n
<\/a><\/figure>Scientists have discovered over 6,000 planets that orbit stars other than our Sun, known as exoplanets. More than half of these planets were discovered thanks to data from NASA\u2019s retired Kepler<\/a> mission and NASA\u2019s current TESS<\/a> (Transiting Exoplanet Survey Satellite) mission. However, the enormous treasure trove of data from these missions still contains many yet-to-be-discovered planets. All of the data from both missions is publicly available in NASA archives, and many teams around the world have used that data to find new planets using a number of techniques.<\/p>\n In 2021, a team from NASA\u2019s Ames Research Center in California\u2019s Silicon Valley created ExoMiner, a piece of open-source software that used artificial intelligence (AI) to validate 370 new exoplanets from Kepler data. Now, the team has created a new version of the model trained on both Kepler and TESS data, called ExoMiner++.<\/p>\n The new algorithm, which is discussed in a recent paper<\/a> published in the Astronomical Journal, identified 7,000 targets as exoplanet candidates from TESS on an initial run. An exoplanet candidate<\/a> is a signal that is likely to be a planet but requires follow-up observations from additional telescopes to confirm.<\/p>\n ExoMiner++ can be freely downloaded from GitHub<\/a>, allowing any researcher to use the tool to hunt for planets in TESS\u2019s growing public data archive.<\/p>\n \u201cOpen-source software like ExoMiner accelerates scientific discovery,\u201d said Kevin Murphy, NASA\u2019s chief science data officer<\/a> at NASA Headquarters in Washington. \u201cWhen researchers freely share the tools they\u2019ve developed, it lets others replicate the results and dig deeper into the data, which is why open data and code are important pillars of gold-standard science.\u201d<\/p>\n ExoMiner++ sifts through observations of possible transits<\/a> to predict which ones are caused by exoplanets and which ones are caused by other astronomical events, such as eclipsing binary stars. \u201cWhen you have hundreds of thousands of signals, like in this case, it\u2019s the ideal place to deploy these deep learning technologies,\u201d said Miguel Martinho, a KBR employee at NASA Ames who serves as the co-investigator for ExoMiner++.<\/p>\n<\/a><\/figure>