AI identifies 50 new planets from old NASA data

British researchers have identified 50 new planets using artificial intelligence, marking technological advances in astronomy. 

Astronomers and computer scientists at the University of Warwick have developed a machine learning algorithm to dig up old NASA data containing thousands of asteroid candidates.

However, it is not clear who these candidates are. When scientists discover exoplanets (planets outside our solar system), they enter light, which refers to the telescope and the planet passing through their star. This dip can also be caused by other factors such as background interference or camera errors.

The research team trained the algorithm using data collected by NASA’s now retired Kepler Space Telescope, which spent nine years in deep space for a global hunting mission. After learning the algorithm to distinguish real planets from false positives, it was used to analyze old data sets that had not yet been verified – it was identified as 50 Explanets.

According to a university news release, these 50 exoplanets will put other stars in orbit, smaller than Earth from Neptune. Some of their revolutions last up to 200 days, while others last up to a day. Now that astronomers know the planets are real, they can prioritize them for further observation.

Now that researchers know it works, they hope to use AI for current and future telescope missions. It provides a consistent and effective verification method as issued; Once properly trained, AI is faster than current technologies and can be automated to perform automatically.

The study suggested that “one can verify thousands of invisible candidates in seconds”. And since it is based on machine learning, it can be further improved and become more effective with each new invention.

The research team argues that astronomers should use a number of validation methods, including this new algorithm, to confirm future exoplanet discoveries. Currently, 30% of all known planets have been verified using the same method, Armstrong says.

“We still have to spend time for algorithm training, but if that happens it will be much easier to implement for future candidates,” he said.

Armstrong said the algorithm could be used with NASA’s Transiting Exoplanet Survey Satellite (TES), an All-Sky survey mission that completed its primary mission on July 4.

By mapping about 75% of the sky, TESS has identified 66 new athletic exoplanets and 2,100 potential candidates. One of the verified exoplanets is Earth-sized habitable, placing a star in orbit 100 light-years away. It also found a planet orbiting two sun-like stars in Star Wars images.

TESS is now on a mission extended to 2022, but scientists are working to verify and verify that any potential candidates are real planets.

Leave a Reply

Your email address will not be published. Required fields are marked *