Home / Others / Artificial intelligence, NASA data used to discover the eighth planet that surrounds a distant star – Astronomy Now

Artificial intelligence, NASA data used to discover the eighth planet that surrounds a distant star – Astronomy Now



With the discovery of an eighth planet, the Kepler-90 system is the first to join with our solar system in number of planets.
Credits: NASA / Wendy Stenzel

Our solar system is now linked to the most number of planets around a single star, with the recent discovery of an eighth planet surrounding Kepler-90, a star similar to the Sun 2,545 light years from Earth. The planet was discovered in data from NASA's Kepler Space Telescope.

The newly discovered Kepler-90i, a hot, rocky planet that orbits its star once every 14.4 days, was found using Google machine learning. Machine learning is an approach to artificial intelligence in which computers "learn". In this case, computers learned to identify planets by finding in Kepler data cases where the telescope recorded signals from planets beyond our solar system, known as exoplanets.

As we expected, there are exciting discoveries lurking in our archived Kepler data, waiting for the right tool or technology to dig them out, "said Paul Hertz, director of NASA's Astrophysics Division in Washington." This finding shows that our data will be a treasure available to innovative researchers in the coming years. "

The discovery came after researchers Christopher Shallue and Andrew Vanderburg trained a computer to learn how to identify exoplanets in light readings recorded by Kepler: tiny change of brightness that occurs when a planet passes in front of a star or travels through it Inspired by the way in which neurons connect in the human brain, this artificial "neural network" filtered through the data of Kepler and he found weak traffic signals from an eighth previously lost planet that orbits around Kepler-90, in the constellation of Draco.

previously used in searches of the Kepler database, this research demonstrates that neural networks are a promising tool to find some of the weakest signals from distant worlds.

Other planetary systems are probably more promising for life than Kepler-90. About 30 percent larger than Earth, Kepler-90i is so close to its star that its average surface temperature is believed to exceed 800 degrees Fahrenheit, on a par with Mercury. Its outermost planet, Kepler-90h, orbits at a distance similar to its star that the Earth to the Sun.

"The Kepler-90 star system is like a mini version of our solar system. planets outside, but everything is wrinkled much closer, "said Vanderburg, a postdoctoral fellow at NASA and an astronomer at the University of Texas at Austin.

Shallue, a senior software engineer with the Google Google AI research team, came up with the idea of ​​applying a neural network to Kepler's data. He became interested in the discovery of exoplanets after learning that astronomy, like other branches of science, is rapidly flooding with data as technology advances for data collection from space.

"In my free time, I started looking for finding exoplanets with large data sets and discovered the Kepler mission and the huge data set available," said Shallue. "Machine learning really shines in situations where there is too much data that humans can not search for themselves."

Kepler's four-year data set consists of 35,000 possible planetary signals. Automated tests, and sometimes human eyes, are used to verify the most promising signals in the data. However, weaker signals are often lost when using these methods. Shallue and Vanderburg thought there might be discoveries of more interesting exoplanets lurking in the data.

First, they trained the neural network to identify exoplanets in transit using a set of 15,000 previously examined signals from the Kepler exoplanet catalog. In the test set, the neural network correctly identified true and false positive planets 96 percent of the time. Then, with the neural network that "learned" to detect the pattern of an exoplanet in transit, the researchers directed their model to look for weaker signals in systems of 670 stars that already had multiple planets known. They assumed that multiple planetary systems would be the best places to search for more exoplanets.

"We got many false positives from planets, but also potentially more real planets," said Vanderburg. "It's like looking at the rocks to find jewelry, if you have a finer sieve, you can catch more rocks, but you could also catch more jewels."

Kepler-90i was not the only jewel that this neural network hovered. In the Kepler-80 system, they found a sixth planet. This, the Kepler-80g the size of the Earth, and four of its neighboring planets form what is called a resonant chain, where the planets are blocked by their mutual gravity in a rhythmic orbital dance. The result is an extremely stable system, similar to the seven planets in the TRAPPIST-1 system.

Their research work informing these findings has been accepted for publication in The Astronomical Journal [https://aj.aas.org]. Shallue and Vanderburg plan to apply their neural network to Kepler's complete set of more than 150,000 stars.

Kepler has produced an unprecedented data set for exoplanet hunting. After observing a patch of space for four years, the spacecraft is now operating on an extended mission and changes its field of vision every 80 days.

"These results demonstrate the enduring value of Kepler's mission," said Jessie Dotson of Kepler. scientific project at the Ames Research Center of NASA in Silicon Valley of California. "New ways of looking at data, like this early-stage research to apply machine learning algorithms, promise to continue to produce significant advances in our understanding of planetary systems around other stars." I'm sure there are more news in the data waiting for people find them. "


Source link

Leave a Reply