The planet was discovered using a machine learning system from Google, which then worked to sort through data from NASA's Kepler spacecraft. Kepler, an extraterrestrial telescope that crossed the Earth in orbit around the sun, has observed 145 thousand stars like the sun for years, looking for signs of distant planets.
Astronomers know about the Kepler-90 solar system, but have not previously detected the planet. It looks like the third planet of a star like the sun, and orbits roughly every 14 days. The temperature on Kepler-90i surface is probably about 1,800 degrees Fahrenheit (980 degrees Celsius). "Kepler-90i is not a place I'd like to visit," said Andrew Vanderburg, an astronomer at the University of Texas at Austin who helped find the planet, told a news conference, quoted by Sciencealert, Friday (15/12).
Vanderburg is assisted by Google's AI software engineer Christopher Shallue. He saw, what they have developed is a tool to help astronomers have a greater impact. Vanderburg and Shallue also discovered a new second planet, called Kepler 80g.
To detect these two new worlds, Google's learning engine learns how to identify signals from exoplanets recorded in Kepler data. The tool processes 14 billion data points from Kepler's imagery for four years, using what is known as a convolution neural network, similar to how the human brain processes information.
NASA and Google say the new technology will help scientists discover more such exoplanets in the future. In fact, Vanderburg believes the Kepler-90 solar system is likely to have more undetected planets.
"It's almost a surprise to me if there is no more.Is the planet solar system like our own eight really incredible? There may be systems out there with so many planets, making our solar system look ordinary," he explained.
Prior to this analysis, a recent examination of Kepler's data confirmed 219 new worlds in more than 4 thousand candidates that Kepler had appeared. The total number of existing exoplanets in space today is 2,525 - and 10 of them may be rocky, Earth-sized, and may be inhabited by alien life.
For those who are wondering if Google's AI system could make astronomers obsolete, NASA says it's nothing to worry about. Jessie Dotson, a Kepler project scientist at NASA's Ames Research Center in Silicon Valley, explains that astronomers will always be required to classify objects before giving information to artificial neural networks so AI can learn to see new data.
'' This will really work with astronomers. You will never take that part, '' he explained.
