NASA’s Kepler space telescope searches for planets beyond our solar system (exoplanets). They do this by measuring how the brightness of a star changes when a planet crosses in front of it. Look at the graph below as an example.
A NASA scientist and a Google Artificial Intelligence researcher worked together to train computers to more closely look at the data from the Kepler Telescope. The alogrithm has been trained to search for dips in their deep space data and have identified a new exoplanet.
They have named it Kepler-90i. This is part of an eight-planet system that orbits a sun-like star Called Kepler-90, 2500 light years away. Our Solar System is now tied with this system to have the largest number of planets revolving around another star.
What’s a light year? A unit of measurement of distance in space. Specifically, it is the distance that light travels in one year. This approximates 10 trillion km!
What’s hot about Kepler-90i? This planet is sizzling hot at a temperature of 800 degrees Fahrenheit. It orbits so close to its star that a year on Kepler-90i lasts only 14 days!
What’s cool about how this was discovered? Google AI, Google’s Artificial Intelligence group took 4 years worth of data from NASA’s Kepler Telescope. This data had already been analysed before but they gave it to their computing systems that use artificial intelligence. They used an application called machine learning, where the machines learn to analyse data differently along the way, as they learn from the inputs and decisions that have been made. Kind of like our human brains do. Without specifically being programmed to perform a particular task, but rather as a logical next step while learning something.
They trained the machines to learn how to identify and analyse faint signals in Kepler’s vast database of space data. Signals that may have been overlooked before. And this is how they identified this exoplanet! This technique is also now being used to search for more planets orbiting around 150,000 known stars.