NASA’s Kepler Space Telescope has been doing a pretty darn good job at discovering exoplanets, even in its somewhat hindered state.
With over 2,500 confirmed discoveries over its 8 year lifespan, NASA has announced an upcoming press conference, this time highlighting more discoveries made with artificial intelligence help from Google.
Mining through tons of data can be quite challenging, but with the advances in AI learning, it’s slowly becoming easier to quickly sift through mountains of content.
“The discovery was made by researchers using machine learning from Google. Machine learning is an approach to artificial intelligence, and demonstrates new ways of analyzing Kepler data.”
The press teleconference will be held on December 14th at 1PM EST, and will feature four major speakers including:
- Paul Hertz, Astrophysics Division director at NASA Headquarters in Washington
- Christopher Shallue, senior software engineer at Google AI in Mountain View, California
- Andrew Vanderburg, astronomer and NASA Sagan Postdoctoral Fellow at The University of Texas, Austin
- Jessie Dotson, Kepler project scientist at NASA’s Ames Research Center in California’s Silicon Valley
Visually, it’s impossible to see planets many light years away, but by observing light fluctuations around distant stars, it’s then pretty easy to determine if there are planets orbiting those stars.
The Kepler Space Telescope works by focusing on a small section of the Milky Way (in the direction of the Lyra and Cygnus constellation). The telescope would observe light fluctuations from stars within those constellations. Software would then be using to determine if those light fluctuations are regular or not, indicating the existence of a potential planet.
When a planet passes in front of a star as viewed from Earth, the event is called a “transit”. On Earth, we can observe an occasional Venus or Mercury transit. These events are seen as a small black dot creeping across the Sun—Venus or Mercury blocks sunlight as the planet moves between the Sun and us. Kepler finds planets by looking for tiny dips in the brightness of a star when a planet crosses in front of it—we say the planet transits the star.
Once detected, the planet’s orbital size can be calculated from the period (how long it takes the planet to orbit once around the star) and the mass of the star using Kepler’s Third Law of planetary motion. The size of the planet is found from the depth of the transit (how much the brightness of the star drops) and the size of the star. From the orbital size and the temperature of the star, the planet’s characteristic temperature can be calculated. From this the question of whether or not the planet is habitable (not necessarily inhabited) can be answered.
Even though Kepler is still on the hunt for new exoplanets, most of its discoveries are still popping up from snapshots taken between 2009 to 2013. Even with 2,500+ confirmed exoplanets, there are at least another 2,500+ designated as ‘candidate’ exoplanets.
However, further data analysis needs to be confirmed before they can be classified as ‘confirmed’.