Google Algorithm Establishes Existence of More Solar Systems Like Ours

Google Algorithm Establishes Existence of More Solar Systems Like Ours

Google might be best known for bringing us the latest technology in the world but recently the tech giant has also been applying itself to other worlds. Nasa’s Kepler space telescope, a central part of the space agency’s planet-hunting mission has been beaming back images of far flung regions of space to Earth from the space observatory on which it is located. The telescope’s most recent discovery is of planetary system around a star that has as many planets as our own solar system.

Until very recently, the Kepler-90 star system, which is 2200 light years away from Earth, was believed to have 7 planets in orbit around its star. However, Google algorithm technology based on machine learning being employed by Nasa to analyse the data send back by Kepler has led to an eighth planet being confirmed. That may not sound particularly significant, with new planets relatively regularly being uncovered. However, it is the first time a star system with at least as many planets as our own has ever been confirmed. While it has always been statistically likely, we now know for certain that the number of planets in our Solar System is not unique.

Like our Solar System, the Kepler-90 star system features small planets on the inside, closer to the sun, and larger planets on the outside. However, Kepler-90 is significantly smaller than our Solar System, with smaller distances between the planets. It may well even turn out that Kepler-90 not only equals our own known record for planets in a solar system but surpasses it.

Commenting on the discovery, Andrew Vanderburg, an astronomer and NASA Sagam Postdoctoral Fellow at Austin’s University of Texas expanded:

“There is a lot of unexplored real estate in Kepler-90 system and it would almost be surprising if there were not more planets in the system.”

So how did the Google technology pinpoint the existence of an eighth planet? A neural network scan was applied by a Google team. This scanned huge amounts of data to isolate weak signals missed by humans. Dips in the brightness of stars suggests an orbiting planet may be passing in front. Google’s algorithm has demonstrated 96% accuracy in identifying planets and discounting false positives.

Google’s Christopher Shallue, a senior software engineer at Google AI (Artificial Intelligence) commented:

“This is really exciting discovery and a successful proof of concept in using neural networks to find planets even in challenging situations where signals are very weak.

“We plan to search all 150,000 stars, we hope using our technique we will be able to find lots of planets including planets like Earth.”

Google have also committed to making the software used to identify the eighth Kepler-90 star freely available so amateur astrologers can also hunt for exoplanets.

Leave a Comment

seven + 16 =