A Sophisticated AI Bot Is Helping Astronomers Search For Extraterrestrial Life
Declaring that there are a lot of stars in our universe is somewhat of an understatement — there are estimated to be about 200 billion stars in the Milky Way galaxy alone.
That makes searching for stars and adjacent planets with distinct specifications an extraordinarily challenging and time-consuming task.
Enter the astronomy bot, a machine learning algorithm designed by astrophysicist Natalie Hinkel.
The bot is helping researchers to identify stars hosting planets similar to Jupiter and Saturn, as this may lead us to clues on the possible location of any existing extraterrestrial life.
Searching for Jupiter and Saturn's twins
The reason astronomers are looking for similar planets to Saturn and Jupiter is that these planets, in our solar system, have protected Earth from asteroids and space debris.
The large size of the two planets means that, over billions of years, large asteroids hurtling into our solar system would have been attracted to their strong gravitational pulls. As such, Saturn and Jupiter can be seen as the protectors of our solar system.
The need for a time-saving algorithm
The problem is that far away planets are hard to view in space, and can often only be spotted by viewing the shadow they cast as they orbit around their stars. On top of this, there are a huge amount of stars to search.
"Searching for planets can be a long and tedious process given the sheer volume of stars we could search," said Stephen Kane, UCR associate professor of planetary astrophysics.
"Eliminating stars unlikely to have planets and pre-selecting those that might will save a ton of time," he said.
The astronomy bot
The astronomy bot will, indeed, save a ton of time. The machine learning algorithm created by Natalie Hinkel, a researcher at the Southwest Research Institute, is already showing results.
On examination of the data provided by the AI bot, Kane and a team of astronomers have already discovered three stars showing strong evidence of harboring giant, Jupiter-like planets about 100 light years away.
Hinkel, Kane and the team were able to train the algorithm to use the findings of spectroscopy — the measuring of light from distant stars to discover their chemical composition — to look for stars that are most likely to harbor planets.
"We found that the most influential elements in predicting planet-hosting stars are carbon, oxygen, iron and sodium," Hinkel told Science Daily.
A paper detailing the team's findings was published this week in the Astrophysical Journal.
Ryan Harne and his team created a material that can "think".