ultra-short-period exoplanets

AI Helps Discover 5 Exoplanets With Record-Breaking Orbits

An international research team claims to have discovered five new USPs, or ultra-short-period exoplanets, by applying AI-based algorithms. The newly detected worlds are smaller than Earth and orbit their host stars in less than a day.

Findings were published in Monthly Notices of the Royal Astronomical Society. It was led by Ge Jian from the Shanghai Astronomical Observatory, which falls under the Chinese Academy of Sciences. To discover these celestial bodies, the team used stellar photometry data available from the Kepler telescope.

A Rare Find

Ultra-short-period exoplanets are planets having orbits with periods of less than one day. They are rare, maybe about 0.5% of all those around stars similar to the Sun. In the five years since their first discovery in 2011, only 145 have been found so far. Fewer than 30 of them have a diameter smaller than Earth’s.

The transit method is one of the most popular techniques used by astronomers to detect exoplanets. It relies on the observation of periodic lowering of the star’s brightness as a planet passes in front of the star in an orbit around it. However, the small size and relatively short orbital periods of ultra-short-period planets make their transit signals too faint to be detected.

AI To The Rescue

The research team developed a new algorithm. It combines GPU phase folding with convolutional neural networks to overcome the challenges they mentioned. Innovation was key in developing this approach. It resulted in a 15-fold increase in search speed and a 7% improvement in accuracy and completeness compared to conventional methods.

When applied to the Kepler dataset, it was able to detect all five ultra-short-period planets. That is a hallmark of success while this algorithm works in the detection of faint transit signals.

A Milestone In AI Astronomy

Team leader Ge Jian said that the research is an important step in the application of AI to astronomical big data. According to him, astronomers must develop innovative AI algorithms and generate large artificial datasets based on the physical characteristics of newly discovered phenomena for groundbreaking discoveries in astronomy.

Understanding Planetary Formation

The ultra-short-period exoplanets discovered in this study offer some good insight into early planetary system evolution, planet-planet interactions, and star-planet interaction dynamics. These are theoretically very important for understanding the consequences that theoretical studies of planetary formation have drawn for further interpretations.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top