First supernova detected and described entirely by bot


A new artificial intelligence tool called BTSbot completely removes humans from the supernova search and discovery process.

Dubbed the Bright Transient Survey Bot, or BTSbot for short, the machine learning algorithm has detected, identified, and classified its first supernova without human interference, Northwestern University announced.

A newly discovered supernova candidate was named SN2023tyk.

The fully automated process was developed by an international team of researchers led by Northwestern. The algorithm underpinning the program was trained on more than 1.4 million historial images from 16,000 astronomical sources.

supernova_1016
A before (left) and after image of the galaxy where SN2023tyk occurred. The upper left region of the galaxy (right) appears bulbous and misshapen, where the star exploded. Image by Northwestern University

“For the first time ever, a series of robots and AI algorithms has observed, then identified, then communicated with another telescope to finally confirm the discovery of a supernova,” Northwestern’s Adam Miller, who led the work, said in a press release.

The discovery will allow the robots to isolate specific subtypes of stellar explosions and represents an “important step forward,” he said.

Miller added: “Ultimately, removing humans from the loop provides more time for the research team to analyze their observations and develop new hypotheses to explain the origin of the cosmic explosions that we observe.”

Supernovas are stars that reach the end of their life cycle and explode, greatly increasing in brightness. The detection and analysis of these stellar explosions was, until now, only partly automated, with humans still involved in the process.

galaxy_1016
A deep-space image of the galaxy where the supernova occurred. Image by Legacy Surveys/Perimeter Institute/NASA/JPL-Caltech

Northwestern estimates that humans have spent approximately 2,200 hours over the past six years visually inspecting and classifying supernova candidates.

The new tool is expected to allow researchers to devote their time to other tasks, thus accelerating the pace of discovery. It officially went online last week.

“We achieved the world's first fully automatic detection, identification and classification of a supernova,” said Nabeel Rehemtulla, who co-led the development with Miller.

“This significantly streamlines large studies of supernovas, helping us better understand the life cycles of stars and the origin of elements supernovae create, like carbon, iron, and gold,” he said.

AI is already helping to accelerate space research and exploration, with NASA planning a mission to the Moon to map out the lunar surface using three autonomous robots, and using the technology in numerous other ways, including spacecraft modeling.


More from Cybernews:

IoT lacking the global standards it needs to progress

Netflix may now raise prices after success of password-sharing crackdown

X becomes first platform to be fined under Australia’s Online Safety Act

Israeli security data for sale at $15,000 – report

YouTube is cracking down on adblock users: pay or disable

Subscribe to our newsletter



Leave a Reply

Your email address will not be published. Required fields are markedmarked