Researchers discover 27,500 asteroids in old telescope images using artificial intelligence

A team of researchers recently announced a remarkable discovery. Using artificial intelligence, they identified 27,500 new possible asteroids in old telescope images, according to Smithsonian Magazine. The images, archived and processed by the Tracklet-less Heliocentric Orbit Recovery (THOR) algorithm, were captured by the US National Optical-Infrared Astronomy Research Laboratory (NOIRLab), covering over 400,000 snapshots.

Astronomers usually discover asteroids by repeatedly observing sections of the sky through telescopic images, in which they appear as moving points of light. But THOR analyzed the archived images, connecting points of light from different telescopes from different days, determining whether they represent the same cosmic object, often an asteroid.

This revolutionary technique has made it possible to discover an impressive number of new possible asteroids, including about 100 near-Earth asteroids that pass close to our planet’s orbit. It should be noted that all other observatories in the world have identified approximately 2,300 asteroids this year, compared to the 27,500 discovered by the THOR algorithm.

This technology not only expands our knowledge of objects in outer space, but can also help identify potentially dangerous asteroids so they can be more closely monitored by our planetary defense system.