In the process of identifying 27,500 newly discovered solar system bodies, scientists have developed a new method for identifying asteroids that have the potential to completely destroy cities on Earth.
The results were released on Tuesday by a collaboration between the Asteroid Institute, Google Cloud technologies, and the University of Washington in the US.
As an alternative to utilizing a telescope to search for asteroids in real time, the technology searches through historical photos of the night sky using an algorithm. Patterns that are invisible to humans or computers can be found by the algorithm.
According to scientists, the work could advance the field of minor planet discovery by enabling solar system mapping and shielding Earth from asteroid collisions.
The Asteroid Institute identified over 100 Near-Earth Asteroids, whose orbits bring them much closer to Earth, even though the bulk of the discoveries were asteroids that orbit the Sun in the region between Mars and Jupiter.
Dr. Ed Lu, Executive Director of the Asteroid Institute, said, “What is exciting is that we are using electrons in data centers, in addition to the usual photons in telescopes, to make astronomical discoveries.”
AI decreasing the need for human labor
The project’s findings—which are not predicated on fresh observations of the night sky—showcase the potential applications of historical data and the potential significance of artificial intelligence (AI) in the future.
According to the research, since verification is now done manually by volunteers from high school, undergraduate and graduate students, scientists, and astronomers, AI automation will be a “crucial step” to scale the job even further.
Because AI automation can process considerably larger datasets than human labor can, it may also result in a reduction in the quantity of labor needed for human discovery.
Massimo Mascaro, Technical Director of Google Cloud’s Office of the CTO, stated, “At Google, we always like hard computational challenges, and Asteroid Institute provided us with complex unstructured data that required heavy computational processing, large tracking requirements, and novel AI capabilities.”
“We are honored to collaborate with the Asteroid Institute in order to promote scientific research and raise global awareness of our lovely solar system neighbors.”
