AI trained by asteroid hunting citizen scientists is helped in identifying more than 1,000 never-before-seen solar objects from Hubble images of 20 years.
After many years of secrecy about Cosmos, more than 1,000 space rocks have been discovered in the solar system that have never been seen before. A combination of artificial intelligence and zitizen scientists helped discover asteroids hidden in Hubble Space Telescope images, according to a new study.
According to NASA, scientists have discovered more than 1.3 million rock bodies, mostly in the Asteroid belt between Mars and Jupiter. There may be hundreds or thousands or even millions of asteroids waiting to be discovered. But these rock bodies are the smallest and therefore faintest objects in the solar system, making them difficult to detect.
In a new study, published on March 15 in the journal Astronomy & Astrophysics, scientists reported that 1,031 uncategorized asteroids were found in the Hubble archive. They were detected by artificial intelligence (AI) trained by thousands of citizen scientists to detect the faint light streak by tiny rocks.
Although these asteroids were discovered by chance, their trajectory showed that most of them are located in the asteroid belt, which makes them very useful to research.
The asteroid streaks in Hubble images are the result of space telescopes orbiting Earth and taking long-exposure images of distant galaxies. Asteroids are often missed in such images because the space rocks are millions of times fainter than the faintest stars in the night. But the streaks make them more visible and allow scientists to study their size and orbital properties.
As of 2019, more than 11,000 citizen scientists are combing the images, looking for these streaks. This project, called Hubble Asteroid Hunter (HAH), provides a great service to scientists who need to examine the images themselves.
In the new study, researchers gave HAH members some Hubble images and then used the results as a training set for AI to help it learn how to detect these hidden Space Rocks. The team used artificial intelligence to sift through 37,000 Hubble images taken over a 19-year period to find the new Asteroids. In total, the AI identified 1,701 candidates, 1,031 of which had never been seen before.
Scientists have been amazed by the AI’s ability to identify stars, and now hoping to use a similar method to analyze different types of data to uncover other hidden gems in the astronomical treasure trove.