How AI Fixed the James Webb Space Telescope's Blurry Vision
The James Webb Space Telescope, a marvel of modern astronomy, has been facing a challenge since its launch: blurry vision. This issue, reminiscent of a flaw in the Hubble Space Telescope's past, threatened to hinder its scientific capabilities. But a team of Australian researchers has developed an innovative solution using artificial intelligence (AI).
The affected instrument, the Aperture Masking Interferometer (API), is designed to enhance the sensitivity and resolution of the Near-InfraRed Imager and Slitless Spectrograph (NIRISS) on the James Webb Space Telescope (JWST). API's unique design, featuring an opaque mask with seven holes, aims to detect small and dim exoplanets around distant stars. However, when astronomers activated the instrument, they encountered a problem: the images were blurry.
This issue traced back to electronic distortions in the infrared camera detector, a challenge similar to the one faced by Hubble in 1990. The Hubble Space Telescope, after reaching orbit, was found to be near-sighted due to imperfections in its primary mirror. The solution for Hubble required a crewed space mission and cost hundreds of millions of dollars. In contrast, the James Webb Space Telescope, orbiting at a much greater distance, made such a mission impractical.
To address the blurriness in API images, former University of Sydney Ph.D. students Max Charles and Louis Desdoigts developed a neural network, an AI algorithm inspired by the human brain. This algorithm, named AMIGO (Aperture Masking Interferometry Generative Observations), detects and corrects pixels affected by electrical charges that distort observations.
The effectiveness of AMIGO was demonstrated through its ability to sharpen images of a dim exoplanet and a low-mass star, located 133 light-years from Earth. In another campaign, API, assisted by AMIGO, produced detailed images of a black hole jet, the volcanic surface of Jupiter's moon Io, and stellar winds from a distant variable star.
"Instead of sending astronauts to bolt on new parts, they managed to fix things with code," said Professor Peter Tuthill, highlighting the innovative approach. Desdoigts, now a postdoctoral researcher at Leiden University, expressed the team's satisfaction, stating, "This work brings JWST's vision into even sharper focus. It's incredibly rewarding to see a software solution extend the telescope's scientific reach."
Since its operational launch in July 2022, the James Webb Space Telescope has revolutionized astronomy, revealing unexpected details about early galaxy formation and black holes. It has also significantly contributed to exoplanet research, making unprecedented measurements of their atmospheric composition. With API now functioning optimally, the telescope is poised for even more groundbreaking discoveries, solidifying its position as a cornerstone of modern astronomy.
