SA scientist uses AI to capture rare astronomical objects
South African scientist, Dr Michelle Lochner has joined other scientists from across the globe to develop fresh techniques to read astronomical data, one such being machine learning.
CAPE TOWN - A South African scientist is part of a group of international researchers who have roped in machine learning to capture an image of a rare astronomical object.
The University of the Western Cape’s, Dr Michelle Lochner explained that with more telescopes such as the meerKAT radio telescope coming online, astronomers have access to vast volumes of new data.
In order for them to analyse the data sets, the researchers are increasingly developing and finetuning automated techniques.
Lochner, who also holds a joint staff position at the South African Radio Astronomy Observatory, said in her research she works on using machine learning, which is a branch of artificial intelligence.
“To discover rare and unusual objects in big data sets, so with this technique that I’ve developed with a colleague of mine, Bruce Bassett, I was able to discover this very strange-looking radio galaxy in a MeerKAT data set.”
Lochner said the image did not resemble anything they had seen before and is something quite unusual.
“We named it Sauron, short for a steep and uneven ring of nonthermal radiation, it seems to bear some similarity to other types of radio galaxies that our Australian colleagues have found, called orcs, hence the name and these are odd radio circles.”