The University of the North-West has developed a new artificial intelligence tool called BTSBOT that eliminates the need for human participation in the process of searching and detecting supernovae. Using a machine learning algorithm, the Bright Transient Survey Bot (BTSBOT) successfully found, identified, and classified its first supernova, SN2023TYK, without any human intervention.
An international group of researchers, led by the University of North-West, developed this fully automated process using more than 1.4 million historical images of 16,000 astronomical sources. Adam Miller, from the University of North-West and project leader, highlighted the significance of this achievement, stating, “For the first time, robots and algorithms observed, determined, and confirmed the detection of a supernova.” This breakthrough allows robots to specialize in specific subtypes of star explosions and provides more time for analyzing observations and developing new hypotheses.
Supernovae are stars that have reached the end of their life cycle and explode, significantly increasing their brightness. Previously, the process of detecting and analyzing these star explosions was only partially automated. Over the past six years, people have spent approximately 2,200 hours on visually verifying and classifying candidates for supernova. However, with the introduction of this new tool, researchers can now allocate more time to other tasks, accelerating the pace of discoveries.
Nabeeel Rehemtulla, co-leader of the development team alongside Miller, expressed pride in achieving the world’s first fully automatic detection, identification, and classification of supernovae. This advancement in artificial intelligence technology marks a significant milestone in the field of astronomy.