Asteroid Institute, a program of B612 Foundation, and Google Cloud today announced the key results of their collaboration to date: identifying 27,500 new high-confidence asteroid candidates. The work, which took place over several weeks, has the potential to map the solar system and protect Earth from collisions, aiding the discovery of small planets.
The project was carried out without new observations of the sky, but by using Google Cloud technology to run advanced algorithms developed by researchers at the Asteroid Institute and the University of Washington, and by collecting historical data sets from the NOIRLab Source Catalog Data Release 2 (NSC DR2). The majority of the new discoveries are “Main Belt” asteroids that orbit the sun between Mars and Jupiter, but the Asteroid Institute has also discovered more than 100 Near-Earth Asteroids that move much closer to Earth due to their orbit.
In collaboration with the University of Washington’s DiRAC Institute, the Asteroid Institute has developed a new algorithm called Tracklet-less Heliocentric Orbit Recovery (THOR), which runs on a cloud-based, open-source astrodynamics platform called Asteroid Discovery Analysis and Mapping (ADAM). THOR projects theoretical orbits over millions of observed moving points of light and links together those points that are consistent with real physical orbits. Google Cloud’s Office of the CTO worked with the Asteroid Institute to help it scale and tune its algorithms on ADAM using Google Cloud.
The Asteroid Institute selected Google Cloud as its cloud provider for its scalability, ease of use, and advanced data and AI products, specifically focused on: Asteroid Institute uses Google Kubernetes Engine (GKE), a managed Kubernetes service to deploy containerized applications at scale and use it to run massive computational workloads of millions of vCPU hours to discover asteroids and predict their movements in the solar system. 5.4 billion observations from various astronomical surveys, representing potential asteroids and other objects, are stored and analyzed in BigQuery, Google’s unified AI-ready data platform. Thanks to its serverless nature, the Asteroid Institute’s system can easily handle the influx of data that this complex process generates.
The Asteroid Institute uses Google Cloud Storage for millions of images from NOIRLab for use in verifying asteroid discoveries, as well as durable storage for billions of data points across the asteroid discovery pipeline. “What’s exciting is that we are using electrons in data centers, in addition to the usual photons in telescopes, to make astronomical discoveries,” says Dr. Ed Lu, executive director of the Asteroid Institute.
The Asteroid Institute is also exploring the use of Google’s AI technologies to automatically examine and verify candidate images identified by the THOR algorithm. Automating the AI is a crucial step to further scale up this work, as the initial verification of likely candidates is a major bottleneck – currently done manually by volunteer high school students, college students, scientists, and astronomers. If successful, the need for human verification could be significantly reduced and the pipelines built for NOIRLab could be adapted to run on much larger datasets, such as those from the Vera C. Rubin Observatory, further fueling new discoveries .
“At Google, we always enjoy tough computational challenges, and the Asteroid Institute provided us with complex unstructured data that required heavy arithmetic, large tracking requirements, and new AI capabilities,” said Massimo Mascaro, technical director of Google Cloud’s Office of the CTO. “We are proud to partner with the Asteroid Institute to advance scientific discovery and raise awareness about our solar system’s beautiful neighbors.” The NSC DR2 catalog is the first of many astronomical data sets that the Asteroid Institute plans to scan. The largest will likely become available in 2025, after the inauguration of the Vera C. Rubin Observatory. With THOR running on ADAM on Google Cloud – and with the help of AI – researchers can scale this work and scan new datasets more efficiently and effectively as they become available. The Asteroid Institute’s goal is to automate this process for the benefit of the astronomical community and the space industry.
“The results from the Asteroid Institute are more than exciting for the Vera C. Rubin Observatory: they can help us optimize our observation strategy and achieve gains for some scientific programs, such as cosmologically important supernovae explosions, which is equivalent to cloning another Rubin observatory,” says Dr. Zeljko Ivezic, Rubin Observatory Construction Director.
The Asteroid Institute and Google Cloud have been working together since 2017 with the goal of transforming asteroid discovery. Read more about the technology that made this discovery possible here.
About B612 Foundation and Asteroid Institute
The Asteroid Institute brings together scientists, researchers and engineers to develop tools and technologies that help us understand, map and navigate our solar system. The Asteroid Institute is a program of the B612 Foundation and uses advances in computer science, instrumentation and astronomy to find and track asteroids. Since 2002, the foundation has supported research and technologies to enable the economic development of space and advance our understanding of the evolution of our solar system. In addition, the foundation supports educational programs, including Asteroid Day and the new Schweickart Prize. Founding Circle and Asteroid Circle members and individual donors from 46 countries provide financial support for the work. For more information, visit B612foundation.org or follow us on social media: Twitter, Facebook, YouTube and LinkedIn.
Source: B612
