Google Cloud accelerates asteroid discovery for US-based non-profit

Google Cloud has revealed details of how its work with a planetary science-focused non-profit has led to the identification of 27,500 new asteroids, accelerating the pace at which the solar system can be mapped.

The public cloud giant has been working with the Asteroid Institute – a programme of work run by the B612 Foundation planetary defence and science initiative – since 2017 to advance its work in mapping the solar system.

The pair recently undertook a project, spanning several weeks, that involved using Google Cloud infrastructure to run algorithms developed in-house by the Asteroid Institute and the University of Washington and combining them with historical datasets.

The work resulted in the discovery of 27,500 asteroids, including 100 that were categorised as being near Earth that pose a risk of colliding with the surface of our planet, without any new observations of the sky being made.

“In partnership with the University of Washington’s DiRAC Institute, the Asteroid Institute developed a novel 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),” the pair said in a joint statement.

“Thor projects theoretical orbits across millions of observed moving points of light and links together those points that are consistent with real physical orbits.”

Google Cloud’s involvement saw the company collaborate with the Asteroid Institute to help scale and fine-tune its algorithms by running Adam on its cloud-based servers.

In terms of the specific Google Cloud technologies the Asteroid Institute uses, they include the Google Kubernetes Engine (GKE) managed service that allows it to carry out large-scale computational workloads needed to track and predict the movements of asteroids in our solar system.

It is also using the Google BigQuery enterprise data warehouse to process the data from 5.4 billion observations from different astronomical surveys, and Google Cloud Storage to retain the millions of images it uses to verify asteroid discoveries.

The institute is also exploring how to use Google’s artificial intelligence (AI) technologies to vet and verify images identified by the Thor algorithm, a process it describes as a “major bottleneck” because it currently relies on volunteers.

“At Google, we always like hard computational challenges, and Asteroid Institute provided us with complex unstructured data that required heavy computational processing, large tracking requirements and novel AI capabilities,” said Massimo Mascaro, technical director of the Google Cloud’s Office of the CTO.

“We’re proud to partner with Asteroid Institute to help further scientific discovery and expand our world’s awareness of the beautiful neighbours we have in our solar system.”