US Space Force Tackles Edge Computing in Space

US Space Force Addresses Edge Computing in Space
Troject aims to tackle in-orbit service, assembly, and manufacturing (OSAM) missions with use cases including satellite life extension, in-orbit refueling, active debris removal, and in-space manufacturing and assembly. use of recycled materials.

As part of a new effort to digitize some reconnaissance and maintenance processes, the USSF hopes to train its personnel (known as Guardians) in artificial intelligence and machine learning (AI/ML) practices.

“The USSF’s mission is to organize, train and equip Guardians to conduct global space operations that improve the way our joint forces and coalitions fight,” said Joel Mozer, director of science. , Spacewerk Technology and Research. “To do this effectively, we need to invest in AI and ML capabilities that can be deployed in the cloud and on-premises.”

New York startup Wallaroo was chosen for its AI/ML platform to help companies evaluate the performance of AI applications.

Using the Wallaroo platform, the USSF will be able to simulate AI and ML algorithms on edge computers in space, addressing the current challenge of the relative lack of computing power onboard spacecraft. compared to Earth’s capacity. “Edge machine learning is a compelling feature for companies to rethink their approach to MLOps, especially when it comes to the last mile of how they deploy, manage and manage observables in their production models,” wrote Vid Jain, CEO of Wallaroo Labs. business blog post.

“Our work with the USSF on use cases for their satellites is certainly an extreme example, but any connected device company that produces sensor data quickly, if not, needs to start rethinking how they take a model that is built on data. a scientist’s laptop and turn it into production-ready software that can run in a computing power-constrained environment.”

Wallaroo Labs expanded its market last year, closing a $25 million funding round in February, led by Microsoft.