NASA SpaceCube Next-Generation Artificial-Intelligence Computing for STP-H9-SCENIC on ISS

Presenter Information

Alessandro Geist, NASA Goddard Space Flight Center
Gary Crum, NASA Goddard Space Flight Center
Cody Brewer, NASA Goddard Space Flight Center
Dennis Afanasev, NASA Goddard Space Flight Center
Sebastian Sabogal, NASA Goddard Space Flight Center
David Wilson, NASA Goddard Space Flight Center
Justin Goodwill, NASA Goddard Space Flight Center
James Marshall, NASA Goddard Space Flight Center
Noah Perryman, NASA Goddard Space Flight Center
Nick Franconi, NASA Goddard Space Flight Center
Timothy Chase Jr., NASA Goddard Space Flight Center
Travis Wise, NASA Goddard Space Flight Center
Sarah Flanagan, NASA Goddard Space Flight Center
Michael Monaghan, NASA Goddard Space Flight Center
Jeff Hosler, NASA Goddard Space Flight Center
Leyland Young, NASA Goddard Space Flight Center
Judy Kuntz, NASA Goddard Space Flight Center
Lindsey Seo, NASA Goddard Space Flight Center
Joseph Schreck, NASA Goddard Space Flight Center
Tracy Price, NASA Goddard Space Flight Center
Munther Hassouneh, NASA Goddard Space Flight Center
Steven Kenyon, NASA Goddard Space Flight Center
Samuel Price, NASA Goddard Space Flight Center
Christopher Wilson, NASA Goddard Space Flight CenterFollow
Nicole Thai, The Aerospace CorporationFollow
Madison Hobbs, The Aerospace Corporation
Andrew Keene, The Aerospace Corporation
Sammy Lin, The Aerospace Corporation
Michael Nemerouf, The Aerospace Corporation
Evan T. Kain, U.S. Space Force
Francisco O. Viramontes, U.S. Space Force
Tyler M. Lovelly, U.S. Space ForceFollow

Session

Technical Poster Session 1

Location

Utah State University, Logan, UT

Abstract

Recently, Artificial Intelligence (AI) and Machine Learning (ML) capabilities have seen an exponential increase in interest from academia and industry that can be a disruptive, transformative development for future missions. Specifically, AI/ML concepts for edge computing can be integrated into future missions for autonomous operation, constellation missions, and onboard data analysis. However, using commercial AI software frameworks onboard spacecraft is challenging because traditional radiation-hardened processors and common spacecraft processors cannot provide the necessary onboard processing capability to effectively deploy complex AI models. Advantageously, embedded AI microchips being developed for the mobile market demonstrate remarkable capability and follow similar size, weight, and power constraints that could be imposed on a space-based system. Unfortunately, many of these devices have not been qualified for use in space. Therefore, Space Test Program - Houston 9 - SpaceCube Edge-Node Intelligent Collaboration (STP-H9-SCENIC) will demonstrate inflight, cutting-edge AI applications on multiple space-based devices for next-generation onboard intelligence. SCENIC will characterize several embedded AI devices in a relevant space environment and will provide NASA and DoD with flight heritage data and lessons learned for developers seeking to enable AI/ML on future missions. Finally, SCENIC also includes new CubeSat form-factor GPS and SDR cards for guidance and navigation.