Adapting On Orbit: Conclusions of the STP-H6 Spacecraft Supercomputing for Image and Video Processing Experiment

Autor: Gretok, Evan W., Roffe, Seth, Gillette, Antony, George, Alan D., Sabogal, Sebastian, Schwarz, Theodore
Rok vydání: 2022
Předmět:
Zdroj: Small Satellite Conference
Popis: Spacecraft Supercomputing for Image and Video Processing (SSIVP) was a payload aboard the Department of Defense Space Test Program – Houston 6 pallet deployed on the International Space Station. SSIVP was designed and constructed by graduate students at the NSF Center for Space, High-Performance, and Resilient Computing (SHREC) at the University of Pittsburgh. The primary objective of this experiment was to evaluate resilient- and parallel-computing capabilities in a small-satellite form factor. Five flight computers, each combining radiation-tolerant and commercial-off-the-shelf technologies, were networked by high-speed interconnects, enabling a reliable space-supercomputing paradigm. Image-processing and computer-vision experiments were conducted on Earth-observation imagery acquired from two five-megapixel cameras. The system operated for 30 months, serving as an adaptable and reconfigurable platform to host academic and industry research. Despite on-orbit challenges with thermal constraints and operations, all mission objectives were completed successfully. SSIVP resulted in a dataset of nearly 20,000 images, radiation-effects data, and an increase in the technology-readiness level for two SHREC flight computers. Its designers and operators hope that SSIVP serves as a model for future reconfigurable and adaptable space computing platforms.
Databáze: OpenAIRE