Spectral Sessions 2022 Playlist

Beyond the Edge: The Future of Space-Based Data Analysis

Dr. James Goodman, HySpeed Computing

Abstract

Never before have the 5 V’s of big data – volume, value, variety, velocity and veracity – been as applicable to Earth Observation as they are today. Consider not just the widespread growth in multi-satellite constellations and the greater variety of sensor modalities, but also the ongoing increases in temporal, spatial, and spectral resolution. And with the industry continuing to expand, never before has the demand for efficiently delivering analysis and insights from this data been greater.

To meet these needs, data providers and analysts alike are now embracing a robust ecosystem of highspeed data downlinks, cloud and edge computing solutions, and applications driven by artificial intelligence. At the same time, capabilities are also rapidly evolving in the domain of onboard computing and space-based data analysis.

By locating increasingly powerful compute resources and software algorithms adjacent to the data, greater amounts of analysis can now be performed at the source, which is enabling new opportunities for on-orbit analytics, advanced data compression, prioritization of high-value data for download, direct downlink of derived products, automated tasking, and systematized multi-sensor tipping and cueing. Together, these various advances are not just expanding the scale and diversity of feasible applications, but also serving to significantly decrease latency and increase the cadence of information delivery, thereby driving even further innovation throughout the geospatial industry.