Expanding data storage in space is crucial for enabling more powerful artificial intelligence systems to process larger volumes of information directly from orbit.
More capable hard drives would facilitate faster and more insightful analysis from space by reducing the need to download data to the ground for processing.
“It’s useful both with AI training and with inference,” said Jon Trantham, principal technologist at data storage specialist Seagate.
“Having a large pool of data available, and then potentially having it augmented with AI processing hardware, we think is … a new application for space.”
Inter-satellite links, which allow spacecraft to relay data among themselves, are also key to accelerating communications with the ground, enabling low Earth orbit (LEO) satellites to transmit information without waiting to pass over an approved ground station.
Trantham is Seagate’s technical lead for a mission testing a four-terabyte, solid-state drive attached to the outside of the International Space Station for the past two months.
The storage device, slightly larger than a standard computer hard disk drive, is housed in a small satellite provided by partner and potential customer BAE Systems.
It is Seagate’s first storage device specifically designed for the rigors of space, although the company supplied hard drives that have been on the ISS for more than two decades.
“The main news out of the telemetry is the environment actually is more benign than what we had feared,” Trantham said.
“For solid state drives, cold temperatures can be really stressful.”
Although stress tests continue, he said the company plans to refine the device next year before its commercial release for LEO operators in 2026.
Seagate aims to strike a balance between a cheap, less rugged, off-the-shelf storage device and more expensive military-grade hardware.
“It’s much larger than most of the alternatives that are out there today,” Trantham added. “A lot of them are [in the] gigabyte capacity sort of range. Terabyte is relatively new here, so that should enable new use cases.”
While those new use cases are uncertain at this point, Trantham anticipates demand for climate data, imagery and video processing applications.
Onboard autonomy
In addition to the real-time use of space data, deploying AI onboard spacecraft is important for missions requiring high levels of autonomy in support of real-time operations, said Michele Faragalli, chief technology officer of Canadian software startup Mission Control.
Onboard autonomy is essential when communication with operators is delayed or limited, and when data volumes exceed affordable downlink capacity. Timely action is necessary for perishable data, and when complex operations demand quick and precise decisions.
“Onboard AI can, therefore, benefit a range of space mission classes, from robotic exploration to disaster response using Earth Observation satellites,” he continued.
Faragalli emphasized that advancements in onboard autonomy, driven by AI, could pave the way for new types of space missions.
These include robotic subsurface exploration of Europa and the autonomous coordination of multiple space-based assets for extensive area observation, using a method known as “tip and cue,” where one sensor identifies an area of interest (tip) and directs other sensors to investigate further (cue).
This article first appeared in the November 2024 issue of SpaceNews Magazine.
