Living on the edge: Satellites adopt powerful computers
The latest Apple Watch has 16 times the memory of the central processor on NASA’s Mars 2020 rover. For the new iPhone, 64 times the car-size rover’s memory comes standard.
For decades, people dismissed comparisons of terrestrial and space-based processors by pointing out the harsh radiation and temperature extremes facing space-based electronics. Only components custom built for spaceflight and proven to function well after many years in orbit were considered resilient enough for multibillion-dollar space agency missions.
While that may still be the best bet for high-profile deep space missions, spacecraft operating closer to Earth are adopting state-of-the-art onboard processors. Upcoming missions will require even greater computing capability.
Satellite sensors produce “an enormous amount of data in the form of scientific research, Earth observation, national security,” Naeem Altaf, IBM distinguished engineer and IBM Space Tech chief technology officer, said by email. “To extract the quick value of data, we will need to bring compute closer to data.”
Consider Earth observation. Traditionally, electro-optical imagery and synthetic-aperture radar data have been sent to the ground for processing. That’s still largely the case, but new Earth-observation sensors continue expanding the volume of data acquired in orbit, sometimes quite dramatically. At the same time, customers are eager for speedy access to insights drawn from various datasets.
Weather observation is a good example. Numerical weather models merge vast quantities of data drawn for space, airborne, maritime and terrestrial sensors. While no one proposes running the forecasting algorithms on satellites, AAC Clyde Space, the Swedish company supplying the core avionics for the European Space Agency’s Arctic Weather Satellite, sees improvements in onboard processing as a way to speed up weather data delivery.
“We see an opportunity in the future to do a lot of processing on board: preparing data, compressing data and starting to fuse data,” said Luis Gomes, AAC Clyde Space CEO. “Our objective is real-time weather observations from space. For that, we need to package the data efficiently and effectively to reduce the amount of time that we are downlinking.”
Hyperspectral sensors also produce huge datasets that make onboard processing “quite critical,” Gomes said.
Some of the new satellite computers will be devoted to crunching sensor data. Others will help spacecraft choreograph complex operations.
Future satellites are likely to operate in swarms, communicating through intersatellite links and working together to capture unique datasets and extend communications networks. Eventually, constellations will employ artificial intelligence to solve problems by, for example, healing or repositioning satellites based on onboard analysis of their health and performance, which will require extensive edge processing, said Chuck Beames, chairman of the SmallSat Alliance, an industry association.
Edge processing, bringing computation closer to data sources, is increasingly popular on Earth. Oil and gas companies, for example, analyze data near sensors that monitor heavy equipment at remote sites to quickly identify equipment problems and to trim communications and data storage expenses.
Companies ranging from IBM and Hewlett Packard Enterprise to startups around the world are positioning themselves to meet what they see as inevitable demand for enhanced space-based edge processing, beginning onboard satellites and extending to data centers in Earth and lunar orbit.
Exodus Orbitals, a Canadian startup that rents satellite services to software application developers, established the Edge Computing in Space Alliance in November. The organization quickly attracted nearly two dozen members.
One of the members, Ramon.Space, advertises “space-resilient supercomputing systems.” While they bear little resemblance to terrestrial supercomputers, they are far different from low capacity spaceflight computers and “a lot closer to the kind of computing capability that we have on Earth,” said Lisa Kuo, vice president of strategic sales for Ramon.Space, an Israeli firm established in 2004 that is expanding internationally. “We go over space computing systems with a very fine-tooth comb and adopt the optimal radiation-hardening technique for each component.”
In contrast to the bespoke approach, startup Exo-Space of Pasadena, California, offers FeatherEdge, a platform that applies artificial intelligence and machine learning to Earth observation data to quickly extract valuable information.
Long term, Exo-Space plans to “adapt the technology to the more general-purpose use cases like constellation management or predictive maintenance,” said CEO Jeremy Allam.
Sydney-based Spiral Blue also applies artificial intelligence to Earth imagery with its Space Edge computer.
“Satellites can capture far more data than they can actually bring down,” said Taofiq Huq, Spiral Blue founder and CEO. With improved onboard processing, satellites can highlight and downlink the most important information, like ship locations for maritime vessel tracking, he added.
BOX IT UP
Other firms specialize in packaging terrestrial computers for spaceflight. OrbitsEdge, for example, works with customers including HPE to provide radiation shielding and thermal management systems that allow computers designed for terrestrial applications to function in orbit.
“By relying on the high-power computation pipeline, we have assurances that whatever we’re flying is the most modern stuff,” said Rick Ward, chief technology officer and founder of the Titusville, Florida-based OrbitsEdge. “When we segue to quantum computing, and we’ve already had conversations with some of the quantum computing companies, we can do that as well.”
Cosmic Shielding Corp. takes a similar approach but instead of focusing on safeguarding processors, the Atlanta startup developed a 3D-printed polymer to protect people and electronics in orbit.
“You can build a satellite bus out of this material, and it will provide significant improvements,” said Yanni Barghouty, Cosmic Shielding founder and CEO. “Right now, we’re seeing around a 60 to 70-percent radiation dose-rate reduction versus traditional materials.”
EXTENDING THE EDGE
In addition to enhancing onboard processing, companies are installing edge processors in ground stations and making plans to launch constellations devoted to data processing.
“Edge computing can be performed at different segments, depending upon the use case and the criticality of data,” said IBM’s Altaf. “We can have dedicated compute satellites, which are tasked to take on the heavy payloads in orbit and perform computation services for other satellites.”
If history is any guide, demand for data processing in orbit will continue to climb. Successive generations of terrestrial applications invariably require additional memory and processing speed.
In space, like on the ground, “you want it faster, you want better networking, and you want more power,” said Mark Fernandez, HPE principal investigator for the Spaceborne Computer-2 on the International Space Station.
This article originally appeared in the January 2022 issue of SpaceNews magazine.