SpaceCube: Meeting On-Board Processing Requirements for Future Science Missions
For nearly two decades, the SpaceCube program at Goddard Space Flight Center (GSFC) has provided a series of processors that have been infused into multiple missions and applications. With funding from several ESTO awards starting as early as 2008, GSFC Internal Research and Development (IRAD) support, and other external partnerships, these reconfigurable spaceflight data processing systems provide 10x to 100x improvements in computing power while lowering power consumption and cost.
The SpaceCube team realized early on that the sophistication of space-based science instruments was quickly outpacing the capabilities of existing space-qualified data processing systems. While radiation-hardened and proven, the available systems could not provide the speed or power needed to handle vast amounts of data. And sensor autonomy, artificial intelligence, and deep learning were beginning to add brand new computing demands. Their approach abandoned the traditional path of developing bespoke radiation-hardened flight processors that were consistently a generation or two behind commercial processors. Instead, they used the latest “radiation-tolerant” processing elements coupled with integrated upset detection and correction architecture to provide both reliability and computing power.
SpaceCube team members are pictured here with some of the family of SpaceCube products. From left to right: Alessandro Geist; Dave Petrick; Tom Flatley; and Gary Crum. The products include (from left to right): SpaceCube 1.0 and 1.5 prototypes; SpaceCube 2.0 and Mini prototypes; and SpaceCube 1.5 flight unit. (Credit: Bill Hrybyk/NASA)
The SpaceCube series of on-board processors – from the original SpaceCube 1.0 through SpaceCube 3.0 and SpaceCube Mini-Z+, along with a full range of SpaceCube Cards for specific applications – have seen enormous success and have generated at least seven patents.
SpaceCube processors have enabled more than 20 missions, including NASA’s Robotic Refueling Mission-3, Restore-L Robotic Servicing Mission, and Asteroid Redirect Robotic Mission. The U.S. Department of Defense (DoD) has also adopted the technology.
Most recently, in March 2023, the SpaceCube Edge Node Intelligent Collaboration (SCENIC) experiment was launched to the International Space Station (ISS) on board the DoD Space Test Program-Houston 9 (STP-H9) mission. One of eight experimental payloads on STP-H9, SCENIC is a testbed to evaluate artificial intelligence and machine learning on FPGAs and custom microchips. On ISS, SCENIC will demonstrate autonomy, decision-making, and sensor collaboration; assess performance and radiation-tolerance; and offer an ongoing testing platform for uploading new algorithms and capabilities.
For more information, visit the SpaceCube Program page at https://spacecube.nasa.gov