Multi-Spacecraft On-orbit Space Computing Testbed

The future of High-Performance Space Computing (HPSC) promises collaborative decision making and dynamic load sharing across distributed spacecraft, but NASA and the HPSC industry lack an off-world platform needed to rapidly develop and test these capabilities at scale. EarthTraq proposes developing a standalone HPSC testbed into our low-cost small satellite clusters, leveraging our fully funded 4-satellite mission launching in early 2026 as an over-the-air development sandbox needed by the industry. Each low-Earth orbit (LEO) spacecraft, flying in tight formation (<100 km apart), will host a dedicated hardware/firmware/software HPSC testbed for digital signal processing (DSP) and artificial intelligence (AI) coprocessors. This will enable first-of-its-kind advancements in performance, energy efficiency, scalability, versatility, and resilience of emerging HPSC technologies in a cooperative multi-spacecraft environment. In this Phase I proposal, EarthTraq will develop & prototype the hardware/firmware/software necessary to provide a "sandboxed" testbed using open interface standards for a flexible platform to meet evolving needs. The testbed will be evaluated and integrated into our flight Avionics & Sensors, likely with a Mentium coprocessor or Microchip PIC64-HPSC MPU, onboard each of the four spacecraft. Planned Phase II activities: proton, vibe, and TVAC testing, providing NASA with over-the-air access to the orbital testbed for rapid software iteration, demo coprocessor-based DSP offloading for primary mission tasks, cluster-wide collaborative AI/DSP processing to optimize thermal & power constraints, cluster-wide “tip and cue” activities. Target markets include NASA Exploration, Science, and Space groups looking to eval AI models & distributed autonomy software, HPSC manufacturers seeking flight heritage, commercial & government teams looking for real-world benchmark opportunities or to demonstrate HPSC superiority.

The proposed on-orbit distributed HPSC testbed will provide first-of-its-kind access to distributed on orbit DSP and AI computing resources onboard four identical spacecraft flying in formation (TX02.1.3, TX02.2.) at an incredible value and time-frame for NASA. Additionally, access to this resource on orbit will be available three to five months after the Phase I period of performance and will continue to grow in size. By building this platform into our existing spacecraft (TX02.2.8) with separate mission and revenue streams, the continued sustainment of this valuable capability is ensured. This provides tremendous value to NASA mission directives accustomed to higher barriers to entry for testing their software on off-world assets, such as the Exploration and Science directorates. For example, the Science directive could finally answer the question: can we use AI to better study or respond to rapidly changing radiation environments by coordinating across multiple spacecraft (TX02.3.2, TX10.1.4)? The NASA Space Weather Science and Observation group could test different AI coprocessing routines for improved/faster total electron content (TEC) measurements through collaborative “tip and cue” across the cluster (TX10.2.2, TX10.3.1). One can imagine distributed HPSC routines developed with this unique platform translating to human exploration, as well. The proposed testbed also benefits the Space Technology directorate by providing an avenue for rapidly testing and iterating on future HPSC architectures to enable new missions and capabilities as well as providing faster and lower cost avenues to evaluating new hardware. The proposed on-orbit distributed HPSC testbed will provide first-of-its-kind access to distributed on orbit digital signal processing (DSP) and artificial intelligence (AI) computing resources onboard four identical spacecraft flying in formation. This allows non-spacecraft builders & operators such as commercial HPSC manufacturers, small emerging coprocessor companies, and AI software companies to demonstrate and validate their hardware/software products in real-world environments that are otherwise prohibitively expensive and difficult to break into. This represents tremendous value-add for companies looking to sell their products to commercial or defense customers that require flight heritage and real-world benchmarking. Therefore, we believe the proposed testbed is one of the tipping points needed in the HPSC industry for future advancements driven by commercial interests – complementary to the methods NASA JPL teams have used to bootstrap commercial HPSC availability.

This is a mature technology (TRL 7+) — the realistic path in is usually NASA's Technology Transfer Program: licensing an existing NASA patent, or a Space Act Agreement to use NASA facilities/expertise directly. NASA also runs a startup licensing program with no upfront fee for companies formed to commercialize a specific NASA technology.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.