It seems from
Solicitation:
SBIR_21_P1
Topic Number: H6.22
that the Boeing High performance Spaceflight Computer (HSPC), due for delivery in December 2022, was conceived before NASA tumbled to the benefits of neuromorphics:
https://legacy.www.sbir.gov/node/1836297
T
he current state of the art (SOA) for in-space processing is the High Performance Spaceflight Computing (HPSC) processor being developed by Boeing for NASA Goddard Space Flight Center (GSFC). The HPSC, called the Chiplet, contains 8 general purpose processing cores in a dual quad-core configuration. Delivery is expected by December 2022. In a submission to the Space Technology Mission Directorate (STMD) Game Changing Development (GCD) program, the highest computational capability required by a typical space mission is 35 to 70 GFLOPS (billion fast logical operations per second).
The current SOA does not address the capabilities required for artificial intelligence and machine learning applications in the space environment. These applications require significant amounts of multiply and accumulate operations, in addition to a substantial amount of memory to store data and retain intermediate states in a neural network computation. Terrestrially, these operations require general-purpose graphics processing units (GP-GPUs), which are capable of teraflops (TFLOPS) each—approximately 3 orders of magnitude above the anticipated capabilities of the HPSC.
Neuromorphic processing offers the potential to bridge this gap through a novel hardware approach. Existing research in the area shows neuromorphic processors to be up to 1,000 times more energy efficient than GP-GPUs in artificial intelligence applications. Obviously, the true performance depends on the application, but nevertheless the architecture has demonstrated characteristics that make it well-adapted to the space environment.
Phase 1 of the project had extraordinarily short deadlines over a holiday period:
Release Date: November 09, 2020
Open Date: November 09, 2020
Application Due Date: January 08, 2021
Close Date: January 08, 2021
... not that we could draw any inferences from that. After all, doesn't everybody have "
concept of operations of the research topic, simulations, and preliminary results. Early development and delivery of prototype hardware/software is encouraged" for a SWaP compliant neuromorphic processor in their back pocket, eady to be produced at a moment's notice?
Phase II will emphasize hardware and/or software development with delivery of specific hardware and/or software products for NASA, targeting demonstration operations on a low-SWaP platform. Phase II deliverables include a working prototype of the proposed product and/or software, along with documentation and tools necessary for NASA to use the product and/or modify and use the software. In order to enable mission deployment, proposed prototypes should include a path, preferably demonstrated, for fault and mission tolerances. Phase II deliverables should include hardware/software necessary to show how the advances made in the development can be applied to a CubeSat, SmallSat, and rover flight demonstration..
I don't know if Phase 2 of any NASA SBIR has ever gone under the radar, but, in retrospect, ANT61 does spring to mind as a cubesat implementation, and one which offered little prospect of near-term commercial viability while absorbing valuable BRN engineering time. Not that that's a dot, but it is a coincidence that the Akida engineering samples and the feature enhanced 4-bit Akida 1000 and a Cubesat implementation did occur during a fairly compressed time period.
Contra-indication is NASA's referenes to the inherent rad-hardness of memristors, which point to a leaning to an analog implementation.
Still, we know that NASA has been playing with Akida for some time. The short submission period suggests that there had been significant pre-match discussions between NASA and their prospective SBIR applicants. One factor to take into consideration is that the "S" in SBIR would exclude the big boys.
So now we come to the recent and not so recent announcements linking Akida to NASA or space applications. Some which spring to mind:
The MOU with EdgX with links to ESA.
There's Frontgrade which links to ESA.
There's Intellisense Neuromorphic Enhanced Cognitive Radio (NERC) which links to a NASA Phase 2 SBIR.
We dabbled in rad-hard processes with Vorago in 2020 for a Phase 1 NASA project.
RTX/Raytheon as the putative sub-contractor for the recent Phase 2 NASA project.
We know some of the big boys, such as IBM, have been dabbling with analog NNs/memristors for some years, so it is an open question as to whether this is at the behest of NASA, but it is very clear that a great deal of the SBIR requirements fit Akida like a glove.
