Adaptable, Scalable, Intelligent Computing for Space (ASICS)

As increasingly ambitious missions are planned for Earth orbit, the Moon, and beyond, the need to leverage the cutting edge of computer technology – advanced artificial intelligence and machine learning (AI/ML) capabilities – becomes ever more apparent. AI/ML applications that can efficiently utilize sensor inputs to rapidly analyze the environment and make decisions autonomously will greatly reduce the risk of many uncrewed mission operations, improve spacecraft safety by elevating situational awareness, and enable high-complexity maneuvers including rendezvous proximity operations and docking (RPOD) operations that rely on the ability to rapidly respond to changes in the environment. On Earth, AI/ML technologies rely on massive cloud computing resources or specialized commercial-grade chipsets. This proposal seeks to develop a new rugged, radiation-tolerant computing solution using a processor tailored specifically for Size, Weight, and Power (SWaP) constrained AI/ML applications: the AMD Versal Adaptive System on Chip (ASoC) Edge Series. The Versal Edge Series is a new space-grade processor with dedicated on-chip AI engine resources that are optimized for power efficient, high bandwidth AI/ML computations. By leveraging this state-of-the-art processor, we can meet the demanding SWaP requirements that are critical for maximizing the efficiency, reliability, and performance of spacecraft. This innovation will overcome the current limitations in AI/ML applications that require significant computational resources, enabling space missions to achieve higher levels of autonomy and performance.

In 2024, NASA Space Science Mission Directorate (STMD), released the Civil Space Shortfall rankings. This ranking identified and prioritized 187 technology areas requiring further development to meet future exploration, science, and mission needs. The ASICS proposal directly addresses the third ranked item in the shortfall list: “1554: High Performance Onboard Computing to Enable Increasingly Complex Operations.” ASICS aims to address this shortfall directly by developing a rugged high performance space grade computer capable of efficiently processing modern AI and other complex algorithms. With embedded AI being an algorithmic approach to solving a particular problem, this project will generate a foundational toolset that can be applied to a vast range of potential problems including TRN, HDA, RPOD, onboard processing for space situational awareness (SSA), spacecraft autonomous anomaly detection, astronaut digital assistants, and many more. Deep neural network models can be loaded by the Versal processor at run-time at different stages of a mission to enable different applications running with one piece of hardware. This reconfigurability feature promises even further optimized hardware utilization for challenging mission profiles. Astrobotic will continue to target both software and hardware sales coming out our high-performance on-board processing systems, with the primary application of near-real time processing of sensing data for in-space missions. The following are all past or current customers (e.g., signed contracts with paid invoices) for sensors or sensing software or related capabilities who would target for the tech resultant from this Phase II: AFRL Space Vehicles, AFRL Center for Rapid Innovation, Astroscale, Draper, and Blue Origin. In addition, we are targeting sales to Advanced Space, Northrop Grumman, and Lockheed Martin.

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.