Fault Tolerant RISC-V Flight Computer with Coprocessor Support

This Phase II project aims to prototype a novel technology that can advance the state-of-the-art in high performance space computing. The base computer technology, called RadPC, brings together a suite of fault recovery mechanisms that enable Commercial-off-the-Shelf (COTS) Field Programmable Gate Arrays (FPGAs) to operate reliably in the presence of space radiation that causes single event effects (SEEs). The RadPC concept has been matured for over 12 years in the research labs of Montana State University (MSU). Through over $6M in NASA research funding, RadPC has been tested on high-altitude balloons (8x), sounding rockets (2x), the International Space Station (3x), and on small satellites (2x). In 2024, the RadPC base computer will be tested on a lunar mission through the NASA Lunar Surface Instruments and Technology Payloads (LSITP) program to stress its SEE recovery mechanisms while passing through the Earths radiation belts and operating outside of the Earths magnetic field. In 2021, RadPC was licensed to Resilient Computing to begin its commercialization efforts. Through a 2021 NASA SBIR Phase I award, the MSU version of RadPC was refined to use a RISC-V architecture to make the computer suitable for real space missions. Through a 2022 NASA SBIR Phase I award, a feasibility study was conducted to evaluate if coprocessors could be embedded within the RadPC architecture in order to accelerate computation while still maintaining the reliability that RadPCs SEE recovery procedures provide. The 2022 Phase I study devised a strategy to successfully include coprocessors into RadPC to perform computationally intense tasks while being able to be repaired through the existing recovery procedures. This Phase II project aims to develop a prototype of the RadPC+coprocessor system to test its computational performance and reliability while conducting object detection algorithms on real-time camera data. ​​​​​​​​​​​​​​Harsh radiation environment of space: Material degradation & intermittent faults. Current “rad-hard” computers: Expensive & lag performance of commercial computers by ~20 years. Cost-prohibitive for short missions, including small satellites. Our RadPC advantages: Implements proven fault-tolerant architecture on a commercial FPGA. Performance & power efficiency at lower price while delivering increased fault tolerance for space applications.  Advances high-performance space computing by enabling novel processing architectures through coprocessor modules. RadPC’s fault-tolerant approach tested on numerous NASA flights: High-altitude balloons (8x) & sounding rockets (2x). International Space Station (3x) & small satellites (2x). Upcoming lunar demonstration through NASA CLPS program (2024). Objectives TO1 – Develop a coprocessor module to accelerate pattern detection in real-time image data using a trained machine learning supervisor. TO2 – Integrate the ML coprocessor into the RadPC fault-tolerant computer system. TO3 – Test the entire prototype computer under laboratory fault injection. TO4 – Refine requirements for mission-ready RadPC+coprocessor product. Deliverables Prototype of the RadPC+coprocessor concept performing pattern detection on real-time camera data. Test report on the prototype’s performance and reliability. Refined requirements for a RadPC+coprocessor product that is ready for mission adoption.

Replacing ~20 year old rad-hard processors with a lower-cost, higher performance, commercial-based technology. Accelerating computationally intense algorithms such as real-time science data processing, autonomy, and navigation using coprocessors. Control & data handling for NASA small satellites. Small satellites needing increased reliability and performance, but at a price-point below current rad-hard computers. Earth image processing (climate monitoring, disaster mitigation, agriculture). Communication networks.

This is early/mid-stage (TRL 4) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

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.