The coming decade of planetary missions will have stringent data requirements. Optimizing limited bandwidth is a major motivation for the use of onboard science data analysis algorithms for deep space missions. Under the traditional mission operations paradigm, all data is sent to Earth before follow-on activities are planned by ground operators. This operational model creates challenges for time and bandwidth constrained mission events. The Near Earth Asteroid Scout (NEAScout) mission is a prime example of a mission with these challenges. NEAScout is an interplanetary CubeSat manifested on Artemis-1. The mission will use a solar sail to navigate to a Near Earth asteroid and perform flyby reconnaissance imaging. Vehicle size, configuration, trajectory and distance from Earth all impose bandwidth constraints. To minimize these limitations, the Project decided to move image processing capabilities onboard the spacecraft. This onboard science software enables image calibration, image coaddition, image subtraction, data compression, downsampling and cropping all onboard the spacecraft. The goal is to reduce the data downlink, without compromising science contents. NEAScout utilizes the SPHINX computing platform, whose processing performance is akin to the RAD750, but tailored for CubeSat resources. Although sufficient to enable these sorts of algorithms, the SPHINX takes multiple minutes per image, which diminishes the prospect of using image content for rapid response reasoning. This effort ported, characterized and benchmarked the performance of the NEAScout science software on the Snapdragon platform, a next-generation higher performing processor. Between $10\mathrm{x}$ and $200\mathrm{x}$ processing time speedup was achieved, depending on the image processing algorithm, with a $50\mathrm{x}$ speedup for a notional image processing pipeline. This work highlights new performance baselines for onboard science data processing algorithms which missions could expect to achieve with modern computing hardware. Future avionics computers with Snapdragon processors would enable onboard science data analysis to be used in more time constrained mission scenarios and with larger onboard data sets.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Benchmarking Onboard Science Data Retrieval Algorithms on the Snapdragon Platform


    Contributors:


    Publication date :

    2023-03-04


    Size :

    9827639 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Benchmarking Space Mission Applications on the Snapdragon Processor Onboard the ISS

    Swope, Jason / Mirza, Faiz / Dunkel, Emily et al. | AIAA | 2023


    Benchmarking Planning Applications on the Qualcomm Snapdragon

    Chien, Steve / Swope, Jason / Mirza, Faiz | NTRS | 2021


    Benchmarking Deep Learning Models on Myriad and Snapdragon Processors for Space Applications

    Dunkel, Emily R. / Swope, Jason / Candela, Alberto et al. | AIAA | 2023



    The "Snapdragon" Family

    Simpson, D. M. / European Space Agency | British Library Conference Proceedings | 2001