Detecting anomalies in spacecraft telemetry channels is a high priority, especially considering the harshness of the spacecraft operating environment. These anomalies often function as precursors for system failure. Currently, telemetry channel monitoring is done manually by domain experts, which is time-consuming and limited in scope. Given that each satellite system has thousands of channels to monitor, an automated approach to anomaly detection would be ideal. Machine learning models have been shown to be effective at detecting the normal behavior of the channels and flagging any abnormalities. However, a unique model needs to be trained for each channel, and high performing models have been shown to require an increased training time. We propose using an ensemble of online sequential extreme learning machines to quickly understand the behavior of a given channel and identify anomalies in near real-time. This greatly reduces the amount of training time and data required to obtain a model for each channel. We present the results of our approach to show that we can achieve performance comparable to state-of-the-art spacecraft anomaly detection methods with minimal training time and data.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Spacecraft Time-Series Online Anomaly Detection Using Extreme Learning Machines


    Contributors:


    Publication date :

    2022-03-05


    Size :

    1684563 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Spacecraft Time-Series Online Anomaly Detection Using Deep Learning

    Baireddy, Sriram / Desai, Sundip R. / Foster, Richard H. et al. | IEEE | 2023



    Semi-Supervised Machine Learning for Spacecraft Anomaly Detection & Diagnosis

    Ramachandran, Sowmya / Rosengarten, Maia / Belardi, Christian | IEEE | 2020



    Spacecraft Anomaly Detection via Transformer Reconstruction Error

    Meng, Hengyu / Zhang, Yuxuan / Li, Yuanxiang et al. | Springer Verlag | 2020