NASA Logo, National Aeronautics and Space Administration

+NASA Home

+Ames Home

Data Sciences Group Team Receives 2021 AIAA Intelligent Systems Best Paper Award
Intelligent Systems Division Banner

Data Sciences Group Team Receives 2021 AIAA Intelligent Systems Best Paper Award

A Data Sciences group team paper entitled “Multi-Class Anomaly Detection in Flight Data Using Semi-Supervised Explainable Deep Learning Model” received the 2021 American Institute of Aeronautics and Astronautics’ (AIAA) Intelligent Systems Best Paper Award. This award is given annually for innovative and impactful contributions to the field and was presented at the virtual AIAA 2021 SciTech Forum and Exposition.

BACKGROUND: The identification of precursors to safety incidents in aviation data is a crucial task, yet extremely challenging. In practice, the main approach leverages domain expertise to define expected tolerances in system behavior and flags exceedances from such safety margins. However, this approach is incapable of identifying unknown risks and vulnerabilities. Various Machine-Learning (ML) approaches have been investigated and deployed to identify anomalies, with the great challenge of procuring enough labeled data to achieve reliable and accurate performance. Building upon recent advancements described in ML literature, we present an explainable deep semi-supervised model for anomaly detection in aviation. Our proposed model combines feature engineering and classification in feature space, while leveraging all available (labeled and unlabeled) data. We validate our approach with case studies of anomaly detection during the take-off and landing phases of commercial aircraft and use those to show that our model outperforms the state-of-the-art supervised anomaly-detection model, reaching significantly higher accuracy and fewer false alarms with only a small amount of labeled data.

NASA PROGRAM FUNDING: NASA System-wide Safety and Assurance Technologies (SSAT) Project, Aviation Safety Program (AvSP), Aeronautics Research Mission Directorate (ARMD). Dr. Milad Memarzadeh is supported by the NASA Academic Mission Services (NAMS) contract NNA16BD14C; and Bryan Matthews is supported by KBRWyle contract 80ARC020D0010.

TEAM: Milad Memarzadeh, Bryan Matthews, and Thomas Templin

POINT OF CONTACT: Milad Memarzadeh, milad.memarzadeh@nasa.gov

First Gov logo
NASA Logo - nasa.gov