A Machine Learning Approach to Aircraft Sensor Error Detection and Correction Conference Paper uri icon


  • 2018, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. The pitot static system provides critical airspeed information and consists of two ports located outside of the aircraft making them vulnerable to interference and failures. If an aircraft has access to redundant sensor output, then it can be trained to autonomously recognize errors in faulty sensors and learn to correct them. In this work, we develop a novel machine learning approach to detecting aircraft sensor failures and predicting corrected flight data using an online/offline paradigm. We demonstrate our methodology on flight data from a four engine commercial jet that contains failures in the pitot static system to show the safety benefits of our system in flight. Autocorrelation of incoming pressure data is used to classify the state of the pitot static system. Feature selection is performed on the high dimensional sensor output to create an offline library for predicting airspeed. This library is used to train a regression model to make real time corrections to airspeed data in the event of a pitot static system failure.

name of conference

  • 2018 AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference

published proceedings

  • 2018 AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference

author list (cited authors)

  • Swischuk, R. C., & Allaire, D. L.

complete list of authors

  • Swischuk, Renee C||Allaire, Douglas L

publication date

  • January 1, 2018 11:11 AM