SENSOR FAULT DETECTION BY TESTING THE GENERALIZED VARIANCE OF THE INNOVATION COVARIANCE

Chingiz Hajiyev, Ulviye Hacizade

Abstract

A new method for testing the covariance matrix of the innovation sequence of the Kalman filter is proposed. The generalized variance (determinant) of the random Wishart matrix is used in this process as a monitoring statistic, and the testing problem is reduced to determination of the asymptotics for Wishart determinants. In the simulations, the longitudinal and lateral dynamics of the F-16 aircraft model is considered, and detection of sensor failures, which affect the covariance matrix of the innovation sequence, are examined.

Keywords

sensor, fault detection, Kalman filter, innovation sequence, generalized variance

Download

IMEKO-WC-2015-TC10-238.pdf

DOI

-

IMEKO TC

TC10 - Measurement for Diagnostics, Optimization and Control

Back to the proceedings