Interval Fuzzy Model for Robust Aircraft IMU Sensors Fault Detection.

Michele Crispoltoni,Marcello Napolitano,Fabio Balzano,Stephane D’Urso,Mario Fravolini

doi:10.3390/s18082488

Michele Crispoltoni, Marcello Napolitano + Show 3 more

Open Access

https://doi.org/10.3390/s18082488

Copy DOI

Abstract

This paper proposes a data-based approach for a robust fault detection (FD) of the inertial measurement unit (IMU) sensors of an aircraft. Fuzzy interval models (FIMs) have been introduced for coping with the significant modeling uncertainties caused by poorly modeled aerodynamics. The proposed FIMs are used to compute robust prediction intervals for the measurements provided by the IMU sensors. Specifically, a nonlinear neural network (NN) model is used as central prediction of the sensor response while the uncertainty around the central estimation is captured by the FIM model. The uncertainty has been also modelled using a conventional linear Interval Model (IM) approach; this allows a quantitative evaluation of the benefits provided by the FIM approach. The identification of the IMs and of the FIMs was formalized as a linear matrix inequality (LMI) optimization problem using as cost function the (mean) amplitude of the prediction interval and as optimization variables the parameters defining the amplitudes of the intervals of the IMs and FIMs. Based on the identified models, FD validation tests have been successfully conducted using actual flight data of a P92 Tecnam aircraft by artificially injecting additive fault signals on the fault free IMU readings.

Highlights

The occurrence of failures on the different classes of sensors in a flight control system is a very critical issue from a safety and reliability point of view
The conventional approach for deriving experimental models for predicting the measurement of sensors as function of correlated signals requires first the selection of a suitable linear or nonlinear in the parameters model structure; this selection is followed by the identification of the model parameters through a specific algorithm using a batch of representative “rich” training data
In a fault detection context, the identified prediction model, at times referred to as “virtual sensor”, is used on-line to compute a diagnostic signal that is the difference between the actual measurement and its estimation, whose abnormal deviation could be attributed to a potential fault

Summary

Introduction

The occurrence of failures on the different classes of sensors in a flight control system is a very critical issue from a safety and reliability point of view. An IMU is the main component of the inertial navigation system (INS) and is used by the flight computer to calculate attitude, angular rates, linear velocity, and position with respect to a global reference frame. Because the guidance system is continually integrating acceleration with respect to time to calculate airspeed and position (dead reckoning), measurement errors, even if small, accumulate substantially over time. This leads to the well-known drifting error, which is essentially an ever-increasing difference between the system “perceived” and actual location. A constant error in attitude rate (gyro) eventually leads to a quadratic error in airspeed and a cubic error growth in position To deal with these issues sensor fusion techniques with

Objectives

Methods

Findings

Conclusion