Experimental Evaluation of Model-Free Hybrid Fault Detection and Isolation

Abstract

This paper deals with a model-free hybrid fault detection and isolation scheme subject to sequential triple faults in inertial measurement sensors. An additional small-sized and lowcost inertial measurement unit is added to the original inertial measurement unit of the UAV system. The parity space method and inline monitoring method are combined to increase the tolerance to sequential multiple faults in the skew-configured inertial measurement system. The first and the second sensor faults are detected and isolated by the parity space method. The third sensor fault is detected and isolated by the parity equation and the inline monitoring method based on the discrete wavelet transform. Hardware in-the-loop simulation test and the flight experiments were performed to verify the performance of the proposed fault diagnosis scheme.