A nonlinear version of the generalized likelihood ratio test

Abstract

In this paper, a new version of the generalized likelihood ratio (GLR) test is proposed to deal with the fault diagnosis problem in nonlinear systems. The unscented Kalman filter (UKF) is utilized for state estimation and innovation generation. The fault signatures used in the hypotheses test are pre-computed by running the fault model and the nominal model simultaneously based on the state estimations. In such a way the intolerable computational problem, which arise from applying a bank of nonlinear filters, is solved. Further more, the fault isolation problem as well as the fault estimation problem of the non-abrupt fault is discussed based on the nonlinear GLR approach. The simulation study of a longitudinal aircraft model is utilized to show the effectiveness of the proposed method.

Reconfiguration control method for non-redundant actuator faults on unmanned aerial vehicle

“…complex than with only possible actuator faults. Investigating the generalized likelihood ratio test 12 may lead to improve the analysis of the signature of residuals. Second, the FTC method should be applied to a more complete model; first, with a linearized model of longitudinal and lateral dynamics, then with a nonlinear model of UAV.…”

Non-redundant high-integrity position estimation robust to sensor bias jumps using MGLR

“…In response to this criticism, further developments of the GLR have been proposed to counter nonlinearities and non-Gaussian noise. In Faurie and Giremus (2010), a detector combining GLR and M-estimation is proposed to detect and estimate range biases, whereas in Liu and Zhong (2014) a nonlinear version of the GLR is proposed and applied to a fault-tolerant flight control system.…”