Kernel Adaptive Filtering Multiple-model Actuator Fault Diagnostic For Multi-effectors Aircraft

In the traditional Multiple Model Adaptive Estimation (MMAE) algorithm, the extended Kalman filter has theoretical limitations, and the establishment of accurate aircraft mathematical model is almost impossible. In this paper, the Kernel Adaptive Filter (KAF) is introduced to replace the Kalman filter, a new multi-model adaptive estimation fault diagnosis method is proposed. Based on the kernel methods, the complex nonlinear system is mapped to the high-dimensional feature space, then the adaptive filter is designed in the high-dimensional feature space without the need to know the system model in advance. After training of KAF using the offline input control signal and output flight states measurement with noise, the estimation of real flight states values and actuator fault detection and isolation can be realized online. The simulation results show good performance of new fault diagnosis method in actuator fault diagnosis.