DUKF-based GTM UAV fault detection and diagnosis with nonlinear and LPV models

Fault tolerant control system (FTCS) is to maintain system stability in the presence of fault occurred in the actuators, sensors or other system components. When a fault/failure occurs either in an actuator, sensor or plant, the fault detection and diagnosis (FDD) as the central part of a FTCS will detect and diagnose the source and the magnitude of the fault. The reconfiguration scheme in the FTCS will design the reconfigurable controller based on the FDD information to balance and adapt the system for being able to tolerate to the faults and failures. This paper presents an applicable procedure of integrated fault detection and diagnosis for the NASA Generic Transport Model (GTM) unmanned aerial vehicle (UAV) aircraft model. The implementation of a FDD scheme in handling partial control effector fault cases with the nonlinear GTM and the linear parameter varying (LPV) representation of the nonlinear GTM have been carried out based on a dual unscented Kalman filter (DUKF) and a Baysian rule for detection and isolation decision making. Simulation results show satisfactory results for detecting and diagnosing the control effectors failures.