UAV attitude Sensor Fault Detection Based On Fuzzy Logic and by Neural Network Model Identification

Fault detection has always been important in aviation systems to prevent many accidents. This process is possible in different ways. In this paper, we first identify the longitudinal axis plane model using neural network approach. Then based on the obtained model and using fuzzy logic, the aircraft status sensor fault detection unit was designed. The simulation results show that the fault detection system is able to work well, with additional alarms averaging 1 alert per 4-hour flight and miss alert rates averaging 1 alert per 2 hours. The results are confirmed by the experts from the UAV system.