| Flight control system is the core of UAV, its reliability is related to the safety of UAV, to diagnose the fault of flight control system rapidly and exactly is the foundation of the security of UAV’s system, so the research of the fault diagnosis technology of flight control system has important research significance and application value. Based on the mathematical model of actuators and UAV, this paper studied the threshold-improved SPRT joint diagnosis algorithm based on the residual, developed the system integrated flight control and fault diagnosis which is based on dual-CPU architecture, to diagnose the fault of of actuators and sensors of UAV’s flight control system.First of all, the paper analyzed the design requirements of fault diagnosis of flight control system, and formulated the overall design strategy of fault diagnosis system.Secondly, based on the analytical model of the actuators, according to the residual error of the output of the model and the actual position of the actuator, the paper researched the threshold-improved SPRT joint diagnosis algorithm based on the residual,to realize the online diagnosis of actuators’ fault quickly and efficiently.Once more, the paper established the mathematical model of UAV and conducted the characteristics simulation of sensors according to the calculating result of the model in order to obtain the residuals between the estimation and the measuring value of sensors, on the basis of the threshold-improved SPRT joint diagnosis algorithm, it synthesized the methods such as expert judgment and analysis of data fusion,to improve the accuracy of the sensors’ fault diagnosis.After that,the paper aimed at the dual-CPU architecture of the platform for fault diagnosis, carried out the function design and software implementation of the flight control and fault diagnosis integrated system. Finally,the paper carried out the simulation verification in the semi physical simulation environment of the system,the verification result shows that the fault diagnosis system based on on-line model can diagnose the failure of the actuators and sensors of flight control system rapidly and accurately. |