Research On Fixed Wing UAV Uncertainty Modeling

The UAV mathematical model reflects the relationship between the control input and the vehicle response,the establishment of the system model can’t be separated from the practical application of flight simulation, flight stability and maneuverability analysis, flight control design and so on.The control performance of the system depends largely on the accuracy of the established model,only the model which can accurately simulates actual UAV flight processhas the application value. In fact, there are some differences between the mathematical model and the real system response,In order to make the mathematical model more accurately reflects the response of the real system,considering the uncertainty factors of the actual system in the nonlinear simulation modelbased on the wind tunnel model isso important. This paper is based on the test data of a certain type of high altitude subsonic UAV,the system identification is completed,combined the aerodynamic parameters of wind tunnel model, the UAV flight dynamics model containing uncertainty factorsis established.The main research contents of this paper are divided into three parts.The first part is the establishment of UAV nonlinear simulation model, including the modeling of aerodynamic model, engine model and the establishment of dynamic equation,providing the basis for modeling and simulation of this work.The second part is the system identification of UAV. First the UAV flight data of three sorties is required to be preprocessed,accomplishing the work including the identificationand elimination of outliers in measurement datausing Observer/Kalman filter identification(OKID) and flight test data compatibility check using Extended Kalman filter(EKF).Subsequently, the flight data were identified by CIFER, according to the results of system identification and the theory of wind tunnel model, the upper and lower bounds and the nominal values of aerodynamic parameters are obtained,and the nonlinear simulation model is updated using the aerodynamic parameter identification,laying the foundation for the uncertainty modeling of aerodynamic model.The third part mainly introduces the methods and procedures of the nonlinear simulation model’s uncertainty modeling. The uncertainty modeling of the aerodynamic model is realized based on the boundary and nominal value of the aerodynamic parameters. The uncertainty modeling of the engine’sforce model, and fuel consumption model as well as the center of gravity’s uncertainty modeling is based on experimental data,and the uncertainty model is completed by considering the a certain range of numerical fluctuation on this basis, then the uncertainty Simulink simulation model of UAV system is established.Afterwards the uncertainty Simulink simulation model is linearized in balancing point linear according to the ulinearize principle, and the M—?Matrix form of LFT model is achieved by the method of Linear fractional transformation(LFT). Finally, the worst-case gain(WCG) analysis is performed to verify the accuracy of the model, and in the case of a given perturbation,the output response of the random samplings modelissimulated and analyzed using the function in the robust control toolbox,the analysis and verification of the validity of the uncertain model is completed.