Ducted Fan Unmanned Aircraft Flight Dynamics Analysis And Control Study

This paper mainly focuses on the modeling and control study of a small ducted fan autonomousvehicle. Ducted fan vehicle is a new type of unmanned VTOL aircraft, which is capable of both hoverand low-speed forward flight. A detailed analysis of the forces and moments acting on the vehicle isoffered in advance, and then the nonlinear simulation model is developed based on the analysis.Ducted fan vehicles are usually highly unstable with complex aerodynamics, therefore, successfulcontroller must be very robust to deal with the uncertainty present in the flight dynamic model. Aneural-network adaptive controller developed to control other UAVs is used to control the ducted fanvehicle, using dynamic inversion method usually require an accurate system model which is difficultto obtain, especially for a ducted fan aircraft. Instead, a very simple linear model of the vehicle is usedin the dynamic inversion and a neural-network is then trained online to adapt for model error. Certaineffects, such as actuator dynamics and nonlinearities, can cause problems for the adaptive element,thus, pseudo-control hedging (PCH) which can eliminate the selected system input characteristicsfrom the tracking error dynamics is used to prevent the neural network from trying to adapt to theseeffects. Finally, the designed flight control law is simulated in the Matlab/Simulink environment andthe effectiveness of this method is verified.