Design And Control Research Of Ducted Fan Micro Air Vehicle

Micro air vehicle (MAV) has a very bright future in military and civilian fields and become ahotspot in scientific and aerospace engineering research. Ducted fan MAV is a very importantbranch of MAV. It can hover, take off and land vertically and fly forward with high security,compact structure and low noise. The primary focus of this paper is design and control of theducted fan MAV, including general design, low Reynolds number aerodynamics research, flightdynamics research and adaptive flight control system design of the ducted fan MAV.The ducted fan MAVs generally can only fly forward slowly in helicopter mode. In this paper,a new tilt-body ducted fan MAV with wings which is capable of high-speed forward flight inwing-borne mode is designed. Pairs of wings are designed and the geometry design of duct andwings reflects a compromise between low and high speed performance requirements. The winglongitudinal location design considers the impact of various motion parameters including propellerspeed, forward speed and angle of attack on the pitch moment of the ducted fan MAV.Numerical simulation and wind tunnel testing are both used to study the aerodynamiccharacteristics of the ducted fan MAV in various modes. The impact of various motion parameterson the aerodynamic characteristics of the ducted fan MAV is analysized. Momentum source itemsare used to replace the propellers in the numerical calculation. The dynamic derivatives of theducted fan MAV are numerically calculated based on unstructured overset grids. The resultsindicate that the aerodynamic characteristics of the ducted fan MAV is remarkably nonlinear andunsteady, it can hover, take off and land vertically, fly slowly in helicopter mode and fly rapidly inwing-borne mode, and it shows good aerodynamic performances throughout the whole flightenvelope.Research on nonlinear flight dynamic mathematical model of Micro Air Vehicles (MAV)plays an important role in MAV design. The flight dynamics of ducted fan MAV is remarkablynonlinear due to its big flight envelope, small size, low speed and special aerodynamicconfiguration. The flight dynamic mathematical model of MAV is established based on itsaerodynamic characteristics, its basic flight performance and trim result is calculated and thenonlinear flight dynamics in the whole flight envelop is studied, which provide the necessaryfoundation for the flight control system design.The traditional control methods are incompatible with the ducted fan MAV due to itsnonlinear and unsteady flight dynamics. In view of this condition, an adaptive flight controlsystem based on dynamic inversion and neural network is employed. The flight control system of the ducted fan MAV is composed of dynamic inversion controller, PD compensators, neuralnetwork compensators and pseudo-control compensators. The atmospheric disturbance is takeninto account in the simulation. Simulation results demonstrate that this control method has strongrobustness, stability and capability of following commands. Compared with the dynamicinversion-PID control system, the adaptive control system designed is more suited for the MAV.