Parameter online identification of a small-scale unmanned aerial vehicle applying unscented kalman filter

To obtain the dynamic aerodynamic derivatives which are difficult to obtain through the wind tunnel experiments, and to solve the issues of the strong nonlinear characteristics of small-scale unmanned aerial vehicle (SUAV), it is proposed that the parameter estimation method based on unscented kalman filter (UKF) utilizing the flight data. The augmented nonlinear state equations are established in terms of parameters which to be identified, and the nonlinear model of SUAV based on the piston engine is built. The UKF formulation is constituted by the augmented nonlinear model. The UKF method is applied to identify the aerodynamic derivatives by flight data. The simulation results show that the UKF estimation method is suitable for the on-line estimation of aerodynamic derivatives within the nonlinear model of SUAV.