Wind disturbance rejection control based on the dynamic parameters estimation of quadrotors UAV

The turbulence or gust in quadrotor flight environment causes drastic changes in the unmanned aerial vehicle(UAV) aerodynamic parameters. Especially, rotor thrust coefficient and reaction torque coefficient of the UAV encounter uncertainty fluctuation, which may undermine the control performance. A real-time estimation strategy for these aerodynamic parameters is proposed to improve the identification on the disturbance. First, the unscented Kalman filter(UKF) algorithm is used to estimate the UAV states and aerodynamic parameters. Then, a double-loop structure, consisting of position and attitude, is designed for the trajectory tracking control. In the outer loop, a proportional-derivative controller is adopted to carry out position tracking and provide Euler angle references for the inner loop, called attitude controller. Moreover, the attitude controller is designed using an inverse dynamic technique. The main contribution of this study is to propose a joint estimation on the aerodynamic parameters with wind disturbance as well as the UAV states. This strategy plays an important role in refining time-varying parameters of wind disturbance. A number of simulations are executed to verify the effectiveness of the proposed method.