Active Disturbance Rejection Flight Control For Quadrotor UAV

With an unprecedented progress in system design and flight control methods,the quadrotor UAV is more and more popular in both military and civil fields in recent years.Aiming at the multi-variable,non-linear,under-driven,and strongly coupled character of the quadrotor UAV,designing high-performance flight controller is the key to achieving stable flight of the quatrotor UAV.PID is the major flight control algorithm of the quadrotor UAV at this stage.Due to the influence of various random disturbances in real flight,the traditional PID control method cannot meet system dynamic performance index,and it is hard to achieve the high-performance flight control effects.Aiming at the anti-interference problem of quadrotor UAV,the flight control system based on Active Disturbance Rejection Control(ADRC)was studied to improve the ability of UAV to suppress interference during flight and improve the robustness and adaptability of the control system.Firstly this paper mainly focuses on the quadrotor UAV QBall2 as the research object.Based on the analysis of flight principles of QBall2,the non-linear system model of QBall2 is established based on Newton-Euler method.A non-linear cascade PID controller was designed and a flight control system MATLAB/Simulink simulation model was set up.The numerical simulation of the controller was performed and the simulation results showed the effectiveness of the cascade PID control method.Aiming at the poor anti-interference ability of PID control,the quadrotor UAV flight control system based on ADRC is studied and compared with the cascade PID control system.Simulation analysis results show that the ADRC controller can not only achieve fast and stable target tracking,but also can well estimate and compensate the internal and external disturbances received by the system in real time,and has strong robustness and adaptability.Secondly,for the complex and intrinsically unstable system of UAV,the adaptive feedback control law of the traditional ADRC controller has poor self-adaptability.The Extended State Observer(ESO)with fixed parameters will have a problem of “disturbance” with low estimation accuracy and compensation effect.For the problem,this paper combines fuzzy control,neural networkcontrol and ADRC to design a fuzzy ADRC flight control system and a neural network ADRC flight control system.Simulation results show that the improved ADRC method can significantly overcome the above problems and make the system have good stability and adaptability.Finally,the cascade PID controller and the ADRC controller were transplanted to the QBall2 UAV experimental platform to perform a semi-physical simulation to verify the effectiveness of the control method studied.At the same time,self-made quadrotor UAV and outdoor flight verification to assess the control method of this study.Flight data shows that the ADRC is more robust to disturbances in the actual flight environment.