Research On Attitude Calculation And Flight Control Of Quad-rotor UAV Based On Anti-External Disturbance

Due to many advantages of small size,flexible maneuverability,strong load capacity,low cost and so on,quad-rotor UAV is widely used in civil and military fields.Quad-rotor UAV is an under-actuated,strongly coupled nonlinear system,when performing special tasks,it will be interfered by motion acceleration and magnetic fields,which will have a serious impact on the accuracy of attitude calculation.Moreover,it is extremely susceptible to wind disturbance during flight,which affects robustness of the system and trajectory tracking accuracy.In order to improve the ability of attitude calculation to suppress motion acceleration and magnetic disturbance and anti-wind performance of flight controller,this paper has conducted immunity research on two aspects of attitude calculation and flight control.The main content of the paper is as follows:(1)For the under-actuated,strong coupling and nonlinear characteristics of quad-rotor UAV system,the Newton-Euler method is used to derive and establish the nonlinear dynamic model and kinematic model.In order to study anti-wind performance of quad-rotor UAV flight controller,the wind shear and Dryden atmospheric turbulence are modeled.In order to obtain dynamic model with wind disturbance factor,the two are superimposed and introduced into the nonlinear dynamic model by aerodynamics.(2)Considering the error of MEMS sensors,error models are established and the error calibration is performed.Due to the inaccurate attitude estimation of single sensor,in order to improve the accuracy of attitude calculation,the attitude calculation algorithms based on principles of complementary filtering and gradient descent are respectively designed.Considering that the fixed step affects the performance of the gradient descent method,Nesterov acceleration gradient is used to fuse 9-axis sensor data,at the same time,in order to improve the ability of attitude calculation to suppress the motion acceleration and magnetic disturbance,combining with fuzzy control,the weight suppression factor is dynamically adjusted according to the attitude error.The results of the four sets of comparative experiments show that the proposed algorithm improves the ability to suppress motion acceleration and magnetic disturbance,and also improves the speed of attitude calculation.(3)To solve the problem of under-actuated,quad-rotor UAV is decomposed into inner and outer ring subsystems.The conditional constraint method is used to simplify the dynamic model,the classic PID control principle is analyzed and the PID controller is designed.Considering the nonlinearity and internal and external disturbance of quad-rotor UAV system,the sliding mode control principle is analyzed and the sliding mode controller is designed.Trajectory tracking simulation results under wind disturbance show that the PID controller has poor robustness,but the control input is small and smooth,and the sliding mode controller has strong robustness,but the system chattering is severe.(4)For the chattering in sliding mode control,a double power approach law with better motion quality is used to suppress chattering.Considering that ordinary sliding mode controller may has steady state errors when subjected to wind disturbance,integral term is introduced into the traditional sliding mode surface to form integral sliding mode surface.In order to improve the ability of system to resist winddisturbance,an extended state observer is introduced to perform real-time observation and feedback compensation of uncertain internal disturbance and external wind disturbance,this also reduces the gain coefficient of the switching function at the same time,further reducing chattering.Trajectory tracking simulation results under wind disturbance show that the designed controller reduces the system chattering,improves the trajectory tracking accuracy and the ability of system to resist wind disturbance.