Research On Nonlinear Attitude Controller For UAV

In recent years, the unmanned aerial vehicle (UAV) has enjoyed high popularity asit is widely used in many areas, such as military and scientific research. Higherstandards have been brought forward to the UAV control which is the basis of efficientperformance. The UAV is a multiple-input multiple output (MIMO) system, and theconventional linear design method of control system can no longer satisfy the needs ofactual needs as the increasing expectations on UAV. The paper studies the nonlinearcontrol method of the UAV.The paper first establishes a six-degrees of freedom mathematical method, and thenconduct simulation analysis on the state changes of the UAV system under uncontrolledconditions, drawing a conclusion that the system is unstable. Furthermore, this paperstudies the nonlinear dynamic inversion method and analyzes the mathematical model,and designs a UAV inside and outside loop dynamic inversion attitude controller.Meanwhile, it simulates the UAV attitude control, and analyzes the multivariabledecoupling effect and robustness of the controller, concluding that precise linearizationcontrol can be achieved under accurate model. But there are still the shortcoming ofpoor robustness for the controller. Faced with the defect of nonlinear dynamicinversion, dynamic inversion combined with fractional order PID control method isproposed. Fractional order PID is the combination of the traditional PID and fractionalorder calculus, and this paper studies the fractional order Oustaloup digital realizationalgorithm, and compare its advantages over traditional PID. Simulation shows that bythe method of the fractional order PID combining with dynamic inverse control, thesystem robustness is improved obviously.In the end, this paper studies ADRC attitude controller for UAV. First，the paperstudies and analyzes the tracking differentiator，the extended state observer and themultivariable decoupling control method. Then, it designs the inner and outer loopcontrol for the UAV and realizes the multivariable decoupling control through thefeedback linearization. Through analyzing the rapidity, accuracy and the robustnesswhich is under interference of wind and parameter perturbation, of the system bysimulation, it shows that the control system performs well in all aspects and meets thecontrol requirements.