Design And Analysis Of Control Algorithms For A Quadrotor UAV

The quadrotor UAV is a kind of the multi-rotor UAVs which has the following advantages:(1)simple mechanical structure;(2)the capability of taking off and landing without runway;(3)the capability of hovering in the air after takeoff,etc..Because of these advantages,the quadrotor UAV is widely used.But the quadrotor UAV is also a complex system,which mainly reflected in the nonlinear and high coupling of the system.During the flight,there are some uncertainties as follows:(1)external perturbation;(2)parameter perturbation;(3)unmodeled dynamics.Therefore,it is of great importance to do further on the controller design by taking these uncertain factors into account.Based on the dynamic model and flight characteristics of the quadrotor UAV,the basic structure and control principle of the four rotor UAV control system are described.For the quadrotor UAV,the robustness of the controller designed with traditional PID algorithm is not satisfied.In view of this problem,feedforward and nonlinear compensation strategy are employed to improve the performance.Simulation results show that the Forward-PID control algorithm has less overshoot,shorter regulation time and faster response speed than the traditional PID control algorithm;On the other hand,during the flight of the quadrotor UAV,there are some uncertain problems.Based on the adaptive control theory,an adaptive backstepping controller is designed for the three channels in the attitude loop by using decentralized control strategy,the simulation results illustrate verify the effectiveness of the proposed control scheme;In addition,in the process of controller design for attitude loop,most exiting algorithms assume that the attitude angle and angular velocity are simple differential relationships,which is only applicable to small attitude angle.In this thesis,we remove this assumption.First,the attitude model is decoupled into three independent subsystems by the dynamic inverse algorithm in this paper.Then,the decentralized control strategy is adopted to design asliding mode variable structure based on reaching law for the decoupled subsystem by which the closed loop can b designed.Simulation results illustrate the effectiveness of the hybrid control scheme.Finally,in order to further verify the tracking performance of the three control schemes under given nonlinear trajectory,a hovering simulation experiment is designed,and the simulation results are compared and analyzed.The results of simulation experiments are compared and analyzed.The results show that the control scheme designed in this paper are effective,the scheme with dynamic inverse algorithm and the sliding mode variable structure based on reaching law has better dynamic performance in this flight scenario.