Swing-Attenuation And Flight Control Methods For Quadrotor UAV With A Cable-Suspended Payload

The transportation by UAV hanging a load is an efficient and very important approach for load transportation because of UAVs have flexible maneuverability,not been restricted by geographical conditions and load shape.However,the swing of the hanging load will have a very adverse effects on the stability of the flight control system,which will bring security problems to the control system.Aiming at sloving these problems,in this thesis,the antiswing flight control methods are researched to ensure the rapidity,stability and safety of the UAV slung flight.Specific research contents include:(1)The dynamics of UAV are obtained by the Newton-Euler method,and the motion equations of hanging load are given by the Lagrange equation,and all the models provide the model basis for controller design.(2)Aiming at the problem that active disturbance rejection control(ADRC)has many parameters to be tuned difficultly,the backpropagation(BP)neural network is used to adjust the parameters of the extended state observer(ESO)online to reduce the difficulty of ADRC parameter adjustment,and improve the estimation accuracy of ESO and the ability of rejecting disturbances of the controller.The simulation results show that,based on the improve ADRC controller BP neural network based,the flight control systemhas not only the stable ability of trajectory tracking,but also the improved ADRC controller is easy to be tuned for there are there parameters less than that of the traditional ADRC controller.When the UAV tracks the way-point,the flight control performances of UAV under the improved ADRC control are better than that of UAV controlled by the traditional ADRC controller.Meanwhile,the improved ADRC also has better the ability of rejecting disturbances than the traditional ADRC controllers.Therefore,by the improved ADRC control,the four-rotor UAV control system has fast response,small overshoot and strong rejecting disturbance abilities.(3)In order to suppress the load oscillation,the antiswing control methods based on input shaping,that is Zero Vibration(ZV)shaper,Zero Vibration and Derivative(ZVD)shaper and Extra Insensitivity(EI)shaper are investigated..With the ZV,ZVD and EI input shaper,when the UAV track the rectangular trajectories,the maximum swing angle suppression ratio are45.33%,54.85% and 59.14%,respectively.In simulation experiments for UAV flying along the way-point,the load mass are 0.3kg,0.5kg and 0.7kg,respectively.The simulation results show that when the load mass is increased,the swing amplitude of load is decreased,and hence the swing time is decreased.In these three input shapers,EI shaper has the strongest anti-swing ability,while the anti-swing ability of ZV shaper is weakest.Changing the rope length from 1m to 2m,the simulation shows that at this process,the load residual swing increases,but the these three input shapers can still effectively suppress the load swing,and EI shaper has better ability in suppress the load swing than ZVD shaper,while the suppression ability of ZV shaper is weakest.(4)The UAV physical platform is built,and based on the open source flight control,the suspended flight control system is designed and the performances of the anti-swing control system is verified by flight experiments.The experimental results show that the maximum load swing angle is reduced by 56.72%,which indicates that the EI input shaper can effectively suppress the load swing.