Research Of Nonlinear Active Disturbance Rejection Control With Applications To Unmanned Aerial Vehicle

UAV is a special air force aircraft, characterized by the major feature “no pilot”.Due to such feature; UAV is known as " wing of the crack troops " in the militaryfield of the21th century, In the local wars of modern times, the optimal choice ofreducing the cost of war is to reduce or even avoid casualties; UAV is graduallyrising in the competation of weapons of the present time due to its outstandingadvantages. In order to meet better the requirement of war under the condition ofinformaliztion; it is extremely urgent to improve the performance of flight controlfor UAV. Considering the fact that UAV conduct autonomous flight throughaerodynamic force to overcome its own weight; UAV is inevitably to suffer from allkinds of disturbance. Obviously, the tradition control method of linearization; oftenused for the elimination of interference errors, no longer apply here. Starting fromthe principles of ADRC, this study mainly concentrate on UAV flight Controlstrategy using ADRC, and try to solve the problem of declining in stability andmaneuverability when using nonllinear ADRC to deal with CONTROL PROBLEMSOF UAV with all kinds of disturbances.Firstly, nonlinear mathematical model of UAV flight control system with sixdegrees of freedom was built on the base of understanding of motion characteristicsof UAV, and the nolinear model is simplified to a linear one based om smalldisturbance assumption. Based on this analysis of kinematics characteristics oflongitudinal direction and Lateral longitudinal direction of UAV; the kinematicsmodel of longitudinal and Lateral longitudinal are built and the key point of thepaper is establised:A reasonable math model of UAV with ideal performance andeffective suppression of interference is bult, and the foundation for further study islaid.Secondly, an in-depth study of the ADRC method is made, and the structure,the function and the working priciple of the controller is clearified. Stability of thesecond order extended state observer (ESO) is demonstrated through method ofLyapunov and the necessary and sufficient condition of the maintainance of thestability is clearified. The evaluation criteria of error convergence of ESO is givenby error anlysis. Design mehtod of ESO and principles for parameter selection isgiven based on filter of feed back structure of fal funcion.Thirdly, aiming at the climb/decline phase of UAV; the reference movement istaken as the non-tilting and non-sliding flight of UAV, the attitude angle trackingloop is designed and a simulation platform is built. Through the design of controlloop and the seeking of reasonable control parameters, the pitch attitude controlproblem under the disturbances of air current and gas stream and is completed anda combination control strategy of ADRC and PI is proposed. Through the analysis of simulation results it is seen that ADRC do not need accurate model of theobjector to be controlled or the external disturbance, and can achieve fast, accuratetracking of the pitch angle of UAV, thus prove ADRC has strong anti-interferenceability and high robustness.Finally, by analyzing the motion characteristics of longitudinal direction flightof UAV, and according to the principle that ESO of ADRC can estimate andcompensate for the uncertainty in the system. On the simulation platform oflongitudinal motion of UAV; and through the decoupling of longitudinal directionloops by ESO of ADRC, the tuning of parameters of ADRC is carried out and thecontrol law is designed. On this based, control of the longitudinal motion of UAV isachieved, and a thorough analysis of the simulation results is given, and advantagesand disadvantages. Of applied ADRC method to control the stability of UAV issummarized...