The Research On Fractional Order PID Type Iterative Learning Control Strategy For Perfect Tracking Of Flexible Manipulator In Two-dimensional Plane

In recent years, many frontier research fields such as aerospace and medical sciences pay dramatically increasing attention on the performance of the manipulators. From a technical point of view, manipulators have been widely used in quite a few domains in practice.From a theoretical point of view, in the light of the characteristics of different manipulator, high complex dynamic system was introduced.With the constant improvement of the performance of the manipulator, the research of lightweight, high overload, high speed, high precision, high adaptability and the intelligent flexible manipulator has become an urgent demand now. The introduction of flexible manipulator meets these requirements.It has the characteristics of lightweight, low cost and high speed.Thus flexible manipulator occupies an important position in the field of robots. Especially in recent years, the study of flexible manipulator gained extensive attention in many fields.This thesis is main about the modeling and control of flexible manipulator as a platform. In order to realize the precise tracking of the control objective,the design of the fractional order PID-ILC controllers are studied. For the implementation and application of a practical flexible manipulator model.this thesis presents a firm-soft coupling model of manipulator in the two-dimensional plane. First of all, this article abstracts the flexible pipe and determines its absolute coordinate system and the relative coordinate system. Then, based on the Hamilton principle, a group of dynamic equations of flexible manipulator are established. By the technique of simplified modeling, we get the approximate dynamics of flexible manipulator dynamic equations. The comparisions of the finite element method and assumed modes method are present as well. Considering the number of degrees of freedom and the design of active control.this paper selects assumed modes method to obtained the approximate solution of dynamic equation. The design of the controller tolerates the parametric errors of models. That has significant importance in the controller design.As mentioned above, this paper establishes the dynamic equations of flexible manipulator,based on which, two strategies are applied to design the controller. Firstly, based on the combination of iterative learning control and PID feedback control to establish the PID type of iterative learning controller(PID-ILC). On the other hand,some fractional order methods are applied to establish discrete incremental fractional order PI controller. These two schemes are shown in Chapters 3 and 4, respectively,and to carry on the design and numerical simulations. Iterative learning method reduces the requirements of system accuracy and makes up for the inadequacy of the model of flexible manipulator. PID controller can effectively reduce uncertainty of the structure and parameters and the influence of unknown disturbances. The combination of both can reduce the system’s error to a great extent. The controller can improve the tracking accuracy on the entire working period. Effectiveness and reliability of the algorithm are proved by using MATLAB/SIMULINK. Discrete incremental FOPI controller of model can reduce the oscillation and uncertainty factors. The system improves the robustness and reduce the vibration.The highlights of this paper:● Fractional order controller has more adjustable parameters. And thus it should deserve more ideal results in multivariable system control.In particular, the theory and experiment have proved that the optimal iterative learning controller is fractional order in order as well.● Compare to the integer order controller system, fractional order iterative learning controller can be effectively implemented without increasing the amount of compulation. In fact, the fractional order iterative learning control algorithm is implemented in small module. That is integer order ones and can be implemented as the transfer function.● Due to the accurate model of flexible manipulator is difficult to obtain, and the design of the fractional order controller must rely on the system structure, this paper applied the flat phase technology for setting the parameters of the fractional order controller.This also proved that the fractional order controller for the parameters of the system model has certain adaptability.● The combination of the fractional order controller and PID iterative learning controller is applied to the trajectory tracking system of the flexible manipulator. When the input torque is equal to zero, the velocity of angular displacement and displacement is also zero. Fundamentally, the system reduces the vibration.