Research On Tracking Control Of Flexible Manipulator Based On Iterative Learning

With the development of science and technology and artificial intelligence,the characteristics of light weight,high speed,precision,and flexibility of flexible manipulators have attracted increasing attention of scholars.However,the flexible manipulator is an infinite-dimensional distributed parameter system,whose too large system dimensionality will make it difficult to establish accurate dynamic model.Moreover,the existence of the flexible structure of the flexible manipulator will cause deformation of the connecting rod during the trajectory tracking control process,and rod deformation and rigid-flexible coupling increase the difficulty of controller design.Therefore,it is of great research significance to design a suitable controller for the trajectory tracking of a flexible manipulator.For this research aim,in this paper,the author designs an iterative learning controller combined with a two-link flexible manipulator system,which is dedicated to solving several defects in the iterative learning control of flexible manipulator trajectory tracking,which are summarized as follows:1.To solve the problem that the flexible manipulator can realize trajectory tracking after multiple iterative learning only under the action of traditional iterative learning control algorithm,a single neuron D-type iterative learning control algorithm was proposed.The specific method is: First,find a correlation function that conforms to the nature of a single neuron in the time interval,and use the inter-correlation nature of single neurons to predict the relationship between the error information generated at each discrete time point in the time interval.The information obtained by the association is superimposed with the control quantity generated by the previous iteration process,and a new control quantity is generated together as the control input of the next iteration.Finally,in the MATLAB simulation environment,compared with the traditional D-type iterative learning control algorithm,the simulation verified the effectiveness of the algorithm,effectively reducing the times of iterative learning of the flexible manipulator.2.To solve the problem that the flexible manipulator caused the failure of the traditional iterative learning controller due to the existence of the flexible link and the change of the initial state of the system,iterative learning control algorithm was put forward under the condition of a random initial state deviation,which can control its adverse effect.The specific method is: on the basis of the traditional iterative learning control algorithm,the time interval is divided into two sub-intervals.One sub-interval effectively suppresses the adverse effects caused by the initial state deviation of the system in the iterative learning control process,and the other one tracks the desired trajectory.Finally,the effectiveness of the algorithm is verified in the MATLAB simulation,which can effectively reduce the influence of the initial state random deviation on the trajectory tracking control of the flexible manipulator.