Application Of Improved Iterative Learning Control In Atomic Force Microscopy System

We can obtain the true three-dimensional morphology of a nano sample according to the interaction between atoms by AFM,At present,commercial AFM systems usually use traditional PI control algorithms to realize the imaging of sample scanning.However,the AFM system has complicated operation and nonlinear characteristics of piezoelectric ceramics.makes it difficult for traditional PI controllers to obtain clear and complete sample images.improving and optimizing the control algorithm is one of the key technologies to achieve reliable high-precision AFM scanning imaging.This paper aims at the shortcomings of the measurement accuracy of PI controller.The simulation platform of AFM system is established by homemade AFM system,the influence of control parameters on AFM system is studied,and set the system experience control parameters.Based on the repetitive scanning motion of the piezoelectric ceramic displacement platform,which meets the requirements of the iterative learning control algorithm,an openloop PID-type iterative learning control algorithm and a closed-loop PID-type iterative learning control algorithm are designed.Two kinds of control are studied through simulation experiments the tracking performance of the algorithm on the desired signal.For non-linear complex AFM systems,the adaptive ability of the conventional PID-type iterative learning algorithm is obviously insufficient.by combining fuzzy adaptive control with closed-loop PID-type iterative learning control and traditional PI control,design fuzzy adaptive closedloop iterative learning PI controller(FAIT-PI controller for short),The simulation results show that the control parameters obtained by the fit of FAIT-PI controller have higher precision for the desired signal scanning and tracking.The experimental results show that the FAIT-PI controller can improve the imaging quality of grating samples,thus improving the accuracy of nano-measurement and manipulation.