|Table of Contents|

[1] Zhang Jinlong, Yu Lingling, Liu Jingnan, et al. Ultra-precision positioning control technique based on neural network [J]. Journal of Southeast University (English Edition), 2006, 22 (1): 130-133. [doi:10.3969/j.issn.1003-7985.2006.01.028]
Copy

Ultra-precision positioning control technique based on neural network()
基于神经网络的超精密定位控制
Share：

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:
22
Issue:
2006 1
Page:
130-133
Research Field:
Automation
Publishing date:
2006-03-20

Info

Title:
Ultra-precision positioning control technique based on neural network
基于神经网络的超精密定位控制
Author(s):
Zhang Jinlong1;  2;  Yu Lingling1;  Liu Jingnan1
1 Department of Automatic Control Engineering, Southeast University, Nanjing 210096, China
2 College of Electrical and Automatic Engineering, Nanjing Normal University, Nanjing 210042, China
张金龙1;  2;  余玲玲1;  刘京南1
1东南大学自动控制系, 南京 210096; 2南京师范大学电气与自动化工程学院, 南京 210042
Keywords:
Moiré signals;  ultra-precision alignment;  neural network;  intelligent control
莫尔信号;  超精密定位;  神经网络;  智能控制
PACS:
TP274+.5
DOI:
10.3969/j.issn.1003-7985.2006.01.028
Abstract:
Due to the non-linearity behavior of the precision positioning system, an accurate mathematical control model is difficult to set up, a novel control method for ultra-precision alignment is presented.This method relies on neural network and alignment marks that are in the form of 100 μm pitch gratings.The 0-th order Moiré signals’ intensity and its intensity rate are chosen as input variables of the neural network.The characteristics of the neural network make it possible to perform self-training and self-adjusting so as to achieve automatic precision alignment.A neural network model for precision positioning is set up.The model is composed of three neural layers, i.e. input layer, hidden layer and output layer.Driving signal is obtained by mapping Moiré signals’ intensity and its intensity rate.The experimental results show that neural network control for precision positioning can effectively improve positioning speed with high accuracy.It has the advantages of fast, stable response and good robustness.The device based on neural network can achieve the positioning accuracy of ±0.5 μm.
针对精密定位装置存在非线性, 精确数学模型难于建立的缺陷, 提出了精密定位的神经网络控制方法.将BP神经网络应用于该控制系统中, 系统以光栅常数100 μm的光栅为定位标记, 以激光衍射产生的莫尔光光强及光强的变化率为神经网络的输入变量, 利用神经网络的自学习功能进行精密定位控制.建立了精密定位的神经网络控制模型, 模型由输入层、隐层和输出层3层神经元组成, 通过对光强及光强变化率的映射, 得到电机驱动信号.实验结果表明, 使用神经网络控制, 控制响应快, 稳定性好, 鲁棒性强, 可有效改善控制质量, 提高定位速度, 系统可获得±0.5 μm的定位精度.

References:

[1] Zhang Jinlong, Cheng Congyan, Yu Lingling.Ultra-precision alignment technique based on modified Moiré signals [J].Journal of Southeast University: English Edition, 2005, 21(1):16-19.
[2] Akihito M, Furuhasi H.Influences of the inclination of gratings on the alignment accuracies in Moiré alignment systems[J].Electrical Engineering in Japan, 2002, 139(2):46-51.
[3] Kanjilal A K, Narain R, Sharma R, et al.Automatic mask alignment without a microscope [J].IEEE Transactions on Instrumentation and Measurement, 1995, 44(3):806-809.
[4] Itoh J, Kanayama T.Optical heterodyne detection of mask to wafer displacement for fine alignment[J].Japanese Journal of Applied Physics, 1986, 25(8):684-686.
[5] Singh B P, Goto T, Sharma R, et al.Tracking and dynamic control of the angular alignment position in a photolithographic mask aligner by Moiré interference technique[J].Review of Scientific Instruments, 1995, 66(3):2658-2660.
[6] Liu Jingnan, Furuhashi H, Uchida Y.Automatic mask alignment in the θ direction using Moiré sensors[J].Nanotechnology, 1995, 6(4):135-138.
[7] Zhou L Z, Furuhasi H, Uchida Y.Sensing characteristics of a precision aligner using Moiré gratings for precision alignment system [J].Chinese Journal of Lasers B: English Edition, 2001, 10(4):257-262.
[8] Liu Jingnan, Furuhashi H.Precision position control system using Moiré signals[C]//Proceedings of the 1995 IEEE IECON Conference.Orlando, USA, 1995:968-972.
[9] Hara K, Nomura T, Kimura S, et al.An alignment technique using diffracted Moiré signals [J].Journal of Vacuum Science and Technology, 1989, B7(6):1977-1979.
[10] Uchida Y, Hattori S, Nomura T.An automatic mask alignment technique using Moiré interference [J].Journal of Vacuum Science and Technology, 1987, B5(1):244-247.
[11] Yan Pingfan, Zhang Changshui.Artificial neural networks and simulated evolution calculation [M].Beijing: Tsinghua University Press, 2000: 10-35.(in Chinese)

Memo

Memo:
Biographies: Zhang Jinlong(1965—), male, graduate, associate professor;Liu Jingnan(corresponding author), male, doctor, professor, liujn@seu.edu.cn.
Last Update: 2006-03-20