Modares Mechanical Engineering

Modares Mechanical Engineering

Design of an Observer for an XY Nano-Positioner using Neural Network.

Document Type : Original Research

Authors
Saeid Bayat 1
Hosein Nejat Pishkenari 2
Hasan Salarieh 3
1 MSC student of mechanical engineering at Sharif University of technology
2 Associate professor of mechanical engineering at Sharif University of technology
3 Associate professor, Mechanical Department, Sharif University of Technology, Tehran, Iran
Abstract
Nowadays, nano-precision positioning stages, have a special position and are used in a variety of applications, such as taking pictures and taking particles of the surface. In this paper,some observers for a nano-precision positioning platform are designed based on three different types of neural networks. The simulated platform was designed at Sharif University of Technology and, based on the system's final requirement for the feedback signal for use in the control rule, neural network observers were designed. In previous studies, the comsol model of the positioning system has been obtained. At this step, the neural network has used the Comsol model and the system has been trained for a sum of a number of sinusoidal functions, and its generalizability has been investigated for ramp input. Neural networks used include, respectively, a multi-layer perceptron network, a radial basis function network and a support vector regression network. By performing simulations, it has been seen that the multi-layer perceptron network and the radial basis function network yielded a good response with low error, but the support vector regression network has a relatively high error.
Keywords
Nano-positioner
observer
Neural Network

Subjects

[1] Y. Qin, Y. Tian, D. Zhang, W. Gao, B. Shirinzadeh, U. Bhagat, et al., Motion control of a 2-DOF decoupled compliant mechanism using H∞ synthesis, in Manipulation, Manufacturing and Measurement on the Nanoscale (3M-NANO), pp. 222-227, 2012.
[2] H. C. Liaw, B. Shirinzadeh, and J. Smith, Robust neural network motion tracking control of piezoelectric actuation systems for micro/nanomanipulation, IEEE Transactions on Neural Networks, vol. 20, pp. 356-367, 2009.
[3] Q. Xu and Y. Li, Radial basis function neural network control of an XY micropositioning stage without exact dynamic model, in Advanced Intelligent Mechatronics, pp. 498-503, 2009.
[4] M. Goldfarb, N. Celanovic, Modeling piezoelectric stack actuators for control of micromanipulation, IEEE Control Systems Magazine, Vol. 17, No. 3, pp. 69-79, 1997.
[5] H. B. Demuth, M. H. Beale, O. De Jess, and M. T. Hagan, Neural network design: Martin Hagan, 2014.
[6] D. W. Ruck, S. K. Rogers, M. Kabrisky, M. E. Oxley, and B. W. Suter, The multilayer perceptron as an approximation to a Bayes optimal discriminant function, IEEE Transactions on Neural Networks, vol. 1, pp. 296-298, 1990.
[7] J. Park and I. W. Sandberg, Universal approximation using radial-basis-function networks, Neural computation, vol. 3, pp. 246-257, 1991.
[8] H. Drucker, C. J. Burges, L. Kaufman, A. J. Smola, and V. Vapnik, "Support vector regression machines," in Advances in neural information processing systems, pp. 155-161, 1997.
Modares Mechanical Engineering
Volume 18, Issue 6
October 2018
Pages 139-143