Abstract
This paper presents a new model for the control of multivariable output probability density function (PDF). Firstly, a Multi-Layer Perceptron (MLP) neural network is adopted to approximate the static output PDF of the MIMO systems. Nonlinear principal component analysis (NLPCA) is then used to reduce the order of the obtained static neural network model and the dynamics of the system is considered based on the lower-order model. After this, an integrated solution is provided to set up the system with lower-order dynamics for the purpose of stochastic distribution control. The controller design is then presented in detail. Finally, a simulation example is given to demonstrate the effectiveness of the method and encouraging results have been obtained.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Aström, K.J.: Introduction to stochastic control theory. Academic Press, New York (1970)
Yue, H., Wang, H.: Recent developments in stochastic distribution control: a review. Journal of Measurement and Control 36(7), 209–215 (2003)
Kárný, M.: Towards fully probabilistic control design. Automatica 32(12), 1719–1722 (1996)
Wang, H.: Bounded dynamic stochastic distributions: modelling and control. Springer, London (2000)
Wang, H., Zhang, J.H.: Bounded stochastic distribution control for pseudo ARMAX systems. IEEE Trans. on Automatic control 46(3), 486–490 (2001)
Girosi, F., Poggio, T.: Networks and the best approximation property Biol. Cybern. 63, 169–176 (1990)
Wang, H.: Robust control of the output probability density functions for multivariable stochastic systems with guaranteed stability. IEEE Trans. on Automatic Control 44(11), 2103–2107 (1999)
Wang, H.: Multivariable output probability density function control for non-Gaussian stochastic systems using simple MLP neural networks. In: Proc. IFAC Int. Conf. on Intelligent Control Systems and Signal Processing Algarve, Portugal (2003)
Kramer, M.A.: Nonlinear principal component analysis using autoassociative neural networks. AIChE Journal 37(2), 233–243 (1991)
William, W.H.: Nonlinear principal component analysis by neural networks. Tellus 53A, 599–615 (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wang, W., Xiong, Y.L., Wang, H., Yue, H. (2008). MIMO Probability Density Function Control Using Simple LOG-MLP Neural Networks. In: Xiong, C., Liu, H., Huang, Y., Xiong, Y. (eds) Intelligent Robotics and Applications. ICIRA 2008. Lecture Notes in Computer Science(), vol 5315. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-88518-4_88
Download citation
DOI: https://doi.org/10.1007/978-3-540-88518-4_88
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-88516-0
Online ISBN: 978-3-540-88518-4
eBook Packages: Computer ScienceComputer Science (R0)