Learning rules for neuro-controller via simultaneous perturbation

作者: Y. Maeda , R.J.P. De Figueiredo

DOI: 10.1109/72.623213

关键词:

摘要: This paper describes learning rules using simultaneous perturbation for a neurocontroller that controls an unknown plant. When we apply direct control scheme by neural network, the network must learn inverse system of In this case, know sensitivity function plant kind gradient method as rule network. On other hand, described here do not require information about function. Some numerical simulations two-link planar arm and tracking problem nonlinear dynamic are shown.

参考文章(23)
Yutaka Maeda, Yakichi Kanata, Learning Rules for Recurrent Neural Networks using Perturbation and their Application to Neuro-Control Ieej Transactions on Electronics, Information and Systems. ,vol. 113, pp. 402- 408 ,(1993) , 10.1541/IEEJEISS1987.113.6_402
Y. Maeda, H. Yamashita, Y. Kanata, Learning rules for multilayer neural networks using a difference approximation international joint conference on neural network. pp. 628- 633 ,(1991) , 10.1109/IJCNN.1991.170470
J.C. Spall, D.C. Chin, A model-free approach to optimal signal light timing for system-wide traffic control conference on decision and control. ,vol. 2, pp. 1868- 1875 ,(1994) , 10.1109/CDC.1994.411110
Wolfram H. Schiffmann, H. Willi Geffers, Original Contribution: Adaptive control of dynamic systems by back propagation networks Neural Networks. ,vol. 6, pp. 517- 524 ,(1993) , 10.1016/S0893-6080(05)80055-3
G. Lightbody, G.W. Irwin, Direct neural model reference adaptive control IEE Proceedings - Control Theory and Applications. ,vol. 142, pp. 31- 43 ,(1995) , 10.1049/IP-CTA:19951613
M. Khalid, S. Omatu, R. Yusof, MIMO furnace control with neural networks IEEE Transactions on Control Systems and Technology. ,vol. 1, pp. 238- 245 ,(1993) , 10.1109/87.260269
Y. Ichikawa, T. Sawa, Neural network application for direct feedback controllers IEEE Transactions on Neural Networks. ,vol. 3, pp. 224- 231 ,(1992) , 10.1109/72.125863