MODIFICATION METHOD OF CONTROLLED PERTURBATION
Keywords:
Neurocontrol, recurrent perceptron, dynamic objectsAbstract
Further development of control perturbation method is proposed. It allows to fully automate a neurocontroller training process and makes possible to control dynamic objects without their prior model development. Experiments that confirm efficiency of the method are described.References
K.S.Narendra, K.Parthasarathy Identification and control of dynamical systems using neural networks. In: IEEE Transactions on Neural Networks, 1, 1990, p. 4 – 27.
P.Werbos, Back propagation through time: what it does and how to do it. In: Proc. IEEE, Vol. 78, No. 10, October 1990.
M.I.Jordan and D.E.Rumelhart. Forward models: Supervised learning with a distal teacher. In: Cognitive Science, Vol. 16, pp.313 – 355, 1990.
H.T.Siegelmann, B.G.Horne, and C.L.Giles. Computational capabilities of recurrent NARX neural networks. In: IEEE Trans. Systems, MAN, and Cybernetics -. Part. B: Cybernetics, 1997, 27(2): 208 – 215.
M.T.Hagan, H.B.Demuth. Neural Networks for Control. In: Proceedings of the 1999 American Control Conference, San Diego, CA, 1999, pp. 1642 – 1656.
D.Prokhorov and D.Wunsch. Adaptive critic designs. In: IEEE Transactions on Neural Networks, 8(5), 1997, p. 997 – 1007.
Д.А.Дзюба, А.Н.Чернодуб. Применение метода контролируемого возмущения для модификации нейроконтроллеров в реальном времени. Математические Машины и Системы, 2011. № 1. с. 20 – 28.
Д.А.Дзюба, А.Н.Чернодуб. Обучение рекуррентной нейронной сети методом контролируемого возмущения для управления динамическими объектами. Proceedings of the 2010 Knowledge – Dialogue – Solution Conference, Sep. 06 – 10, Kyiv (Ukraine).
M.Casini, D.Prattichizzo, A.Vicino. The Automatic Control Telelab. A Web-based technology for Distance Learning // IEEE Control Systems Magazine. 2004. N 24 (3). – P. 36 – 44.
