Dynamic systems control and identification using VSC-based learning algorithms for perceptron networks

In this paper a set of Variable Structure Control (VSC)-based on-line learning algorithms for continuous time two layer and three layer perceptron networks with non-linear and linear activation functions are presented. The proposed algorithms result in a temporal learning capabilities of a neural network with dynamically adjusted weights, and zero convergence of the learning error in a finite time. These learning algorithms are used with identification and control schemes for linear and non linear dynamic systems.