<title>Neural variable structure controller for telescope pointing and tracking improvement</title>

Abstract

Recently, neural network models (NN), such as the multilayer perceptron (MLP), have emerged as important components for applications of adaptive control theories. Their intrinsic generalization capability, based on acquired knowledge, together with execution rapidity and correlation ability between input stimula, are basic attributes to consider MLP as an extremely powerful tool for on-line control of complex systems. By a control system point of view, not only accuracy and speed, but also, in some cases, a high level of adaptation capability is required in order to match all working phases of the whole system during its lifetime. This is particularly remarkable for a telescope control system. In fact, strong changes in terms of system speed and instantaneous position error tolerance are necessary. In this paper we introduce the idea of a new approach (NVSPI, neural variable structure PI) related to the implementation of a MLP network in an Alt-Az telescope control system to improve the PI adaptive capability in terms of flexibility and accuracy of the dynamic response range.