ABSTRACT
The term artificial neural network (ANN) has come to mean any computer architecture that has massively parallel interconnections of simple processing elements. As an area of research it is of great interest due to its potential for providing insights into the kind of highly parallel computation performed by physiological nervous systems. Research in the area of artificial neural networks has had a long and interesting history, marked by periods of great activity followed by years of fading interest and revival due to new engineering insights [1]–[8], technological developments, and advances in biology. The latest period of explosive growth in pure and applied research in both real and artificial neural networks started in the 1980s, when investigators from across the scientific spectrum were attracted to the field by the prospect of drawing ideas and perspectives from many different disciplines. Many of them also believed that an integration of the knowledge acquired in the different areas was possible. Among these were control theorists like the first author, who were inspired by the ability of biological systems to retrieve contextually important information from memory, and process such information to interact efficiently with uncertain environments. They came to the field with expectations of building controllers based on artificial neural networks with similar information processing capabilities. At the same time they were also convinced that the design of such controllers should be rooted in the theoretical research in
of Complex
progress in different areas of mathematical control theory, such as adaptive, learning, stochastic, nonlinear, hierarchical, and decentralized control.
