Chapter 1 - Introduction: Neural Networks and Automatic Control

Abstract

This chapter discusses the relationship between control systems and neural networks. In order to use mathematical or computational methods to improve the control system's response to its input command, the plant and the feedback controller are modeled mathematically by differential equations, difference equations, or by a neural network with internal time lags. The motivation for control system design is often to optimize a cost, such as the energy used or the time taken for a control action. Control designed for minimum cost is called optimal control. This can be achieved by the use of artificial neural networks. They can efficiently approximate and interpolate multivariate data that might otherwise require huge databases. One of the most promising current applications of neural network technology is to “intelligent sensors,” or “virtual instruments.” Other techniques of intelligent control, such as fuzzy logic, can be combined with neural networks as in reconfigurable control. Most control applications of neural networks currently use high-speed microcomputers, often with coprocessor boards that provide single-instruction, multiple-data parallel computing well suited to the rapid functional evaluations needed to provide control action.