Intelligent control

Adaptive control and system identification of the standard types usually reduce to parameter estimation. Since the control law is ultimately a mapping from the measurement history to commands to the plant, it is intriguing to consider attempting to establish an appropriate such mapping without using the specialized structure that the standard methods use. Several techniques have been suggested which are motivated by biological notions and which, to a greater or lesser extent, result in self-learned or taught control algorithms. Among these are artificial neural networks, based on a simple brain-like structure;evolutionary computation, motivated by models of biological evolution;expert systems, which attempt to incorporate the operational rules used by knowledgeable controllers; andfuzzy systems, which try to represent inexact knowledge in a computer-compatible form.