Architecture for intelligent control systems

Abstract

Intelligent control, encompassing the theory and applications of both artificial intelligence and automatic control, is discussed. Three types of software environment and architecture have been developed: single expert systems that only process symbolic information and assist control engineers in the decision-making process for design and offline monitoring; coupling systems that simply combine numerical computation programs with an expert system so that it can be used to solve some engineering problems; and large knowledge-integration environments that can integrate different expert systems and numerical packages to solve complex problems. The authors point out that building an expert system is not just a translation from the existing knowledge into a computer program. It is a process in which new expert knowledge can be generated and acquired. >