Tuning rough controllers by genetic algorithms

Abstract

A design process of intelligent control systems, called rough logic government, consists of a sequence of transformations of mathematical models of control systems. It is a modification of the design process of fuzzy controllers using rough sets, rough logic, and evolutionary computing. Rough logic government starts with a symbolic model, called rough linguistic model, and concludes with tedious experiments. The final mathematical structure is called rough verification and validation model. In this paper, genetic algorithm is used to partially automate the final step, and experiments from various angles are conducted and analyzed.