Design of fuzzy controller with synthesized rule base

Although the requirement of completeness is not necessary, most engineers always build up a rule base comprehensively due to the lack of knowledge about the physical properties of systems. This kind of rule base for fuzzy systems is therefore redundant and usually causes needless calculations in fuzzy inference. It is worthwhile, therefore, to refine the complete rule base by some systematic and reasonable strategy. In this paper, a rule base reduction method is proposed which is inspired by the methods of gates minimization in logic design. An experiment is conducted on an inverted pendulum for its apparent property of nonlinearity. The experimental results show that the control behavior of the fuzzy controller with a simplified rule base is still similar to that of a comprehensive rule base