Designing fuzzy logic controllers by genetic algorithms considering their characteristics
Seihwan Park, Hyung Lee-Kwang · 2002
Fuzzy control has been used successfully in many practical applications. In traditional methods, experience and control knowledge of human experts are needed to design fuzzy controllers. However, it takes much time and cost. In this paper, an automatic design method for fuzzy controllers using genetic algorithms, which considers their characteristics is proposed. Our method consists of two stages: rough designing stage and fine tuning stage. In rough designing stage, we determine the number and rough center positions of membership functions, and rule table. In fine tuning stage, exact positions of each membership function are obtained and the rule table is refined. Also, we proposed an effective encoding scheme and new genetic operators. The proposed genetic operators maintain the characteristics, such as the correspondence between membership functions and control rules. The proposed method is applied to a cart centering problem. The result of the experiment has been satisfactory compared with other design methods using genetic algorithms.