Development of a fuzzy rule base design algorithm using a genetic algorithm and an inverted pendulum
Nikita Marushchenko, Становов Владимир Вадимович · ITM Web of Conferences · 2025
This paper presents the development of a fuzzy rule base construction algorithm that leverages a genetic algorithm to optimize control parameters for complex dynamic systems. The algorithm was tested on an inverted pendulum model, a classical nonlinear control problem, to identify chromosome configurations capable of stabilizing the pendulum in an upright position. Through extensive experimentation, the program successfully discovered such chromosomes that stabilized the pendulum from a variety of initial positions, including inverted (top-down) states, by forming the needed fuzzy rules base and fuzzy input and output terms. The results demonstrate the efficiency of the proposed approach in automating the design of fuzzy logic controllers using evolutionary methods.