Design and optimization of fuzzy controllers based on the operator's knowledge

Industrial processes are normally operated by skilled humans who have cumulative and logical information about the systems. Fuzzy control has been investigated for many applications. Intelligent control approaches based on fuzzy logic might be able to include human thinking. This article describes a modeling approach based upon an operator's knowledge without a mathematical model of the system and optimization of the controller. The test system applied was constructed to send a ball into the goal position using wind from two DC motors in a predefined path. A vision camera to mimic human eyes detects the ball's position. The system used in this experiment was difficult to model by mathematical methods, and could not easily be controlled by conventional methods. The controller is designed based on the input-output data as well as experimental knowledge obtained by trials, and optimized under predefined performance criteria.