Abstract
In this research, a fuzzy controller system is used as a key technique to design a machine controller. A fuzzy system can be designed using fuzzy sets, which can be represented using linguistic descriptions of natural language. These natural languages are able to capture human knowledge in fuzzy rules. A fuzzy controller system consists of fuzzy rules and the fuzzy inference techniques. Recently, research on fuzzy control has demonstrated the benefits of machine control using fuzzy rules. Through our research, we were able to verify that fuzzy control can be applicable to a line tracing car, which is running at a high speed. When carrying out our research, the adjustments and setup of the fuzzy controller required much time. In order to apply another Fuzzy controller to our line tracing car, we designed a simple inference system, and setup a Differential Evolution (DE) as a search algorithm. The DE was able to search a candidate solution of the fuzzy label. In addition, the auto-tuned fuzzy inference system can be controlled smoothly. In this paper, we present the search results of the simulation experiment and compare the results between the fuzzy controller and the auto-tuned controller using a DE.
