Effect of cell map granularity on fuzzy control system analysis

Cell mapping is a simulation technique for analyzing the global performance characteristics of control systems. It simplifies the task of analyzing a continuous phase space by partitioning it into a finite number of disjoint cells. Each cell comprising an infinite number of states is abstracted by its center-point. Continuous system trajectories are approximated by cell transition sequences. Lumping system states into cells and abstracting cells by their center-points can produce significant errors. This paper investigates the effect of cell granularity on various types of cell mapping errors. The results provide useful guidelines for applying cell mapping techniques to fuzzy controller design and evaluation.