A systematic design method for fuzzy logic control with application to automotive idle speed control

A systematic design procedure for automatic rule generation for dynamic systems such as a nonlinear engine dynamic model is investigated. By 'automatic rule generation' the authors mean clustering or collection of such meaningful transitional relations from one conditional subspace to another. Elements that result from such transitions are anticipated according to the applied action of each rule. Data required for this transitional set of rules can be collected via (i) information such as experimental results, (ii) numerical simulation runs based on dynamic models, and (iii) heuristics. However, the main advantage of the automatic rule generation scheme is that reliability can be potentially increased even in the presence of large-grained uncertainty in the system investigated. Specifications-accuracy and precision-of the system tolerances can be arbitrarily adjusted and are a function of the resolution of the design parameters.<
>