A fuzzy processor is programmed to provide anoptimum output for solving a given problem. It could theoretically solve any problem (from a static point of view) if it is an universal approximator. This paper addresses the design of fuzzy processors aiming at a twofold objective: efficient adaptive approximation of different and even dynamically changing surfaces and hardware simplicity. Adequate programmable parameters and a fully-parallel architecture are selected. Mixed-signal blocks based on digitally programmed current mirrors are employed. Error-descent learning algorithms for tuning are discussed. Adaptive behavior is illustrated with an application to the on-line identification of a nonlinear plant.

Peer reviewed