Hardware implementation issues of multivariable fuzzy control systems

We show two models for synchronous finite state machines (FSM) based on fuzzy logic: the crisp-state-fuzzy-output (CSFO) FSM, and the fuzzy-state-fuzzy-output (CSFO) FSM. As a result of the introduction of the FSM models, improved architectures for fuzzy logic controller have been defined. The new method for linguistic fuzzy model building of multi-input-multi-output controllers is proposed. Due to the applied theoretic approach to the fuzzy model building, the fuzzy logic controller architecture features massively parallel fuzzy data inferencing is obtained. This architecture provides almost constant performance, independent on the number of inputs, on the number of outputs, and on the number of processed rules. The approach can be utilised for logic controller hardware to work in the real-time environment