Performance analysis of Cognitive Map-Fuzzy Logic Controller model for adaptive control application

Cognitive map and fuzzy logic controller hybrid model is presented in this paper. Sample control applications are included to demonstrate incorporation of analytical and empirical knowledge on the fuzzy cognitive map construction with the purpose of generating fuzzy logic controller (FLC) design on-line. Cognitive map state vector includes all FLC-defining concepts. Controller performance concepts are also included in order to automate FLC parameters adjustments. This model targets control applications when only incomplete plant model is available, or for cases where plant parameters values are not known with certainty, or for cases when plant parameters are expected to change while in operation. On these cases, the CM portion designs the FLC, and fine-tunes controller parameters whenever controller performance is unsatisfactory until user-defined control objectives are achieved; this model emulates expert empirical FLC design and fine-tuning within cognitive map knowledge domain.