Fuzzified RBF network-based learning control: structure and self-construction

An example of how fuzzy systems can integrate with neural networks and what benefits can be obtained from the combination is described. By drawing some equivalence between a simplified fuzzy control algorithm (SFCA) and radial basis functions (RBF) networks it is concluded that the RBF network can be interpreted in the context of fuzzy systems and can be naturally fuzzified into a class of more general networks, referred to as FBFN. The FBFN is used as multivariable rule-based controller with the ability of self-constructing its own rule-base by incorporating an iterative learning control algorithm into the system. The approach is applied to a problem of multivariable blood pressure control with a FBFN-based controller having six inputs and two outputs, representing a complicated control structure.<>