A New ANFIS Based Learning Algorithm for CMOS Neuro-Fuzzy Controllers

This paper presents a new learning procedure for ANFIS (Adaptive-Network-based Fuzzy Inference System), a fuzzy inference system implemented in the framework of adaptive networks. By using this new algorithm, the ANFIS can construct an input-output mapping based on both human knowledge (in the form of fuzzy if-then rules) and stipulated input-output data pairs. This algorithm is a combination of perceptron neural network and hybrid learning algorithm, but this is convenient than hybrid learning procedure. The main purpose of this method is to provide a powerful algorithm to program CMOS fuzzy controllers, considering CMOS implementation limits. Simulation results are provided to demonstrate the capability of proposed algorithm.