Adaptive design of a fuzzy cerebellar model arithmetic controller neural network

Abstract

Adaptation fuzzy cerebellar model arithmetic controller (CMAC) neural networks are considered. Adaptation mechanisms for a fuzzy CMAC neural network are proposed to enable the construction of indirect and direct control laws. These control laws are then used to enhance the robustness of a closed-loop control system. It is shown that the fuzzy CMAC's can cope with the system's uncertainties using adaptation with no preliminary off-line learning phase being required. The adaptation laws are derived using a Lyapunov stability analysis, so that both system tracking stability and error convergence can be guaranteed in the closed-loop system. Simulation results from the two systems show a satisfactory performance of the proposed control schemes even in the presence of modelling uncertainties.