Adaptive Self-Organizing Fuzzy Sliding-Mode Radial Basis-Function Neural-Network Controller for Robotic Systems

Abstract

A self-organizing fuzzy radial basis-function neural-network controller (SFRBNC) has been proposed to control robotic systems. The SFRBNC uses a radial basis-function neural-network (RBFN) to regulate the parameters of a self-organizing fuzzy controller (SOFC) to appropriate values in real time. This method solves the problem caused by the inappropriate selection of parameters in an SOFC. It also eliminates the dynamic coupling effects between degrees of freedom (DOFs) for robotic system control because the RBFN has coupling weighting regulation capabilities. However, its stability is difficult to demonstrate. To overcome the stability issue, this study developed an adaptive self-organizing fuzzy sliding-mode radial basis-function neural-network controller (ASFSRBNC) for robotic systems. The ASFSRBNC solves the problem of an SFRBNC implementation in determining the stability of the system control. It also applies an adaptive law to modify the fuzzy consequent parameter of a fuzzy logic controller to manipulate a robotic system to improve its control performance. The stability of the ASFSRBNC was proven using the Lyapunov stability theorem. From the experimental results of 6-DOF robotic control tests, the ASFSRBNC achieved better control performance than the SFRBNC as well as the SOFC.