A hierarchical fuzzy-neural-PD controller for robot manipulators

Abstract

An original multivariable hierarchical controller for N degrees-of-freedom robot manipulators for control tracking problems is presented in this paper. The system is composed of a coordinator implemented as a fuzzy-neural network, whose purpose is to select activation levels for local regulators implemented as PD controllers. A systematic design method is developed and a stability analysis included to provide the necessary formalism. The controller is tested in simulation and a comparison is made with the relay-type control algorithm commonly used in industrial robots. A disturbance rejection test is also included.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>