Fuzzy self-adaptive radial basis function neural network-based control of a seven-link redundant industrial manipulator

This paper proposes a method for the identification of dynamics and control of a multi-link industrial robot manipulator using Runge-Kutta-Gill neural networks (RKGNNs). RKGNNs are used to identify an ordinary differential eq uation of the dynamics of the robot manipulator. A structured function neur al network (NN) with sub-networks to represent the components of the dynami cs is used in the RKGNNs. The sub-networks consist of shape adaptive radial basis function (RBF) NNs. An evolutionary algorithm is used to optimize th e shape parameters and the weights of the RBFNNs. Due to the fact that the RKGNNs can accurately grasp the changing rates of the states. this method c an effectively be used for long-term prediction of the states of the rebut manipulator dynamics. Unlike in conventional methods, the proposed method c an even be used without input torque information because a torque network i s part of the functional network. This method can be proposed as an effecti ve option for the dynamics identification of manipulators with high degrees -off-freedom, as opposed to the derivation of dynamic equations and making additional hardware changes as in the case of statistical parameter identif ication such as linear least-squares method. Experiments were carried out u sing a seven-link industrial manipulator. The manipulator was controlled fo r a given trajectory, using adaptive fuzzy selection of nonlinear dynamic m odels identified previously. Promising experimental results are obtained to prove the ability of the proposed method in capturing nonlinear dynamics o f a multi-link manipulator in an effective manner.