GMDH TYPE NEURAL NETWORKS AND THEIR APPLICATION TO THE IDENTIFICATION OF THE INVERSE KINEMATIC EQUATIONS OF ROBOTIC MANIPULATORS

In this paper, it is shown that group method of data handling (gmdh)-type neural networks and their application can be effectively used to acquire the inverse kinematic equations of a puma760 robot manipulator based on the numerical data of its motion. the aim of such modeling is to show the accuracy of gmdh-type neural networks. for evaluating the accuracy of the obtained equations, a new trajectory is employed to demonstrate whether the models are still valid or not. finally, the best results are used to define the inverse kinematic equations.