NXT* SCARA Model Based on an Adaptive Neural Network Controller

This paper describes experimental results by Adaptive Neural Networks by performing the position control of a real SCARA manipulator robot. A proposed controller is derived to control and adapt weights as supervised learning, which compensating and eliminating errors of a neural net. The adaptive controller is used to solve trajectory tracking problems of robotic manipulators with uncertainties. To produce a controller using Adaptive Linear Element (ADALINE) and Radial Basis Function based (RBF) based controllers don't require mathematical modeling. The comparison gives that the adaptive neural network reduces the error tracking and reaches quickly than the neural network.