Stability Analysis Of Variable Structure Controlled Industrial Robotic Manipulator With Single Term Haar Wavelet Series Method

Abstract

In this paper, stability analysis has been made using Neural Tool to validate the obtained errors from experimental work. A neural network prediction has been done by selecting appropriate factors which played vital role in increasing errors during continuous working of the robot. In NN prediction, two simulations were carried out, one for predicting variation from existing tested data and the other for finding out the unknown values of non-tested input values. From neural network analysis it has been observed that the neural network predictions very closely match with the variation obtained by experimental validation. Also the neural network predictions yield better results for non-tested input values and also it is very close to the actual values of variation obtained by experimental validation. The conclusion is that the proposed mathematical tool with variable structure control is stable and can be implemented to industrial robots for the robust control of positions.