Neural network-based adaptive controller design of robotic manipulators with an observer

A neural network (NN)-based adaptive controller with an observer is propose d in this paper for the trajectory tracking of robotic manipulators with un known dynamics nonlinearities. It is assumed that the robotic manipulator h as only joint angle position measurements, A linear observer is used to est imate the robot joint angle velocity, while NNs are employed to further imp rove the control performance of the controlled system through approximating the modified robot dynamics function. The adaptive controller for robots w ith an observer can guarantee the uniform ultimate bounds of the tracking e rrors and the observer errors as well as the bounds of the NN weights. For performance comparisons, the conventional adaptive algorithm with an observ er using linearity in parameters of the robot dynamics is also developed in the same control framework as the NN approach for online approximating unk nown nonlinearities of the robot dynamics. Main theoretical results for des igning such an observer-based adaptive controller with the NN approach usin g multilayer NNs with sigmoidal activation functions, as well as with the c onventional adaptive approach using linearity in parameters of the robot dy namics are given. The performance comparisons between the NN approach and t he conventional adaptation approach with an observer is carried out to show the advantages of the proposed control approaches through simulation studi es.