Abstract: Neural Network (NN) and Fuzzy Inference System (FIS) had been successfully employed in many engineering applications and were adopted in identification and designing controllers for robotic manipulators, the former because of its model-free feature and the other for its high flexibility. In this study, we focus on the application of a neuro-fuzzy technique to bring certain advantages over neural networks and fuzzy logic control for identification and tracking control of a robot manipulator which is a complicated multivariable nonlinear dynamical system. Neural network has the ability to learn by adjusting the interconnections between layers while fuzzy inference system is a computing framework based on the concept of fuzzy sets, fuzzy if-then rules and fuzzy reasoning. A Fuzzy Logic Controller (FLC) is combined with the neural network plant model trained on-line by the backpropagation algorithm using an adaptive learning rate.