This paper presents a robust adaptive neural-fuzzy network control (RANFNC) system for an n-link robot manipulator to achieve the highprecision position tracking. Initially, the model dynamic of an n-link robot manipulator is introduced. However, it is difficult to design a conformable model-based control scheme, for instance, external disturbances, friction forces and parameter variations. In order to deal with this problem, the RANFNC system is investigated to the joint position control of an n-link robot manipulator. In this control scheme, a four-layer neural-fuzzy-network (NFN) is used for the main role, and the adaptive tuning laws of network parameters are derived in the sense of a projection algorithm and the Lyapunov stability theorem to ensure network convergence as well as stable control performance. The merits of this model-free control scheme are that not only the stable position tracking performance can be guaranteed but also unknown system information and auxiliary control design are required in the control process. The simulation results are provided to verify the effectiveness of the proposed RANFNC methodology.

In this paper, adaptive iterative learning control (AILC) of uncertain robot manipulators in task space is considered for trajectory tracking in an iterative operation mode. The control scheme incluces a PD controller with a gain switching technique plus a learning feedforward term, is exploited to predict the desired actuator torque. By using Lyapunov method, an adaptive iterative learning control scheme is presented for robotic system with both structured and unstructured uncertainty, and the overall stability of the closed-loop system in the iterative domain is established. The validity of the scheme is confirmed through a numerical simulation.