Neural adaptive control of two-link manipulator with sliding mode compensation

In this paper we developed a new neuro controller for robot manipulators. A simple dynamic neural network is used to estimate the unknown robot manipulators, then the direct linearization controller is derived via this neuro identifier. Because the approximation capability is limited, another robust sliding mode compensator is addressed. Our main contributions are: first we give a bound for the identification error of the parallel neuro identifier; second we establish a bound for the tracking error of the hybrid controller.