A nonlinear adaptive H∞ tracking control design in robotic systems via neural networks

An adaptive neural-network tracking control with a guaranteed H ^∞ performance is proposed for robotic systems with plant uncertainties and external disturbances. A neural-network system is introduced to learn these unknown (or uncertain) dynamics by an adaptive algorithm, Moreover, the effects on the tracking error due to the approximation error via the adaptive neural network must be attenuated to a prescribed level, i.e. an H^∞ tracking performance is achieved. Hence, in this study, both the H^∞ tracking theory and adaptive neural-network control scheme are combined together to achieve the nonlinear adaptive H^∞ tracking control design for uncertain or unknown robotic systems. The developed control scheme is smooth and semiglobal as well as very simple and computationally efficient, since it does not require a knowledge of either the mathematical model or the parameterization of the robotic dynamics. Finally, extensive simulations are given to illustrate the tracking performance of a two-link robotic manipulator with the proposed adaptive neural H^∞ control design