H∞ tracking of unknown nonlinear systems usingneural network

摘要

A stable neural adaptive control scheme, is proposed to achieve H∞ performance for a class of unknown nonlinear SISOsystems with external disturbances. In the control design, thecontroller comprises a certainty equivalence control term and anH∞ compensating term. The neural network is used toapproximate the unknown nonlinear functions for the design of theequivalence controller, and the Lyapunov method is used for the updateof the parameters of the neural network, the stability of the proposedcontrol algorithm can be guaranteed. The performance of the proposedmethod is demonstrated through the control of the inverted pendulumsystem