Robust control of nonlinear uncertain systems under generalized matching conditions using neuro-compensator

The robust adaptive control scheme with neural-compensator is developed for a class of uncertain systems satisfied the generalized matching condition. In the control procedure, the neural compensator is combined with the feedback linearization method, and the neural network is implemented to learn and compensate the uncertainty, rescale the uncertainty of nonlinear systems to the reconstruction error of the network, and a sliding mode controller is used to deal with the uncertainty due to the reconstruction of the network. No property of the uncertainty is used for the design, it is more practical than the existing ones in the literature