Stability Analysis of A Neuro-Identification Scheme with Asymptotic Convergence

This paper focuses on the stability and convergence analysis of a neuro-identification scheme for uncertainnonlinear systems. Based on linearly parameterized neural networks and the previous knowledge of upperbounds for the approximation error and disturbances, a robust modification of the descent gradientalgorithm is proposed to make the overall identification process stable, and, in addition, the on-lineresidual prediction error asymptotically null, despite the presence of approximation error anddisturbances. A simulation study to show the application and comparative performance of the proposedalgorithm is presented