Nonlinear system identification using discrete-time recurrent neural networks with stable learning algorithms

In general, neural networks cannot match nonlinear systems exactly. Neuro identifier has to include robust modification in order to guarantee Lyapunov stability. In this paper input-to-state stabil...