Neural Network-Based Practical/Ideal Integral Sliding Mode Control

This paper deals with the design of a novel neural network based integral sliding mode (NN-ISM) control for nonlinear systems uncertain drift term and effectiveness matrix. Specifically, this extends classical law to case unknown nominal model. The latter is indeed reconstructed by two deep networks capable approximating terms, which are instrumental so-called manifold. In paper, ultimate boundedness system state formally proved using Lyapunov stability arguments, thus providing conditions enforce practical modes. possible generation ideal modes also discussed. Moreover, proposed NN-ISM assessed in simulation relying on Duffing oscillator.