Design of a Deep Neural Network-Based Integral Sliding Mode Control for Nonlinear Systems Under Fully Unknown Dynamics

被引：12

作者：

Vacchini, Edoardo ^{[1
]}

Sacchi, Nikolas ^{[1
]}

Incremona, Gian Paolo ^{[2
]}

Ferrara, Antonella ^{[1
]}

机构：

[1] Univ Pavia, Dipartimento Ingn Ind & Informaz, I-27100 Pavia, Italy

[2] Politecn Milan, Dipartimento Elettron Informaz & Bioingn, I-20133 Milan, Italy

来源：

IEEE CONTROL SYSTEMS LETTERS | 2023年 / 7卷

关键词：

Nonlinear dynamical systems; Artificial neural networks; Manifolds; Sliding mode control; Robustness; Perturbation methods; Instruments; deep neural networks; uncertain systems;

D O I：

10.1109/LCSYS.2023.3281288

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

In this letter a novel deep neural network based integral sliding mode (DNN-ISM) control is proposed for controlling perturbed systems with fully unknown dynamics. In particular, two DNNs with an arbitrary number of hidden layers are exploited to estimate the unknown drift term and the control effectiveness matrix of the system, which are instrumental to design the ISM controller. The DNNs weights are adjusted according to adaptation laws derived directly from Lyapunov stability analysis, and the proposal is satisfactorily assessed in simulation relying on benchmark examples.

引用

页码：1789 / 1794

页数：6

共 28 条

[1]

Bartolini G, 1999, LECT NOTES CONTR INF, V247, P329

[2] A combined backstepping/second order sliding mode approach to control a class of nonlinear systems [J].