A Neuro-augmented Observer for a Class of Nonlinear Systems

A Neuro-augmented Observer for a Class of Nonlinear Systems

A new type of state observer for nonlinear systems is presented in this paper. This observer is a hybrid of linear and nonlinear parts: it is based on a conventional linear observer design, and augmented by a neural network. The neural network approximates only the nonlinear part of the system. The...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Huajun Gong, Hao Xu, Chowdhury, F.N.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Convergence
Design methodology
Fault detection
NASA
Neural networks
Nonlinear dynamical systems
Nonlinear systems
Observers
State estimation
State feedback
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A new type of state observer for nonlinear systems is presented in this paper. This observer is a hybrid of linear and nonlinear parts: it is based on a conventional linear observer design, and augmented by a neural network. The neural network approximates only the nonlinear part of the system. The state estimation error is proved to approach zero asymptotically.
ISSN:2161-4393
2161-4407
DOI:10.1109/IJCNN.2006.247100