On realizability of neural networks-based input-output models in the classical state-space form

Authors:
í. Kotta;F. N. Chowdhury;S. NíMm
Affiliations:
Institute of Cybernetics, Tallinn University of Technology, Akadeemia tee 21, Tallinn 12618, Estonia;University of Louisiana at Lafayette, P.O. Box 43890, Lafayette, LA 70504-3890, USA;Institute of Cybernetics, Tallinn University of Technology, Akadeemia tee 21, Tallinn 12618, Estonia
Venue:
Automatica (Journal of IFAC)
Year:
2006

Citing 3
Cited 3

State observability in recurrent neural networks

Systems & Control Letters
Linearization of Discrete-Time Systems

SIAM Journal on Control and Optimization
Brief Transfer equivalence and realization of nonlinear higher order input-output difference equations

Automatica (Journal of IFAC)

Online identification of the system order with ANARX structure

ICAIS'11 Proceedings of the Second international conference on Adaptive and intelligent systems
Evolutionary design of the closed loop control on the basis of NN-ANARX model using genetic algorithm

ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part I
Neural networks based system for the supervision of therapeutic exercises

ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part IV

Quantified Score

Hi-index	22.14

Visualization

Abstract

This paper proves that the typical neural network-based input/output model does not have a state-space realization and suggests the Additive Nonlinear Auto-Regressive with eXogenous input (ANARX) structure as an excellent choice for neural-network-based input-output models. The advantage of the ANARX model is that the time-steps in the argument are pair-wise decomposed, which allows the ANARX model to be realized in state space, and to be linearized via dynamic output feedback. Moreover, accessibility of the state-space realization has been proved.