Grammar-based representation and identification of dynamical systems

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Downloads (Pure)

Abstract

In this paper we propose a novel approach to identify dynamical systems. The method estimates the model structure and the parameters of the model simultaneously, automating the critical decisions involved in identification such as model structure and complexity selection. In order to solve the combined model structure and model parameter estimation problem, a new representation of dynamical systems is proposed. The proposed representation is based on Tree Adjoining Grammar, a formalism that was developed from linguistic considerations. Using the proposed representation, the identification problem can be interpreted as a multiobjective optimization problem and we propose an Evolutionary Algorithm-based approach to solve it. A benchmark example is used to demonstrate the proposed approach. The achieved performance of the proposed method, without making use of knowledge of the system description, was comparable to that obtained by state-of-the-art non-linear system identification methods that do take advantage of correct selection of model structure and complexity based on a priori information.

Original languageEnglish
Title of host publication2019 18th European Control Conference, ECC 2019
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages1318-1323
Number of pages6
ISBN (Electronic)978-3-907144-00-8
DOIs
Publication statusPublished - 1 Jun 2019
Event18th European Control Conference, ECC 2019 - Naples, Italy
Duration: 25 Jun 201928 Jun 2019

Conference

Conference18th European Control Conference, ECC 2019
CountryItaly
CityNaples
Period25/06/1928/06/19

Fingerprint

grammars
Model structures
Grammar
dynamical systems
Dynamical systems
Dynamical system
Identification (control systems)
Multiobjective optimization
Linguistics
Evolutionary algorithms
Model
Parameter estimation
Nonlinear System Identification
Nonlinear systems
linguistics
system identification
Identification Problem
Multiobjective Optimization Problems
nonlinear systems
Parameter Estimation

Cite this

Khandelwal, D., Schoukens, M., & Toth, R. (2019). Grammar-based representation and identification of dynamical systems. In 2019 18th European Control Conference, ECC 2019 (pp. 1318-1323). [8795719] Piscataway: Institute of Electrical and Electronics Engineers. https://doi.org/10.23919/ECC.2019.8795719
Khandelwal, Dhruv ; Schoukens, Maarten ; Toth, Roland. / Grammar-based representation and identification of dynamical systems. 2019 18th European Control Conference, ECC 2019. Piscataway : Institute of Electrical and Electronics Engineers, 2019. pp. 1318-1323
@inproceedings{be4c09c0a3d5454489c525990c83c265,
title = "Grammar-based representation and identification of dynamical systems",
abstract = "In this paper we propose a novel approach to identify dynamical systems. The method estimates the model structure and the parameters of the model simultaneously, automating the critical decisions involved in identification such as model structure and complexity selection. In order to solve the combined model structure and model parameter estimation problem, a new representation of dynamical systems is proposed. The proposed representation is based on Tree Adjoining Grammar, a formalism that was developed from linguistic considerations. Using the proposed representation, the identification problem can be interpreted as a multiobjective optimization problem and we propose an Evolutionary Algorithm-based approach to solve it. A benchmark example is used to demonstrate the proposed approach. The achieved performance of the proposed method, without making use of knowledge of the system description, was comparable to that obtained by state-of-the-art non-linear system identification methods that do take advantage of correct selection of model structure and complexity based on a priori information.",
author = "Dhruv Khandelwal and Maarten Schoukens and Roland Toth",
year = "2019",
month = "6",
day = "1",
doi = "10.23919/ECC.2019.8795719",
language = "English",
pages = "1318--1323",
booktitle = "2019 18th European Control Conference, ECC 2019",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Khandelwal, D, Schoukens, M & Toth, R 2019, Grammar-based representation and identification of dynamical systems. in 2019 18th European Control Conference, ECC 2019., 8795719, Institute of Electrical and Electronics Engineers, Piscataway, pp. 1318-1323, 18th European Control Conference, ECC 2019, Naples, Italy, 25/06/19. https://doi.org/10.23919/ECC.2019.8795719

Grammar-based representation and identification of dynamical systems. / Khandelwal, Dhruv; Schoukens, Maarten; Toth, Roland.

2019 18th European Control Conference, ECC 2019. Piscataway : Institute of Electrical and Electronics Engineers, 2019. p. 1318-1323 8795719.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Grammar-based representation and identification of dynamical systems

AU - Khandelwal, Dhruv

AU - Schoukens, Maarten

AU - Toth, Roland

PY - 2019/6/1

Y1 - 2019/6/1

N2 - In this paper we propose a novel approach to identify dynamical systems. The method estimates the model structure and the parameters of the model simultaneously, automating the critical decisions involved in identification such as model structure and complexity selection. In order to solve the combined model structure and model parameter estimation problem, a new representation of dynamical systems is proposed. The proposed representation is based on Tree Adjoining Grammar, a formalism that was developed from linguistic considerations. Using the proposed representation, the identification problem can be interpreted as a multiobjective optimization problem and we propose an Evolutionary Algorithm-based approach to solve it. A benchmark example is used to demonstrate the proposed approach. The achieved performance of the proposed method, without making use of knowledge of the system description, was comparable to that obtained by state-of-the-art non-linear system identification methods that do take advantage of correct selection of model structure and complexity based on a priori information.

AB - In this paper we propose a novel approach to identify dynamical systems. The method estimates the model structure and the parameters of the model simultaneously, automating the critical decisions involved in identification such as model structure and complexity selection. In order to solve the combined model structure and model parameter estimation problem, a new representation of dynamical systems is proposed. The proposed representation is based on Tree Adjoining Grammar, a formalism that was developed from linguistic considerations. Using the proposed representation, the identification problem can be interpreted as a multiobjective optimization problem and we propose an Evolutionary Algorithm-based approach to solve it. A benchmark example is used to demonstrate the proposed approach. The achieved performance of the proposed method, without making use of knowledge of the system description, was comparable to that obtained by state-of-the-art non-linear system identification methods that do take advantage of correct selection of model structure and complexity based on a priori information.

UR - http://www.scopus.com/inward/record.url?scp=85071566277&partnerID=8YFLogxK

U2 - 10.23919/ECC.2019.8795719

DO - 10.23919/ECC.2019.8795719

M3 - Conference contribution

AN - SCOPUS:85071566277

SP - 1318

EP - 1323

BT - 2019 18th European Control Conference, ECC 2019

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Khandelwal D, Schoukens M, Toth R. Grammar-based representation and identification of dynamical systems. In 2019 18th European Control Conference, ECC 2019. Piscataway: Institute of Electrical and Electronics Engineers. 2019. p. 1318-1323. 8795719 https://doi.org/10.23919/ECC.2019.8795719