Recursive Identification of Continuous Time Variant Dynamical Systems with the Extended Kalman Filter and the Recursive Least Squares State-Variable Filter

Flávio Luiz Rossini, Guilherme Santos Martins, João Paulo Silva Gonçalves, Mateus Giesbrecht

2018

Abstract

In this paper, a method for the continuous time varying dynamical systems identification is presented. The study is based on the integration of the State-Variable Filter (SVF), the Extended Kalman Filter (EKF) and the Recursive Least Squares State-Variable Filter (RLSSVF). The main contribution of the algorithm applied in this paper is that a state space continuous time model can be estimated based on the system sampled inputs and outputs. To validate the method, a continuous time varying benchmark system is simulated and the benchmark parameters are compared to the estimated model parameters. The benchmark outputs are also compared to the model outputs to verify the accuracy of the proposed method. The results obtained show that the model reproduces the benchmark behavior accurately.

Download


Paper Citation


in Harvard Style

Santos Martins G., Paulo Silva Gonçalves J. and Giesbrecht M. (2018). Recursive Identification of Continuous Time Variant Dynamical Systems with the Extended Kalman Filter and the Recursive Least Squares State-Variable Filter.In Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO, ISBN 978-989-758-321-6, pages 458-465. DOI: 10.5220/0006865504580465


in Bibtex Style

@conference{icinco18,
author={Guilherme Santos Martins and João Paulo Silva Gonçalves and Mateus Giesbrecht},
title={Recursive Identification of Continuous Time Variant Dynamical Systems with the Extended Kalman Filter and the Recursive Least Squares State-Variable Filter},
booktitle={Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,},
year={2018},
pages={458-465},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006865504580465},
isbn={978-989-758-321-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - Volume 1: ICINCO,
TI - Recursive Identification of Continuous Time Variant Dynamical Systems with the Extended Kalman Filter and the Recursive Least Squares State-Variable Filter
SN - 978-989-758-321-6
AU - Santos Martins G.
AU - Paulo Silva Gonçalves J.
AU - Giesbrecht M.
PY - 2018
SP - 458
EP - 465
DO - 10.5220/0006865504580465