Robust parametric identification procedure of stochastic nonlinear continuous-discrete systems

V. M. Chubich, S. O. Kulabukhova

Novosibirsk State Technical University, pr. Karla Marksa 20, Novosibirsk, 630073 Russia

Abstract: Based on the weighted maximum likelihood method and the correntropy cubature Kalman filter, software to solve the problem of parametric estimation of models of stochastic nonlinear continuous-discrete systems in the presence of anomalous observations in the measurement data is developed. The effectiveness of the proposed procedure is demonstrated on two model structures for the stochastic and the grouped ordering of anomalous data.

Keywords: stochastic nonlinear continuous-discrete system, parametric identification, anomalous observations, robust filtering, cubature Kalman filter, weighted maximum likelihood estimation. .

UDC: 681.5.015

Received: 12.05.2020
Revised: 20.05.2021
Accepted: 24.06.2021

DOI: 10.33048/SIBJIM.2021.24.310

 English version: Journal of Applied and Industrial Mathematics, 2021, 15:3, 384–392