Bayesian Filters for Parameter Identification of Duffing Oscillator

Abstract

In this paper, a joint state and parameter estimation problem of Duffing oscillator is explored using Bayesian filters, where the parameter to be identified is considered as an additional state variable. From a variety of Bayesian filters, the unscented Kalman filter (UKF), cubature Kalman filter (CKF) and Gauss-Hermite filter (GHF) are chosen for solving this problem. The performance of these filters are compared in terms of the root mean square error (RMSE) calculated over a specified number of Monte-Carlo runs. From simulation results, it is found that the accuracy of CKF and GHF are almost same while the computational time for GHF is almost three times higher.