Parametric Identification of Nonlinear Dynamical Systems Using Filtering Techniques

Khalil, M., Sarkar, A. and Adhikari, S..
1st International Conference on Uncertainty in Structural Dynamics, University of Sheffield, Sheffield, UK, June 2007.

Combined state and parameter estimation of dynamical systems plays a crucial role in extracting system response from noisy measurements. A wide variety of methods have been developed to deal with the joint state-parameter estimation of nonlinear dynamical systems. The Extended Kalman Filter method is a popular approach for the joint system-parameter estimation of nonlinear systems. This method combines the traditional Kalman filtering techniques with the linearisation tools to tackle nonlinear problems and its formulation is based on the assumption that the probability density function of the state vector can be reasonably approximated to be Gaussian. Recent research has been focused on non-Gaussian models. Of particular interest is the Ensemble Kalman Filter and the Particle Filter. These methods are capable of handling various forms of nonlinearities as well as non-Gaussian noise models. This paper examines and contrasts the feasibility of joint state and parameter identification in non-linear dynamical systems using the Extended Kalman, Ensemble Kalman and Particle filters.
BiBTeX Entry
@INPROCEEDINGS{cp43,
    AUTHOR={M. Khalil and A. Sarkar and S. Adhikari},
    TITLE={Parametric identification of non-linear dynamical systems using filtering techniques},
    BOOKTITLE={Proceedings of the 1st International Conference on Uncertainty in Structural Dynamics},
    YEAR={2007},
    Address={University of Sheffield, Sheffield, UK},
    Month={June},
    Note={}
}

by Sondipon Adhikari