Visible to the public An augmented cubature Kalman filter for nonlinear dynamical systems with random parameters

TitleAn augmented cubature Kalman filter for nonlinear dynamical systems with random parameters
Publication TypeConference Paper
Year of Publication2017
AuthorsQu, X., Mu, L.
Conference Name2017 36th Chinese Control Conference (CCC)
Keywordsaugmented cubature Kalman filter, augmented system, Bayes methods, Bayesian filtering problem, CKF, composability, computational complexity, Cubature Kalman filter, cubature point, discrete nonlinear dynamical systems, Dynamical Systems, Estimation, estimation accuracy, Kalman filters, Metrics, Mobile communication, mobile source localization, Noise measurement, nominal values, Nonlinear dynamical systems, Probability density function, pubcrawl, random parameters, random sensor positions, resilience, Resiliency, state vector, TDOA measurements, time difference of arrival, time-of-arrival estimation

In this paper, we investigate the Bayesian filtering problem for discrete nonlinear dynamical systems which contain random parameters. An augmented cubature Kalman filter (CKF) is developed to deal with the random parameters, where the state vector is enlarged by incorporating the random parameters. The corresponding number of cubature points is increased, so the augmented CKF method requires more computational complexity. However, the estimation accuracy is improved in comparison with that of the classical CKF method which uses the nominal values of the random parameters. An application to the mobile source localization with time difference of arrival (TDOA) measurements and random sensor positions is provided where the simulation results illustrate that the augmented CKF method leads to a superior performance in comparison with the classical CKF method.

Citation Keyqu_augmented_2017