Approximate recursive Bayesian estimation of state space model with uniform noise

0492001 - ÚTIA 2019 RIV PT eng C - Konferenční příspěvek (zahraniční konf.)
Pavelková, Lenka - Jirsa, Ladislav
Approximate recursive Bayesian estimation of state space model with uniform noise.
ICINCO 2018 : Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics. Setubal: INSTICC, SCITEPRESS., 2018 - (Madani, K.; Gusikhin, O.), s. 388-394. ISBN 978-989-758-321-6. ISSN 2184-2809.
[15th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2018). Porto (PT), 29.07.2018-31.07.2018]
Grant CEP: GA ČR(CZ) GA18-15970S
Institucionální podpora: RVO:67985556
Klíčová slova: probabilistic state-space model * approximate state estimation * linear systems * bounded noise * Bayesian estimation
Obor OECD: Automation and control systems
http://library.utia.cas.cz/separaty/2018/AS/pavelkova-0492001.pdf

This paper proposes a recursive algorithm for the state estimation of a linear stochastic state space model. A model with discrete-time inputs, outputs and states is considered. The model matrices are supposed to be known. A noise of the involved model is described by a uniform distribution. The states are estimated using Bayesian approach. Without using an approximation, the complexity of the posterior probability density function (pdf) increases with time. The paper proposes an approximation of this complex pdf so that a feasible support of the posterior pdf is kept during the estimation. The state estimation consists of two stages, namely the time and data update including the mentioned approximation. The behaviour of the proposed algorithm is illustrated by simulations and compared with other methods.
Trvalý link: http://hdl.handle.net/11104/0285675