Bayesian estimation of traffic lane state

0411034 - UTIA-B 20030021 RIV GB eng J - Článek v odborném periodiku
Nagy, Ivan - Kárný, Miroslav - Nedoma, Petr - Voráčová, Š.
Bayesian estimation of traffic lane state.
International Journal of Adaptive Control and Signal Processing. Roč. 17, č. 1 (2003), s. 51-65. ISSN 0890-6327. E-ISSN 1099-1115
Grant CEP: GA ČR GA102/03/0049; GA AV ČR IBS1075351
Výzkumný záměr: CEZ:AV0Z1075907
Klíčová slova: mixture models * estimation * Bayesian approach
Kód oboru RIV: BB - Aplikovaná statistika, operační výzkum
Impakt faktor: 0.602, rok: 2003
http://library.utia.cas.cz/prace/20030021.ps

The paper deals with modelling and estimation by a model described as a mixture of distributions. In this case, the exact application of the Bayes theory adopted is not feasible and its approximation is used. The general algorithm is specified for mixtures with components from exponential family. The theory is demonstrated on estimation of the basic relation between density and intensity of traffic flow in a single point of a vehicular communication and it can provides us with state classification.
Trvalý link: http://hdl.handle.net/11104/0131121