Documents disponibles écrits par cet auteur

Prior checking and moving horizon smoothing for improved particle filtering / Lixin Lang in Industrial & engineering chemistry research, Vol. 49 N° 9 (Mai 2010) 

Public ISBD [article]  in Industrial & engineering chemistry research > Vol. 49 N° 9 (Mai 2010) . - pp. 4197-4209 Titre : Prior checking and moving horizon smoothing for improved particle filtering Type de document : texte imprimé Auteurs : Lixin Lang, Auteur ; Prem K. Goel, Auteur ; Bhavik R. Bakshi, Auteur Année de publication : 2010 Article en page(s) : pp. 4197-4209 Note générale : Industrial chemistry Langues : Anglais (eng) Mots-clés : Smoothing  Surveillance Résumé : Particle filtering, also known as sequential Monte Carlo (SMC) sampling, has been successfully applied to general state-space models for Bayesian inference. Being a simulation method, its performance relies to some extent on the generated samples or particles. For a poor initial guess a large fraction of particles is usually less representative of the underlying state's distribution and could even cause SMC to diverge. In this paper, an intuitive statistic, predictive density is proposed to monitor the particles' performance. When below a statistically controlled threshold value, our approach triggers smoothing for obtaining a better estimate of the initial state in the case of a poor prior. We find that combining a moving horizon smoother with SMC is very effective for recovering from a poor prior and develop an integrated practical approach that combines these two powerful tools. ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=22732912 [article] Prior checking and moving horizon smoothing for improved particle filtering [texte imprimé] / Lixin Lang, Auteur ; Prem K. Goel, Auteur ; Bhavik R. Bakshi, Auteur . - 2010 . - pp. 4197-4209. Industrial chemistry Langues : Anglais (eng) in Industrial & engineering chemistry research > Vol. 49 N° 9 (Mai 2010) . - pp. 4197-4209 Mots-clés : Smoothing  Surveillance Résumé : Particle filtering, also known as sequential Monte Carlo (SMC) sampling, has been successfully applied to general state-space models for Bayesian inference. Being a simulation method, its performance relies to some extent on the generated samples or particles. For a poor initial guess a large fraction of particles is usually less representative of the underlying state's distribution and could even cause SMC to diverge. In this paper, an intuitive statistic, predictive density is proposed to monitor the particles' performance. When below a statistically controlled threshold value, our approach triggers smoothing for obtaining a better estimate of the initial state in the case of a poor prior. We find that combining a moving horizon smoother with SMC is very effective for recovering from a poor prior and develop an integrated practical approach that combines these two powerful tools. ISSN : 0888-5885 En ligne : http://cat.inist.fr/?aModele=afficheN&cpsidt=22732912