Improving the particle filter in high dimensions using conjugate artificial process noise. Issue 15 (2018)

• Record Type: Journal Article
• Authors: Wigren, Anna; Murray, Lawrence; Lindsten, Fredrik
• Is Part Of: IFAC-PapersOnLine. Volume 51:Issue 15(2018)
• Page Start: 670
• Page End: 675
• Publication Date: 2018
• Subjects: Data assimilation -- Sequential Monte Carlo -- Estimation -- filtering -- State-space models -- Nonlinear system identification; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2018.09.207
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: The particle filter is one of the most successful methods for state inference and identification of general non-linear and non-Gaussian models. However, standard particle filters suffer from degeneracy of the particle weights, in particular for high-dimensional problems. We propose a method for improving the performance of the particle filter for certain challenging state space models, with implications for high-dimensional inference. First we approximate the model by adding artificial process noise in an additional state update, then we design a proposal that combines the standard and the locally optimal proposal. This results in a bias-variance tradeoff, where adding more noise reduces the variance of the estimate but increases the model bias. The performance of the proposed method is empirically evaluated on a linear-Gaussian state space model and on the non-linear Lorenz'96 model. For both models we observe a significant improvement in performance over the standard particle filter.