论文信息 - A kalman-particle kernel filter and its application to terrain navigation

A kalman-particle kernel filter and its application to terrain navigation

A new nonlinear filter, the KalmanParticle Kernel Filter (KPKF) is proposed. Compared with other particle filters like Regularized Particle Filter (RPF), it adds a local linearization in a kernel representation of the conditional density. Therefore, it strongly reduces the number of redistributions which causes undesirable Monte Carlo fluctuations. This new filter is applied to terrain navigation, which is a nonlinear and multimodal problem. Simulations show that the KPKF outperforms the classical particle filter.

Dinh-Tuan Pham | C. Musso | K. Dahia

[1] B. Silverman. Density estimation for statistics and data analysis , 1986 .

[2] Carlos H. Muravchik,et al. Posterior Cramer-Rao bounds for discrete-time nonlinear filtering , 1998, IEEE Trans. Signal Process..

[3] C. D. Kemp,et al. Density Estimation for Statistics and Data Analysis , 1987 .

[4] Simon J. Godsill,et al. On sequential simulation-based methods for Bayesian filtering , 1998 .

[5] Christian Musso,et al. Improving Regularised Particle Filters , 2001, Sequential Monte Carlo Methods in Practice.

[6] Fredrik Gustafsson,et al. Terrain navigation using Bayesian statistics , 1999 .

[7] D. Pham. Stochastic Methods for Sequential Data Assimilation in Strongly Nonlinear Systems , 2001 .

[8] Niclas Bergman,et al. Recursive Bayesian Estimation : Navigation and Tracking Applications , 1999 .

[9] Simon J. Godsill,et al. On sequential Monte Carlo sampling methods for Bayesian filtering , 2000, Stat. Comput..

[10] Nando de Freitas,et al. Sequential Monte Carlo Methods in Practice , 2001, Statistics for Engineering and Information Science.

[11] Neil J. Gordon,et al. Group and extended object tracking , 1999 .

[12] 李幼升,et al. Ph , 1989 .