Порівняння методів
Переглядайте обрані методи поруч; рядки з відмінностями підсвічено.
| Просторовий фільтр Калмана× | Фільтр частинок (послідовний Монте-Карло)× | |
|---|---|---|
| Галузь | Баєсові методи | Баєсові методи |
| Родина | Bayesian methods | Bayesian methods |
| Рік появи≠ | 1960 (base); spatial extensions 1990s–2000s | 1993 |
| Автор методу≠ | R. E. Kalman (base filter, 1960); extended to spatial settings by Cressie, Wikle and colleagues | Gordon, Salmond & Smith |
| Тип≠ | Bayesian state-space model | Sequential Monte Carlo estimator |
| Основоположне джерело≠ | Cressie, N. & Wikle, C. K. (2011). Statistics for Spatio-Temporal Data. Wiley. ISBN: 978-0-471-69274-4 | Gordon, N. J., Salmond, D. J., & Smith, A. F. M. (1993). Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE Proceedings F (Radar and Signal Processing), 140(2), 107–113. DOI ↗ |
| Інші назви≠ | spatial state-space filter, spatio-temporal Kalman filter, SKF, spatial dynamic linear model | SMC, sequential Monte Carlo, bootstrap filter, condensation algorithm |
| Пов'язані≠ | 6 | 4 |
| Підсумок≠ | The spatial Kalman filter applies classical Kalman filtering to spatio-temporal state-space models, treating a spatially distributed latent field as the hidden state that evolves over time. At each time step, the filter recursively predicts the spatial field forward and then updates the prediction with new spatial observations, producing optimal linear estimates of the field and its uncertainty across all locations. | The particle filter, introduced by Gordon, Salmond, and Smith in 1993, is a sequential Monte Carlo algorithm that approximates the Bayesian filtering distribution for nonlinear and non-Gaussian state-space models. Rather than tracking a single best estimate, it maintains a cloud of N weighted random samples — particles — that collectively represent the full posterior distribution of a hidden state at each point in time as new observations arrive. |
| ScholarGateНабір даних ↗ |
|
|