In this chapter, sequential sampling techniques are considered. Kalman filtering is viewed in terms of probabilistic arguments as a special case of a linear dynamic system, where the involved variables follow Gaussian distributions. Particle filtering techniques are then considered as a vehicle to treat more general nonlinear models and/or non-Gaussian random variables. They are introduced as a special instance of the more general family of sequential sampling methods. Different schemes are discussed such as the generic particle and the auxiliary particle filtering algorithms.
Sequential importance sampling
Generic particle filters
Get Machine Learning now with O’Reilly online learning.
O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.