Sampling Densities of Particle Filter: A Survey and Comparison

The paper deals with the particle filter in discrete-time nonlinear non-Gaussian system state estimation. One of the key parameters affecting estimation quality of the particle filter is the sampling density (also called importance function or proposal density). In the literature, there are many sampling density proposals based on various ideas. The goal of the paper is to provide a survey of sampling densities, to classify them and to compare estimation quality of the particle filter with various sampling densities in an illustration example.