Resampling algorithms for particle filters: a computational complexity perspective (Q1773796)

Particle (sample) filters (PF) are used for tracking states of plants described by means of discrete-time dynamic state-space models. The main goal of this paper is the development of resampling methods that allow for increased speed of PF, require less memory, achieve timings regardless of the statistics of the samples, and that are computationally less complex. The resampling algorithms, replicating samples having large weights and removing samples with negligible weights, are based on stratified resampling. The computational complexity of the proposed resampling algorithms is evaluated.