Resampling algorithms and architectures for distributed particle filters

Abstract

In this paper, we propose novel resampling algorithms with architectures for efficient distributed implementation of particle filters. The proposed algorithms improve the scalability of the filter architectures affected by the resampling process. Problems in the particle filter implementation due to resampling are described and appropriate modifications of the resampling algorithms are proposed so that distributed implementations are developed and studied. Distributed resampling algorithms with proportional allocation (RPA) and non-proportional allocation (RNA) of particles are considered. The components of the filter architectures are the processing elements (PEs), a central unit (CU) and an interconnection network. One of the main advantages of the new resampling algorithms is that communication through the interconnection network is reduced and made deterministic, which results in simpler network structure and increased sampling frequency. Particle filter performances are estimated for the bearings-only tracking applications. In the architectural part of the analysis, the area and speed of the particle filter implementation are estimated for different number of particles and different level of parallelism with FPGA implementation. In this paper only sampling importance resampling (SIR) particle filters are considered, but the analysis can be extended to any particle filters with resampling.

Cooperative parallel particle filters for online model selection and applications to urban mobility

“…In general, the resampling steps are not applied at each iterations, but only when some statistical criterion is satisfied [8,13,15] (e.g., see Section 4). The recursive expression of the weights for SIR becomes…”

“…Indeed, each IS scheme can compute independentlyw (m) k,t and Z k,t , and then they merge all the information for calculatingρ k,t (see Figure 1). Therefore, we consider K parallel particle filters (as in [8,12,13], for instance) using the transition model as proposal pdf 7 φ k,t (x k,t |x k,1:t−1 ) = q k,t (x k,t |x k,t−1 ), each one tracking a different states-space model M k , for k = 1, . .…”

“…. = M K = T , then MAPF described a distributed PF scheme [8,12,15] which K parallel PFs cooperate for providing a global estimator, I 1:t = K k=1ρ k,t I k,1:t . The computational effort is distributed in order to foster the filters that are providing the best performance, in the specific run.…”

An adaptive particle filter based on PSO and fuzzy inference system for nonlinear state systems

“…It is known that the objective of resampling steps is to eliminate particle degeneracy problem [26][27][28]. The idea of resampling is that particles with large weights are duplicated while those with small weights are abandoned.…”

“…In addition, in UPF, an inherent rounding error in the numerical calculation may cause negative definiteness of the state covariance, which will lead to filter divergence and influence the estimation precision [23][24][25]. Finally, the resampling step in UPF leads to a great loss of diversity in particles [26,27]. In this paper, APF is proposed to overcome these problems.…”