Gaussian sum particle flow filter

Pal, Soumyasundar; Coates, Mark

doi:10.1109/camsap.2017.8313189

Cited by 6 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is thus unsuitable for problems that involve multimodal pdfs. For problems where the measurement noise follows a Gaussian mixture pdf, [43] introduces the Gaussian sum PFL filter. Here, the means of the Gaussian mixture components are updated by performing an update step similar to the LEDH.…”

Section: A State-of-the-artmentioning

confidence: 99%

“…On the other hand, the covariance matrices of the components are updated by extended Kalman filters that also run in parallel. An extension of [43] to the case where both driving noise and measurement noise are distributed by a Gaussian mixture pdf is presented in [44]. Here, invertible flow is used for particle weight update in an importance sampling step.…”

Section: A State-of-the-artmentioning

confidence: 99%

See 1 more Smart Citation

Multisensor Multiobject Tracking With Improved Sampling Efficiency

Zhang,

Meyer

2024

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

Passive monitoring of acoustic or radio sources has important applications in modern convenience, public safety, and surveillance. A key task in passive monitoring is multiobject tracking (MOT). This paper presents a Bayesian method for multisensor MOT for challenging tracking problems where the object states are high-dimensional, and the measurements follow a nonlinear model. Our method is developed in the framework of factor graphs and the sum-product algorithm (SPA) and implemented using random samples or "particles". The multimodal probability density functions (pdfs) provided by the SPA are effectively represented by a Gaussian mixture model (GMM). To perform the operations of the SPA with improved sample efficiency, we make use of particle flow (PFL). Here, particles are migrated towards regions of high likelihood based on the solution of a partial differential equation. This makes it possible to obtain good object detection and tracking performance even in challenging multisensor MOT scenarios with single sensor measurements that have a lower dimension than the object positions. We perform a numerical evaluation in a passive acoustic monitoring scenario where multiple sources are tracked in 3-D from 1-D time-difference-of-arrival (TDOA) measurements provided by pairs of hydrophones. Our numerical results demonstrate favorable detection and estimation accuracy compared to state-of-the-art reference techniques.

show abstract

Section: A State-of-the-artmentioning

confidence: 99%

Section: A State-of-the-artmentioning

confidence: 99%

Multisensor Multiobject Tracking With Improved Sampling Efficiency

Zhang,

Meyer

2024

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

show abstract

“…In many Bayesian inference problems, it is very important to estimate the normalizing constant Z k . For a general HMM, analytical evaluation of (41) and (42) is not possible, as π k−1 is not tractable. However, in a linear Gaussian filtering problem, the posterior distribution can be analytically computed from a Kalman filter, which allows exact estimation of Z k .…”

Section: B Estimation Of Normalizing Constants From a Linear Gaussiamentioning

confidence: 99%

“…The PF-GMM [41], which uses a separate LEDH filter to track each component of the posterior, performs reasonably well, whereas the particle flow algorithms LEDH, EDH and the particle flow particle filters perform poorly as they are better suited for uni-modal posterior distributions. The Gaussian sum particle filter (GSPF) [42], which approximates each component of the predictive and posterior densities by a Gaussian distribution by performing importance sampling exhibits poor representation capability in higher dimensions.…”

Section: Nonlinear Model With Gmm Process and Measurement Noisesmentioning

confidence: 99%

Invertible Particle-Flow-Based Sequential MCMC With Extension to Gaussian Mixture Noise Models

Pal

Coates

2019

IEEE Trans. Signal Process.

Self Cite

View full text Add to dashboard Cite

Sequential state estimation in non-linear and non-Gaussian state spaces has a wide range of applications in statistics and signal processing. One of the most effective non-linear filtering approaches, particle filtering, suffers from weight degeneracy in high-dimensional filtering scenarios. Several avenues have been pursued to address high-dimensionality. Among these, particle flow particle filters construct effective proposal distributions by using invertible flow to migrate particles continuously from the prior distribution to the posterior, and sequential Markov chain Monte Carlo (SMCMC) methods use a Metropolis-Hastings (MH) accept-reject approach to improve filtering performance. In this paper, we propose to combine the strengths of invertible particle flow and SMCMC by constructing a composite Metropolis-Hastings (MH) kernel within the SMCMC framework using invertible particle flow. In addition, we propose a Gaussian mixture model (GMM)-based particle flow algorithm to construct effective MH kernels for multi-modal distributions. Simulation results show that for high-dimensional state estimation example problems the proposed kernels significantly increase the acceptance rate with minimal additional computational overhead and improve estimation accuracy compared with state-of-the-art filtering algorithms.Y. Li is with the

show abstract

“…The computational requirements of this filter scale poorly with the state dimension. As an improvement, [21] proposed an approximate particle flow where both prior and likelihood are modelled as Gaussian mixture distributions. Particle flow filters involve various model assumptions for tractable solutions or approximations in numerical implementations, which leads to statistical inconsistency.…”

Section: Introductionmentioning

confidence: 99%

Particle Flow Gaussian Sum Particle Filter

Comandur¹,

Liu²,

Nannuru³

2022

Preprint

View full text Add to dashboard Cite

Particle flow Gaussian particle flow (PFGPF) uses an invertible particle flow to generate a proposal density. It approximates the predictive and posterior distributions as Gaussian densities. In this paper, we use bank of PFGPF filters to construct a Particle flow Gaussian sum particle filter (PFGSPF), which approximates the predictive and posterior as Gaussian mixture model. This approximation is useful in complex estimation problems where a single Gaussian approximation is not sufficient. We compare the performance of this proposed filter with PFGPF and others in challenging numerical simulations.

show abstract

Gaussian sum particle flow filter

Cited by 6 publications

References 30 publications

Multisensor Multiobject Tracking With Improved Sampling Efficiency

Multisensor Multiobject Tracking With Improved Sampling Efficiency

Invertible Particle-Flow-Based Sequential MCMC With Extension to Gaussian Mixture Noise Models

Particle Flow Gaussian Sum Particle Filter

Contact Info

Product

Resources

About