scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

On adaptive probability hypothesis density filter for multi-target tracking

01 Jul 2017-pp 5424-5428
TL;DR: In this paper, an adaptive probability hypothesis density (PHD) filter was proposed for multi-target tracking, where both the completed clutter process and the single-target measurement likelihood were improved based on the Bayesian theory.
Abstract: In order to improve tracking performance of the existing probability hypothesis density (PHD) filters, we present an adaptive filter for multi-target tracking in this paper. At first, both the completed clutter process and the single-target measurement likelihood are improved based on the Bayesian theory. Then, the target cardinality is corrected using the adaptive detection gate. What's more, a novel particle implementation is explored step by step. Numerical study results have been carried out to confirm the promising tracking performance of the proposed PHD filter.
Citations
More filters
Journal Article
TL;DR: In this paper, an improved multiple model probability hypothesis density (PHD) filter is proposed to solve the particle degenerate problem and the Poisson assumption for the target number distribution.

5 citations

Proceedings ArticleDOI
25 Jul 2018
TL;DR: The numeric simulation indicated that the proposed PHD filter can effectively track occluded target in cluttered environment with stable track estimation and accurate cardinality statistics.
Abstract: To deal with the problem of unsatisfactory performance of close target tracking in cluttered environmental by using the standard probability hypothesis density (PHD) filter, this paper presented an improved PHD filter. First, the track management scheme was proposed to correct the number of targets and clutters. As a result, the number of required particles was reduced. In the measurement update step, the matrix consisted of both particle weight and clutter intensity was built to reassign weight proportion under the unchanged sum of all elements based on the row-column operation. The numeric simulation indicated that the proposed filter can effectively track occluded target in cluttered environment with stable track estimation and accurate cardinality statistics.

Cites background from "On adaptive probability hypothesis ..."

  • ...However, the filters have unsatisfactory tracking performance when the target has nonlinear motion state and the clutter occurs in the surveillance region [6,11]....

    [...]

References
More filters
Journal ArticleDOI
TL;DR: In this paper, a sequential Monte Carlo (SMC) multitarget filter is proposed and demonstrated on a number of simulated scenarios, which is suitable for problems involving nonlinear nonGaussian dynamics.
Abstract: Random finite sets (RFSs) are natural representations of multitarget states and observations that allow multisensor multitarget filtering to fit in the unifying random set framework for data fusion. Although the foundation has been established in the form of finite set statistics (FISST), its relationship to conventional probability is not clear. Furthermore, optimal Bayesian multitarget filtering is not yet practical due to the inherent computational hurdle. Even the probability hypothesis density (PHD) filter, which propagates only the first moment (or PHD) instead of the full multitarget posterior, still involves multiple integrals with no closed forms in general. This article establishes the relationship between FISST and conventional probability that leads to the development of a sequential Monte Carlo (SMC) multitarget filter. In addition, an SMC implementation of the PHD filter is proposed and demonstrated on a number of simulated scenarios. Both of the proposed filters are suitable for problems involving nonlinear nonGaussian dynamics. Convergence results for these filters are also established.

1,248 citations


"On adaptive probability hypothesis ..." refers background or methods in this paper

  • ...Further, the point spread function in (10) is given by [1,11]:...

    [...]

  • ...After defining the true set =1 g i i X x and the estimated track set respectively, we write the OSPA distance as [1,4,11]: =1 ˆ ˆ e i i X x...

    [...]

Book
24 Oct 2014
TL;DR: In this paper, Finite-Set Statistics RFS Filters for Nonstandard Measurement Models Sensor, Platform, and Weapons Management (SPMM) Filters: Standard Measurement Model (SMM) and Non-standard Model (NMM)
Abstract: Introduction Elements of Finite-Set Statistics RFS Filters: Standard Measurement Model RFS Filters for Unknown Backgrounds RFS Filters for Nonstandard Measurement Models Sensor, Platform, and Weapons Management

461 citations


"On adaptive probability hypothesis ..." refers background or methods in this paper

  • ...Further, the point spread function in (10) is given by [1,11]:...

    [...]

  • ...where , n k I is the source intensity, 2 is the variance of measurement error, and is the time–updated position coordinate [1]....

    [...]

  • ...After defining the true set =1 g i i X x and the estimated track set respectively, we write the OSPA distance as [1,4,11]: =1 ˆ ˆ e i i X x...

    [...]

  • ...1 Introduction According to the stochastic–geometry formulation of random finite set (RFS) theory, the probability hypothesis density (PHD) filter has been applied in the target state and measurement spaces [1]....

    [...]

  • ...2 Problem Statements Considering the 2–dimension (2D) state vector with the position coordinate T , , , , k k k k k k x x y y x , k k x y , the velocity coordinate , k k x y , and the turn rate k , where T denotes the transpose of matrix, we have the stochastically dynamic equations at time k [1,4,8]:...

    [...]

01 Jan 2007
TL;DR: Multi-Target Tracking Using 1st Moment of Random Finite Sets using 1st moment of random Finite sets for multi- target tracking.
Abstract: Multi-Target Tracking Using 1st Moment of Random Finite Sets

23 citations


"On adaptive probability hypothesis ..." refers background in this paper

  • ...By propagating the PHD of multi–target posterior density, the scalar product of the PHD in state space represents the estimated cardinality of targets, and the associated peaks define the state estimates [2]....

    [...]

Journal ArticleDOI
TL;DR: In this paper, the performance of particle methods for the implementation of the class of Bayes filters formulated using the random finite set formalism is described in the context of bearings-only target tracking.

23 citations

Posted Content
TL;DR: This overview paper describes the particle methods developed for the implementation of the class of Bayes filters formulated using the random finite set formalism and focuses in on the Bernoulli particle filter, the probability hypothesis density (PHD) particle filter and the generalised labelled multi-Bernoulli (GLMB) particle filters.
Abstract: This overview paper describes the particle methods developed for the implementation of the a class of Bayes filters formulated using the random finite set formalism. It is primarily intended for the readership already familiar with the particle methods in the context of the standard Bayes filter. The focus in on the Bernoulli particle filter, the probability hypothesis density (PHD) particle filter and the generalised labelled multi-Bernoulli (GLMB) particle filter. The performance of the described filters is demonstrated in the context of bearings-only target tracking application.

22 citations