Multi-target Bayes filter with the target detection

Liu, Zongxiang; Zou, Yanni; Xie, Weixin; Li, Liang-qun

doi:10.1016/j.sigpro.2017.05.016

Cited by 12 publications

(10 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we reveal the performance of the proposed AM-MDMTB filter by comparing this filter with the GM-PHD filter [7], CBMeMber filter [9], LMB filter [28], R-GLMB filter [30] and MDMTB filter [10], and exploit the OSPA distance [35] with c = 100 m and p = 2 as the assessment criteria.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…The MDMTB filter was proposed in [10] and [11] to propagate the state distribution (i.e., probability density) and existence probability of each target in filter recursion. It assumes that each target evolves and generates the measurement independently of one another.…”

Section: Problem Formation a Mdmtb Filtermentioning

confidence: 99%

“…where N k is the number of hypothesized targets at time step k. For more details, we refer the reader to [10] and [11].…”

Section: Problem Formation a Mdmtb Filtermentioning

confidence: 99%

“…In the linear Gaussian system, transition probability f (x i,k |x i,k−1 ) in (3) and probability density f (z j,k |x i,k ) in (10) and (12) are given by…”

Section: Am-mdmtb Filter For a Linear Gaussian Systemmentioning

confidence: 99%

“…However, they have a weak memory and become inefficient in case of low detecting probability because they fail to provide state estimation of an existing target if this target is undetected at a time. Instead of maintaining the joint posterior probability density of multiple targets, the marginal distribution multi-target Bayes (MDMTB) filter was proposed in [10] and [11] to propagate the state distribution and existence probability of each target in the filter recursion. The advantage of the MDMTB filter over the PHD filter and CBMeMber filter is that it may provide the state estimations of missed targets in case of low detecting probability due to its strong memory.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Marginal Distribution Multi-Target Bayes Filter With Assignment of Measurements

Liu

Gan

2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

This work proposes a marginal distribution multi-target Bayes filter with assignment of measurements to track multiple targets in the presence of an unknown and variable number of targets, clutter, and missed detections. Mathematically, the association of the measurements with either a target or clutter may be established by maximizing the joint likelihood function of the measurement partition, which leads to a two-dimensional assignment problem. By the introduction of detecting label, a handling approach for missed detections is also developed and is applied to the proposed filter. This filter greatly reduces the number of hypothesized targets or Gaussian terms by selecting the predicted probability density of a target or one of its multiple updated probability densities as its state distribution at each time step. Experimental results indicate that the proposed filter requires a less computational load than the existing filters and performs better than the efficient implementations of the δ-generalized labeled multi-Bernoulli filter for multi-target tracking at low and moderate clutter densities. INDEX TERMS multi-target tracking, marginal distribution multi-target Bayes filter, generalized labeled multi-Bernoulli filter, two-dimensional assignment.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Section: Problem Formation a Mdmtb Filtermentioning

confidence: 99%

“…where N k is the number of hypothesized targets at time step k. For more details, we refer the reader to [10] and [11].…”

Section: Problem Formation a Mdmtb Filtermentioning

confidence: 99%

“…In the linear Gaussian system, transition probability f (x i,k |x i,k−1 ) in (3) and probability density f (z j,k |x i,k ) in (10) and (12) are given by…”

Section: Am-mdmtb Filter For a Linear Gaussian Systemmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations