"Spooky Action at a Distance" in the Cardinalized Probability Hypothesis Density Filter

Fränken, Dietrich; Schmidt, Maximilian; Ulmke, Martin

doi:10.1109/taes.2009.5310327

Cited by 73 publications

(55 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is worth noting that when the third target disappears at the 50th minute, the OSPA distance of the MMP-CPHD algorithm is larger than those of the other algorithms. This is due to that the missed detection problem [28] has been cured in the CPHD method, which is beneficial when missed detection occurs but is harmful when targets really disappear.…”

Section: Simulationsmentioning

confidence: 99%

Multi‐model particle cardinality‐balanced multi‐target multi‐Bernoulli algorithm for multiple manoeuvring target tracking

Yang

2013

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

Multiple manoeuvring target tracking is an extremely difficult problem in the target tracking field, especially under the non-linear systems. The probability hypothesis density (PHD) and cardinalised PHD (CPHD) algorithms based on the particle filter have proved to be promising algorithms for multi-target tracking. However, they have a heavy computational burden because of the particle clustering in the stage of state extraction. Especially, the additional calculation is added to the CPHD algorithm because of the estimation of the cardinality distribution. To solve the problem, the authors propose a novel multiple manoeuvring target tracking algorithm by extending the multi-model method to the cardinality-balanced multi-target multiBernoulli filter and then using the sequential Monte Carlo implementation. Moreover, in order to obtain the individual target tracks, the particle labelling technique is introduced in the proposed algorithm. Simulation results show that the proposed algorithm can effectively achieve the track continuity for the multiple manoeuvring target tracking and has a higher accuracy of state estimates than the multiple model particle PHD and CPHD algorithm with a better computational efficiency.

show abstract

Section: Simulationsmentioning

confidence: 99%

Multi‐model particle cardinality‐balanced multi‐target multi‐Bernoulli algorithm for multiple manoeuvring target tracking

Yang

2013

IET Radar, Sonar & Navigation

View full text Add to dashboard Cite

show abstract

“…Hence the Lemma holds by induction. Proposition 2: Suppose that the multi-target posterior density is an LMB RFS with state space , (finite) label space , and parameter set , and that the multi-target birth model is an LMB RFS with state space , (finite) label space , and parameter set , then the multi-target predicted density is also an LMB RFS with state space and finite label space (with ) given by the parameter set (36) where (37) (38) Proof: Enumerate the labels and rewrite the weight in (29), in the following form:…”

Section: A Predictionmentioning

confidence: 99%

“…Remark 3: The multi-target prediction for an LMB process actually coincides with performing the prediction on the unlabeled process and interpreting the component indices as track labels. Thus to perform the LMB filter prediction it is sufficient to predict the parameters forward according to (36) which is identical to the prediction for the multi-Bernoulli filter. This result is used later in implementations.…”

Section: A Predictionmentioning

confidence: 99%

“…On the other hand, by utilizing the labeled RFS formulation [34], [35], the proposed LMB filter formally estimates target tracks and does not require high signal to noise ratio. The proposed LMB filter also does not exhibit the so called "spooky effect" [36], a phenomenon present in the CPHD filter due to its use of an i.i.d. cluster approximation, which causes a redistribution of target masses within the PHD when missed detections are encountered.…”

Section: Introductionmentioning

confidence: 96%

See 1 more Smart Citation

The Labeled Multi-Bernoulli Filter

Reuter,

Vo,

et al. 2014

IEEE Trans. Signal Process.

564

View full text Add to dashboard Cite

This paper proposes a generalization of the multiBernoulli filter called the labeled multi-Bernoulli filter that outputs target tracks. Moreover, the labeled multi-Bernoulli filter does not exhibit a cardinality bias due to a more accurate update approximation compared to the multi-Bernoulli filter by exploiting the conjugate prior form for labeled Random Finite Sets. The proposed filter can be interpreted as an efficient approximation of the -Generalized Labeled Multi-Bernoulli filter. It inherits the advantages of the multi-Bernoulli filter in regards to particle implementation and state estimation. It also inherits advantages of the -Generalized Labeled Multi-Bernoulli filter in that it outputs (labeled) target tracks and achieves better performance.Index Terms-Bayesian estimation, conjugate prior, marked point process, random finite set, target tracking.

show abstract

“…The cardinalized PHD (CPHD) filter [8] offers a nice cure to the problem of premature target death through adding memory to the target-number process, and although it introduces an artificial linkage between all targets' existence probabilities (a significant effect observed as "spooky action at a distance" in [9]), it seems to work very effectively in most applications we have seen. It is a PHD filter with a hierarchically "supervising" hidden Markov model (HMM) that describes the number of targets in the scene; that is, the cardinality of the Poisson RFS is not restricted to be Poisson distributed, and can in fact be arbitrary.…”

Section: Cardinalized Probability Hypothesis Density Filtermentioning

confidence: 99%

The Multiple Model CPHD Tracker

Georgescu

Willett

2012

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

The probability hypothesis density (PHD) is a practical approximation to the full Bayesian multi-target filter. The cardinalized PHD (CPHD) filter was proposed to deal with the "target death" problem of the PHD filter. A multiple-model PHD exists; in this work, a multiple model version of the considerably more complex CPHD filter is derived. It is implemented using Gaussian mixtures, and a track management (for display and scoring) strategy is developed.Index Terms-Cardinalized probability hypothesis density filter, CPHD, MMCPHD, MMPHD, multiple model, PHD.

show abstract

"Spooky Action at a Distance" in the Cardinalized Probability Hypothesis Density Filter

Cited by 73 publications

References 10 publications

Multi‐model particle cardinality‐balanced multi‐target multi‐Bernoulli algorithm for multiple manoeuvring target tracking

Multi‐model particle cardinality‐balanced multi‐target multi‐Bernoulli algorithm for multiple manoeuvring target tracking

The Labeled Multi-Bernoulli Filter

The Multiple Model CPHD Tracker

Contact Info

Product

Resources

About