On the ordering of the sensors in the iterated-corrector probability hypothesis density (PHD) filter

Nagappa, Sharad; Clark, Dan E.

doi:10.1117/12.884618

Cited by 36 publications

(38 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the IC-PHD filter, the updated PHD of each sensor is the predicted PHD of the next one, and the updated PHD of the last sensor is considered as tracking results. In [32], it is pointed out that the IC-PHD filter is affected by the sensor order and sensitive to the probability of detection. The product multi-sensor PHD (PM-PHD) filter [33], an improved version of the IC-PHD filter, is proposed to deal with this problem.…”

Section: Introductionmentioning

confidence: 99%

A cardinality modified product multi-sensor PHD

Ling

Fan

2016

Information Fusion

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

A cardinality modified product multi-sensor PHD

Ling

Fan

2016

Information Fusion

View full text Add to dashboard Cite

“…It is shown that the PHD estimation performance differs significantly with different permutations of sensor ordering, and the sensor with the lower detectability should be updated in earlier stages [53]. As an alternative implementation, the product multi-sensor PHD filter provides more robust estimation performance against sensor ordering [55].…”

Section: Update Of Daughter Processmentioning

confidence: 99%

“…In this work, an iterated multi-sensor implementation, referring to the iterated-corrector PHD update, is utilized as in [47,53]. Starting with the predicted PHD intensity Therefore, the multi-sensor PHD update is iteratively performed over N stages, referring to total N BS nodes.…”

Section: Update Of Daughter Processmentioning

confidence: 99%

“…Conditioning on each particle of the parent process, the random finite set of the scatter state is denoted as the daughter process, and its PHD intensity is represented using a Gaussian Mixture structure. Therefore, the distribution of parent process is updated using a particle filter; and for each particle, the associated PHD surface is updated via a multi-sensor iterated Gaussian Mixture PHD filter [47,53]. The…”

mentioning

confidence: 99%

See 1 more Smart Citation

Simultaneous Target and Multipath Positioning

Krolik

2014

IEEE J. Sel. Top. Signal Process.

View full text Add to dashboard Cite

In this work, we present the Simultaneous Target and Multipath Positioning (STAMP) technique to jointly estimate the unknown target position and uncertain multipath channel parameters. We illustrate the applications of STAMP for target tracking/geolocation problems using single-station hybrid TOA/AOA system, monostatic MIMO radar and multistatic range-based/AOA based localization systems.The STAMP algorithm is derived using a recursive Bayesian framework by including the target state and multipath channel parameters as a single random vector, and the unknown correspondence between observations and signal propagation channels is solved using the multi-scan multi-hypothesis data association. In the presence of the unknown time-varying number of multipath propagation modes, the STAMP algorithm is modified based on the single-cluster PHD filtering by modeling the multipath parameter state as a random finite set. In this case, the target state is defined as the parent process, which is updated by using a particle filter or multi-hypothesis Kalman filter. The multipath channel parameter is defined as the daughter process and updated based on an explicit Gaussian mixture PHD filter. Moreover, the idenfiability analysis of the joint estimation problem is provided in terms of Cramér-Rao lower bound (CRLB).The Fisher information contributed by each propagation mode is investigated, and the v effect of Fisher information loss caused by the measurement origin uncertainty is also studied. The proposed STAMP algorithms are evaluated based on a set of illustrative numeric simulations and real data experiments with an indoor multi-channel radar testbed. Substantial improvement in target localization accuracy is observed.vi

show abstract

“…Iterating the update for all synchronously received sensor measurements, yields to the PHD updated by multiple sensors. However, in [NC11] the limitations of the iterated-corrector approximation are demonstrated. It is shown that the value of the updated PHD significantly depends on the order in which the sensors' measurement sets are provided and the ratio between the sensors' detection probabilities p S D .…”

Section: The Iterated-corrector Approximation Of the Multi-sensor Phdmentioning

confidence: 99%

Intersection-Based Road User Tracking Using a Classifying Multiple-Model PHD Filter

Meißner

Reuter

Strigel

et al. 2014

IEEE Intell. Transport. Syst. Mag.

View full text Add to dashboard Cite

Bibliografische Information der Deutschen NationalbibliothekDie Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.dnb.de abrufbar. AbstractDue to an increasing number of traffic accident fatalities at intersections, the prevention of traffic accidents at intersection attracts more and more attention in public as well as in research. One example therefore, is the joint-project Ko-PER of the research initiative Ko-FAS which aims to increase the preventive traffic safety. This is realized by an preferably exhaustive recognition of the vehicles' environments based on locally gathered and communicated perception data. Therefore, especially at urban intersections, an intersection-based perception system which communicates the current traffic situation to vehicles in the intersection's surrounding is essential. Thus, a busy public intersection in Aschaffenburg, Germany was equipped with an intersection perception system, which contains multiple laserscanners and video cameras as well as data processing units. Beside the planning of the perception system, the core of this thesis is the laserscanner-based detection and classification of road users and the tracking of them by fusing detections of the laserscanners and video cameras.Since the laserscanners at the intersection are mounted at posts of lamps, traffic-lights as well as surrounding buildings, segmentation algorithms are introduced which utilize the knowledge about the consistent environment to segment the laserscanner measurements in background and foreground. Subsequently, the object detection in the foreground data relies on position as well as motion information. The 3D measurement points are clustered with a real-time capable implementation of the density based spatial clustering for applications with noise (DBSCAN) algorithm. Nevertheless, particularly large objects are sometimes not entirely detected due to occlusions, poor reflection characteristics and measurement uncertainties. This may lead to the separation of one object in multiple clusters. To avoid the overclustering, a combination of 2D tracking and 3D clustering methods is proposed. Beside the benefit of the considerably more robust object detection, additionally the determination of the objects' orientations is improved. Finally, the detected objects are classified by means of features of their laserscanner point clusters.The focus of this thesis lays on the tracking of road users at traffic intersections. Thus, a tracking approach is required which is able to handle multiple objects detected by multiple classifying sensors with different field of views. Due to the topdown modeling of the multi-object tracking problem and the availability of efficient implementations, a Gaussian mixture probability hypothesis density (GM-PHD) filter is used. Based on this the classifying multiple-model probability hypothesis density (CMMPHD) filter is developed which facilitates to track road user...

show abstract

On the ordering of the sensors in the iterated-corrector probability hypothesis density (PHD) filter

Cited by 36 publications

References 0 publications

A cardinality modified product multi-sensor PHD

A cardinality modified product multi-sensor PHD

Simultaneous Target and Multipath Positioning

Intersection-Based Road User Tracking Using a Classifying Multiple-Model PHD Filter

Contact Info

Product

Resources

About