Consensus CPHD Filter for Distributed Multitarget Tracking

Battistelli, Giorgio; Chisci, Luigi; Fantacci, Claudio; Farina, A.; Graziano, A.

doi:10.1109/jstsp.2013.2250911

Cited by 295 publications

(259 citation statements)

References 25 publications

Supporting

Mentioning

259

Contrasting

Order By: Relevance

“…Examples include the fusion of Poisson multi-object posteriors of multiple local PHD filters [33], i.d.d. clusters densities of several local CPHD filters [34], multi-Bernoulli densities of local multi-Bernoulli filters [35], and LMB or Vo-Vo densities of several local LMB or Vo-Vo filters [6]. The problem of multi-sensor control for labeled random set filters is recently considered by Meng et al [3].…”

Section: Sensor Fusion and Optimal Controlmentioning

confidence: 99%

Multi-sensor control for multi-object Bayes filters

et al. 2018

View full text Add to dashboard Cite

Sensor management in multi-object stochastic systems is a theoretically and computationally challenging problem. This paper presents a novel approach to the multi-target multi-sensor control problem within the partially observed Markov decision process (POMDP) framework. We model the multi-object state as a labeled multi-Bernoulli random finite set (RFS), and use the labeled multi-Bernoulli filter in conjunction with minimizing a task-driven control objective function: posterior expected error of cardinality and state (PEECS). A major contribution is a guided search for multi-dimensional optimization in the multi-sensor control command space, using coordinate descent method. In conjunction with the Generalized Covariance Intersection method for multi-sensor fusion, a fast multi-sensor algorithm is achieved. Numerical studies are presented in several scenarios where numerous controllable (mobile) sensors track multiple moving targets with different levels of observability. The results show that our method works significantly faster than the approach taken by a state of art method, with similar tracking errors. Index Termspartially observed Markov decision process, multi-target tracking, random finite sets, labeled multi-Bernoulli filter, coordinate descent.

show abstract

Section: Sensor Fusion and Optimal Controlmentioning

confidence: 99%

Multi-sensor control for multi-object Bayes filters

et al. 2018

View full text Add to dashboard Cite

show abstract

“…A distributed fusion of SMC-CPHD filters via exponential mixture densities (EMD) has been presented by Uney et al in [8]. Battistelli et al used EMD for distributed fusion of GM-CPHD densities [9]. Common for all three works is the assumption that all PHD filters work in the same domain, i.e., that all agents have entirely overlapping FOV in which objects are sensed.…”

Section: Firstnamelastname@epflchmentioning

confidence: 99%

“…It is shown in [9] that GCI can be approximated by applying Covariance Intersection (CI) pairwise to components from the two intensities. But this approach does not work if the vehicle FOVs do not overlap entirely, because an object seen by only one vehicle does not have its corresponding component in the set of another vehicle, so it likely gets discarded.…”

Section: B Cooperative Fusionmentioning

confidence: 99%

A system implementation and evaluation of a cooperative fusion and tracking algorithm based on a Gaussian Mixture PHD filter

Vasić

Mansolino

Martinoli

2016

2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Abstract-This paper focuses on a real system implementation, analysis, and evaluation of a cooperative sensor fusion algorithm based on a Gaussian Mixture Probability Hypothesis Density (GM-PHD) filter, using simulated and real vehicles endowed with automotive-grade sensors. We have extended our previously presented cooperative sensor fusion algorithm with a fusion weight optimization method and implemented it on a vehicle that we denote as the ego vehicle. The algorithm fuses information obtained from one or more vehicles located within a certain range (that we call cooperative), which are running a multi-object tracking PHD filter, and which are sharing their object estimates. The algorithm is evaluated on two Citroën C-ZERO prototype vehicles equipped with Mobileye cameras for object tracking and lidar sensors from which the ground truth positions of the tracked objects are extracted. Moreover, the algorithm is evaluated in simulation using simulated C-ZERO vehicles and simulated Mobileye cameras. The ground truth positions of tracked objects are in this case provided by the simulator. Multiple experimental runs are conducted in both simulated and real-world conditions in which a few legacy vehicles were tracked. Results show that the cooperative fusion algorithm allows for extending the sensing field of view, while keeping the tracking accuracy and errors similar to the case in which the vehicles act alone.

show abstract

“…Its two well known implementations are based on a Gaussian Mixture (GM) model [6], and a Sequential Monte Carlo (SMC) model [7]. Works on fusing together the PHD intensities originating from different sources exist: an approach for GM-PHD intensities is given in [8] and for SMC-PHD intensities in [9]. Common for these works is the assumption that there exist a unique FOV (area) that is covered by all participating sensors simultaneously.…”

Section: Related Workmentioning

confidence: 99%

“…In order to avoid data incest problem, we use a General Covariance Intersection (GCI) algorithm, which offers a conservative way of fusing two Gaussian mixtures [12]. It is shown in [8] that GCI can be approximated to applying Covariance Intersection (CI) pairwise to components from the two intensities. To address the fact that the FOVs may not overlap entirely, we apply CI intersection only to components whose Mahalanobis distance from each other is less than T F .…”

Section: B Cooperative Fusion Of Phd Intensitiesmentioning

confidence: 99%

An overtaking decision algorithm for networked intelligent vehicles based on cooperative perception

Vasić

Lederrey

Navarro

et al. 2016

2016 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Abstract-This paper presents an overtaking decision algorithm for networked intelligent vehicles. The algorithm is based on a cooperative tracking and sensor fusion algorithm that we previously developed. The ego vehicle is equipped with lane keeping and lane changing capabilities, as well as a forward-looking lidar sensor. The lidar data are fed to the tracking module which detects other vehicles, such as the vehicle that is to be overtaken (leading) and the oncoming traffic. Based on the estimated distances to the leading and the oncoming vehicles and their speeds, a risk is calculated and a corresponding overtaking decision is made. We compare the performance of the overtaking algorithm between the case when the ego vehicle only relies on its lidar sensor, and the case in which it fuses object estimates received from the leading car which also has a forward-looking lidar. Systematic evaluations are performed in Webots, a calibrated high-fidelity simulator.

show abstract

Consensus CPHD Filter for Distributed Multitarget Tracking

Cited by 295 publications

References 25 publications

Multi-sensor control for multi-object Bayes filters

Multi-sensor control for multi-object Bayes filters

A system implementation and evaluation of a cooperative fusion and tracking algorithm based on a Gaussian Mixture PHD filter

An overtaking decision algorithm for networked intelligent vehicles based on cooperative perception

Contact Info

Product

Resources

About