Accelerating the Single Cluster PHD Filter with a GPU implementation

The SC-PHD filter is an algorithm which was designed to solve a class of multiple object estimation problems where it is necessary to estimate the state of a single-target parent process, in addition to estimating the state of a multi-object population which is conditioned on it. The filtering process usually employs a number of particles to represent the parent process, coupled each with a conditional PHD filter, which is computationally burdensome. In this article, an implementation is described which exploits the parallel nature of the filter to obtain considerable speed-up with the help of a GPU. Several considerations need to be taken into account to make efficient use of the GPU, and these are also described here.

[1]  Joaquim Salvi,et al.  SLAM with single cluster PHD filters , 2012, 2012 IEEE International Conference on Robotics and Automation.

[2]  R. Mahler.  Multitarget Bayes filtering via first-order multitarget moments , 2003 .

[3]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[4]  Branko Ristic,et al.  Calibration of Multi-Target Tracking Algorithms Using Non-Cooperative Targets , 2013, IEEE Journal of Selected Topics in Signal Processing.

[5]  Daniel E. Clark,et al.  Bayesian Estimation of the Intensity for Independent Cluster Point Processes: An analytic solution , 2011 .

[6]  Jeremie Houssineau,et al.  PHD filter with diffuse spatial prior on the birth process with applications to GM-PHD filter , 2010, 2010 13th International Conference on Information Fusion.

[7]  Daniel E. Clark,et al.  First-moment filters for spatial independent cluster processes , 2010, Defense + Commercial Sensing.

[8]  Daniel E. Clark,et al.  Simultaneous tracking of multiple particles and sensor position estimation in fluorescence microscopy images , 2013, 2013 International Conference on Control, Automation and Information Sciences (ICCAIS).

[9]  Ba-Ngu Vo,et al.  The Gaussian Mixture Probability Hypothesis Density Filter , 2006, IEEE Transactions on Signal Processing.

[10]  Bernard Mulgrew,et al.  Cooperative sensor localisation in distributed fusion networks by exploiting non-cooperative targets , 2014, 2014 IEEE Workshop on Statistical Signal Processing (SSP).

[11]  Joaquim Salvi,et al.  SLAM With Dynamic Targets via Single-Cluster PHD Filtering , 2013, IEEE Journal of Selected Topics in Signal Processing.

[12]  Ronald P. S. Mahler,et al.  Statistical Multisource-Multitarget Information Fusion , 2007 .