The Gaussian Mixture Probability Hypothesis Density Filter英文文献资料.pdfVIP

IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 54, NO. 11, NOVEMBER 2006 4091 The Gaussian Mixture Probability Hypothesis Density Filter Ba-Ngu Vo and Wing-Kin Ma, Member, IEEE Abstract—A new recursive algorithm is proposed for jointly An intrinsic problem in multiple-target tracking is the un- estimating the time-varying number of targets and their states known association of measurements with appropriate targets from a sequence of observation sets in the presence of data associ- [1], [2], [6], [7]. Due to its combinatorial nature, the data ation uncertainty, detection uncertainty, noise, and false alarms. The approach involves modelling the respective collections of association problem makes up the bulk of the computational targets and measurements as random finite sets and applying the load in multiple-target tracking algorithms. Most traditional probability hypothesis density (PHD) recursion to propagate the multiple-target tracking formulations involve explicit associa- posterior intensity, which is a first-order statistic of the random tions between measurements and targets. Multiple hypotheses finite set of targets, in time. At present, there is no closed-form tracking (MHT) and its variations concern the propagation of solution to the PHD recursion. This paper shows that under linear, Gaussian assumptions on the target dynamics and birth process, association hypotheses in time [2], [6], [7]. The joint proba- the posterior intensity at any time step is a Gaussian mixture. bilistic data association filter (JPDAF) [1], [8], the probabilistic M