Generalized CPHD filter modeling spawning targets

Bearings-only multitarget tracking based onRao-Blackwellized particle CPHD filter

“…It can be observed that Eq. (15) describes a linear Gaussian model and the states can be estimated by the Kalman filter (KF). The nonlinear states can be estimated by utilizing the sequential Monte Carlo (SMC) method.…”

“…Many scholars have carried out relevant research [10][11][12][13][14]. In [15], based on the famous Faà di Bruno determinant, a tractable recursion computation technique of the general cardinality prediction equation is presented. In [16], a robust CPHD based on interacting multiple model (RCPHD-IMM) is proposed for multiple maneuvering targets tracking under the Doppler blind zone of airborne pulse Doppler radar.…”

The CPHD Filter With Target Spawning

“…Another work deals with CPHD spawning [16] that cites and forms comparisons with an earlier version of the current article [17]. The most significant difference between presented results is that a conclusive analytical expression for the predicted cardinality distribution that includes arbitrary birth and spawning is not provided in [16], i.e., quantities remain requiring further derivation by the reader, whereas the current article presents conclusive, analytical and tractable expressions in Section III-B.…”

“…Another work deals with CPHD spawning [16] that cites and forms comparisons with an earlier version of the current article [17]. The most significant difference between presented results is that a conclusive analytical expression for the predicted cardinality distribution that includes arbitrary birth and spawning is not provided in [16], i.e., quantities remain requiring further derivation by the reader, whereas the current article presents conclusive, analytical and tractable expressions in Section III-B. Additionally, while derivations in this article include the application of Faà di Bruno's formula to probability generating functionals (p.g.fl.s) describing point processes, the approach in [16] is more comparable to that of the technical report [15] in that both apply the formula to probability generating functions (p.g.fs) describing probability mass functions.…”

Information Fusion for Industrial Mobile Platform Safety via Track-Before-Detect Labeled Multi-Bernoulli Filter

“…Many theoretical and algorithmic improvements for these algorithms are proposed in the literature. Such methods include: a GM-PHD filter with false alarm detection using an irregular window [25], a variable structure multiple model GMCPHD filter [26], multiple modal PHD filters for tracking maneuvering targets [27,28,29], joint underwater target detection and tracking using the Bernoulli filter [30] and a generalized CPHD filter with a model for spawning targets [31].…”