Centralized Cooperative Sensor Fusion for Dynamic Sensor Network With Limited Field-of-View via Labeled Multi-Bernoulli Filter

Heterogeneous Multi-Sensor Fusion With Random Finite Set Multi-Object Densities

“…It should be noted that, after implementing union find set, a subset M i,g can contain more than one Bernoulli component when potential objects are in close proximity. In this case, one can sequentially examine the Bernoulli components in subset M i,g and retain only the best one according to (41).…”

“…This problem is especially evident in the fusion among sensors having different fields of view (FoVs), since the relative information confidence is different in different regions due to the non-overlapping FoVs [34]. Recently, the different FoV issue has been intensively studied, and effective countermeasures have been proposed for both unlabel MODs [34]- [38] and labelled RFS MODs [39]- [41]. However, these methods do not change the fundamental problem of the fusion mechanism.…”

Cooperative Localization and Multitarget Tracking in Agent Networks with the Sum-Product Algorithm

“…It follows that the development of robust, reliable, scalable and efficient multitarget tracking algorithms becomes of paramount importance, as safety issues are involved. Abundant literature on multitarget tracking is available, starting from the pioneering works in [36]- [39] through very recent studies such as [40]- [48]. Approaches based on SPA have been proposed as well, both with stationary sensors -whose location is either known [35], [49] or unknown [50] -and with mobile ones [18], [51]- [53].…”

“…Focusing on multitarget tracking algorithms with mobile sensors, the cited works are affected by the following limitations: in [51] sensors do not localize themselves cooperatively; in [52] the maximum number of targets that can be tracked simultaneously is limited and needs to be set a priori; in [18] and [53] the number of targets is time-invariant and known, and, in addition, in [18] neither false alarms nor missed detections are considered, and in [53] the association between targets and measurements is assumed known. Random finite sets (RFSs) constitute an alternative framework for the development of multitarget tracking methods 1 both with stationary [40], [41], [48] and mobile sensors [45]. In particular, in [45] the authors develop a Poisson multi-Bernoulli multitarget tracking filter that jointly estimates the uncertain mobile sensor states and target states using two types of measurements: sensor state measurements, e.g., global navigation satellite system (GNSS) measurements, and target measurements.…”

An efficient multi-objective optimization approach for sensor management via multi-Bernoulli filtering

“…Compared with the single-sensor system, the multi-sensor system produces much more accurate estimation by using the spatial diversity. There are three major multi-sensor system architectures, namely centralized [19][20][21], distributed [22][23][24][25], and decentralized [26,27]. However, the multi-sensor MTT problem is challenging.…”