Asynchronous Non–uniform Distributed Multi–target Tracking Filter Based on Asymmetric Alpha–divergence Consensus
Yuqin Zhou, Liping Yan, Hui Li, Yuanqing Xia · IEEE Transactions on Aerospace and Electronic Systems · 2022
With more and more extensive application of target tracking, distributed multi–target tracking (DMT) becomes an important research direction. However, the synchronization characteristics of some sensor networks can not be guaranteed due to communication delay, non–uniform sampling and so on. Therefore, a DMT algorithm for asynchronous non–uniform multi–sensor networks is studied in this paper. Firstly, by using the cardinalized probability hypothesis density (CPHD) based on continuous–discrete multi–target dynamic (CD) model (CD-CPHD), a CD–CPHD algorithm with multi–step birth process and time–triggered structure (TCD–CPHD) is proposed; Secondly, an asymmetric alpha–divergence (AAD) based adaptive computing structure of time–trigger index is designed to reduce the communication rate of TCD–CPHD; Thirdly, the fusion rule of AAD consensus is derived to construct the fusion process of the proposed algorithm; Finally, by combining the above structures and the fusion rule, the implementation process of the proposed algorithm is given. Theoretical analysis and exhaustive experimental analysis show the effectiveness of the proposed algorithm.