Research on measurement set partitioning method for tracking multiple extended targets
Hongyan Zhu, Pandeng Zhang, Tingting Ma · 2015
In extended target tracking, it is a critical step to implement the measurement set partitioning, which aims to partition the whole measurement set into several distinct clusters. When multiple extended targets are in close proximity, there are great challenges. To deal with this problem, we introduce the DBSCAN-FCM partition to the Gaussian inverse Wishart PHD (GIW-PHD) filter, which combines the density based spatial clustering of application with noise (DBSCAN) and the Fuzzy C means (FCM) clustering. Selected simulation results are provided to demonstrate the performance superiority of the proposed method compared with other competing algorithms.