Joint detection, tracking and classification algorithm for multiple maneuvering targets based on LGJMS-GMPHDF

Abstract

The Gaussian Mixture implementation of Probability Hypothesis Density Filter in Linear Gaussian Jump Markov multi-target System model (LGJMS-GMPHDF) is proved to be an effective tool for tracking an unknown and time-varying number of targets with uncertain target dynamics in clutter. This paper further integrates the class information into LGJMS-GMPHDF and proposes a recursive Joint Detection Tracking and Classification (JDTC) algorithm for multiple maneuvering targets in dense clutter. The main idea is to augment the kinematic state vector with the target class vector, and then use their combined measurement likelihood to integrating the target classification information into the update process of LGJMS-GMPHDF. The combined target kinematic state and class measurement likelihood improves the discrimination of different class targets and clutter, so better detection and tracking performance can be expected compared with the original LGJMS- GMPHDF. The classification probabilities and state vectors are updated synchronously. The proposed JDTC algorithm can simultaneously estimate the time-varying number of maneuvering target, their corresponding kinematic states and classes. The algorithm is demonstrated via a simulation example involving tracking of two closely spaced parallel moving targets and two crossing moving targets from different classes, where targets can appear and disappear.