Abstract
Multitarget tracking in clutter using bearings-only measurements is a challenging problem. In this paper, a performance improved nonlinear filter is proposed on the basis of the Random Finite Set (RFS) theory and is named as Gaussian mixture measurements-based cardinality probability hypothesis density (GMMbCPHD) filter. The GMMbCPHD filter enables to address two main issues: measurement-origin-uncertainty and measurement nonlinearity, which constitutes the key problems in bearings-only multitarget tracking in clutter. For the measurement-origin-uncertainty issue, the proposed filter estimates the intensity of RFS of multiple targets as well as propagates the posterior cardinality distribution. For the measurement-origin-nonlinearity issue, the GMMbCPHD approximates the measurement likelihood function using a Gaussian mixture rather than a single Gaussian distribution as used in extended Kalman filter (EKF). The superiority of the proposed GMMbCPHD are validated by comparing with several state-of-the-art algorithms via intensive simulation studies.
Highlights
Multitarget tracking in clutter [1] is an interesting but difficult problem needed to be investigated, especially when only bearings-only measurements are available
The improvements of cardinality and localization estimation of GMMbCPHD are presented by comparing with GMbPHDwEKF, GMbPHDwUKF, Gaussian mixture measurements-based probability hypothesis density (GMMbPHD), GMbCPHDwEKF as well as GMbCPHDwUKF
The initial target range information is barely known, for simplicity, the initial target range information required in the UKF-based and extended Kalman filter (EKF)-based algorithm is assumed to follow a Gaussian distribution in this simulation, with known mean r = 12,000 m and standard deviation σr = 4000 m and, the minimum and maximum target range information required in the GMMbPHD and GMMbCPHD are set to be [300 m, 18, 000 m], respectively
Summary
Multitarget tracking in clutter [1] is an interesting but difficult problem needed to be investigated, especially when only bearings-only measurements are available. In order to address multitarget tracking in clutter using bearings-only measurements, an improved RFS-based filter, called the Gaussian mixture measurements-based probability hypothesis density (GMMbPHD), was proposed in [17]. In order to fully improve the tracking performance on both position and cardinality accuracy, an further improved filter, termed as the Gaussian mixture measurements-based cardinality probability hypothesis density (GMMbCPHD), is proposed in this paper. Inspired from the idea of the Gaussian mixture measurement methodology, the GMMbCPHD algorithm approximates the likelihood function of the bearings-only measurements by a Gaussian mixture of numbers of refined Gaussian probability density functions, thereafter, rederives the update procedure of the cardinality and intensity density of the multitarget state set, eventually, the improved formulations on both the localization and position intensity update are obtained.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
