Measurement Preprocessing-Based Variable Dimension PDAF for Maneuvering Target Tracking in Clutter

摘要

This paper presents a measurement preprocessing based variable dimension probabilistic data association filter (PDAF) for tracking a single maneuvering target in clutter. The measurement preprocessing scheme is derived from a maximum likelihood method and utilizes validated measurements for yielding target's estimated accelerations and velocities which lead to preprocessed measurements. The proposed preprocessing method is integrated into a variable dimension PDAF which uses preprocessed measurements as its inputs. Simulation results show that the proposed tracking algorithm can achieve improved performance.