Converted Measurements Bayesian Extended Target Tracking Applied to X-band Marine Radar Data

摘要

In this paper we exploit the random matrix framework to trackextended targets. Proper measurement models to deal with theradar’s measurement noise and its conversion into Cartesian coordinatesare presented here. Benefits of the proposed extended targettracking using converted measurements can be mainly related to theproblem of the targets’ size estimation, while advantages on estimationof the targets’ kinematic features can be considered negligible.The validity of the proposed approach has been demonstrated byusing both simulated and real data. Gains up to 70% for the targets’width estimation accuracy and around 65% for the length areobserved on real data. The integration of the proposed model intothe gamma Gaussian inverseWishart probability hypothesis densitytracker is also provided and tested on real data.