Road-map assisted convoy track maintenance using random matrices

Collectively moving object clusters are of particular interest in certain applications and have to be tracked as separate aggregated entities consisting of an unknown number of individuals. Tracking of convoys or larger vehicles in wide area ground surveillance are important examples. The objects of interest are usually considered as point source objects, i. e., compared to the sensor resolution their spatial extension is neglected, though increasing resolution capabilities of modern sensors may give rise to distinct or several detections. In this sense also, collectively moving object groups can be considered as extended objects. Due to the resulting data association and resolution conflicts, any attempt of tracking individual objects within the group is no longer reasonable because of inaccuracy and waste of computer resources. In this paper ellipsoidal object extension is modeled by a symmetric positive definite random matrix, which is treated as an additional state variable to be estimated or tracked. An important aspect is the incorporating of context information into the Bayesian formalism. Herein, kinematical constraints such as road maps are considered to improve tracking results.