Multi-sensor tracking of a vehicle on a grid-II

Abstract

A hybrid model integrates a family of local representations to create a global description of a system operating in a heterogeneous environment. The common hybrid models have been developed for aerial applications and are not well suited for ground motions restricted to a road grid. This paper extends the work presented at last year's conference and presents a provisional version of the Gaussian wavelet estimator, modified to accept road map information. It is shown by example that map-enhancement improves the accuracy of the location estimates and boosts the quality of the calculated uncertainty region