Title :
On the observability of indirect filtering in vehicle tracking and localization using a fixed camera
Author :
Perera, L.D.L. ; Elinas, P.
Author_Institution :
Australian Center for Field Robot., Univ. of Sydney, Sydney, NSW, Australia
Abstract :
In several vehicle tracking and localization applications, the initial position of a vehicle may be given by GPS measurements or other means. However, the information required for accurate tracking after initialization may only be available intermittently or not at all. In this paper, we demonstrate that the indirect or error form of state variables can be used in accurate bearing only tracking of a vehicle when the GPS measurements of its location are discontinued for some reason. Using Piece-wise Constant Systems Theory of Observability Analysis and the indirect form of the state variables for a constant velocity model we show that an object moving in a two-dimensional environment tracked by bearing only measurements using a fixed monocular camera is fully observable. We experimentally verify the theoretical results with simulations and real data from a fixed monocular camera tracking a pedestrian consistently when GPS measurements are discontinued.
Keywords :
Global Positioning System; cameras; filtering theory; road vehicles; sensor fusion; traffic engineering computing; GPS measurements; Global Positioning System; constant velocity model; fixed monocular camera; indirect filtering; observability analysis theory; piecewise constant systems theory; vehicle localization; vehicle tracking; Calibration; Cameras; Global Positioning System; Observability; Pixel; Radar tracking; Vehicles; Tracking; localization; observability;
Conference_Titel :
Information Fusion (FUSION), 2010 13th Conference on
Conference_Location :
Edinburgh
Print_ISBN :
978-0-9824438-1-1
DOI :
10.1109/ICIF.2010.5711901