Talks and Poster Presentations (with Proceedings-Entry):

S. Niedermayr, A. Wieser:
"Combination of feature-based and geometric methods for positioning";
Talk: MCG Stuttgart 2012, Stuttgart (invited); 2012-03-27 - 2012-03-29; in: "3rd International Conference on Machine Control & Guidance Proceedings", V. Schwieger, S. Böttinger, B. Zheng (ed.); (2012), ISBN: 9783000372957; 301 - 310.

English abstract:
The location of mobile objects is usually based on angles, distances or distance differences, and on analytical geometry. It typically requires unobstructed line-of-sight between the mobile object and known reference points for "absolute positioning" or integrating measurements of position changes for "relative positioning". The main drawback of the relative methods is the rapid decrease of spatial accuracy over time and distance. The main drawback of the absolute methods is that a direct line-of-sight can often be established only with considerable effort or not at all e.g., in densely built-up areas or with distinctive topography. We show that the established geometric methods can be augmented by feature-based positioning which is based on the comparison of observed feature vectors with given reference features. Feature-based positioning does not require any line-of-sight, yields "absolute" position with accuracy independent of time and distance travelled, and allows using the inherent location information of a variety of observable features. We give a brief overview of feature-based positioning techniques and present an approach using Bayesian estimation for the combination with classic geometric methods. The real data used for demonstration have been obtained at an airport where a combination of WLAN signal strength measurements and GPS-observations is used to locate vehicles for real-time display in an airport management system.

Feature-based positioning, location fingerprinting, Particle Filter, WLAN, GNSS

Created from the Publication Database of the Vienna University of Technology.