[Back]

M. Thienelt, A. Eichhorn, A. Reiterer, G. Roberts:
"Pedestrian Positioning without Map Matching Methods - Prototype of a Knowledge-Based Kalman Filter";
Talk: International Workshop on Ubiquitous, Pervasive and Internet Mapping 2006, Seoul, South Korea; 2006-10-23 - 2006-10-25; in: "SMF/UPIMap 2006 (II) Proceedings of the Second International Workshop on Ubiquitous, Pervasive and Internet Mapping", Chamber of Commerce & Indu (ed.); The Chamber of Commerce & Industry, Band II (2006), 148 - 156.

The utilization of Location Based Services for pedestrian navigation requires the derivation of accurate, reliable and always available position information. In contrast to common car navigation systems it is required to navigate within scenarios with no fixed trajectories available from digital maps. In this contribution the prototype of a map-independent positioning system for pedestrians is presented. It consists of a knowledge-based component for outlier detection in the measurement data, quality analysis and calibration of the multi-sensor system as soon as a positioning component with a central Kalman-filter for the optimal estimation of the kinematic state of the pedestrian. Two typical test scenarios for Indoor and Outdoor scenarios from the city of Vienna are presented. The results can be evaluated as very promising.

http://aleph.ub.tuwien.ac.at/F?base=tuw01&func=find-c&ccl_term=AC06586714