![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
Abstract
GPS-based technology has served the positioning and navigation industry for decades with outstanding reliability and accuracy. However, limitations for location-based services in indoor scenarios remain where GPS signal reception suffers from severe degradation or even outages. Wi-Fi-based positioning is currently the most popular indoor location solution, with an availability and time-to-first-fix that are significantly better than GPS. However, the achievable position accuracy is only at the level of tens of meters on average depending on database density and signal reception quality. In order to improve positioning accuracy and availability, motion sensors like accelerometers, gyros, and magnetic compasses are increasingly integrated into smart devices. However, their position solutions suffer from the effects of significant accumulative errors. In this paper, a vision-based indoor positioning method is developed to overcome the limitations above. The proposed vision-based system relies on a single camera, widely available on smart phones and tablets. The derivation of the absolute 3D position from 2D snapshots of a single camera requires the use of an external geo-reference database. In this research, a ubiquitous floor plan database has been used to provide accurate geodetic information. Unlike other popular geo-reference databases, the database used in this work can easily be generated with existing resources. The proposed system has been developed as an iOS App and was tested on iPad for various indoor scenarios. The results show that the performance of the proposed system is superior to Wi-Fi-based positioning systems.
Acknowledgements
This research is financially supported by the Natural Sciences and Engineering Research Council (NSERC) of Canada and Tecterra.