Indoor Positioning System using Sensor and Crowdsourcing Landmark Map Update

Abstract

The indoor positioning technology has been attracting the attention of researchers due to the increasing pervasiveness of smartphones and the development of sensor technology along with the increase of time that modern people spend indoor. Sensor technology, which is the most commonly used data source for indoor positioning, has an advantage that sensors can receive data from smartphone without installing the device separately. However, indoor positioning technology which depends only on sensor data, has the disadvantage of cumulatively accumulating errors. To solve this problem, techniques using landmarks which make a singular point of indoor localization are studied. In this paper, we provide a system that combines crowd sourcing technique and landmarks to receive data from many unspecified users. Our proposed system updates the indoor landmark map by automatically removing the unused landmarks for a long time and adding more landmarks through crowd sourcing. The system calculates the location of the user through the SLAM algorithm using the improved landmark map.