Indoor pedestrian navigation based on hybrid route planning and location modeling

Abstract

This paper introduces methods and services called PerPosNav for development of custom indoor pedestrian navigation applications to be deployed on a variety of platforms. PerPosNav is built on top of the PerPos positioning middleware [8] that fusions GPS, WiFi and inertial tracking into indoor positioning with high accuracy in many types of buildings. The challenges of indoor navigation are discussed and the PerPosNav services are introduced. PerPosNav combines symbolic and geometry based modeling of buildings, and in turn combines graph-based and geometric route computation. The paper argues why these hybrid approaches are necessary to handle the challenges of indoor pedestrian navigation. Furthermore, a fluent navigation is maintained via route tracking and navigation services that generate instructions based on how the user moves in relation to the prescribed route. The viability of PerPosNav has been proven by implementation of support for multiple modes of pedestrian indoor navigation: 1) augmented signs, 2) map based navigation on smartphones, 3) auditory navigation on smartphones solely via earbuds, and 4) augmented reality navigation. Experiences from the use of the PerPosNav services are discussed and compared to other indoor pedestrian navigation approaches. © 2012 Springer-Verlag Berlin Heidelberg.