Academia.eduAcademia.edu

Indoor Navigation Enhanced by Map-Matching

2005, European Journal of …

Abstract
sparkles

AI

This paper presents algorithms for indoor navigation, focusing on the creation of a 3D topological model to enhance the localization process within a user-oriented pedestrian navigation system. It discusses the integration of a dedicated map-matching routine that combines user position data with a link/node model to evaluate navigation algorithms. An emphasis is placed on the unique requirements of indoor navigation, particularly for physically-challenged users, highlighting the necessity of specific data models and algorithms.

Key takeaways

  • The pedestrian navigation system provides the user's position, which is combined with the link/node model.
  • The digital database for vehicle navigation system is based on a link/node model of the street network, which has significant advantages in supporting navigation.
  • A map database can be used to control, to correct and to update the position estimated by the navigation system.
  • As input to our MM process we have two data sources: the map database (links, nodes) and the list of positions recorded by the navigation system.
  • The basic idea of this process is based on the comparison of the raw trajectory with the link/node view of the map database.This visual analysis is shown on Figure 9, which combines raw positions (red crosses) and the navigation network.