Multi-modal probabilistic indoor localization on a smartphone

Dümbgen, Frederike (School of Computer and Communication Sciences, EPFL, Lausanne, Switzerland) ; Oeschger, Cynthia (School of Engineering and Architecture (HEIA-FR), HES-SO // University of Applied Sciences Western Switzerland ; School of Computer and Communication Sciences, EPFL, Lausanne, Switzerland) ; Kolundzija, Mihailo (School of Computer and Communciation Sciences, EPFL, Lausanne, Switzerland) ; Scholefield, Adam (School of Computer and Communication Sciences, EPFL, Lausanne, Switzerland) ; Girardin, Emmanuel (School of Engineering and Architecture (HEIA-FR), HES-SO // University of Applied Sciences Western Switzerland) ; Leuenberger, Johan (Vidinoti, Fribourg, Switzerland) ; Ayer, Serge (School of Engineering and Architecture (HEIA-FR), HES-SO // University of Applied Sciences Western Switzerland)

The satellite-based Global Positioning System (GPS) provides robust localization on smartphones outdoors. In indoor environments, however, no system is close to achieving a similar level of ubiquity, with existing solutions offering different trade-offs in terms of accuracy, robustness and cost. In this paper, we develop a multi-modal positioning system, targeted at smartphones, which aims to get the best out of each of its constituent modalities. More precisely, we combine Bluetooth low energy (BLE) beacons, round-trip-time (RTT) enabled WiFi access points and the smartphone’s inertial measurement unit (IMU) to provide a cheap robust localization system that, unlike fingerprinting methods, requires no pre-training. To do this, we use a probabilistic algorithm based on a conditional random field (CRF). We show how to incorporate sparse visual information to improve the accuracy of our system, using pose estimation from pre-scanned visual landmarks, to calibrate the system online. Our method achieves an accuracy of around 2 meters on two realistic datasets, outperforming other distance-based localization approaches. We also compare our approach with an ultra-wideband (UWB) system. While we do not match the performance of UWB, our system is cheap, smartphone compatible and provides satisfactory performance for many applications.


Keywords:
Conference Type:
full paper
Faculty:
Ingénierie et Architecture
School:
HEIA-FR
Institute:
Energy - Institut de recheche appliquée en systèmes énergétiques
Publisher:
Pisa, Italy, 30 Sept.-3 Oct. 2019
Date:
2019-07
Pisa, Italy
30 Sept.-3 Oct. 2019
Pagination:
8 p.
Published in:
Proceedings of 2019 International Conference on Indoor Positioning and Indoor Navigation (IPIN), 30 Sept.-3 Oct. 2019, Pisa, Italy
ISBN:
978-1-7281-1788-1
External resources:
Appears in Collection:



 Record created 2019-11-05, last modified 2019-11-05

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)