Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Accurate Indoor Positioning Based on Learned Absolute and Relative Models

Kjellson, Christoffer ; Larsson, Martin LU orcid ; Astrom, Kalle LU orcid and Oskarsson, Magnus LU orcid (2021) 2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021
Abstract

To improve the accuracy of indoor positioning systems it can be useful to combine different types of sensor data. This paper describes deep learning methods both for estimating absolute positions and for performing pedestrian dead reckoning, and then how to combine the resulting estimates using weighted least squares optimization. The positioning model is based on a custom neural network which uses measurements of received signal strength indication from one instant of time as input. The model for estimating relative positions is on the other hand based on inertial sensors, the accelerometer, magnetometer and gyroscope. The position estimates are then combined using a least squares approach with weights based on the standard deviations... (More)

To improve the accuracy of indoor positioning systems it can be useful to combine different types of sensor data. This paper describes deep learning methods both for estimating absolute positions and for performing pedestrian dead reckoning, and then how to combine the resulting estimates using weighted least squares optimization. The positioning model is based on a custom neural network which uses measurements of received signal strength indication from one instant of time as input. The model for estimating relative positions is on the other hand based on inertial sensors, the accelerometer, magnetometer and gyroscope. The position estimates are then combined using a least squares approach with weights based on the standard deviations of errors in predictions from the used models.

(Less)
Please use this url to cite or link to this publication:
author
; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
deep learning, fingerprinting, indoor positioning, optimization, PDR, radio beacons, sensor fusion, smartphone
host publication
2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
conference name
2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021
conference location
Lloret de Mar, Spain
conference dates
2021-11-29 - 2021-12-02
external identifiers
  • scopus:85124797938
ISBN
9781665404020
DOI
10.1109/IPIN51156.2021.9662534
language
English
LU publication?
yes
id
1760b9a8-6e27-4eea-8d96-b8d75a9b44be
date added to LUP
2022-04-12 09:09:54
date last changed
2025-04-04 13:56:32
@inproceedings{1760b9a8-6e27-4eea-8d96-b8d75a9b44be,
  abstract     = {{<p>To improve the accuracy of indoor positioning systems it can be useful to combine different types of sensor data. This paper describes deep learning methods both for estimating absolute positions and for performing pedestrian dead reckoning, and then how to combine the resulting estimates using weighted least squares optimization. The positioning model is based on a custom neural network which uses measurements of received signal strength indication from one instant of time as input. The model for estimating relative positions is on the other hand based on inertial sensors, the accelerometer, magnetometer and gyroscope. The position estimates are then combined using a least squares approach with weights based on the standard deviations of errors in predictions from the used models. </p>}},
  author       = {{Kjellson, Christoffer and Larsson, Martin and Astrom, Kalle and Oskarsson, Magnus}},
  booktitle    = {{2021 International Conference on Indoor Positioning and Indoor Navigation, IPIN 2021}},
  isbn         = {{9781665404020}},
  keywords     = {{deep learning; fingerprinting; indoor positioning; optimization; PDR; radio beacons; sensor fusion; smartphone}},
  language     = {{eng}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  title        = {{Accurate Indoor Positioning Based on Learned Absolute and Relative Models}},
  url          = {{http://dx.doi.org/10.1109/IPIN51156.2021.9662534}},
  doi          = {{10.1109/IPIN51156.2021.9662534}},
  year         = {{2021}},
}