Collaborative infrastructure-free multi-person indoor navigation with shared magnetic fingerprinting

The practical deployment of multi-person indoor navigation in Global Navigation Satellite System-denied environments remains challenged by cumulative inertial sensor drift and dependence on pre-installed infrastructure. This study presents a collaborative and infrastructure-free inertial navigation framework with three core algorithmic contributions. First, an online collaborative magnetic fingerprint map construction mechanism enables individuals to collectively build and refine a shared magnetic fingerprint pool during traversal, fundamentally eliminating the need for offline surveying. Second, a confidence-weighted cross-individual correction scheme allows observations from other individuals that exceed a predefined confidence threshold to serve as virtual anchors for trajectory refinement. This mechanism is distinct from standard Extended Kalman Filter updates in that correction signals originate from peer individuals rather than fixed references. Third, a region-of-interest identification scheme selectively extracts repeatable magnetic fingerprint signatures at key navigational locations, improving matching robustness while reducing computational overhead. Unlike existing approaches that rely on pre-built fingerprint maps or limit collaboration to position-level fusion, the proposed framework achieves map-free, trajectory-level collaborative correction through a unified online architecture. Experimental validation on two indoor environments produces an average root mean squared error of 1.5 m, substantially outperforming conventional individual inertial navigation system approaches. These results demonstrate a robust and scalable solution for emergency response, elderly monitoring and industrial worker tracking.

Authors

Institutions

Publication Details

Journal
Engineering Applications of Artificial Intelligence
Published
2026-11-18
DOI
https://doi.org/10.1016/j.engappai.2026.116773
Primary Topic
Indoor and Outdoor Localization Technologies
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Collaborative infrastructure-free multi-person indoor navigation with shared magnetic fingerprinting

Linzhen Zhu, Craig Matthew Hancock, Matthew Pike, Xiaoqing Chai et al.
5 links
Engineering Applications of Artificial Intelligence
Indoor and Outdoor Localization Technologies
article

Collaborative infrastructure-free multi-person indoor navigation with shared magnetic fingerprinting

Linzhen Zhu, Craig Matthew Hancock, Matthew Pike, Xiaoqing Chai, Boon Giin Lee, Renjie Wu
article en

Abstract

The practical deployment of multi-person indoor navigation in Global Navigation Satellite System-denied environments remains challenged by cumulative inertial sensor drift and dependence on pre-installed infrastructure. This study presents a collaborative and infrastructure-free inertial navigation framework with three core algorithmic contributions. First, an online collaborative magnetic fingerprint map construction mechanism enables individuals to collectively build and refine a shared magnetic fingerprint pool during traversal, fundamentally eliminating the need for offline surveying. Second, a confidence-weighted cross-individual correction scheme allows observations from other individuals that exceed a predefined confidence threshold to serve as virtual anchors for trajectory refinement. This mechanism is distinct from standard Extended Kalman Filter updates in that correction signals originate from peer individuals rather than fixed references. Third, a region-of-interest identification scheme selectively extracts repeatable magnetic fingerprint signatures at key navigational locations, improving matching robustness while reducing computational overhead. Unlike existing approaches that rely on pre-built fingerprint maps or limit collaboration to position-level fusion, the proposed framework achieves map-free, trajectory-level collaborative correction through a unified online architecture. Experimental validation on two indoor environments produces an average root mean squared error of 1.5 m, substantially outperforming conventional individual inertial navigation system approaches. These results demonstrate a robust and scalable solution for emergency response, elderly monitoring and industrial worker tracking.

Engineering Applications of Artificial IntelligenceVol. 188
University of Nottingham Ningbo China (CN), University of Nottingham (GB), Loughborough University (GB), City University of Hong Kong (HK), University of Michigan (US), NingboTech University (CN), University of Hong Kong (HK)
University of Nottingham Ningbo China
Industry, innovation and infrastructure
Indoor and Outdoor Localization Technologies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon