Hybrid 3d Modelling Framework For Indoor Navigation Using Federated Learning And Internet Of Things-Enabled Edge Devices

Document Type

Article

Source of Publication

Peerj Computer Science

Publication Date

2-18-2026

Abstract

Background There has been a recent trend towards using three-dimensional (3D) models to enhance spatial awareness and maximize resource utilization in complex environments. 3D building model can be used in various applications, such as real-time guidance and tracking the positions of individuals in multi storied buildings. Cities are now being modeled and studied in three dimensions as an improved method of urban planning. Method This study proposes an advanced indoor navigation framework that combines 3D modelling, federated learning (FL), and Internet of Things (IoT) integration to deliver reliable floor-level localization and real-time guidance. In Phase 1, highly accurate 3D models were created using Quantum Geographic Information System (QGIS) software and validated against Light Detection and Ranging (LiDAR) sensor measurements, achieving an error percentage of approximately 0.99%. In Phase 2, a federated learning approach using a hierarchical recurrent neural network (H-RNN) was developed to improve indoor positioning accuracy based on the 3D model data. In Phase 3, a dedicated navigation application was integrated with IoT devices strategically placed throughout multi-story buildings to enable real-time routing. Results The error percentage was very low (approximately 0.99%), demonstrating excellent agreement between the LiDAR and QGIS models. An RNN method using FL was implemented in the second part of the project to enhance the location accuracy of the system using a three-dimensional model. During Phase 3, a navigation application was integrated into IoT devices placed at strategic places within the building. This work is novel in its integration of accurate 3D modelling, federated learning, and IoT-based routing in a single system that achieves over 99% positioning accuracy and 98.7% routing accuracy across multi-story buildings. Current indoor navigation systems lack scalability, privacy preservation, and adaptability across complex multi-story buildings. This study proposes a novel three-phase approach integrating 3D modelling in QGIS, federated learning-based positioning, and IoT-enabled routing.

ISSN

2376-5992

Publisher

PeerJ

Volume

12

Disciplines

Computer Sciences

Keywords

Federated learning, Machine learning, Edge computing, Three-dimension model, Indoor positioning

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Indexed in Scopus

no

Open Access

yes

Open Access Type

Gold: This publication is openly available in an open access journal/series

This document is currently not available here.

Share

COinS