Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Framework for Devising, Evaluating and Fine-tuning Indoor Tracking Algorithms

Alpha Diallo, Benoit Garbinato

TL;DR

This work presents MobiXIM, a complete framework to devise, evaluate, and fine-tune indoor tracking algorithms on consumer devices. It integrates a plugin-based architecture, an orchestration platform with execution replay, a mobile data-collection app, and a flexible floorplan representation to standardize end-to-end ITS development across wireless, inertial, and collaborative approaches. The authors demonstrate substantial accuracy gains by combining PDR with opportunistic collaboration and BLE beacon cues in a real-world deployment, achieving an average corrected trajectory error around 4 m. By providing preloaded datasets, beacons, and execution-replay facilities, MobiXIM promotes reproducibility and rapid prototyping, addressing major reproducibility challenges in indoor localization research. The framework lays a foundation for future integration of advanced sensing like Angle-of-Arrival and broader collaborative scenarios.

Abstract

In recent years, we have observed a growing interest in Indoor Tracking Systems (ITS) for providing location-based services indoors. This is due to the limitations of Global Navigation and Satellite Systems, which do not operate in non-line-of-sight environments. Depending on their architecture, ITS can rely on expensive infrastructure, accumulate errors, or be challenging to evaluate in real-life environments. Building an ITS is a complex process that involves devising, evaluating and fine-tuning tracking algorithms. This process is not yet standard, as researchers use different types of equipment, deployment environments, and evaluation metrics. Therefore, it is challenging for researchers to build novel tracking algorithms and for the research community to reproduce the experiments. To address these challenges, we propose MobiXIM, a framework that provides a set of tools for devising, evaluating and fine-tuning tracking algorithms in a structured manner. For devising tracking algorithms, MobiXIM introduces a novel plugin architecture, allowing researchers to collaborate and extend existing algorithms. We assess our framework by building an ITS encompassing the key elements of wireless, inertial, and collaborative ITS. The proposed ITS achieves a positioning accuracy of 4 m, which is an improvement of up to 33% compared to a baseline Pedestrian Dead Reckoning algorithm.

A Framework for Devising, Evaluating and Fine-tuning Indoor Tracking Algorithms

TL;DR

This work presents MobiXIM, a complete framework to devise, evaluate, and fine-tune indoor tracking algorithms on consumer devices. It integrates a plugin-based architecture, an orchestration platform with execution replay, a mobile data-collection app, and a flexible floorplan representation to standardize end-to-end ITS development across wireless, inertial, and collaborative approaches. The authors demonstrate substantial accuracy gains by combining PDR with opportunistic collaboration and BLE beacon cues in a real-world deployment, achieving an average corrected trajectory error around 4 m. By providing preloaded datasets, beacons, and execution-replay facilities, MobiXIM promotes reproducibility and rapid prototyping, addressing major reproducibility challenges in indoor localization research. The framework lays a foundation for future integration of advanced sensing like Angle-of-Arrival and broader collaborative scenarios.

Abstract

In recent years, we have observed a growing interest in Indoor Tracking Systems (ITS) for providing location-based services indoors. This is due to the limitations of Global Navigation and Satellite Systems, which do not operate in non-line-of-sight environments. Depending on their architecture, ITS can rely on expensive infrastructure, accumulate errors, or be challenging to evaluate in real-life environments. Building an ITS is a complex process that involves devising, evaluating and fine-tuning tracking algorithms. This process is not yet standard, as researchers use different types of equipment, deployment environments, and evaluation metrics. Therefore, it is challenging for researchers to build novel tracking algorithms and for the research community to reproduce the experiments. To address these challenges, we propose MobiXIM, a framework that provides a set of tools for devising, evaluating and fine-tuning tracking algorithms in a structured manner. For devising tracking algorithms, MobiXIM introduces a novel plugin architecture, allowing researchers to collaborate and extend existing algorithms. We assess our framework by building an ITS encompassing the key elements of wireless, inertial, and collaborative ITS. The proposed ITS achieves a positioning accuracy of 4 m, which is an improvement of up to 33% compared to a baseline Pedestrian Dead Reckoning algorithm.
Paper Structure (33 sections, 4 equations, 7 figures, 1 table, 1 algorithm)

This paper contains 33 sections, 4 equations, 7 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Scope and Methodology
  3. Problem statement
  4. Roadmap
  5. Related work
  6. System model
  7. Methodology
  8. Devising tracking algorithms
  9. Evaluating tracking algorithms
  10. Evaluating tracking algorithms on existing data.
  11. Evaluating tracking algorithms on collected data.
  12. Designing the floor plans.
  13. Cleaning the data.
  14. Adjusting parameters.
  15. MobiXIM Architecture
  16. ...and 18 more sections

Figures (7)

  • Figure 1: Architecture and process flow of MobiXIM
  • Figure 2: User interfaces of the Mobile Companion App and the Orchestrator Platform
  • Figure 3: Devices within a collaboration range
  • Figure 4: Third quartile of localization errors of all the devices
  • Figure 5: DFD score for all the devices
  • ...and 2 more figures