Skip to content

HuMoLiRe is a Pedestrian Dead-Reckoning Particle Filter Map-aided system that leverages human motion likelihood in indoor spaces to estimate their position. This repository is the dataset and software published with its paper.

Notifications You must be signed in to change notification settings

anisghaoui/humolire

Repository files navigation

HuMoLiRe

humolire

Example of trajectory

trajectory

Citation

This dataset and software are related to the following publication in the IEEE Sensors journal. Please cite using the following:

@ARTICLE{ghaouihumolire,
author={Ghaoui, Mohamed Anis and Vincke, Bastien and Reynaud, Roger}, 
journal={IEEE Sensors Journal},   
title={Human Motion Likelihood Representation Map-Aided PDR Particle Filter},  
year={2023},  
volume={23},  
number={1},  
pages={484-494},  
doi={10.1109/JSEN.2022.3222639}}

IEEE page

Introduction

This program runs on python3.8+. It is recommanded to use PyCharm. (it can easily be turned into older version by editing every print(f"{variable=})" call)

Entry point is main.py. generate_figures.py is used to recreate the figures mentionned in the article.

Requirements

requirements.txt lists:

  • numpy~=1.21
  • matplotlib~=3.3.3
  • scipy~=1.5.4
  • AHRS~=0.3.0
  • imageio~=2.9.0
  • tqdm~=4.51.0
  • opencv-python~=4.5.1.48

Optional:

  • adjustText~=0.7.3 , is used to place the purple numbers automatically
  • requests~=2.22.0 , is used to send an SMS when the program is finished

Folder structure:

.
├── data
├── docs
├── humolire
├── map_editor
├── README.md
├── requirements.txt
└── tests

Documentation:

There are many README.MD files in the folders about. The main entry point is main.py. There is a beginning of documentation at read the docs. I don't have much time. If you want to help in documentation, I would be immensely grateful.

Contribution:

  • If you have a research question, please reach me by email: [email protected]
  • If you have a question about the code, open an issue
  • If you wanna help documenting (I lack the time and skills to do that correctly), open an issue. Thank you!

LICENSE

Humolire Dataset and Software by Anis GHAOUI is licensed under a Creative Commons Attribution 4.0 International License.

About

HuMoLiRe is a Pedestrian Dead-Reckoning Particle Filter Map-aided system that leverages human motion likelihood in indoor spaces to estimate their position. This repository is the dataset and software published with its paper.

Topics

Resources

Stars

Watchers

Forks

Languages