WP4 Emerald: T4.1 - Trajectory Data / Travel Time Analysis -- Trajectools

Description

This repository contains the first version of QGIS Processing Trajectools plugin for the Trajectory Data / Travel Time Analysis Emerald, developed as part of T4.1 of WP4 within EMERALDS project. The Trajectools plugin adds trajectory analysis algorithms to the QGIS Processing toolbox. This README provides essential information for setting up this Emerald.

Related repositories include:

• T4.1 - Trajectory Data / Travel Time Analysis -- No-code Model
• T4.1 - Trajectory Data / Travel Time Analysis -- Cartoblog Post

Table of Contents

• Requirements
• Examples
• Authors

Requirements

Running these models requires QGIS (a popular open source geographic information system) with MovingPandas (a Python library for movement data analysis) and the QGIS Trajectools plugin.

Trajectools requires MovingPandas (a Python library for movement data analysis) and optionally integrates scikit-mobility (for privacy tests), stonesoup (for smoothing), and gtfs_functions (for GTFS data support).

The recommended way to install these dependencies is through conda/mamba:

(base) conda create -n qgis -c conda-forge python=3.9 
(base) conda activate qgis
(qgis) mamba install -c conda-forge qgis movingpandas scikit-mobility stonesoup
(qgis) pip install gtfs_functions h3==3.7.7

(More details: https://anitagraser.com/2023/01/21/pyqgis-jupyter-notebooks-on-windows-using-conda/)

Pip install

If you cannot use conda, you may try installing from the QGIS Python Console:

import pip
pip.main(['install', 'movingpandas'])
pip.main(['install', 'scikit-mobility'])
pip.main(['install', 'stonesoup'])
pip.main(['install', 'gtfs_functions'])

Plugin installation

The Trajectools plugin can be installed directly in QGIS using the built-in Plugin Manager:

Figure 1: QGIS Plugin Manager with Trajectools plugin installed.

Figure 2: Trajectools (v2.4) algorithms in the QGIS Processing toolbox

Examples

The Trajectools plugin is used in EMERALDS Use Case 3 to analyze travel time on public transport network segments, as documented in the corresponding repo.

The individual Trajectools algorithms are flexible and modular and can therefore be used on a wide array on input datasets, including, for example, the open Microsoft Geolife dataset a sample of which is included in the plugin repo:

Authors

AIT

Citation information

Please cite [0] when using Trajectools in your research and reference the appropriate release version using the Zenodo DOI: https://doi.org/10.5281/zenodo.13847642

[0] Graser, A., & Dragaschnig, M. (2024, June). Trajectools Demo: Towards No-Code Solutions for Movement Data Analytics. In 2024 25th IEEE International Conference on Mobile Data Management (MDM) (pp. 235-238). IEEE.

@inproceedings{graser2024trajectools,
  title = {Trajectools Demo: Towards No-Code Solutions for Movement Data Analytics},
  author = {Graser, Anita and Dragaschnig, Melitta},
  booktitle = {2024 25th IEEE International Conference on Mobile Data Management (MDM)},
  pages = {235--238},
  year = {2024},
  organization = {IEEE},
  doi = {10.1109/MDM61037.2024.00048},
}