Supporting code and data for: Dynamic capacity balancing in urban airspace: comparing historical and real-time aggregate flow data

This repository contains some supporting data and code for the 2024 SESAR Innovation Days Conference paper "Dynamic capacity balancing in urban airspace: comparing historical and real-time aggregate flow data."<br>Note that this is a continuation of the work published here:<br><strong>Morfin Veytia, Andres; Ellerbroek, Joost; Hoekstra, Jacco (2024): Supporting data and code for Decentralised Traffic Management for Constrained Urban Airspace: Dynamically Generating and Acting Upon Aggregate Flow Data. Version 2. 4TU.ResearchData. dataset. https://doi.org/10.4121/54825f14-8743-447d-8346-3afa46d319d6.v2</strong><br>Therefore, much of the data and code is similar. However, this work provides some additional code and scenarios.<br>The main components need to reproduce the results are:<br>BlueSky simulator code to reproduce the simulations.Post processing code for the scenarios to generate the plots in the paper and logs.Voronoi creation codePython environment description.<br><strong>1. BlueSky Simulator code</strong>This includes the BlueSky code for simulating the scenarios. This is the bluesky.zip folder. Note that the code provided is a condensed version of the one in https://github.com/amorfinv/bluesky/tree/rotterdam. The plugins and scenarios are also provided in the simulator code. The plugins are based on those based in the following repository, https://github.com/amorfinv/bluesky_plugins.<br>Refer to the README.md file provided to learn how to run the scenarios. Also, make sure to install a compatible python environment.<br><strong>2. Post-processing code, plots, and logs</strong>This includes the code to generate the plots seen in the paper and the logs of the simulations. It also includes some additional plots not shown in the paper. Read the README.md file for recreating the plots. This information can be found in main_experiment_results.zip. Some of the logs come from the previous work. The previous logs are labelled as real-time data labelling in this paper.<br><strong>3. Voronoi creation code</strong>The file called generate_voronois.zip includes the code to generate the voronoi code used for the historical data concept in the paper. Note that to generate the voronoi a more recent version of geopandas is necessary, so a different python environment is required. All you need is python=3.12, geopandas=1.0.1 and scikit-learn=1.5.1.<br><strong>6. Python environment description</strong>This includes the python environment used to simulate, post-process, and generate the plots. This work used conda environments. The main packages used are those required by BlueSky in addition to geopandas, osmnx, and seaborn. Note that the voronoi creation requires a different python environment.

本仓库包含2024年SESAR创新日会议论文《城市空域动态容量平衡：历史与实时聚合流量数据对比》的配套数据与代码。
请注意，本项目为此前已发表工作的延续，相关工作信息如下：
<strong>Morfin Veytia, Andres; Ellerbroek, Joost; Hoekstra, Jacco (2024): 《受限城市空域去中心化交通管理的配套数据与代码：动态生成并基于聚合流量数据执行决策》，版本2。4TU.ResearchData，数据集。https://doi.org/10.4121/54825f14-8743-447d-8346-3afa46d319d6.v2</strong>
因此，本项目的大部分数据与代码与此前工作重合，但新增了部分代码与场景。
可用于复现论文结果的核心组件如下：
<strong>1. BlueSky仿真器代码（BlueSky Simulator code）</strong>
本部分包含用于复现仿真场景的BlueSky仿真器代码，即bluesky.zip压缩包。需注意，本代码为https://github.com/amorfinv/bluesky/tree/rotterdam 仓库中代码的精简版本，仿真器中同时附带了插件与场景文件，其插件基于https://github.com/amorfinv/bluesky_plugins 仓库中的相关插件。请参考附带的README.md文件了解场景运行方法，并确保安装兼容的Python环境。
<strong>2. 后处理代码、绘图文件与仿真日志</strong>
本部分包含生成论文中图表所需的代码与仿真日志，同时还附带了部分未在论文中展示的额外图表。相关文件存储于main_experiment_results.zip压缩包中，可通过其中的README.md文件复现绘图结果。部分日志源自此前的研究工作，本文中将其标注为实时数据标注环节。
<strong>3. Voronoi图生成代码（Voronoi creation code）</strong>
名为generate_voronois.zip的压缩包包含了生成本文中历史数据概念所用Voronoi图（Voronoi）的代码。需注意，生成Voronoi图需要较新版本的geopandas库，因此需使用独立的Python环境，所需依赖为python=3.12、geopandas=1.0.1与scikit-learn=1.5.1。
<strong>6. Python环境配置说明</strong>
本部分包含本项目仿真、后处理与绘图环节所用的conda环境配置。本项目所用核心包除BlueSky所需依赖外，还包括geopandas、osmnx与seaborn。需注意，Voronoi图生成环节需使用独立的Python环境。