Data pertaining to Chapter 4 "Identification of Driving Heterogeneity using Action-chains"

This dataset accompanies the paper <em>“Identification of Driving Heterogeneity using Action-chains”</em> (Chapter 4 of the PhD dissertation). The research introduces a comprehensive framework for identifying driving heterogeneity from an action perspective. Driving trajectories are identified into Action phases with physical meanings based on rule-based segmentation techniques. The Action chain concept is then introduced by implementing the Action phase transition probability. Evaluating using a naturalistic dataset indicates that this approach effectively identifies driving heterogeneity while providing clear interpretations. The research includes data preprocessing to clean data, rule-based segmentation to extract Action phases, driving behaviour map establishment, Action chain modelling, and heterogeneous traffic flow evaluation. The dataset is provided as a zipped folder containing four Jupyter notebooks (<code>.ipynb</code>) and supporting files in <code>.xlsx</code>, <code>.csv</code>, <code>.mat</code>, <code>.m</code>, <code>.txt</code>, and <code>.pdf</code> formats. A <code>ch4_Readme.txt</code> file is included to guide users on the structure, usage, and purpose of the data.本数据集配套于学术论文《基于动作链（Action-chains）的驾驶异质性（driving heterogeneity）识别》（该文为博士学位论文第4章）。该项研究提出了一套从动作视角识别驾驶异质性的完整分析框架。研究基于规则分割技术，将驾驶轨迹划分为具备明确物理意义的动作阶段（Action phases）。随后通过引入动作阶段转移概率，正式提出了动作链的概念。基于自然驾驶数据集的评估结果表明，该方法可有效识别驾驶异质性，同时具备清晰的可解释性。该项研究涵盖数据预处理与清洗、基于规则的动作阶段提取、驾驶行为图谱构建、动作链建模以及异质交通流评估等核心研究内容。本数据集以压缩包形式提供，内含4个Jupyter Notebook（.ipynb）文件，以及.xlsx、.csv、.mat、.m、.txt与.pdf格式的配套支撑文件。压缩包中还包含ch4_Readme.txt文件，用于指导用户了解数据集的结构、使用方法与应用目的。