Artifacts for the paper "Automated and Complete Generation of Traffic Scenarios at Road Junctions Using a Multi-level Danger Definition"

This deposit contains measurement data and additional artifacts pertaining to the "Automated and Complete Generation of Traffic Scenarios at Road Junctions Using a Multi-level Danger Definition" paper. The deposit is structured as follows: Data pertaining to RQ1 is found in the baseline-comparison/ directory, which contains (1) statistics measured during scenario generation and (2) data analysis results for both our proposed approach (which we name complete) and the baseline Scenic approach, including statistical significance data. Data pertaining to RQ2-4 and to the following Discussion is found in the following three directories. 0-generated-scenarios/ contains (1) the generated abstract scenario specifications (i.e. maneuver instance and path region assignments) and (2) concrete scenarios represented in an `xml` format that are executable through the CARLA Scenario Runner framework. 1-simulation-results/ contains the simulation results for our measurement runs. We include simulation traces, which show the exact position of each actor at each frame, in a human-readable, textual format. We also include result analysis (i.e. pertaining to simulation runtime, outcome, preventive maneuvers, etc.) within `measurements.json` files. 2-generated-figures/ contains figures derived from the contents of 1-simulation-results/ (including additional figures not included in the publication). All the code of the proposed scenario generation approach is implemented as extensions to the Concretize framework (for scenario generation and analysis), and to the Transfuser repository (for simulation). Concretize is available under the Eclipse Public License - v 2.0, while Transfuser is available under the MIT License.本数据集包含与论文《基于多层危险度定义的道路交叉口交通场景全自动完整生成（"Automated and Complete Generation of Traffic Scenarios at Road Junctions Using a Multi-level Danger Definition"）》相关的实测数据与附加衍生制品。数据集结构如下： 对应研究问题1（RQ1）的数据存放于baseline-comparison/目录下，其中涵盖（1）场景生成过程中测得的统计数据，以及（2）本文提出的完整生成方法（下称"完整方法"）与基线Scenic方法的数据分析结果，包含统计学显著性检验数据。 对应研究问题2至4（RQ2-4）以及后续讨论部分的数据，存放于以下三个目录中。 0-generated-scenarios/ 目录包含（1）生成的抽象场景规范（即机动实例与路径区域分配结果），以及（2）以xml格式表示的可通过CARLA Scenario Runner框架执行的具体场景。 1-simulation-results/ 目录包含本次实测运行的仿真结果。我们提供了以人类可读的文本格式存储的仿真轨迹，该轨迹记录了每一帧中所有交通参与者的精确位置。此外，我们还在measurements.json文件中收录了结果分析内容，涉及仿真运行时长、输出结果、预防性机动等相关分析。 2-generated-figures/ 目录包含从1-simulation-results/目录内容衍生得到的图表，其中包括论文未收录的额外图表。 本文提出的场景生成方法的全部代码均作为扩展模块集成至Concretize框架（用于场景生成与分析）以及Transfuser仓库（用于仿真）。Concretize采用Eclipse Public License - v 2.0进行授权，而Transfuser则采用MIT License进行授权。