DeepScenario: An Open Driving Scenario Dataset for Autonomous Driving System Testing

With the rapid development of autonomous driving systems (ADSs), testing ADSs under various driving conditions has become a key method to ensure the successful deployment of ADS in the real-world. However, it is impossible to test all the scenarios due to the inherent complexity and uncertainty of ADSs and the driving tasks. Further, testing of ADSs is expensive regarding time and computational resources. Therefore, a large-scale driving scenario dataset consisting of various driving conditions is needed. To this end, we present an open driving scenario dataset DeepScenario, containing over 30K executable driving scenarios, which are collected by 2880 test executions of three driving scenario generation strategies. Each scenario in the dataset is labeled with six attributes characterizing test results. We further show the attribute statistics and distribution of driving scenarios. For example, there are 1050 collision scenarios, in 917 scenarios there were collisions with other vehicles, 105 and 28 with pedestrians and static obstacles, respectively.

This dataset contains:

deepscenario-dataset - DeepScenario dataset, which includes driving scenarios generated by executing three scenario generation strategies: Reinforcement Learning (RL)-based Strategy, Random-based Strategy, Greedy-based Strategy;

deepscenario-toolset - The toolset for DeepScenario dataset, including ScenarioCollector that can automatically collect driving scenarios, and ScenarioRunner that can support replaying driving scenarios. We also provide source code and usage examples for the toolset. 

More information about DeepScenario dataset is available in our Github repository: https://github.com/Simula-COMPLEX/DeepScenario.

随着自动驾驶系统（Autonomous Driving Systems, ADSs）的快速发展，在各类驾驶场景下对其进行测试已成为保障其在真实世界中成功部署的核心手段。然而，由于自动驾驶系统本身的复杂性与不确定性，以及驾驶任务的固有特性，无法穷尽所有场景开展测试。此外，自动驾驶系统的测试在时间与计算资源层面均耗费高昂。为此，亟需构建涵盖各类驾驶场景的大规模驾驶场景数据集。基于此，我们开源了深度场景（DeepScenario）数据集，其包含超过3万个可执行驾驶场景，这些场景通过三种驾驶场景生成策略的2880次测试执行采集得到。数据集中的每个场景均标注了六个表征测试结果的属性。我们还给出了各属性的统计数据与驾驶场景的分布情况：例如，数据集中共包含1050个碰撞场景，其中917个场景涉及与其他车辆的碰撞，另有105个场景涉及与行人的碰撞，28个场景涉及与静态障碍物的碰撞。

本数据集包含以下内容：
- deepscenario-dataset：深度场景（DeepScenario）数据集，包含通过三种场景生成策略生成的驾驶场景，分别为基于强化学习（Reinforcement Learning, RL）的策略、基于随机采样的策略以及基于贪心算法的策略；
- deepscenario-toolset：深度场景（DeepScenario）数据集配套工具集，包含可自动采集驾驶场景的ScenarioCollector（场景采集器），以及支持驾驶场景回放的ScenarioRunner（场景运行器）。我们同时提供了该工具集的源代码与使用示例。

有关深度场景（DeepScenario）数据集的更多详细信息，可访问我们的GitHub仓库：https://github.com/Simula-COMPLEX/DeepScenario.