Analyzing Real-world Accidents for Test Scenario Generation for Automated Vehicles
Analyzing Real-world Accidents for Test Scenario Generation for Automated Vehicles
复制标题
分析现实世界的事故以生成自动驾驶汽车的测试场景
DOI：
10.1109/iv48863.2021.9576007
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Esenturk E
中科院分区：
文献类型：
--
作者：
Esenturk E
自动驾驶系统(ADSS)测试场景的识别仍然是自动驾驶系统验证和确认的关键挑战。已经提出了各种方法，包括基于数据的方法和基于知识的方法来生成情景。识别导致严重交通事故的条件不仅可以帮助我们确定ADS的测试情景，还可以实施拯救生命和基础设施资源的措施。本文采用基于数据的方法，提出了一种新的用于生成测试场景的事故数据分析方法，通过分析英国的Stats19事故数据来识别高严重性事故的趋势，以生成测试场景。本文首先研究事故的严重程度，综合考虑道路、环境条件、车辆特性和驾驶员特征等可操作设计领域(ODD)特性，将其与静态和时间相关的内外部因素联系起来。为此，本文采用了一种数据分组策略(粗粒化)，并在传统回归模型的基础上建立了一种Logistic回归方法，其中新出现的特征变得更加明显，而不感兴趣的特征和噪声被削弱。该方法使因素和结果变量之间的关系更加明显，因此非常适合于严重性分析。与普通Logistic模型相比，该方法表现出更好的性能，以拟合优度衡量，并考虑了模型变化(对于普通模型，对于当前模型)。然后，该模型被用于解决构建高风险的坠机前条件作为基于模拟的ADSS测试的测试场景的反问题。
Identification of test scenarios for Automated Driving Systems (ADSs) remains a key challenge for the Verification & Validation of ADSs. Various approaches including data based approaches and knowledge based approaches have been proposed for scenario generation. Identifying the conditions that lead to high severity traffic accidents can help us not only identify test scenarios for ADSs, but also implement measures to save lives and infrastructure resources. Taking a data based approach, in this paper, we introduce a novel accident data analysis method for generating test scenarios where we analyze UK's Stats19 accident data to identify trends in high severity accidents for test scenario generation. This paper first focuses on the severity of the accidents with the goal of relating it to static and time-dependent internal and external factors in a comprehensive way taking into account Operational Design Domain (ODD) properties, e.g. road, environmental conditions, and vehicle properties and driver characteristics. For this purpose, the paper utilizes a data grouping strategy (coarse-graining) and builds a logistic regression approach, derived from conventional regression models, in which emerging features become more pronounced, while uninteresting features and noise weaken. The approach makes the relationship between the factors and outcome variable more visible and hence well suited for the severity analysis. The method shows superior performance as compared to ordinary logistic models measured by goodness of fit and accounting for model variancefor the ordinary model,for the current model). The model is then used to solve the inverse problem of constructing high-risk pre-crash conditions as test scenarios for simulation based testing of ADSs.
登录
查看更多内容
单车碰撞事故中侧翻和事件序列的伤害影响
DOI：
--
发表时间：
2000
期刊：
影响因子：
--
作者：
K. A. Krull;A. Khattak;F. Council
通讯作者：
F. Council
影响因子：
5.9
作者：
Imprialou, Marianna;Quddus, Mohammed
通讯作者：
Quddus, Mohammed
DOI：
10.1016/j.trc.2018.07.001
发表时间：
2018-11-01
影响因子：
8.3
作者：
Khastgir, Siddartha;Birrell, Stewart;Jennings, Paul
通讯作者：
Jennings, Paul
DOI：
10.1016/j.ress.2021.107610
发表时间：
2021-03
期刊：
Reliab. Eng. Syst. Saf.
影响因子：
--
作者：
S. Khastgir;S. Brewerton;John Thomas;P. Jennings
通讯作者：
S. Khastgir;S. Brewerton;John Thomas;P. Jennings
确定智能汽车系统现有验证方法的差距：引入 3xD 模拟器
DOI：
--
发表时间：
2015
期刊：
2015 IEEE Intelligent Vehicles Symposium (IV)
影响因子：
--
作者：
S. Khastgir;S. Birrell;G. Dhadyalla;P. Jennings
通讯作者：
P. Jennings