人机混驾条件下的车辆纵向交互安全影响因素分析

王艺贇; 余荣杰

doi:10.3963/j.jssn.1674-4861.2024.03.002

人机混驾条件下的车辆纵向交互安全影响因素分析

doi: 10.3963/j.jssn.1674-4861.2024.03.002

王艺贇,
余荣杰^,

同济大学道路与交通工程教育部重点实验室上海 201804

基金项目:

国家自然科学基金项目 52172349

详细信息

作者简介:
王艺贇（1995—），博士研究生. 研究方向：交通安全. E-mail: wangyiyun@tongji.edu.cn

通讯作者:
余荣杰（1989—），博士，教授. 研究方向：交通安全等. E-mail: yurongjie@tongji.edu.cn

中图分类号: U491
计量
- 文章访问数: 167
- HTML全文浏览量: 97
- PDF下载量: 18
- 被引次数: 0
出版历程
- 收稿日期: 2023-08-16
- 网络出版日期: 2024-10-21

An Analysis of Safety Influencing Factors for Longitudinal Interaction Between Vehicles in Human-machine Mixed Traffic Driving Conditions

WANG Yiyun,
YU Rongjie^,

Key Laboratory of Road and Traffic Engineering of the Ministry of Education, Tongji University, Shanghai 208104, China

摘要

摘要: 自动驾驶汽车正向现有交通运行环境中逐步渗透，形成了与人工驾驶汽车混合运行的人机混驾交通流。有研究表明：自动驾驶汽车的百公里事故率为9.1，高出人工驾驶汽车（4.1）的1倍多；另外，人机纵向交互造成的追尾事故形态占所有事故形态的57.5%，远超过人类驾驶的27.9%，因此亟需研究人机纵向交互安全影响机理。现有研究通常采用驾驶模拟实验，分析虚拟仿真环境下人工驾驶汽车驾驶人与自动驾驶汽车的纵向交互行为与安全性，但模拟环境与实际道路场景差异较大，难以准确反映人机混驾交通流中的真实车辆交互行为。通过自动驾驶汽车开放道路测试数据，获取真实混驾条件下的车辆纵向交互场景，对车辆类型、行驶环境等影响因素与纵向交互行为及安全的影响机理开展研究。具体针对筛选后的人工驾驶汽车驾驶人分别跟驰人工驾驶汽车和跟驰自动驾驶汽车的场景数据，利用结构方程模型，构建了前车驾驶行为、前车车辆类型、路段运行速度水平与交互安全替代指标之间的链式作用关系。模型结果表明：前车车辆类型是否为自动驾驶汽车是影响纵向交互安全的显著影响因素之一，其他变量保持不变时，人工驾驶汽车驾驶人与自动驾驶前车的交互安全性相较于人类驾驶前车降低。
- 交通安全 /
- 自动驾驶 /
- 人机混驾 /
- 影响因素分析 /
- 结构方程模型 /
- Waymo数据
Abstract: Autonomous vehicles are gradually introduced to the existing traffic environment, leading to a mixed flow of both autonomous vehicles and human-driven vehicles. Studies show that the crash rate per-kilometer for autonomous vehicles is 9.1, which is more than twice that of human-driven vehicles (4.1). The ratio of the rear-end crash pattern between autonomous vehicles and human-driven vehicles is 57.5%, which exceeds 27.9% of among human-driven vehicles. Therefore, there is an urgent need to investigate the safety mechanisms of longitudinal interactions of autonomous vehicles and human-driven vehicles. Existing studies typically employ driving simulation experiments to analyze the longitudinal interaction and safety between human-driven and autonomous vehicles in virtual environments. However, the differences between simulated environments and real-world road scenarios make it challenging to accurately capture the interaction behavior between vehicles in mixed human-autonomous traffic flows. In this study, public road-testing dataset of autonomous vehicles are utilized to extract longitudinal interacting scenarios, and the influencing factors and the impact mechanisms of longitudinal interaction behavior and safety are investigated. Specifically, scenarios of human-driven vehicles following the other human-driven vehicle, and following an autonomous vehicle are studied, Structural equation model is applied to construct a chained relationship among driving behavior of leading vehicle, type of leading vehicle (whether it is an autonomous vehicle or not), speed level of vehicles on the roadway, and the safety surrogate measure. The modelling results revealed the type of leading vehicle is identified as an influencing factor in longitudinal interaction safety. When other variables remain constant, the safety of interactions between human drivers and autonomous vehicles as leading vehicles decreased compared to interactions with other human-driven vehicles as leading vehicles.
- traffic safety /
- autonomous driving /
- mixed driving /
- influencing factors analysis /
- Structural equation model /
- Waymo dataset

HTML全文

图 1 交通运行状态提取流程图

Figure 1. Traffic operation status extraction flowchart

下载: 全尺寸图片幻灯片

图 2 人机纵向交互安全结构方程模型

Figure 2. Equation model of the interactive safety between human and automated vehicles

下载: 全尺寸图片幻灯片

图 3 驾驶行为表征变量相关性矩阵图

Figure 3. Correlation matrix plot of driver behavior characterization variables

下载: 全尺寸图片幻灯片

图 4 修正后的结构方程模型标准化路径系数

Figure 4. Standardized path coefficients of the modified structural equation model

下载: 全尺寸图片幻灯片

表 1 数据变量及描述

Table 1. Data variables and descriptions

变量	描述
Segment_id	场景编号
Local_time	场景时间/ (0.1 s)
Local_id	本车编号
Leader_id	前车编号
Processed_position	处理后的位置信息/(m
Length	车长/m

下载: 导出CSV

表 2 变量描述性统计

Table 2. Descriptive analysis of variables

特征变量	前车类型
特征变量	AV	HDV
平均速度（/m/s）	8.47（6.66）	6.83（5.84）
速度变异系数	42.51（40.05）	49.07（34.93）
速度时间波动性	11.93（19.00）	14.10（24.19）
MTTC/s	6.19（7.29）	4.68（3.29）
路段速度平均水平（/m/s）	7.75（6.01）
注：值为平均值（标准差）。

下载: 导出CSV

表 3 AV和HDV前车类型驾驶行为特征差异性检验

Table 3. Significance test for driving behavioral characteristics between AVs and HDVs

驾驶行为特征变量	P值
平均速度	0.000 2
速度变异系数	0.015 5
速度时间波动性	0.120 9

下载: 导出CSV

表 4 结构方程模型的拟合指标

Table 4. Fitting goodness of structural model

指标	判别标准	拟合值
比较拟合指数（CFI）	≥0.90	0.980
调整适配度（AGFI）	≥0.80	0.925
适配度（GFI）	≥0.90	0.975
估计误差均方根（RMSEA）	＜0.10	0.097

下载: 导出CSV

表 5 前车驾驶行为测量模型的标准化路径系数

Table 5. Factor loads of the driving behaviors of the leading vehicles obtained by measurement model

潜在变量	观测变量	标准化路径系数
前车驾驶行为	平均速度	-0.411
	速度变异系数	0.863
	速度时间波动性	0.589
	车辆类型	0.402

下载: 导出CSV

表 6 结构方程模型的标准化路径系数

Table 6. Standardized path coefficients for structural equation model

路径	路径系数	S.E.	P
路段速度水平→前车驾驶行为	-0.54	0.032	***
路段速度水平→MTTC	-0.46	0.043	***
前车驾驶行为→MTTC	-0.44	0.044	***
前车车辆类型→MTTC	0.19	0.008	***
注：S. E. 为标准化误差；P为显著性：***表示p＜0. 001。

下载: 导出CSV

参考文献(27)

[1]	WAYMO L L C. We're building the world's most experienced driver[R/OL]. (2019-09-05)[2023-08-16]. https://waymo.com.
[2]	魏文. 华为预测2030年自动驾驶新车渗透率达20%, 智能汽车时代临近?[R/OL]. (2021-09-23)[2023-08-16]. https://m.yicai.com/news/101181289.html. WEI W. The penetration rate of new autonomous vehicles in China is expected to reach 20% by 2030. [EB/OL]. (2021-09-23)[2023-08-16]. https://m.yicai.com/news/101181289.html.
[3]	TAHIR Z, ALEXANDER R. Coverage based testing for V&V and safety assurance of self-driving autonomous vehicles: a systematic literature review[C]. IEEE International Conference on Artificial Intelligence Testing(AITest), Oxford, UK: IEEE, 2020.
[4]	MAHDINIA I, MOHAMMADNAZAR A, ARVIN R, et al. Integration of automated vehicles in mixed traffic: evaluating changes in performance of following human-driven vehicles[J]. Accident Analysis & Prevention, 2021, 152: 106006.
[5]	SCHOETTLE B, SIVAK M. A preliminary analysis of real-world crashes involving self-driving vehicles[R]. Ann Arbor, USA: University of Michigan Transportation Research Institute, 2015.
[6]	XU C, DING Z, WANG C, et al. Statistical analysis of the patterns and characteristics of connected and autonomous vehicle involved crashes[J]. Journal of Safety Research, 2019, 71: 41-47. doi: 10.1016/j.jsr.2019.09.001
[7]	MA Z, ZHANG Y. Driver-automated Vehicle interaction in mixed traffic: types of interaction and drivers' driving styles[J]. Human Factors, 2024, 66(2): 544-561. doi: 10.1177/00187208221088358
[8]	STANGE V, KUHN M, VOLLRATH M. Safety at first sight? - manual drivers' experience and driving behavior at first contact with Level 3 vehicles in mixed traffic on the highway[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2022, 87: 327-346. doi: 10.1016/j.trf.2022.04.004
[9]	REDDY N, HOOGENDOORN S P, FARAH H. How do the recognizability and driving styles of automated vehicles affect human drivers' gap acceptance at T-Intersections?[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2022, 90: 451-465. doi: 10.1016/j.trf.2022.09.018
[10]	HUANG Y, YE Y, SUN J, et al. Characterizing the impact of autonomous vehicles on macroscopic fundamental diagrams[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(6): 6530-6541. doi: 10.1109/TITS.2023.3265647
[11]	WEN X, CUI Z, JIAN S. Characterizing car-following behaviors of human drivers when following automated vehicles using the real-world dataset[J]. Accident Analysis & Prevention, 2022, 172: 106689.
[12]	WANG Y, FARAH H, YU R, et al. Characterizing behavioral differences of autonomous vehicles and human-driven vehicles at signalized intersections based on Waymo open dataset[J]. Transportation Research Record, 2023, 2677(11): 324-337. doi: 10.1177/03611981231165783
[13]	HU X, ZHENG Z, CHEN D, et al. Autonomous vehicle's impact on traffic: empirical evidence from Waymo open dataset and implications from modelling[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24 (6) : 6711-6724. doi: 10.1109/TITS.2023.3258145
[14]	YU R, ZHENG Y, QIN Y, et al. Utilizing partial least-squares path modeling to analyze crash risk contributing factors for Shanghai urban expressway system[J]. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, 2019, 5(4): 05019001. doi: 10.1061/AJRUA6.0001022
[15]	WAYMO L L C. Waymo open dataset[R/OL]. (2019)[2023-08-16]. https://waymo.com/open/.
[16]	HOUSTON J, ZUIDHOF G, BERGAMINI L, et al. One thousand and one hours: self-driving motion prediction dataset[C]. Conference on Robot Learning, London, UK: PMLR, 2021.
[17]	CAESAR H, BANKITI V, LANG A, et al. Nuscenes: a multimodal dataset for autonomous driving[C]. IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, USA: IEEE, 2020.
[18]	PUNZO V, MONTANINO M, CIUFFO B. On the assessment of vehicle trajectory data accuracy and application to the next generation simulation(NGSIM)program data[J]. Transportation Research Part C: Emerging Technologies, 2011, 19(6): 1243-1262. doi: 10.1016/j.trc.2010.12.007
[19]	SUN P, KRETZSCHMAR H, DOTIWALLA X, et al. Scalability in perception for autonomous driving: Waymo open dataset[C]. IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, USA: IEEE, 2020.
[20]	ETTINGER S, CHENG S, CAINE B, et al. Large scale interactive motion forecasting for autonomous driving: the waymo open motion dataset[C]. IEEE/CVF International Conference on Computer Vision. Montreal, Canada: IEEE, 2021.
[21]	SUN Z, YAO X, QIN Z, et al. Modeling car-following heterogeneities by considering leader-follower compositions and driving style differences[J]. Transportation Research Record, 2021, 2675(11): 851-864. doi: 10.1177/03611981211020006
[22]	HU X, ZHENG Z, CHEN D, et al. Processing, assessing, and enhancing the Waymo autonomous vehicle open dataset for driving behavior research[J]. Transportation Research Part C: Emerging Technologies, 2022, 134: 103490. doi: 10.1016/j.trc.2021.103490
[23]	WANG C, XIE Y, HUANG H, et al. A review of surrogate safety measures and their applications in connected and automated vehicles safety modeling[J]. Accident Analysis & Prevention, 2021, 157: 106157.
[24]	鄢云珠, 傅忠宁, 岳金田. 车辆怠速起停系统使用意愿结构方程模型及影响因素分析[J]. 交通信息与安全, 2023, 41(6): 161-170. YAN Y Z, FU Z N, YUE J T. An analysis of the influence factors on using intention of vehicle idle start-stop system with a structural equation model[J]. Journal of Transport Information and Safety, 2023, 41(6): 161-170. (in Chinese)
[25]	王永岗, 张衡, 彭志鹏, 等. 基于结构方程模型的出租车事故影响因素分析[J]. 重庆交通大学学报(自然科学版), 2021, 40(6): 36. doi: 10.3969/j.issn.1674-0696.2021.06.06 WANG Y G, ZHANG H, PENG Z P, et al. Analysis of influencing factors of taxi accidents based on structural equation model[J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(6): 36. (in Chinese) doi: 10.3969/j.issn.1674-0696.2021.06.06
[26]	陈春. 道路交通事故的影响因素研究: 基于结构方程模型的实证研究[J]. 中国安全生产科学技术, 2014, 10(5): 110-116. CHEN C. Research on influencing factors of road traffic accidents: empirical study based on structural equation model[J]. Journal of Safety Science and Technology, 2014, 10 (5): 110-116. (in Chinese)
[27]	姚荣涵, 祁文彦, 郭伟伟. 自动驾驶环境下驾驶人接管行为结构方程模型[J]. 交通运输工程学报, 2021, 21(2): 209-221. YAO R H, QI W Y, GUO W W. Structural equation model of drivers' takeover behaviors in autonomous driving environment[J]. Journal of Traffic and Transportation Engineering, 2021, 21(2): 209-221. (in Chinese)