网联环境下基于精简车头时距特性的驾驶风格分类

吕能超; 高谨谨; 王维锋; 王玉刚

doi:10.3963/j.jssn.1674-4861.2022.01.014

网联环境下基于精简车头时距特性的驾驶风格分类

doi: 10.3963/j.jssn.1674-4861.2022.01.014

1.
武汉理工大学智能交通系统研究中心武汉 430063
2.
河海大学土木与交通学院南京 210098

基金项目:

国家自然科学基金面上项目 52072290

国家重点研发计划项目 2020YFB1600302

湖北省杰出青年基金项目 2020CFA081

详细信息

作者简介:
吕能超（1982—），博士、教授. 研究方向：辅助驾驶系统、智能网联汽车、交通安全评价.E-mail：lnc@whut.edu.cn

通讯作者:
王维锋（1979—），博士、教授. 研究方向：智慧公路、车路协同、交通大数据与人工智能.E-mail：Wangweifeng@hhu.edu.cn

中图分类号: U491
计量
- 文章访问数: 802
- HTML全文浏览量: 394
- PDF下载量: 53
- 被引次数: 0
出版历程
- 收稿日期: 2021-09-28
- 网络出版日期: 2022-03-31

Classification of Driving Style Using Simplified Features of Headway Under the Connected Vehicles Environment

1.
Intelligent Transportation Systems Research Center, Wuhan University of Technology, Wuhan 430063, China
2.
College of Civil and Transportation Engineering, Hohai University, Nanjing 210098, Chinz

摘要

摘要:
基于现有网联数据获取技术与条件，从车联网系统提取车头时距参数并将3 s内的车头时距特征值定义为驾驶模式，根据驾驶模式进而对驾驶风格（即驾驶人的驾驶行为习惯）进行分类。通过车头时距特性对驾驶模式进行量化分类，根据标定好的驾驶风格结果，辨识每种驾驶风格包含的典型驾驶模式；运用模糊分类方法赋予典型驾驶模式相应分值，通过计算每位驾驶人分值并结合已标定的驾驶风格结果设定每种驾驶风格的阈值；利用该阈值对测试集中的驾驶人风格进行识别，以验证识别准确率。采集了44名驾驶人网联环境行车数据将驾驶人标定为激进型、普通（即既不保守也不激进）型和保守型。按上述方法设置各驾驶风格阈值，结果表明：各驾驶风格的阈值分别为：S < 64.67为保守型，64.67 ≤ S < 181.20为普通型，S ≥ 181.20为激进型；使用所提方法来识别驾驶人风格，总体准确率为85.7%。所提出的基于车头时距的驾驶风格分类方法，使用了极精简的驾驶行为参数，为驾驶风格分类应用提供了新思路。
- 交通工程 /
- 车联网 /
- 驾驶风格分类 /
- 车头时距 /
- 驾驶模式 /
- 模糊隶属度
Abstract:
Based on current data collection techniques from connected vehicles, this paper aims to classify driving styles(i.e., driving habits or behavior)by analyzing driving modes which is defined as time headway in 3 s. Specifically, driving modes are quantitatively classified by time headway and typical driving modes reflecting each driving style are identified according to the calibrated driving styles. Evaluation score is assigned to each typical driving mode using a fuzzy classification method and the thresholds of each driving style is proposed based on the evaluation scores and the calibrated driving styles. The thresholds are applied to a test data set, which includes driving behavior data of 44 drivers, to verify the accuracy of the proposed method. In summary, three types of driving styles are identified: the evaluation score S < 64.67 is seen as the conservative driving style(CDS), the score 64.67 ≤ S < 181.20 is classified as the"regular"(that is, neither-conservation-nor-aggressive(NCNA))driving style; and the score S ≥ 181.20 is grouped into the aggressive driving style(ADS). Study results show that the accuracy of the proposed method against the testing data set is 85.7%. The proposed method uses simplified driving parameters (headway)for driving-style classification, which provides a new way for driving-style classification.
- traffic engineering /
- Connected Vehicles /
- driving style classification /
- time headway /
- driving mode /
- fuzzy membership degree

HTML全文

图 1 基于网联数据的驾驶风格分类建模思路流程

Figure 1. Procedure of driving style classification modeling based on connected data

下载: 全尺寸图片幻灯片

图 2 基于车头时距的典型驾驶模式提取方法

Figure 2. The method of typical driving patterns based on THW

下载: 全尺寸图片幻灯片

图 3 车头时距分级及编码

Figure 3. Grading and coding of THW

下载: 全尺寸图片幻灯片

图 4 驾驶模式提取及编码过程

Figure 4. The process of car driving patterns extracting and coding

下载: 全尺寸图片幻灯片

图 5 实验平台

Figure 5. Experimental platform

下载: 全尺寸图片幻灯片

图 6 实验路线

Figure 6. Experimental routes

下载: 全尺寸图片幻灯片

图 7 训练集驾驶人分值分布情况

Figure 7. Drivers'scoring of training set

下载: 全尺寸图片幻灯片

图 8 驾驶风格识别结果

Figure 8. The result of driving style recognition.

下载: 全尺寸图片幻灯片

表 1 车头时距等级划分情况

Table 1. Classification of THW class

车头时距/s	> 0~0.6	> 0.6~1.0	> 1.0~1.5	> 1.5~2.0	> 2.0~2.5	> 2.5~4.0	> 4.0~6.0	> 6.0~∞
等级	1	2	3	4	5	6	7	8

下载: 导出CSV

表 2 自然驾驶平台采集的原始数据

Table 2. Raw data from a natural flight platform

数据来源	变量名
	油门踏板开度/%
	制动压力/MPa
车载CAN总线	转向盘角度/(°)
	转向盘角速度/((°)/s)
	本车车速/(km/h)
	本车横向加速度/(m/s²)
	本车纵向加速度(m/s²)
惯导系统	横摆角速度(Os)
	经度/(°)
	纬度/(°)
	距右车道线距离/m
	距左车道线距离/m
Mobileye	同车道前车车头间距(DHW)/m
	同车道前车车头时距(THW)/s
	同车道前车碰撞时间(TTC) /s

下载: 导出CSV

表 3 驾驶风格标定结果

Table 3. The result of driving style calibration

驾驶风格	激进型	普通型	保守型
人数	6	22	16

下载: 导出CSV

表 4 典型驾驶模式初始分值

Table 4. Typical initial values of the following patterns

驾驶风格	激进型	普通型	保守型
初始分值	15	5	-5

下载: 导出CSV

表 5 部分典型的驾驶模式隶属度及对应分值

Table 5. A part of the membership and the corresponding values of typical car following patterns

激进型			普通型			保守型
驾驶模式	隶属度	分值	驾驶模式	隶属度	分值	驾驶模式	隶属度	分值
222	1.000 0	15.00	333	1.000 0	5.00	666	1.000 0	-5.00
433	0.103 5	1.56	444	0.823 3	4.12	777	0.390 8	-1.95
332	0.100 9	1.52	544	0.085 3	0.43	555	0.380 5	-1.90
443	0.094 0	1.41	554	0.073 3	0.37	665	0.050 6	-0.25
322	0.090 7	1.36	654	0	0	566	0.043 5	-0.22

下载: 导出CSV

表 6 训练集驾驶人得分情况

Table 6. Training set the driver of the score

分值	驾驶人风格类型
分值	激进型	普通型	保守型
平均分值	399.03	123.11	-15.03
分值标准差	96.63	85.02	135.38
分值极大值	436.99	238.38	222.81
分值极小值	273.94	-180.03	-81.21

下载: 导出CSV

表 7 测试集驾驶人得分情况

Table 7. Drivers'scoring of test set

序号	专家1评分	专家2评分	专家3评分	三分制评价结果	模型分值	识别结果
1	1	2	2	2	-13.67	1
2	1	1	1	1	-13.74	1
3	2	3	3	3	309.83	3
4	2	1	2	2	41.14	1
5	1	2	2	2	86.68	2
6	1	1	1	1	64.09	1
7	2	2	2	2	66.00	2
8	2	2	2	2	75.84	2
9	2	1	1	1	-25.02	1
10	2	3	2	2	82.89	2
11	2	1	1	1	27.98	1
12	1	2	2	2	133.89	2
13	2	1	1	1	-27.56	1
14	2	3	3	3	244.58	3
注：1-保守型；2-普通型；3-激进型。

下载: 导出CSV

参考文献(27)

[1]	FRENCH D J, WEST R J, Elander J, et al. Decision-making style, driving style, and self-reported involvement in road traffic accidents[J]. Ergonomics, 1993(36): 627-644. http://www.researchgate.net/profile/John_Wilding/publication/14685091_Decision-making_style_driving_style_and_self-reported_involvement_in_road_traffic_accidents/links/0912f50b9260ae158c000000.pdf
[2]	SAGBERG F, SELPI S, PICCININI G F B, et al. A Review of research on driving styles and road safety[J]. Human Factors: The Journal of the Human Factors Society, 2015, 57(7): 1248-1275. doi: 10.1177/0018720815591313
[3]	万豫, 黄妙华, 王思楚. 基于改进DBSCAN算法的驾驶风格识别方法研究[J]. 合肥工业大学学报(自然科学版), 2020, 43(10): 23-30. https://www.cnki.com.cn/Article/CJFDTOTAL-HEFE202010003.htm WAN Y, HUANG M H, WANG S C. Study on driving style recognition based on improved DBSCAN algorithm[J]. Journal of Hefei University of Technology(Natural Science), 2020, 43(10): 23-30. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HEFE202010003.htm
[4]	DENG C, WU C Z, LYU N C, et al. Driving style recognition method using braking characteristics based on hidden Markov model[J]. Plos One, 2017, 12(8): 0182419. http://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0182419&type=printable
[5]	余荣杰, 龙晓捷, 涂颖菲, 等. 基于低频轨迹数据的分时租赁驾驶人驾驶风格分析[J]. 同济大学学报(自然科学版), 2019, 47(10): 1463-1469. doi: 10.11908/j.issn.0253-374x.2019.10.011 YU R J, LONG X J, TU Y F, et al. Driving style analysis of time-sharing rental drivers based on low frequency trajectory data[J]. Journal of Tongji University(Natural Science), 2019, 47(10): 1463-1469. (in Chinese) doi: 10.11908/j.issn.0253-374x.2019.10.011
[6]	AMIN M, RAMIN A, ASAD J K. Classifying travelers'driving style using basic safety messages generated by connected vehicles: Application of unsupervised machine learning[J]. Transportation Research Part C: Emerging Technologies, 2021(122): 102917. http://www.sciencedirect.com/science/article/pii/S0968090X20308160
[7]	ALJAAFREH A, ALSHABATAT N, AL-DIN M. Driving style recognition using fuzzy logic[C]. 2012 IEEE International Conference on Vehicular Electronics and Safety (ICVES 2012), Istanbul, Turkey: IEEE, 2012.
[8]	CARSTEN O, KIRCHER K, JAMSON S. Vehicle-based studies of driving in the real world: The hard truth?[J]. Accident Analysis & Prevention, 2013, 58(5): 162-174. http://pdfs.semanticscholar.org/0808/72a6480f631994522fc71c01d0eb33ff639f.pdf
[9]	邓天民, 朱杰, 朱凯家. 基于iForest+Biscting K-means的驾驶风格分类识别方法研究[J]. 重庆交通大学学报(自然科学版), 2021, 40(6): 1-6. doi: 10.3969/j.issn.1674-0696.2021.06.01 DENG T M, ZHU J, ZHU K J, et al. Study on classification and recognition of driving style based on iForest + Biscting K-means[J]. Journal of Chongqing Jiaotong University(Natural Science)2021, 40(6): 1-6. (in Chinese) doi: 10.3969/j.issn.1674-0696.2021.06.01
[10]	PADILLA J L, CASTRO C, DONCEL P, et al. Adaptation of the multidimensional driving styles inventory for Spanish drivers: Convergent and predictive validity evidence for detecting safe and unsafe driving styles[J]. Accident Analysis & Prevention, 2020(136): 1045413. http://www.sciencedirect.com/science/article/pii/S0001457519300752
[11]	张雪榆, 潘存书, 徐进. 城市道路多车道交织区的微观驾驶行为特性[J]. 交通信息与安全, 2020, 38(3): 138-147. doi: 10.3963/j.jssn.1674-4861.2020.03.018 ZHANG X Y, PAN C S, XU J. Micro-interleaved multi-lane area of urban road driving behaviors[J]. Traffic Information and security, 2020, 38(3): 138-147.(in Chinese) doi: 10.3963/j.jssn.1674-4861.2020.03.018
[12]	TAUBMAN-BEN-ARI O, MIKULINCER M, GILLATH O. The multidimensional driving style inventory-scale construct and validation[J]. Accident Analysis & Prevention, 2004, 36 (3): 323-332. http://www.researchgate.net/profile/Mario_Mikulincer/publication/7487488_The_multidimensional_driving_style_inventory_-_Scale_construct_and_validation/links/573d6ffb08ae298602e5a523.pdf
[13]	孙龙, 杨程程, 常若松. 多维度驾驶风格量表的修订及初步应用[J]. 人类工效学, 2014, 20(2): 6-9. https://www.cnki.com.cn/Article/CJFDTOTAL-XIAO201402003.htm SUN L, YANG C C, CHANG R S. Revision and preliminary application of multidimensional driving style scale[J]. Ergonomics, 2014, 20(2): 6-9. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XIAO201402003.htm
[14]	LI G F, LI S E, CHENG B, et al. Estimation of driving style in naturalistic highway traffic using maneuver transition probabilities[J]. Transportation Research Part C: Emerging Technologies, 2017(74): 113-125. http://smartsearch.nstl.gov.cn/paper_detail.html?id=b8c1be8ae08c94cdf69a08976cab645e
[15]	MARIA V N Z, MENG F L, SU J Y, et al. Dynamic clustering analysis for driving styles identification[J]. Engineering Applications of Artificial Intelligence, 2021(97): 104096. http://www.sciencedirect.com/science/article/pii/S0952197620303468
[16]	王畅, 付锐, 彭金柱, 等. 应用于换道预警的驾驶风格分类方法[J]. 交通运输系统工程与信息, 2014, 14(3): 187-193. doi: 10.3969/j.issn.1009-6744.2014.03.029 WANG C, FU R, PENG J Z, et al. Driving style classification method applied to lane change warning[J]. Systems Engineering, 2014, 14(3): 187-193. (in Chinese) doi: 10.3969/j.issn.1009-6744.2014.03.029
[17]	GUO Q Y, ZHAO Z G, SHEN P H, et al. Adaptive optimal control based on driving style recognition for plugin hybrid electric vehicle[J]. Energy, 2019(186): 115824. http://econpapers.repec.org/article/eeeenergy/v_3a186_3ay_3a2019_3ai_3ac_3as0360544219314963.htm
[18]	刘通, 付锐, 张名芳, 等. 融合K-means与高斯混合模型的驾驶风格聚类研究[J]. 中国安全科学学报, 2019, 29 (12): 40-45. https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201912007.htm LIU T, FU R, ZHANG M F, et al. Driving style clustering based on hybrid model of K-means and Gauss[J]. Chinese Journal of Security Sciences, 2019, 29(12): 40-45. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201912007.htm
[19]	王鑫鹏, 陈志军, 吴超仲, 等. 考虑驾驶风格的智能车自主驾驶决策方法[J]. 交通信息与安全, 2020, 38(2): 37-46. doi: 10.3963/j.jssn.1674-4861.2020.02.005 WANG X P, CHEN Z J, WU C Z, et al. Consider intelligent autonomous decision method of driving the car driving style[J]. Journal of Transport Information and safety, 2020, 38(2): 37-46. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2020.02.005
[20]	李明俊, 张正豪, 宋晓琳, 等. 基于一种多分类半监督学习算法的驾驶风格分类模型[J]. 湖南大学学报(自然科学版), 2020, 47(4): 10-15. https://www.cnki.com.cn/Article/CJFDTOTAL-HNDX202004002.htm LI M J, ZHANG Z H, SONG X L, et al. Driving style classification model based on a multi-classification semi-supervised learning algorithm[J]. Journal of the Hunan University (Natural Science), 2020, 47(4): 10-15. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HNDX202004002.htm
[21]	MACADAM C, BAREKET Z, FANCHER P, et al. Using neural networks to identify driving style and headway control behavior of drivers[J]. Vehicle System Dynamics, 1998, 29(1): 143-160. http://www.researchgate.net/profile/Zevi_Bareket/publication/245309079_USING_NEURAL_NETWORKS_TO_IDENTIFY_DRIVING_STYLE_AND_HEADWAY_CONTROL_BEHAVIOR_OF_DRIVERS/links/54bd5d5f0cf27c8f2814b6b9.pdf
[22]	MARTINEZ C M, HEUCKE M, WANG F Y, et al. Driving style recognition for intelligent vehicle control and advanced driver assistance: A survey[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 19(3): 666-676. doi: 10.1109/TITS.2017.2706978
[23]	LIN J, KEOGH E J, LONARDI S, et al. A Symbolic Representation of Time Series, with Implications for Streaming Algorithms[C]. 8^th ACM SIGMOD Workshop on Research Issues in Data Mining and Knowledge Discovery, San Diego, California: USA, 2003.
[24]	LYU N C, DENG C, XIE L, et al. A field operational test in China: Exploring the effect of an advanced driver assistance system on driving performance and braking behavior[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2019(65): 730-747. http://www.sciencedirect.com/science?_ob=ShoppingCartURL&_method=add&_eid=1-s2.0-S1369847816306672&originContentFamily=serial&_origin=article&_ts=1520521217&md5=205f499431f7c021df09c13c97e32c06
[25]	CHEN C, ZHAO X, ZHANG Y L, et al. A graphical modeling method for individual driving behavior and its application in driving safety analysis using GPS data[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2019(63): 118-134. http://www.onacademic.com/detail/journal_1000042281617099_d096.html
[26]	CHOW S. Sample size calculations in clinical research[M]. London: Taylor & Francis, 2007.
[27]	刘通, 付锐, 马勇, 等. 考虑驾驶人风格的跟车预警规则研究[J]. 中国公路学报, 2020, 33(2): 170-180. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202002016.htm LIU T, FU R, MA Y, et al. Interleaved multi-lane urban roads into account research on driving style come with alert rules[J]. China Journal of Highway and Transport, 2020, 33 (2): 170-180. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202002016.htm