Online Driving Style Recognition Method Considering Lane-Changing Game

doi:10.12141/j.issn.1000-565X.230159

Abstract

Abstract:

Driving style is the external expression of driving behavior. Drivers with aggressive style tend to engage in more frequent risky driving operations, intensifying interactions between vehicles and affecting lane-changing safety. Identifying a driver’s driving style before executing a lane-changing can effectively constrain driver’s behavior through personalized warning information. This paper proposed the SHAP-XGBoost method, which considers lane-changing game in a connected environment, aiming to achieve the real-time recognition of driving styles during the lane-changing intention phase. Firstly, the fluctuation degree of individual behavior and gaming behavior during the lane-changing intention was used as input feature variables, and the driving style was marked by correlation analysis, principal component analysis, and four different clustering methods. Next, the proposed SHAP-XGBoost model was used to select key features for training the driving style recognition model, and online recognition was completed through a sliding window. Finally, experiments were conducted using the HighD dataset. Results show that: compared with clustering methods based on centroid distance, connectivity and density distribution, spectral clustering based on graph theory principles can better label driving styles based on the morphology of the input feature variables; using the proposed SHAP-XGBoost model with 14 key features for driving style recognition can improve online recognition efficiency without loss of accuracy, and the driving style recognition accuracy is up to 99%; simultaneously incorporating individual features and gaming features as inputs to the model can improve the accuracy of driving style labeling and recognition. The research results can be used to support personalized lane-changing decisions and early warnings.

Key words: intelligent transportation, driving style recognition, extreme gradient boosting tree, lane-changing safety

CLC Number:

U491.31

ZHANG Yunchao, HUANG Jianling, LI Yongxing, et al. Online Driving Style Recognition Method Considering Lane-Changing Game[J]. Journal of South China University of Technology(Natural Science Edition), 2024, 52(4): 126-137.

Figures/Tables 17

Fig.1

Fig.2

Fig.3

Table 1

Table 2

Fig.4

Fig.5

Fig.6

Table 3

Fig.7

Fig.8

Table 4

Table 5

Fig.9

Table 6

Table 7

Fig.10

References 28

1	ISHIBASHI M， OKUWA M，DOI S，et al ．Indices for characterizing driving style and their relevance to car following behavior［C］∥ Proceedings of SICE Annual Conference 2007．Takamatsu：IEEE，2007：1132-1137.
2	CARSTEN O， KIRCHER K， JAMSON S ．Vehicle-based studies of driving in the real world：the hard truth？［J］．Accident Analysis & Prevention，2013，58：162-174.
3	SHANGGUAN Q， FU T， WANG J，et al ．A proactive lane-changing risk prediction framework considering driving intention recognition and different lane-changing patterns［J］．Accident Analysis & Prevention，2022，164：106500/1-14.
4	公安部交通管理局．中国道路交通事故统计年鉴（2019）［R］．北京：公安部交通管理局，2020.
5	TAUBMAN-BEN-ARI O， SKVIRSKY V ．The multidimensional driving style inventory a decade later：review of the literature and re-evaluation of the scale［J］．Accident Analysis & Prevention，2016，93：179-188.
6	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.
7	MOHAMMADNAZAR A， ARVIN R， KHATTAK A J ．Classifying travelers’ driving style using basic safety messages generated by connected vehicles：application of unsupervised machine learning［J］．Transportation Research Part C：Merging Technologies，2021，122：102917/1-18.
8	BROMBACHER P， MASINO J， FREY M，et al ．Driving event detection and driving style classification using artificial neural networks［C］∥ Proceedings of 2017 IEEE International Conference on Industrial Technology．Toronto：IEEE，2017：997-1002.
9	De RANGO F， TROPEA M， SERIANNI A，et al ．Fuzzy inference system design for promoting an eco-friendly driving style in IoV domain［J］．Vehicular Communications，2022，34：100415/1-17.
10	刘冠颖，郭凤香，申江卫，等．基于数据特征的驾驶风格分类与识别方法研究［J］．昆明理工大学学报（自然科学版），2023，48（3）：165-173.
	LIU Guanying， GUO Fengxiang， SHEN Jiangwei，et al ．Driving style classification and recognition method based on data features［J］．Journal of Kunming University of Science and Technology （Natural Science Edition），2023，48（3）：165-173.
11	柳祖鹏，罗陈怡，严运兵．考虑车辆跟车及换道交互参数的驾驶风格识别［J］．武汉理工大学学报（交通科学与工程版），2023，47（2）：209-213.
	LIU Zupeng， LUO Chenyi， YAN Yunbing ．Driving style recognition considering vehicle following vehicle and lane-changing interaction parameters［J］．Journal of Wuhan University of Technology （Transportation Science and Engineering），2023，47（2）：209-213.
12	朱兴林，姚亮，刘泓君，等．考虑驾驶风格差异的高原公路危险路段识别研究［J］．交通运输系统工程与信息，2022，22（6）：172-182.
	ZHU Xing-lin， YAO Liang， LIU Hong-jun，et al ．Identification of dangerous sections of highland roads considering different driving behaviors［J］．Journal of Transportation Systems Engineering and Information Technology，2022，22（6）：172-182.
13	LI G， 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.
14	LI Y， GU R， LEE J，et al ．The dynamic tradeoff between safety and efficiency in discretionary lane-changing behavior：a random parameters logit approach with heterogeneity in means and variances［J］．Accident Analysis & Prevention，2021，153：106036/1-7.
15	SUZDALEVA E， NAGY I ．An online estimation of driving style using data-dependent pointer model［J］．Transportation Research Part C：Emerging Technologies，2018，86：23-36.
16	KRAJEWSKI R， BOCK J， KLOEKER L，et al ．The highD dataset：a drone dataset of naturalistic vehicle trajectories on German highways for validation of highly automated driving systems［C］∥ Proceedings of 2018 the 21st International Conference on Intelligent Transportation Systems．Maui：IEEE，2018：2118-2125.
17	HAN T， JING J， ÖZGÜNER Ü ．Driving intention recognition and lane change prediction on the highway［C］∥ Proceedings of 2019 IEEE Intelligent Vehicles Symposium （Ⅳ）．Paris ：IEEE，2019：957-962.
18	ARVIN R， KAMRANI M， KHATTAK A J ．The role of pre-crash driving instability in contributing to crash intensity using naturalistic driving data［J］．Accident Analysis & Prevention，2019，132：105226/1-13.
19	KAMRANI M， ARVIN R， KHATTAK A J ．Extracting useful information from basic safety message data：an empirical study of driving volatility measures and crash frequency at intersections［J］．Transportation Research Record，2018，2672（38）：290-301.
20	ARVIN R， KHATTAK A J， QI H ．Safety critical event prediction through unified analysis of driver and vehicle volatilities：application of deep learning methods［J］．Accident Analysis & Prevention，2021，11：105949/1-12.
21	JAIN A K ．Data clustering：50 years beyond K-means ［J］．Pattern Recognition Letters，2010，31（8）：651-666.
22	ABIRAMI K， MAYILVAHANAN P ．Performance analysis of K-means and bisecting K-means algorithms in Weblog data［J］．International Journal of Emerging Technologies in Engineering Research，2016，4（8）：119-124.
23	NG A， JORDAN M， WEISS Y ．On spectral clustering：analysis and an algorithm［C］∥ Proceedings of the 14th International Conference on Neural Information Processing Systems：Natural and Synthetic．Vancouver：ACM，2001：849-856.
24	LUNDBERG S M， LEE S ．A unified approach to interpreting model predictions［C］∥ Proceedings of the 31st International Conference on Neural Information Processing Systems．Long Beach：ACM，2017，30：4768-4777.
25	CHEN T， GUESTRIN C ．XGBoost：a scalable tree boosting system［C］∥ Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining．San Francisco：ACM，2016：785-794.
26	CHAWLA N V， BOWYER K W， HALL L O，et al ．SMOTE：synthetic minority over-sampling technique［J］．Journal of Artificial Intelligence Research，2002，16（1）：321-357.
27	赵建东，赵志敏，屈云超，等．轨迹数据驱动的车辆换道意图识别研究［J］．交通运输系统工程与信息，2022，22（4）：63-71.
	ZHAO Jian-dong， ZHAO Zhi-min， QU Yun-chao，et al ．Vehicle lane change intention recognition driven by trajectory data［J］．Journal of Transportation Systems Engineering and Information Technology，2022，22（4）：63-71.
28	赵晓华，董文慧，李佳，等．基于驾驶行为的隧道交通标志影响特征及作用机理［J］．华南理工大学学报（自然科学版），2023，51（4）：88-100.
	ZHAO Xiaohua， DONG Wenhui， LI Jia，et al ．Influence characteristics and action mechanism of tunnel traffic signs based on driving behavior［J］．Journal of South China University of Technology （Natural Science Edition），2023，51（4）：88-100.

时间窗/s	换道样本数		换道样本数
1.0	4 417	3.5	2 750
1.5	4 098	4.0	2 471
2.0	3 813	4.5	2 146
2.5	3 459	5.0	1 884
3.0	3 136

特征类型	特征序号	特征名称	特征含义	单位
个体操作行为特征	1	xVelocity	车辆在X轴方向上的速度	m/s
	2	yVelocity	车辆在Y轴方向上的速度	m/s
	3	xAcceleration	车辆在X轴方向上的加速度	m/s²
	4	yAcceleration	车辆在Y轴方向上的加速度	m/s²
博弈行为特征	5	SV&CLV_vd_x	SV&CLV在X轴方向的速度差	m/s
	6	SV&CLV_vd_y	SV&CLV在Y轴方向的速度差	m/s
	7	SV&CLV_ad_x	SV&CLV在X轴方向的加速度差	m/s²
	8	SV&CLV_ad_y	SV&CLV在Y轴方向的加速度差	m/s²
	9	SV&CLV_gd_x	SV&CLV在X轴方向的距离差	m
	10	SV&CLV_gd_y	SV&CLV在Y轴方向的距离差	m
	11	SV&TFV_vd_x	SV&TFV在X轴方向的速度差	m/s
	12	SV&TFV_vd_y	SV&TFV在Y轴方向的速度差	m/s
	13	SV&TFV_ad_x	SV&TFV在X轴方向的加速度差	m/s²
	14	SV&TFV_ad_y	SV&TFV在Y轴方向的加速度差	m/s²
	15	SV&TFV_gd_x	SV&TFV在X轴方向的距离差	m
	16	SV&TFV_gd_y	SV&TFV在Y轴方向的距离差	m
	17	SV&TLV_vd_x	SV&TLV在X轴方向的速度差	m/s
	18	SV&TLV_vd_y	SV&TLV在Y轴方向的速度差	m/s
	19	SV&TLV_ad_x	SV&TLV在X轴方向的加速度差	m/s²
	20	SV&TLV_ad_y	SV&TLV在Y轴方向的加速度差	m/s²
	21	SV&TLV_gd_x	SV&TLV在X轴方向的距离差	m
	22	SV&TLV_gd_y	SV&TLV在Y轴方向的距离差	m

聚类算法	聚类原理	SC		DB
聚类算法	聚类原理	个体特征	个体+博弈特征	个体特征	个体+博弈特征
K-均值	质心距离	0.19	0.22	1.82	1.73
高斯混合	概率分度	0.20	0.22	1.76	1.73
层次聚类	连通性	0.21	0.23	1.73	1.72
谱聚类	图论	0.39	0.41	0.97	0.90

特征	均值
特征	类别1	类别2	类别3
A_v_SV&CLV_vd_x	2.46	3.02	3.69
S_dev_SV&CLV_gd_x	3.41	5.02	5.21
S_dev_SV&TLV_vd_x	0.53	0.84	1.10
S_dev_SV&TLV_gd_x	9.76	15.51	18.72
S_dev_SV&TLV_gd_y	0.16	0.18	0.22
C_v_SV&TFV_vd_x	0.22	1.04	2.29
C_v_SV&TFV_ad_y	0.76	1.20	2.10
C_v_SV&TFV_gd_x	0.40	1.24	2.45
C_v_SV&TFV_gd_y	0.11	1.01	2.42

参数名称	参数含义	最优取值
colsample_bytree	采样比例	0.8
gamma	惩罚项系数	0.3
learning_rate	学习率	0.08
max_depth	树最大深度	3
min_child_weight	叶子最小样本数	2
n_estimators	弱分类器数量	600
reg_alpha	L1正则化权重项	0.01
reg_lambda	L2正则化权重项	0.1
subsample	采样率	0.7