基于车辆位置与速度特征的驾驶行为模式分类方法

张苇冲; 杨涛; 吕能超

doi:10.3963/j.jssn.1674-4861.2023.01.009

基于车辆位置与速度特征的驾驶行为模式分类方法

doi: 10.3963/j.jssn.1674-4861.2023.01.009

张苇冲^1,,
杨涛²,
吕能超^1, ,

1.
武汉理工大学智能交通系统研究中心武汉 430063
2.
武汉中交交通工程有限责任公司武汉 430000

基金项目:

国家自然科学基金项目 52072290

湖北省杰青项目 2020CFA081

详细信息

作者简介:
张苇冲（1998—），硕士研究生.研究方向：交通信息与安全. E-mail: zhangweichong@whut.edu.cn

通讯作者:
吕能超（1982—），博士，研究员. 研究方向：道路交通安全评价. E-mail: lnc@whut.edu.cn

中图分类号: U492.8+4
计量
- 文章访问数: 582
- HTML全文浏览量: 263
- PDF下载量: 38
- 被引次数: 0
出版历程
- 收稿日期: 2022-09-28
- 网络出版日期: 2023-05-13

A Method for Classifying Driving Behavior Based on Vehicle Position and Speed

1.
Intelligent Transportation Systems Research Center, Wuhan University of Technology, Wuhan 430063, China
2.
Wuhan China Communication Transportation Engineering Company, Wuhan 430000, China

摘要

摘要: 精细车辆轨迹中包含连续的时间戳、位置，以及速度等信息。通过对车辆轨迹数据进行量化表达与挖掘分析，可以实现对车辆行为模式的分类。现有研究大多关注对位置的聚类，很少对车速、加速度等特征进行研究分析，而车速等是反映驾驶行为模式的重要特征。为了将轨迹多维信息纳入分析框架，研究了基于位置与速度特征的车辆轨迹行为模式分类方法。为克服现有行为模式分类方法的维度单一性，运用豪斯多夫轨迹距离算法计算出位置和速度特征的综合距离矩阵，针对豪斯多夫距离算法鲁棒性差的缺点，采用单向豪斯多夫距离90%分位值对算法进行了改进，降低噪声影响。同时，引入了车辆位置和速度来进一步提高分类的准确性，运用多次分层聚类算法依次对位置与速度轨迹图进行分类，得到车辆位置和速度上的行为模式。以HighD数据集为样本，提取了三车道上的行车轨迹，验证了基于速度与位置特征的车辆行为模式分类方法。结果表明：①本方法可以得到位置和速度的综合行为模式，聚类平均准确率达到94.8%，优于DBTCAN准确率89.3%和t-Cluster准确率86.4%；②基于换道模式轨迹偏移率曲线的分析，得到了4种互异的典型车辆换道模式。该方法可利用多维轨迹数据对行车模式进行分类及行为辨识，在车辆轨迹分类与不良行为辨识方面具有应用潜力。
- 智能交通 /
- 车辆轨迹 /
- 多维特征 /
- 轨迹分类 /
- 豪斯多夫距离
Abstract: Vehicle trajectory data contains vehicle movement information, including time stamp, vehicle position, speed, etc. By analyzing vehicle trajectory data, driving patterns can be classified. As important features from such data reflect driving behavior, vehicle positioning characteristics have been widely studied, but the others such as speed and acceleration are rarely analyzed. In order to incorporate the multi-dimensional information from vehicle trajectory data into the analysis framework, a method for classifying driving patterns based on the characteristics of vehicle position and speed is studied. To overcome the issue of a single dimension of the existing classification methods, the algorithm for Hausdorff trajectory distance is applied to calculate a comprehensive distance matrix of vehicle position and speed. Given the fact that the robustness of the Hausdorff distance algorithm is low, the algorithm is improved by using 90% percentile value of the one-way Hausdorff distance to reduce the influence of noise. At the same time, vehicle position and speed are introduced to further improve the accuracy of classification, and a multiple hierarchical clustering algorithm is used to classify the trajectory diagrams of position and trajectory diagrams of speed in sequence. At the end, the driving patterns based on vehicle position and speed are obtained. The HighD dataset is used as a sample, the vehicle trajectories on three lanes are extracted to verify the proposed classification method. Study results show that ①the proposed method can provide the comprehensive driving patterns of vehicle position and speed, and the average accuracy of clustering is 94.8%, which is higher than the accuracy of DBTCAN (89.3%) and t-Cluster (86.4%). ②Based on the analysis of trajectory deviation curve of lane changing, four typical driving patterns are obtained. The proposed method can use multidimensional trajectory data to classify driving patterns, which has potentials in trajectory classification and identifying abnormal behavior.
- intelligent transportation /
- vehicle trajectory /
- multidimensional characteristics /
- track classification /
- Hausdorff distance

HTML全文

图 1 聚合层次算法流程图

Figure 1. Aggregation hierarchy Algorithm

下载: 全尺寸图片幻灯片

图 2 轨迹聚类分析流程图

Figure 2. Flow chart of trajectory clustering analysis

下载: 全尺寸图片幻灯片

图 3 highD数据集道路图

Figure 3. Highd Dataset Road Map

下载: 全尺寸图片幻灯片

图 4 车辆轨迹分布示意图

Figure 4. Schematic diagram of vehicle track distribution

下载: 全尺寸图片幻灯片

图 5 轨迹距离矩阵示意图

Figure 5. Trajectory Distance Matrix

下载: 全尺寸图片幻灯片

图 6 轨迹相似性度量层次聚类树状图

Figure 6. Hierarchical clustering tree for trajectory similarity measurement

下载: 全尺寸图片幻灯片

图 7 轮廓系数随聚类数目的变化

Figure 7. Variation of contour coefficient with the number of cluster

下载: 全尺寸图片幻灯片

图 8 空间轨迹聚类图

Figure 8. Spatial Trajectory Clustering Diagram

下载: 全尺寸图片幻灯片

图 9 不同换道模式下的轨迹偏移率

Figure 9. Track Deviation Rate Under Different Changing Modes

下载: 全尺寸图片幻灯片

图 10 不同换道模式下的轨迹偏移率

Figure 10. Track Deviation Rate Under Different Changing Modes

下载: 全尺寸图片幻灯片

图 11 位置轨迹模式第1类的速度模式聚类结果

Figure 11. Speed pattern clustering of the first type of position trajectory pattern

下载: 全尺寸图片幻灯片

表 1 空间轨迹的聚类结果

Table 1. Clustering results of spatial trajectories

轨迹类别	轨迹总条数	正确分类条数	正确率/%
第一类	136	124	91.2
第二类	61	59	96.7
第三类	33	32	96.9
第四类	202	202	100
第五类	259	238	91.9
总数		691	655
平均正确率/%		94.8

下载: 导出CSV

表 2 各算法的聚类结果

Table 2. Clustering Results of Each Algorithm

方法名称	分类数	轨迹总条数	正确分类条数	正确率/%
本文算法	5	691	655	94.8
DBTCAN	5	691	617	89.3
t-Cluster	3	691	597	86.4

下载: 导出CSV

参考文献(30)

[1]	ZHAO T, XU Y, MONFORT M, et al. Multi-agent tensor fusion for contextual trajectory prediction[C]. The IEEE/CVF Conference on Computer Vision and Pattern Recognition, New York: IEEE, 2019.
[2]	陆德彪, 郭子明, 蔡伯根, 等. 基于深度数据的车辆目标检测与跟踪方法[J]. 交通运输系统工程与信息, 2018, 18(3): 55-62. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201803009.htm LU D B, GUO Z M, CAI B G, et al. A vehicle detection and tracking method based on range data[J]. Journal of Transportation Systems Engineering and Information Technology, 2018, 18(3): 55-62. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201803009.htm
[3]	杨红红, 曲仕茹. 基于压缩感知尺度自适应的多示例交通目标跟踪算法[J]. 中国公路学报, 2018, 31(6): 281-290, 316. doi: 10.3969/j.issn.1001-7372.2018.06.015 YANG H H, QU S R. Traffic target tracking algorithm based on scale adaptive multiple instance learning with compressive sensing[J]. China Journal of Highway and Transport, 2018, 31(6): 281-290, 316. (in Chinese) doi: 10.3969/j.issn.1001-7372.2018.06.015
[4]	冯汝怡, 李志斌, 吴启范, 等. 航拍视频车辆检测目标关联与时空轨迹匹配[J]. 交通信息与安全, 2021, 39(2): 61-69, 77. doi: 10.3963/j.jssn.1674-4861.2021.02.008 FENG R Y, LI Z B, WU Q F, et al. Association of vehicle object detection and the time-space trajectory matching from aerial videos[J]. Journal of Transport Information and Safety, 2021, 39(2): 61-69, 77. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2021.02.008
[5]	CHOI S, KIM J, YEO H. Trajgail: generating urban vehicle trajectories using generative adversarial imitation learning[J]. Transportation Research Part C: Emerging Technologies, 2021(128): 91-113.
[6]	LI X C, ZHAO K Q, C G, et al. Deep representation learning for trajectory similarity computation[C]. 34^th International Conference on Data Engineering (ICDE), Paris, France: IEEE, 2018.
[7]	马文耀, 吴兆麟, 杨家轩, 等. 基于单向距离的谱聚类船舶运动模式辨识[J]. 重庆交通大学学报(自然科学版), 2015, 34 (5): 130-134. https://www.cnki.com.cn/Article/CJFDTOTAL-CQJT201505026.htm MA W Y, WU Z L, YANG J X, et al. Vessel motion pattern recognition based on one-way distance spectral clustering algorithm[J]. Journal of Chongqing Jiaotong University(Natural Science), 2015, 34(5): 130-134. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-CQJT201505026.htm
[8]	王培, 江南, 万幼, 等. 应用Hausdorff距离的时空轨迹相似性度量方法[J]. 计算机辅助设计与图形学学报, 2019, 31 (4): 647-658. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJF201904016.htm WANG P, JIANG N, WAN Y, et al. Measuring similarity of spatio-temporal trajectory using hausdorff distance[J]. Journal of Computer-Aided Design & Computer Graphics, 2019, 31(4): 647-658. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSJF201904016.htm
[9]	CHOONG M Y, ANGELINE L, CHIN R K Y, et al. Modeling of vehicle trajectory clustering based on LCSS for traffic pattern extraction[C]. 2^nd International Conference on Automatic Control and Intelligent Systems, Kota Kinabalu, Malaysia: IEEE, 2017.
[10]	李颖, 赵莉, 赵祥模, 等. 基于大货车GPS数据的轨迹相似性度量有效性研究[J]. 中国公路学报, 2020, 33(2): 146-157. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202002014.htm LI Y, ZHAO L, ZHAO X M, et al. Effectiveness of trajectory similarity measures based on truck GPS data[J]. China Journal of Highway and Transport, 2020, 33(2): 146-157. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL202002014.htm
[11]	ANDRE S F, DESPINA K, LUIS O A, et al. Multidimensional similarity measuring for semantic trajectories[J]. Transactions in Gis, 2016, 20(2): 280-298. doi: 10.1111/tgis.12156
[12]	FHWA, Department of Transportation, America. NGSIM: Next generation simulation[EB/OL]. (2007-5-5)[2023-02-15]. http://www.ngsim-com-munity.org/.
[13]	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]. 21^st International Conference on Intelligent Transportation Systems(ITSC), Hawaii: IEEE, 2018.
[14]	WANG J H, FU T, XUE J T, et al. Realtime wide-area vehicle trajectory tracking using millimeter-wave radar sensors and the open TJRD TS dataset, [EB/OL]. (2021)[2023-02-15]. https://www.tjrdts.com.
[15]	潘晓, 马昂, 郭景峰, 等. 基于时间序列的轨迹数据相似性度量方法研究及应用综述[J]. 燕山大学学报, 2019, 43 (6): 531-545. PAN X, MA A, GUO J F, et al. A survey of research and application of track data similarity measurement based on time series[J]. Journal of Yanshan University, 2019, 43(6): 531-545. (in Chinese)
[16]	丁华, 杨文杰, 姜超. 考虑轨迹分析的车辆异常行为辨识[J]. 重庆理工大学学报(自然科学版), 2022, 36(7): 62-69. DING H, YANG W J, JIANG C. Vehicle abnormal behavior identification considering trajectory analysis[J]. Journal of Chongqing University of Technology(Natural Science), 2022, 36(7): 63-69. (in Chinese)
[17]	LI H P. Typical trajectory extraction method for ships based on AIS data and trajectory clustering[C]. 2^nd International Conference on Artificial Intelligence and Information Systems, chongqing: ACM, 2021.
[18]	杨家轩, 刘元. 基于DBTCAN算法的船舶轨迹聚类与航路识别[J]. 上海海事大学学报, 2022, 43(3): 7-12. YANG J X, LIU Y. Ship trajectory clustering and route recognition based on DBTCAN algorithm[J]. Journal of Shanghai Maritime University, 2022, 43(3): 7-12. (in Chinese)
[19]	姜乔文, 刘瑜, 潭大宁, 等. 时空轨迹多维特征融合的行为规律挖掘算法[J/OL]. (2021-11-23)[2023-02-15]. https://kns.cnki.net/kcms/detail/11.1929.v.20211123.1336.014.html JIANG Q W, LIU Y, TAN D N, et al. Behavior rule mining algorithm based on multi-dimensional feature fusion of spatio-temporal trajectory[J/OL]. (2021-11-23)[2023-02-15]. https://kns.cnki.net/kcms/detail/11.1929. v. 20211123.1336.014. html. (in Chinese)
[20]	ARIF A S, DEBI P D, SAMARJIT K, et al. Video trajectory analysis using unsupervised clustering and multi-criteria ranking[J]. Soft Computing, 2020, 24(21): 16643-16654.
[21]	ZHAO P X, LIU X T, SHEN J W, et al. A network distance and graph-partitioning-based clustering method for improving the accuracy of urban hotspot detection[J]. Geocarto International, 2019, 34(3): 293-315.
[22]	SHI Y, WANG D, TANG J B, et al. Detecting spatiotemporal extents of traffic congestion: A density-based moving object clustering approach[J]. International Journal of Geographical Information Science, 2021, 35(7): 1449-1473.
[23]	YUE H Q, GUAN Q F, PAN Y T, et al. Detecting clusters over intercity transportation networks using k-shortest paths and hierarchical clustering: a case study of mainland china[J]. International Journal of Geographical Information Science, 2019, 33(5): 1082-1105.
[24]	赵怀鑫, 邓然然, 张英杰, 等. 1种用于高速公路通行情况分析的收费数据挖掘方法[J]. 中国公路学报, 2018, 31(8): 155-164. ZHAO H X, DENG R R, ZHANG Y J, et al. Method of mining fee data for expressway traffic analysis[J]. China Journal of Highway and Transport, 2018, 31(8): 155-164. (in Chinese)
[25]	RODRIGUEZ M, COMIN C, CASANOVA D, et al. Clustering algorithms: a comparative approach[J]. PLoS ONE, 2019, 14(1): 107-141.
[26]	王灵丽, 黄敏, 高亮. 基于聚类算法的交通网络节点重要性评价方法研究[J]. 交通信息与安全, 2020, 38(2): 80-88. doi: 10.3963/j.jssn.1674-4861.2020.02.010 WANG L L, HUANG M, GAO L. Methods of importance evaluation of traffic network node based on clustering algorithms[J]. Journal of Transport Information and Safety, 2020, 38(2): 80-88. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2020.02.010
[27]	YANG J X, LIU Y, MA L Q, et al. Maritime traffic flow clustering analysis by density based trajectory clustering with noise[J]. Ocean Engineering, 2022(249): 111001.
[28]	JIANG R Q, CHEN L L. Driving stress estimation in physiological signals based on hierarchical clustering and multi-view intact space learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(8): 13141-13154.
[29]	徐进, 陈莹, 陈海源, 等. 回头曲线路段的轨迹行为模式与事故风险[J]. 东南大学学报(自然科学版), 2020, 50(5): 973-982. https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX202005025.htm XU J, CHEN Y, CHEN H Y, et al. Vehicle track patterns and accident risk on hairpin curves[J]. Journal of Southeast University(Natural Science Edition), 2020, 50(5): 973-982. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX202005025.htm
[30]	PU Z Y, CUI Z Y, TANG J J, et al. Multimodal traffic speed monitoring: A real-time system based on passive wi-fi and bluetooth sensing technology[J]. IEEE Internet of Things Journal, 2022, 9(14): 12413-12424.