基于时空特征序列匹配的交通流状态估计方法

陈佳良; 胡钊政; 李飞

doi:10.3963/j.jssn.1674-4861.2021.03.009

基于时空特征序列匹配的交通流状态估计方法

doi: 10.3963/j.jssn.1674-4861.2021.03.009

武汉理工大学智能交通系统研究中心武汉 430063

基金项目:

国家重点研发计划项目 2018YFB1600801

详细信息

作者简介:
陈佳良（1996—），硕士研究生.研究方向：智能交通、交通数据分析.E-mail：chenjl@whut.edu.cn

通讯作者:
胡钊政（1979—），博士，教授.研究方向：智能交通系统、智能车路系统、交通数据分析.E-mail：zzhu@whut.edu.cn

中图分类号: U491.2
计量
- 文章访问数: 490
- HTML全文浏览量: 250
- PDF下载量: 33
- 被引次数: 0
出版历程
- 收稿日期: 2021-01-25

An Estimation Method of Traffic Flow State Based on Matching of Temporal-spatial Feature Sequences

Intelligent Transportation Systems Research Center, Wuhan University of Technology, Wuhan 430063, China

摘要

摘要: 为了针对无交通流检测器路段更好地进行交通流状态估计，提高估计精度，研究了基于时空特征序列匹配的交通流状态估计模型。通过交通运行指数的计算方法预设城市道路中有交通流参数路段的交通流状态；分析影响城市道路运行条件的各项因素，引入交通流参数与道路参数、路网拓扑参数等时空多维度参数特征，提取3个维度8个特征1个附加维度组成交通流时空特征，构建城市道路交通流DNA特征序列对交通流状态进行描述；将各个特征的值归一化处理，利用WH-KNN匹配方法，得到全路网中与待估计路段最近的交通流状态。实验选取武汉市中环快速路编号为10468、10483以及8816的路段1周数据，假定路段数据缺失，通过所述方法进行交通流状态估计，将估计结果与原始数据结果进行对比。研究表明，模型不仅能够得到无检测数据路段的交通流状态，其状态估计结果的准确率保持在88%以上，且误判结果在1个运行指数等级之内。
- 智能交通 /
- 交通流状态 /
- 时空特征序列 /
- 特征匹配 /
- 交通流DNA /
- WH-KNN
Abstract: An estimation model of the traffic flow state based on matching of temporal-spatial feature sequences is studied to better estimate the traffic flow state for the road section without a traffic flow detector and improve the estimation accuracy.The model firstly uses the calculation method of the traffic-operation index to preset the traffic-flow state of the urban-road section with traffic flow data.Various factors affecting the operating conditions of urban roads are analyzed, with the introduction of the characteristics of time and space multi-dimensional parameters such as traffic flow parameters, road parameters, and road network topology parameters.The temporal and spatial characteristics of traffic flow form by extracting 3 dimensions, 8 features, and 1 additional dimension, thus constructing the DNA feature sequence of urban-road traffic flow.After normalizing the value of each feature, the WH-KNN matching method is used to obtain the traffic-flow state closest to the road section to be estimated in the whole road network.The experiment selects the data of one week in road sections 10468, 10483, and 8816 of Wuhan Zhonghuan Expressway.Assuming that the road section data is missing, the traffic flow state is estimated by the method described above, and the estimated results are compared with the original data results.The results shows that the model can obtain the traffic flow state of the road section without detection data as well as maintain the accuracy rate of the state estimation result above 88%. The misjudgment is within a performance-index level.
- Intelligent transportation /
- traffic-flow state /
- temporal-spatial feature sequence /
- feature matching /
- traffic flow DNA /
- WH-KNN

HTML全文

图 1 技术路线图

Figure 1. Technology roadmap

下载: 全尺寸图片幻灯片

图 2 城市道路网络时空特征序列划分

Figure 2. Division of temporal-spatial characteristic sequences of urban road network

下载: 全尺寸图片幻灯片

图 3 城市道路交通流DNA序列

Figure 3. DNA sequence of urban road traffic flow

下载: 全尺寸图片幻灯片

图 4 路段向量与特征空间示意图

Figure 4. Link vector and feature space

下载: 全尺寸图片幻灯片

图 5 Link 10483流量-时间曲线

Figure 5. Flow-time curve of Link10483

下载: 全尺寸图片幻灯片

图 6 Link 10483及其上下游直行路段典型流量关系

Figure 6. Typical-flow relationship between Link10483 and its upstream and downstream straight sections

下载: 全尺寸图片幻灯片

图 7 Link 10483典型工作日与非工作日1 d交通流状态

Figure 7. One-day traffic flow status in typical working day and non-working day of Link 10483

下载: 全尺寸图片幻灯片

图 8 路网交通流状态匹配示意图

Figure 8. Matching of the road-network traffic-flow state

下载: 全尺寸图片幻灯片

图 9 Link 10483的1周匹配结果与正确率对比

Figure 9. Comparison of one-week matching result and correct rates of Link 10483

下载: 全尺寸图片幻灯片

表 1 道路交通运行指数分级

Table 1. Classification of traffic circulation indices

拥堵等级	畅通	基本畅通	缓行	轻度拥堵	拥堵
指数范围	0~1	1~2	2~3	3~4	4~5

下载: 导出CSV

表 2 特征结构及可能性数量表

Table 2. Feature structure and possibility

特征	取值	对应拓扑结构	对应数量	子特征可能性	组合可能性
邻接层级	1		1	1	1
邻接路段数	0, 1, 2, 3, 4	上游	0，1，2，3，4	5	5²
邻接路段数	0, 1, 2, 3, 4	下游	0，1，2，3，4	5	5²
道路等级	高速路、快速路、主干道、次干道、支路	上游	1，2，3，4，5	5⁵	5¹¹
		中游	1，2，3，4，5	5
		下游	1，2，3，4，5	5⁵
路段车道数	0，1，2，3，4，5	上游	0，1，2，3，4，5	6⁵	6¹¹
		中游	0，1，2，3，4，5	6
		下游	0，1，2，3，4，5	6⁵
路段交通状态	畅通、基本畅通、缓行、轻度拥堵、拥堵	上游	1，2，3，4，5	5⁵	5¹¹
		中游	1，2，3，4，5	5
		下游	1，2，3，4，5	5⁵

下载: 导出CSV

表 3 时空特征归一化表

Table 3. Normalization of spatio-temporal features

特征	取值	对应拓扑结构	归一化取值
数据时间	—		—
邻接层级	1		1
邻接路段数	0，1，2，3，4	上游	0，0.25，0.5，0.75，1
邻接路段数	0，1，2，3，4	下游	0，0.25，0.5，0.75，1
道路等级	高速路、快速路、主干道、次干道、支路	上游	0，0.25，0.5，0.75，1
		中游
		下游
路段长度	—	上游	—
		中游
		下游
路段车道数	0，1，2，3，4，5	上游	0，0.2，0.4，0.6，0.8，1
		中游
		下游
路段流量	—	上游	以0为区间下限，最大通行能力为区间上限，采用线性方法归一化
		中游
		下游
路段平均速度	—	上游	以0为区间下限，道路限速为区间上限，采用线性方法归一化
		中游
		下游
路段交通状态	畅通、基本畅通、缓行、轻度拥堵、拥堵	上游	0，0.25，0.5，0.75，1
		中游
		下游

下载: 导出CSV

表 4 车道数修正系数取值表

Table 4. Correction factors of lane numbers

车道数	修正系数
1	1
2	1.87
3	2.6
4	3.2

下载: 导出CSV

表 5 BPR函数参数及道路通行能力值

Table 5. BPR function parameters and road-capacity values

道路等级	设计速度/（km/h）	C₀/（pcu/h）	γ	η	μ	n'	α	β
高速路	80	1500	1	1	1		0.115	1.156
快速路	70	1500	1	1	1	见车	0.59	1.921
主干道	60	1500	1	0.75	0.75	道数	0.71	1.504
次干道	50	1500	0.8	0.75	0.75	修正	0.65	1.763
支路	40	1500	0.5	0.75	0.5	系数表	0.59	1.921
快速路辅路	40	1500	0.5	0.75	0.5		0.118	1.038

下载: 导出CSV

参考文献(23)

[1]	吕北岳. 基于浮动车的深圳市道路交通运行评价研究[D]. 武汉: 武汉大学, 2013. LYU Beiyue. Research on evaluation of road traffic operations based on floating car for Shenzhen[D]. Wuhan: Wuhan University, 2013. (in Chinese)
[2]	WANG Jingfeng, WANG Chao, LYU Jiarun, et al. Modeling travel time reliability of road network considering connected vehicle guidance characteristics indexes[J]. Journal of Advanced Transportation, 2017(2017): 1-9. http://www.researchgate.net/publication/315922627_Modeling_Travel_Time_Reliability_of_Road_Network_Considering_Connected_Vehicle_Guidance_Characteristics_Indexes
[3]	韦伟, 毛保华, 陈绍宽, 等. 基于时空自相关的道路交通状态聚类方法[J]. 交通运输系统工程与信息, 2016, 16(2): 57-63. doi: 10.3969/j.issn.1009-6744.2016.02.011 WEI Wei, MAO Baohua, CHEN Shaokuan, et al. Urban traffic status clustering method based on spatio-temporal autocorrelation[J]. Journal of Transportation Systems Engineering and Information Technology, 2016, 16(2): 57-63. (in Chinese) doi: 10.3969/j.issn.1009-6744.2016.02.011
[4]	张婧. 城市道路交通拥堵判别、疏导与仿真[D]. 南京: 东南大学, 2016. ZHANG Jing. The study on identification, dispersion and simulation of urban traffic congestion[D]. Nanjing: Southeast university, 2016. (in Chinese)
[5]	庄广新. 基于多源检测器的交通流数据融合方法研究[D]. 北京: 北京交通大学, 2017. ZHUANG Guangxin. Research on the traffic information fusion with the multi-source detectors[D]. Beijing: Beijing Jiaotong University, 2017. (in Chinese)
[6]	WANG Xiangxue, XU Lunhui, CHEN Kaixun. Data-driven short-term forecasting for urban road network traffic based on data processing and LSTM-RNN[J]. Arabian Journal for Science and Engineering, 2019, 44(4): 3043-3060. http://www.zhangqiaokeyan.com/academic-journal-foreign_other_thesis/0204112953625.html
[7]	崔玮. 高速路网交通状态判别与预测的研究[D]. 淄博: 山东理工大学, 2016. CUI Wei. Research on traffic state identification and prediction of highway network[D]. Zibo: Shandong University of Technology, 2016. (in Chinese)
[8]	XU Dongwei, WANG Yongdong, JIA Limin, et al. Real-time road traffic states measurement based on Kernel-KNN matching of regional traffic attractors[J]. Measurement, 2016(94): 862-872. http://www.sciencedirect.com/science/article/pii/S0263224116304985
[9]	YANG Senyan, WU Jianping, QI Geqi, et al. Analysis of traffic state variation patterns for urban road network based on spectral clustering[J]. Advances in Mechanical Engineering, 2017(9): 1624-1639. http://www.researchgate.net/publication/319975367_Analysis_of_traffic_state_variation_patterns_for_urban_road_network_based_on_spectral_clustering
[10]	商强, 林赐云, 杨兆升, 等. 基于谱聚类与RS-KNN的城市快速路交通状态判别[J]. 华南理工大学学报(自然科学版), 2017, 45(6): 52-58. doi: 10.3969/j.issn.1000-565X.2017.06.009 SHANG Qiang, LIN Ciyun, YANG Zhaosheng, et al. Traffic state identification for urban expressway based on spectral clustering and RS-KNN[J]. Journal of South China University of Technology(Natural Science Edition), 2017, 45(6): 52-58. (in Chinese) doi: 10.3969/j.issn.1000-565X.2017.06.009
[11]	LIN Xiaohui. A road Network traffic state identification method based on macroscopic fundamental diagram and spectral clustering and support vector machine[J]. Mathematical Problems in Engineering, 2019(4): 1-10. http://www.researchgate.net/publication/332562853_A_Road_Network_Traffic_State_Identification_Method_Based_on_Macroscopic_Fundamental_Diagram_and_Spectral_Clustering_and_Support_Vector_Machine
[12]	DAI Yihong, LU Weike, HUANG Hao, et al. Threshold division of urban road network traffic state based on macroscopic fundamental diagram and k-means clustering[C]. 2019 International Conference on Transportation Engineering, Chengdu: Southwest Jiaotong University, 2020.
[13]	马勇. 城市快速路实时交通状态估计方法研究[D]. 北京: 北京工业大学, 2014. MA Yong. Real-time traffic state estimation method for urban expressway[D]. Beijing: Beijing University of Technology, 2014. (in Chinese)
[14]	ZHAO Shuxu, ZHANG Baohua. Traffic flow prediction of urban road network based on LSTM-RF model[J]. Journal of Measurement Science and Instrumentation, 2020, 11(2): 135-142.
[15]	黄振盛, 汪玉美, 韩江洪, 等. 基于MLS-SVM和时空特性的短时交通流量预测方法[J]. 合肥工业大学学报(自然科学版), 2020, 43(1): 57-63. https://www.cnki.com.cn/Article/CJFDTOTAL-HEFE202001010.htm HUANG Zhensheng, WANG Yumei, HAN Jianhong, et al. Short-term urban traffic flow prediction based on MLS-SVMand spatiotemporal features[J]. Journal of Hefei University of Technology(Natural Science), 2020, 43(1): 57-63. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HEFE202001010.htm
[16]	唐智慧, 郑伟皓, 董维, 等. 基于交互式BP-UKF模型的短时交通流预测方法[J]. 公路交通科技, 2019, 36(4): 117-124+134. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK201904017.htm TANG Zhihui, ZHENG Weihao, DONG Wei, et al. A method for predicting short-term traffic flow based on interactive IMM-BP-UKF model[J]. Journal of Highway and Transportation Research and Development, 2019, 36(4): 117-124+134. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK201904017.htm
[17]	常丽君, 郑黎黎, 杨帆. 基于(SAGA-FCM)-PNN的交通状态判别方法研究[J]. 交通信息与安全, 2019, 37(2): 120-127. https://www.cnki.com.cn/Article/CJFDTOTAL-JTJS201902017.htm CHANG Lijun, ZHENG Lili, YANG Fan. A method of discrimination for traffic state based on(SAGA-FCM)-PNN[J]. Journal of Transport Information and Safety, 2019, 37(2): 120-127. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JTJS201902017.htm
[18]	FOWE A J, CHAN Y P. A microstate spatial-inference model for network-traffic estimation[J]. Transportation Research Part C: Emerging Technologies, 2013(36): 245-260. http://www.sciencedirect.com/science/article/pii/S0968090X13001757
[19]	张婧, 任刚. 城市道路交通拥堵状态时空相关性分析[J]. 交通运输系统工程与信息, 2015, 15(2): 175-181. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201502027.htm ZHANG Jing, REN Gang. Spatio-temporal correlation analysis of urban traffic congestion diffusion[J]. Journal of Transportation Systems Engineering and Information Technology, 2015, 15(2): 175-181. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201502027.htm
[20]	MA Shiyong, AN Shi, YU Hang. Urban traffic congestion discrimination algorithm based on the ordered decision theory[J]. Advances in Information Sciences & Service Sciences, 2012, 4(23): 814-820. http://www.researchgate.net/publication/276004646_Urban_Traffic_Congestion_Discrimination_Algorithm_Based_on_the_Ordered_Decision_Theory
[21]	LIU Zhe, ZHOU Shunbo, SUO Chuanzhe, et al. LPD-Net: 3D point cloud learning for large-scale place recognition and environment analysis[C]. 2019 IEEE/CVF International Conference on Computer Vision, Seoul, Korea: IEEE, 2019.
[22]	刘国忠, 胡钊政. 基于SURF和ORB全局特征的快速闭环检测[J]. 机器人, 2017, 39(1): 36-45. https://www.cnki.com.cn/Article/CJFDTOTAL-JQRR201701005.htm LIU Guozhong, HU Zhaozheng. Fast loop closure detection based on holistic features from SURF and ORB[J]. Robot, 2017, 39(1): 36-45. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JQRR201701005.htm
[23]	张巧. 混合交通流条件下城市路段BPR函数参数标定研究[D]. 长沙: 中南大学, 2013. ZHANG Qiao. Calibrating BPR function under urban mixed traffic flow condition[D]. Changsha: Central South University, 2013. (in Chinese)