Analysis of Crossing Behavior of Non-Motor Vehicle at Overlap Phase Signal Intersections

doi:10.12141/j.issn.1000-565X.220779

Abstract

Abstract:

In order to improve the safety level of non-motor vehicle traffic at signalized intersections, based on the survey data of signalized intersections with overlapping phase control in Guangzhou, this study analyzed the influencing factors of non-motor vehicle crossing behavior based on the C5.0 decision tree algorithm. Considering the influence of different periods on the crossing behavior of non-motor vehicles in the signal cycle, the study divided a complete signal cycle into four risk periods according to the risk conflict of non-motorized vehicles crossing the street, namely, the opposite green light risk period, the same direction green light safety period, the same direction green light risk period and the vertical direction risk period. And it divided the crossing behavior into three categories according to the waiting selection of non-motor vehicles at the intersection and whether or not to run red-light, namely, risky, opportunistic and law-abiding. It studied the influencing factors of the three types of crossing behavior by constructing a C5.0 decision tree model and analyzed and evaluated the classification effect of the model. The results show that the overall accuracy of the model classification results is greater than 83.04%, the AUC is greater than 0.880, and the model prediction accuracy is good. The crossing behavior of non-motorized vehicles at signalized intersections with overlapping phase control is mainly significantly related to the traffic environment, while the factors related to the rider’s behavior are less significant. The arrival risk period, non-motor vehicle signal light facilities, conflicting motor traffic flow, number of lanes and crossing risk have significant impacts on the occurrence of risk-taking crossing behavior, among which the arrival risk period is the most important influencing factor. The number of lanes, red-light time and arrival risk period have significant impacts on the occurrence of opportunistic crossing behavior, among which the number of lanes is the most important influencing factor. The conflicting motor traffic flow flow, signal period, number of lanes, crossing area and crossing risk have significant impacts on the occurrence of law-obeying crossing behavior, among which the conflicting motor traffic flow is the most important influencing factor.

Key words: non-motor vehicle, crossing behavior, analysis of influencing factor, overlap phase, C5.0 decision tree

CLC Number:

U491

WEN Huiying, LIU Hao, DU Yingxin, et al. Analysis of Crossing Behavior of Non-Motor Vehicle at Overlap Phase Signal Intersections[J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(8): 1-11.

Figures/Tables 10

Fig.1

Table 1

Fig.2

Table 2

Fig.3

Table 3

Fig.4

Fig.5

Fig.6

Fig.7

References 15

1	中华人民共和国国家统计局．国家数据-2021年度数据［EB/OL］．［2022-11-20］．.
2	张凡．信号交叉口外卖电动车违规过街行为研究［D］．西安：长安大学，2020.
3	FRABONI F， PUCHADES V M， de ANGELIS M，et al. Red-light running behavior of cyclists in Italy：an observational study［J］．Accident Analysis & Prevention，2018，120（11）：219-232.
4	WU C， YAO L， ZHANG K ．The red-light running behavior of electric bike riders and cyclists at urban intersections in China：an observational study［J］．Accident Analysis & Prevention，2012，49（11）：186-192.
5	高纯．信号交叉口电动自行车闯红灯行为影响因素研究［D］．北京：中国人民公安大学，2020.
6	BAI L，SZE N ．Red light running behavior of bicyclists in urban area：effects of bicycle type and bicycle group size［J］．Travel Behaviour & Society，2020，21（10）：226-234.
7	PAI C，JOU R ．Cyclists’ red-light running behaviours：an examination of risk-taking，opportunistic，and law-obeying behaviours［J］．Accident Analysis & Prevention，2014，62（1）：191-198.
8	刘微．多相位交叉口行人和非机动车的违规行为分析［D］．北京：北京交通大学，2020.
9	王进，魏星，刘玮．基于等饱和度的交叉口信号搭接相位设计［J］．交通与运输，2020，36（1）：16-20.
	WANG Jin， WEI Xing， LIU Wei ．Overlap phase design for equal saturation［J］．Traffic & Transportation，2020，36（1）：16-20.
10	苏灿航，高圣涵，郭建钢．基于潮汐交通流特性的交叉口信号配时研究［J］．哈尔滨商业大学学报（自然科学版），2022，38（2）：186-191.
	SU Canhang， GAO Shenghan， GUO Jiangang ．Research on intersection signal timing based on characteristics of tidal traffic flow［J］．Journal of Harbin University of Commerce （Natural Sciences Edition），2022，38（2）：186-191.
11	WANG W， YUAN Z， LIU Y，et al ．A random parameter logit model of immediate red-light running behavior of pedestrians and cyclists at major-major intersections［J］．Journal of Advanced Transportation，2019，2019（7）：2345903/1-13.
12	李行，杨小宝，郑天宇．交通信号灯转换期间的违规穿越行为及其影响因素分析［J］．综合运输，2019，41（1）：59-64.
	LI Hang， YANG Xiaobao， ZHENG Tianyu ．Crossing behavior and influencing factors at the conversion periods of traffic signal［J］．China Transportation Review，2019，41（1）：59-64.
13	胡小霞．多模式交通出行动态决策行为研究［D］．西安：长安大学，2019.
14	FAWCETT T ．An introduction to ROC analysis［J］．Pattern Recognition Letters，2006，27（8）：861-874.
15	GAJOWNICZEK K， ZĄBKOWSKI T ．ImbTreeAUC：an R package for building classification trees using the area under the ROC curve （AUC） on imbalanced datasets［J］．SoftwareX，2021，15（7）：100755/1-7.

过街行为类型	是否等待	是否闯红灯	过街方式	过街特点
冒险型	否	是	先减速观察，再加速穿越	速度快、耗时短
机会型	是	是	先停车等待，再伺机穿越	速度快、耗时短
守法型	是	否	停车等待直至绿灯再穿越	速度慢、耗时长

属性	特征因素	特征变量及其编码	频数	百分比/%
个体属性	性别	0=女性	2 426	22.60
	性别	1=男性	8 307	77.40
	年龄	0=18岁及以下	93	0.87
		1=19~35岁	2 956	27.54
		2=36岁及以上	7 684	71.59
	车辆类型	0=个人所有的普通自行车	452	4.21
		1=个人租赁的共享单车	1 994	18.58
		2=两轮式电动自行车	8 055	75.05
		3=其他类型非机动车	232	2.16
	职业身份	0=非专职骑手	8 408	78.34
	职业身份	1=专职骑手	2 325	21.66
出行行为	头盔佩戴	0=骑行者与乘客均有佩戴头盔	7 653	71.30
		1=骑行者与乘客均无佩戴头盔	2 953	27.51
		2=骑行者与乘客有一方未佩戴头盔	127	1.18
	是否载人	0=不载人	9 975	9294
	是否载人	1=载人	758	7.06
	是否拉货	0=不拉货	7 992	74.46
	是否拉货	1=拉货	2 741	25.54
	过街区域	0=非机动车道	7 732	72.04
		1=机动车道	2 347	21.87
		2=非机动车道与机动车道混合使用	654	6.09
	是否跟随过街	0=不跟随	5 938	55.32
	是否跟随过街	1=跟随	4 795	44.68
交通环境	到达时期	0=红灯初期	885	8.25
		1=红灯稳定期	9 519	88.69
		2=红灯末期	329	3.07
	到达风险期	0=对向绿灯风险期	2 229	20.77
		1=同向绿灯风险期	3 404	31.72
		2=垂直方向风险期	5 100	47.51
	过街风险	0=2 s内无冲突车辆到达	9 079	84.59
	过街风险	1=2 s内有冲突车辆到达	1 654	15.41
	车道数	1=双向四车道及以下	6 484	60.41
	车道数	2=双向车道数大于4	4 249	39.59
	冲突机动车流量	1=30辆及以下	5 289	49.28
		2=31~60辆	3 270	30.47
		3=61辆及以上	2 174	20.26
	非机动车信号灯	0=未设置	6 193	57.70
	非机动车信号灯	1=设置	4 540	42.30
	信号周期	1=150 s及以下	7 254	67.59
	信号周期	2=大于150 s	3 479	32.41
	红灯时间	1=90 s及以下	1 423	13.26
		2=91~120 s	5 388	50.20
		3=121 s及以上	3 922	36.54

过街行为类型	准确率/%	精确率/%		召回率/%
过街行为类型	准确率/%	0	1	0	1
冒险型	84.29	86.54	81.56	85.00	83.39
机会型	83.04	92.14	62.99	84.58	78.44
守法型	90.43	90.33	90.75	96.55	76.57

[1]	. Review of the Impact of Spatiotemporal Constraints on Active Travel [J]. Journal of South China University of Technology(Natural Science Edition), 2024, 52(2): 113-123.
[2]	CAO Ningbo, ZHAO Liying. Management Method Based on Maximum Pressure Control of Autonomous Vehicles and Pedestrians at Intersections [J]. Journal of South China University of Technology(Natural Science Edition), 2024, 52(2): 124-135.
[3]	ZHAO Xiaomei, HAO Guoyu, NIU Xiaojing, ZHOU Zhiqian. Traffic Boundary Control Strategy Based on Macroscopic Fundamental Diagram Under Different Rainfall [J]. Journal of South China University of Technology(Natural Science Edition), 2024, 52(1): 72-82.
[4]	ZHUANG Yan, DONG Chunjiao, MI Xueyu, WANG Jing, YANG Miaoyan. Identification of Accident Black Spots Based on Improved Network Kernel Density and Negative Binomial Regression [J]. Journal of South China University of Technology(Natural Science Edition), 2024, 52(1): 119-126.
[5]	DONG Chunjiao, LU Yuxiao, MA Sheqiang, LI Penghui, ZHUANG Yan. Analyzing and Modeling of Multi-Class E-Bikes Violation Behaviors at Signalized Intersection [J]. Journal of South China University of Technology(Natural Science Edition), 2024, 52(1): 83-89.
[6]	ZHANG Jianhua, ZHAO Xiaohua, OU Jushang, LUAN Sen, DAI Yibo. Evaluation of Utility and Optimal of Variable Speed Limit Value for Bridge in Foggy Condition [J]. Journal of South China University of Technology(Natural Science Edition), 2024, 52(1): 127-138.
[7]	ZHANG Wenhui, SU Jiaqi, HA Zihong, et al. Location and Capacity Optimization Model of Battery-Swapped Electric Bus Charging Station [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 126-134.
[8]	HU Xinghua, CHEN Xinghui, WANG Ran, et al. Optimization Model of Bus Priority Control Considering Carbon Emissions with Stochastic Characteristics [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 160-170.
[9]	XU Lunhui, YU Jiaxin, PEI Mingyang, et al. Repositioning Strategy for Ride-Hailing Vehicles Based on Geometric Road Network Structure and Reinforcement Learning [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 99-109.
[10]	LIN Yongjie, CHEN Ning, LU Kai. Vehicle Trajectory Tracking at Intersections Based on Millimeter Wave Radar Point Cloud [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 110-125.
[11]	BIE Yiming, ZHU Aoze, CONG Yuan. Electric Bus Scheduling Method Considering Differences in the State of Health of Batteries [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 11-21.
[12]	YANG Min, CHEN Shantao, JIANG Ruiyu, et al. Flexible Bus Scheduling Optimization for Integrated Hub Connections in the Context of MaaS [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 22-30.
[13]	PEI Mingyang, ZHU Hongyu. Modeling Method for Optimizing Dynamic Wireless Charging Lane of Electric Vehicles [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 135-151.
[14]	XU Zhihang, YAO Xinpeng, XU Zhigang, et al. Review of Research on Road Traffic Detectors and Its Optimized Deployment Methods [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(10): 68-88.
[15]	ZHAO Xiaomei, ZHU Xiangyuan, WANG Qin, et al. Operational Reliability Optimization Strategies of Multi-type Bus Lines [J]. Journal of South China University of Technology(Natural Science Edition), 2023, 51(8): 32-39.