信号交叉口左转非机动车行为特性及安全性分析

doi:10.16265/j.cnki.issn1003-3033.2026.07.1629

摘要/Abstract

摘要：

为探究信号交叉口左转非机动车行为特性及安全性影响因素,采集武汉市3个信号交叉口的视频数据,利用Tracker软件提取左转非机动车运动信息和交通冲突数据;对比分析3种左转方式下的轨迹、速度和加速度差异;基于后侵入时间(PET)和冲突当量速度,采用K-means聚类算法划分交通冲突严重程度;从车辆、道路和环境特征中选取10个潜在影响因素,构建Ordered Probit模型,识别左转非机动车交通冲突严重程度的显著影响因素,并通过边际效应定量分析各显著因素的影响程度。结果表明:3种左转方式的非机动车行为特性差异显著;非机动车左转方式、左转时期、机动车运动方向、机动车类型、非机动车类型、机非通过情况和左直比为交通冲突严重程度的显著影响因素;绿灯后期对严重冲突正向影响程度最大(边际效应为9.81%),逆时针二次过街反向影响程度最大(边际效应为-17.41%)。

关键词: 信号交叉口, 左转非机动车, 行为特性, 安全性, 交通冲突, 边际效应

Abstract:

To investigate the behavioral characteristics of left-turning non-motorized vehicles under signalized intersections and the influence of various factors on the safety of left-turning non-motorized vehicles, firstly, video data from three signalized intersections in Wuhan were collected. Left-turning non-motorized vehicles motion information and traffic conflict data were extracted by using Tracker software. Then, the variability of three different left-turning modes was analyzed from the distribution of non-motorized vehicle trajectories, speeds, and accelerations. Based on the extracted traffic conflict indicators post-encroachment time (PET) and conflict equivalent speed (V), the traffic conflict severity was classified into 3 categories by using the K-means clustering algorithm. Finally, 10 potential influencing factors were selected as variables from vehicle, road and environmental characteristics to construct an Ordered Probit model to find out the significant influences of the left-turning non-motorized vehicle traffic conflict severity factors. The influence degree of each significant factor was quantitatively analyzed by marginal effect. The results of the study show that: there are large differences in the behavioral characteristics of non-motorized vehicles in three different left-turning modes. The non-motorized vehicle left-turning mode, left-turning period, motorized vehicle motion direction, motorized vehicle type, non-motorized vehicle type, vehicle passing order, and the left-to-straight ratio are the significant factors for the severity of the traffic conflict. Among them, the factor with the greatest positive influence on the severity of the conflict is the late green phase (marginal effect 9.81%). The factor with the greatest negative influence is counterclockwise second crossing (marginal effect -17.41%).

Key words: signalized intersections, left-turning non-motorized vehicles, behavioral characteristics, safety, traffic conflicts, marginal effects

中图分类号:

X951交通运输安全

李涛, 张存保, 符鼎俊, 高天昊, 张梦岩. 信号交叉口左转非机动车行为特性及安全性分析[J]. 中国安全科学学报, 2026, 36(7): 225-233.

Li Tao, Zhang Cunbao, Fu Dingjun, Gao Tianhao, Zhang Mengyan. Analysis of behavioral characteristics and safety of left-turning non-motorized vehicles at signalized intersections[J]. China Safety Science Journal, 2026, 36(7): 225-233.

图/表 13

图1

图2

图3

图4

图5

图6

表1

表2

图7

图8

表3

表4

表5

参考文献 18

[1]	国家数据-2024年度数据[EB/OL]. [2025-06-20]. https://data.stats.gov.cn/dg/website/page.html#/pc/national/search?s=%E4%BA%A4%E9%80%9A%E4%BA%8B%E6%95%85.
[2]	葛婷, 胡俊艳. 无信号控制路段电动自行车过街特性[J]. 科学技术与工程, 2023, 23(23):10120-10 126.
	Ge Ting, Hu Junyan. Crossing characteristics of electric bicycle in unsignal control section[J]. Science Technology and Engineering, 2023, 23(23): 10 120-10 126.
[3]	后旗旸. 基于骑行实验的自行车微观行为研究[D]. 北京: 北京交通大学, 2019.
	Hou Qiyang. Study on micro-behavior of cyclist under experiment condition[D]. Beijing: Beijing Jiaotong University, 2019.
[4]	周旦. 城市基本路段混合非机动车交通流特性研究[D]. 杭州: 浙江大学, 2016.
	Zhou Dan. Study on characterisitcs of mixed bicycle traffic flow in basic sections of urban road[D]. Hangzhou: Zhejiang University, 2016.
[5]	潘晓东, 马小翔, 赵晓翠. 信号交叉口非机动车骑行特性及安全性实验研究[J]. 交通科学与工程, 2010, 26(4):60-64, 69.
	Pan Xiaodong, Ma Xiaoxiang, Zhao Xiaocui. Experimental research on the safety and riding characteristics of non-motorized vehicle at signalized intersection[J]. Journal of Transport Science and Engineering, 2010, 26(4):60-64, 69.
[6]	Kumar A, Paul M, Ghosh I. Analysis of pedestrian conflict with right-turning vehicles at signalized intersections in India[J]. Journal of Transportation Engineering, Part A: Systems, 2019, 145(6): DOI: 10.1061/JTEPBS.0000239.
[7]	Liang Xinyu, Meng Xianghai, Zheng Lai. Investigating conflict behaviours and characteristics in shared space for pedestrians, conventional bicycles and e-bikes[J]. Accident Analysis & Prevention, 2021, 158: 106167.DOI:10.1016/j.aap.2021.106167.
[8]	Zhang Yidan, Zhang Xiangjun, Zhang Wanting, et al. Analysis of influencing factors on safety of right-turn motor vehicles and straight non-motorized vehicles at signalized intersections[C]. COTA International Conference of Transportation Professionals, 2020: 3575-3585.
[9]	卢艺. 信号交叉口右转机动车与非机动车交通冲突分析[D]. 北京: 中国人民公安大学, 2021.
	Lu Yi. Analysis of traffic conflict between right-turning motor vehicles and non-motor vehicles at signalized intersections[D]. Beijing: People's Public Security University of China, 2021.
[10]	王正武, 谢静怡, 王杰, 等. 基于轨迹数据的非机动车左转信号灯安全效应评估[J]. 长沙理工大学学报(自然科学版), 2024, 21(1):164-173.
	Wang Zhengwu, Xie Jingyi, Wang Jie, et al. Traffic safety evaluation of left-turning non-motor vehicle signal lights at intersections based on trajectory data[J]. Journal of Changsha University of Science and Technology (Natural Science), 2024, 21(1):164-173.
[11]	于瑞康. 信号交叉口右转机动车与非机动车冲突行为分析[D]. 北京: 北京交通大学, 2018.
	Yu Ruikang. The analysis on conflict behaviors of right-turn motor vehicles and non-motor vehicles at signalized intersections[D]. Beijing: Beijing Jiaotong University, 2018.
[12]	张燕, 龙科军, 谷健. 信号交叉口机非交通冲突安全评估[J]. 交通科学与工程, 2021, 37(3):98-106.
	Zhang Yan, Long Kejun, Gu Jian. Safety assessment of vehicle-non-motor traffic conflict at signalized intersection[J]. Journal of Transport Science and Engineering, 2021, 37(3):98-106.
[13]	潘福全, 王满福, 孙伟, 等. 考虑非机动车过街的移位左转交叉口延误建模[J]. 广西大学学报(自然科学版), 2023, 48(4):857-866.
	Pan Fuquan, Wang Manfu, Sun Wei, et al. Construction of a delay evaluation model for displaced left-turn intersections considering non-motorized crossing modes[J]. Journal of Guangxi University (Natural Science Edition), 2023, 48(4):857-866.
[14]	顾晨阳. 二相位交叉口左转非机动车交通组织评价与方案比选研究[D]. 成都: 西南交通大学, 2020.
	Gu Chenyang. Study on evaluation and scheme selection of left-turn non-motor vehicle traffic organization at two-phase intersections[D]. Chengdu: Southwest Jiaotong University, 2020.
[15]	裴玉龙, 杜小敏, 沈威宇. 基于Ordered Probit模型的人车冲突安全影响因素研究[J]. 中国安全生产科学技术, 2024, 20(3):181-187.
	Pei Yulong, Du Xiaomin, Shen Weiyu. Research on influencing factors of pedestrian-vehicle conflict safety based on Ordered Probit model[J]. Journal of Safety Science and Technology, 2024, 20(3):181-187.
[16]	江玉杰, 万征, 陈继红. 基于有序选择模型的船舶事故严重度致因分析[J]. 中国安全科学学报, 2023, 33(12):131-139. doi: 10.16265/j.cnki.issn1003-3033.2023.12.0245
	Jiang Yujie, Wan Zheng, Chen Jihong. Causation analysis of ship accident severity based on ordered choice model[J]. China Safety Science Journal, 2023, 33(12):131-139. doi: 10.16265/j.cnki.issn1003-3033.2023.12.0245
[17]	张名芳, 马艳华, 马勇. 无信号交叉口人车冲突严重程度影响因素分析[J]. 中国安全科学学报, 2023, 33(8):190-197.
	Zhang Mingfang, Ma Yanhua, Ma Yong. Influencing factors on severity of vehicle-pedestrian conflict at unsignalized intersections[J]. China Safety Science Journal, 2023, 33(8):190-197. doi: 10.16265/j.cnki.issn1003-3033.2023.08.1084
[18]	周智文, 马健霄, 王冠森. 信号交叉口左转非机动车交通冲突分析[J]. 合肥工业大学学报(自然科学版), 2021, 44(1):95-100.
	Zhou Zhiwen, Ma Jianxiao, Wang Guansen. Analysis of traffic conflicts of left-turn non-motor vehicles in signalized intersection[J]. Journal of Hefei University of Technology (Natural Science), 2021, 44(1):95-100.

左转方式组合	统计量F值	P值
直接左转与顺时针二次过街	142.97	0.00
直接左转与逆时针二次过街	65.09	0.00
顺时针二次过街与逆时针二次过街	27.54	0.00

左转方式组合	统计量F值	P值
直接左转与顺时针二次过街	18.22	0.00
直接左转与逆时针二次过街	40.99	0.00
顺时针二次过街与逆时针二次过街	4.38	0.04

左转方式	潜在冲突/次	一般冲突/次	严重冲突/次	总计	个体平均冲突数/ (次/辆)
直接左转	34	83	89	206	0.95
顺时针二次过街	67	133	88	288	0.84
逆时针二次过街	57	47	32	136	0.77
总计	158	263	209	630	—

变量名称	变量类型	编码及描述	频数	占比/ %
左转时期	分类变量	0:绿灯前期(最初5 s)	265	42.1
		1:绿灯中期	206	32.7
		2:绿灯后期(最后5 s)	159	25.2
机动车运动方向	分类变量	0:直行	222	35.2
		1:左转	248	39.4
		2:右转	160	25.4
左直比	连续变量	左转非机动车与直行机动车流量之比	—	—
左左比	连续变量	左转非机动车与左转机动车流量之比	—	—
左右比	连续变量	左转非机动车与右转机动车流量之比	—	—
机动车类型	分类变量	0:小汽车	593	94.1
机动车类型	分类变量	1:公交车	37	5.9
交通冲突类型	分类变量	0:追尾冲突	109	17.3
		1:侧面冲突	295	46.8
		2:正面冲突	226	35.9
非机动车类型	分类变量	0:电动自行车	591	93.8
非机动车类型	分类变量	1:人力自行车	39	6.2
机非通过情况	分类变量	0:机动车先通过冲突点	378	60
机非通过情况	分类变量	1:非机动车先通过冲突点	252	40
非机动车左转方式	分类变量	0:直接左转	206	32.7
		1:顺时针二次过街	288	45.7
		2:逆时针二次过街	136	21.6

变量	参数估计	取值	P值	边际值/%
变量	参数估计	取值	P值	潜在冲突	一般冲突	严重冲突
左转方式	-0.268	顺时针二次过街	0.030	7.81	1.14	-8.95
左转方式	-0.522	逆时针二次过街	<0.01	15.19	2.22	-17.41
左转时期	0.294	绿灯后期	0.015	-8.56	-1.25	9.81
机动车运动方向	-0.286	左转	0.019	8.32	1.22	-9.53
机动车类型	-0.458	公交车	0.021	13.32	1.95	-15.27
非机动车类型	-0.405	自行车	0.034	11.77	1.72	-13.51
机非通过情况	-0.365	非机动车先通过冲突点	<0.01	10.63	1.55	-12.18
左直比	-0.066	—	0.019	1.93	0.28	-2.21
λ₁	-1.286	—	—	—	—	—
λ₂	-0.081	—	—	—	—	—