车辆在高速公路合流区内跟驰行驶时，会受到其他车辆加减速、频繁变道及匝道车辆并入主线等行为的影响。为探究此情况下跟驰行为表现出的特征及受不同干扰而呈现出的行为差异，基于在高速公路合流区域拍摄的视频，处理出48辆车的轨迹数据，从中提取出11个可表征车辆跟驰状态的特征参数。利用粒子群优化的模糊聚类（Particle Swarm Optimization Fuzzy C-Means, PSO-FCM）算法，从车速稳定性、跟驰行为倾向性和横向位置稳定性3方面对跟驰行为特征进行聚类分析。结果表明：不同车道内驾驶员的跟驰行为受到的影响及表现出的行为特征都呈现出显著的差异，外侧车道驾驶员受到的影响最大，行驶状态不稳定且易表现出较为激进的跟驰行为；中间车道次之；内侧车道相对稳定，受区域交通流干扰相对较小。可推断，高速公路合流区交通流的不稳定性是该区域不同车道内驾驶员跟驰行为存在明显差异的直接诱因，与路段内交通事故频发的现状也存在一定的内在联系。
When the vehicles were following in the merging area of freeway, they would be affected by the behavior of other vehicles, such as acceleration and deceleration, frequent lane changing and merging of ramp vehicles into the main line. In order to explore the characteristics of the following behavior under the influence and the behavior difference caused by different interferences, the trajectory data of 48 vehicles were processed from the video taken in freeway merging area. 11 characteristic parameters which could represent the vehicle following state were extracted from the trajectory data. The PSO-FCM (Particle Swarm Optimization Fuzzy C-Means) algorithm was used to analyze the following behavior characteristics from three aspects: speed stability, tendency of following behavior and lateral position stability. The results show that there was significant difference of the drivers′ following behaviors in different lanes. The drivers in the outer lane were most affected, their driving state was unstable and they were prone to more aggressive following behaviors; the middle lane was the second; their inner lane was relatively stable, and the interference of regional traffic flow was relatively small. It could be inferred that the instability of traffic flow in the merging area of freeway is the direct cause to the obvious difference of vehicle following behavior in different lanes of this area, and it also has a certain internal relationship with the current situation of frequent traffic accidents in the road section.
[1]阎莹，盛彦婷，袁华智，等. 高速公路出入口区域行车风险评价及车速控制[J]. 交通运输工程学报，2011，11（2）：90-96.
[2]梁夏，郭忠印，方守恩. 道路线形与道路安全性关系的统计分析[J]. 同济大学学报（自然科学版），2002（2）：203-206.
[3]王殿海，金盛. 车辆跟驰行为建模的回顾与展望[J]. 中国公路学报，2012（1）：115-127.
[4]Gipps P G. A Behavioural Car-Following Model for Computer Simulation[J]. Transportation Research Part B: Methodological, 1981, 15(2): 105-111.
[5]Krauss S, Wagner P, Gawron C. Metastable States in a Microscopic Model of Traffic Flow[J]. Physical Review E, 1997, 55(5): 5597-5602.
[6]Ossen S, Hoogendoorn S P. Heterogeneity in Car-Following Behavior: Theory and Empirics[J]. Transportation Research Part C: Emerging Technologies, 2011, 19(2): 182-195.
[7]Higgs B, Abbas M. Segmentation and Clustering of Car-Following Behavior: Recognition of Driving Patterns[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(1): 81-90.
[8]彭勇，刘世洁，Dennis Z Yu. 前后多车影响的跟驰模型及稳定性分析[J]. 交通运输系统工程与信息，2018，18（6）：41-47.
[9]陶鹏飞. 考虑侧向车影响的跟驰行为分析与建模[D]. 长春：吉林大学，2009.
[10]曲大义，陈文娇，郭涛，等. 考虑横向干扰的交叉跟驰模型研究[J]. 公路交通科技，2015，32（1）：122-127.
[11]王波，惠宏伟，李永建. 基于期望安全车距的车辆跟驰行为分析与仿真[J]. 中国安全科学学报，2015，25（9）：91-95.
[12]金勇，刘岩. 环形交叉口入口道车辆跟驰行为研究[J]. 大连交通大学学报，2012，33（1）：9-12.
[13]李修海，于少伟. 信号交叉口集聚车辆的跟驰行为分析与建模[J]. 交通运输系统工程与信息，2015，15（2）：102-108.
[14]刘志强，王兆华，钱卫东. 基于速度的交通事故分析[J]. 中国安全科学学报，2005，15（11）： 35-38.
[15]Li W, Liu Y S, Zhao X X, et al. Particle Swarm Optimization for Fuzzy C-Means Clustering[C]// 2006 6th World Congress on Intelligent Control and Automation. Dalian: IEEE, 2006: 6055-6058.
[16]Ahmadyfard A, Modares H. Combining PSO and K-Means to Enhance Data Clustering[C]// 2008 International Symposium on Telecommunications. Tehran: IEEE, 2008: 688-691.
[17]Chen J S, Pi D C. A Cluster Validity Index for Fuzzy Clustering Based on Non-Distance[C]// 2013 International Conference on Computational and Information Sciences. Shiyang: IEEE, 2013: 880-883.
京公网安备 11010102003417号
京ICP备17006742号-3