基于自然驾驶数据的城市交叉口纵向驾驶特征分析

为满足驾驶辅助系统在城市交叉口对类人驾驶能力的更高需求,本文中研究了实际交通中的驾驶人在该区域的纵向驾驶行为特征。从自然驾驶数据中提取了778条驾驶人接近城市交叉口的样本数据,应用YOLOv4识别了交通场景中的各类道路使用者,采用方差分析研究了反应特性在不同运动类型和交通密度中的差异,建立分层回归模型分析了制动特性与运动状态、运动类型和道路使用者的关系。结果表明：高密度交通显著降低接近速度;与右转驾驶人相比,停车驾驶人的反应距离更长;当接近速度较高或反应距离较短时,会在更短的时间内以更高的减速度和制动强度接近交叉口,且提前4.46 s开始制动;不同道路使用者对制动特性产生了不同影响,停车驾驶人主要关注同向行驶的他车,右转驾驶人主要关注行人和骑车人。     

无信号控制交叉口老年驾驶人转向行为图谱研究

考虑到老年驾驶人存在的视觉弱化、反应能力下降等问题,有必要分析其转向行为特征。基于此,本文分析了在无信号控制交叉口情景下老年驾驶人的转向行为特征,并绘制图谱描述行为特征的时序变化。基于无信号控制交叉口的现实场景调查,搭建了具备常见冲突类型的虚拟仿真驾驶场景（包含6个无信号控制交叉口）,招募符合要求的老年驾驶人与中青年驾驶人进行驾驶模拟实验,分别采集车辆行驶数据（驾驶模拟器）、眼动数据和生理心理数据分析老年驾驶人与中青年驾驶人在不同转向场景下的行为特征差异,应用图谱理论构建驾驶人转向行为图谱描述老年驾驶人与中青年驾驶人的转向行为特征时序变化。实验结果表明:老年驾驶人的速度均值为20.4 km/h、注视持续时间均值为289.47 ms、扫视幅度均值为3.51°;中青年驾驶人的速度均值为35.79 km/h、注视持续时间均值为247.94 ms、扫视幅度均值为4.56°。老年驾驶人的心率变异性时域指标（SDNN和RMSSD）与频域指标（LF/HF和TP）的值更低,表明老年驾驶人在转向过程中更加紧张。图谱显示老年驾驶人的紧张持续时间更长,并在信息获取广度上弱于中青年驾驶人。结合图谱时空差异性指标发现,这2类驾驶人的驾驶行为在左转向场景下存在显著性差异,老年驾驶人驾驶操作的稳定性与安全性较低。      

城市交叉口车路网联信息对青年驾驶人驾驶行为的影响分析

基于驾驶模拟器设计了城市道路信号和无信号交叉口场景下的模拟驾驶实验,研究网联信息的存在和内容对青年驾驶人工作负荷和操纵行为的影响.实验共包括被试26人,均为22～30岁的青年驾驶人.结果表明:对无信号交叉口(区分次干路直行车辆和主路对向左转车辆)或信号交叉口红灯时间即将结束时,网联信息可以显著降低青年驾驶人的工作负荷,...     

基于驾驶模拟器的性别对信控铁路道口抵近行为的影响研究

为了分析不同性别驾驶人抵近信号灯控制铁路道口的驾驶行为特性,设计了5个具有不同红灯触发时距的信号灯控制铁路道口的虚拟场景,并进行了驾驶模拟试验,采集24名男性驾驶人和20名女性驾驶人抵近道口的微观驾驶行为数据。基于时空特性将抵近行为划分为月白灯期间、红灯进入时刻、红灯期间3个阶段,利用混合线性模型和二元Logistics线性回归模型对比分析了2组驾驶人抵近道口的制动行为、加速行为和闯红灯行为特性,并探索性别与信号灯控制铁路道口抵近行为的相关关系。结果表明：男性驾驶人更倾向于在月白灯期间提前制动,其提前制动率是女性驾驶人的1.57倍;男性驾驶人在红灯进入时刻和红灯期间的车辆操控行为特性与女性驾驶人无显著性差异;男性驾驶人的总体违法率是女性驾驶人的1.5倍,尤其在红灯触发时距较大（4~6 s）时,女性驾驶人几乎无人违法,而男性驾驶人在3个场景下的违法率仍然达到25.0%、12.5%和16.7%;男性驾驶人闯红灯的严重性更高,驾驶风格更加冒险激进;另外,驾驶人闯红灯行为还受红灯进入时刻位置、制动时刻位置和加速时刻位置显著影响,说明驾驶人在走停决策的制定上更依赖于物理空间位置,与车辆运行速度和驾驶人反应能力无关。研究结果可应用于铁路道口安全管理与防护对策设计。     

公交车驾驶人组织认同感和工作倦怠对不良驾驶行为的影响

公共交通是城市道路交通运输系统中的重要组成部分,公交车驾驶人存在的不良驾驶行为已经成为影响交通安全的最大制约因素之一,由其引发的道路交通事故给个人和社会都造成了严重损失,探索更为安全的公共交通环境显得极为重要。以公交车驾驶人为对象,探究影响公交车驾驶人驾驶行为特征的个人心理因素和组织环境因素,及其对不良驾驶行为的内在影响机制。选用工作倦怠量表（MBI-GS）、组织认同感问卷（OIQ）及驾驶行为问卷（DBQ）对844名城市公交车驾驶人展开问卷调查研究,并使用回归分析和中介检验来探究工作倦怠和组织认同感对驾驶行为的影响机制。结果表明：①32.8%的公交车驾驶人存在不良驾驶行为,错误驾驶行为显著高于违法驾驶行为。其中,在违法维度中,侵略性违法显著高于普通违法。②公交车驾驶人存在较高的组织认同感,组织认同感对不良驾驶行为具有显著的负向预测作用,对工作倦怠程度具有显著的负向预测作用;工作倦怠程度对不良驾驶行为具有显著的正向预测作用。③公交车驾驶人的工作倦怠水平在组织认同感对不良驾驶行为的影响中起到部分中介作用。该结果厘清了公交车驾驶人的工作倦怠和组织认同感对不良驾驶行为的影响机制,为进一步探究不良驾驶行为的心理干预方法提供了思路,从而达到改善城市道路公共交通安全环境的目标。     

老年驾驶人在交叉口的视觉特性（双语出版）

机动车驾驶人主要通过视觉搜索行为获取交通信息,为了分析老年驾驶人进入道路交叉口的视觉特性,建立了5个道路交叉口的虚拟场景,并进行了驾驶模拟试验,采集了21名老年驾驶人和17名中青年驾驶人进入交叉口的眼动数据。应用动态聚类法将驾驶人的视野平面划分为5个注视区域;在此基础上对比分析了2组驾驶人进入交叉口的注视特性、扫视特性、注视转移概率和注视平稳分布概率;并通过灰色关联度矩阵分析了影响注意转移行为的因素。分析结果表明：老年驾驶人对非行驶方向上的注视区域的注意都相对较差,2组驾驶人的注视持续时间、扫视幅度和扫视平均速度存在较大的差异,其中老年驾驶人的扫视幅度和扫视速度明显低于中青年驾驶人,而且老年驾驶人的注视点分布在车辆正前方区域的平稳分布概率为80.7%,比中青年驾驶人的69.3%高11.4%,并且具有显著性差异,说明老年驾驶人注视转移模式的灵活性相对中青年驾驶人较差。     

预警时间对信控交叉口避撞驾驶行为的影响研究

研究不同预警信息发布时间下,不同性别驾驶人对于交叉口迎头侧面避撞情景驾驶行为影响规律,为提高车辆避撞预警系统功效性提供理论依据.基于汽车驾驶模拟器设计实验,招募具有稳定驾驶能力的驾驶人45名,采集7种预警信息发布时间(2.5～5.5 s),将无预警作为控制组,采用C#编程提取能够表征驾驶行为的变量.用方差分析和线性回归方法分析在交叉口迎头侧面碰撞情景下不同预警信息发布时间对驾驶人的制动时间、最大减速度、测试车辆与冲突车辆的最小距离的影响.结果表明,预警信息发布越早,驾驶人的制动时间越长,最大减速度越小,说明较早发布预警信息可以减缓驾驶人采取制动措施的剧烈程度.同时,预警信息发布较早,可以增大车辆间的最小间距,降低碰撞事故发生的可能性.此外,女性驾驶人的驾驶行为比男性驾驶人更加保守.     

不同信号过渡方式对交叉口驾驶行为选择的影响

为掌握黄灯、绿闪和倒计时等不同信号过渡方式对驾驶行为选择的影响,通过对驾驶人黄灯期间交叉口驾驶行为决策过程及其因素的分析,采用现场视频摄录观测方法,对苏锡常地区的若干信号交叉口黄灯期间的机动车驾驶行为选择进行深度调查,对比了闯黄灯车辆行驶速度和初始位置,分析了前3辆机动车的驾驶行为选择.结果表明,信号交叉口采用绿闪和倒计时提示对驾驶行为选择影响较大,其闯黄灯行为发生率明显低于黄灯提示方式,但是倒计时提示会诱发机动车驾驶人加速通过交叉口并增加安全隐患.     

驾驶经验和气质类型对驾驶行为的影响分析

驾驶人的生理心理因素会对驾驶人的驾驶行为产生一定的影响已经得到了较多认同,但是通过科学实验进行定量化实证分析的研究还较少。论文首先采用量表测量与仪器测试相结合的实验方法,以专业和非专业驾驶人为实验对象,利用气质类型量表测量驾驶人的气质类型;利用实验车在实际道路交通环境下行驶以获得跟驰车辆对前后车的行驶轨迹,并计算得到影响驾驶行为的关键参数。基于实验数据分析了驾驶经验和气质类型对驾驶人的行驶速度、加速度、跟驰距离、时间间隔等方面的影响。在跟驰时驾驶人的速度和加速度选择方面,专业驾驶人对速度和加速度的选择更加具有自主性,并且内向型性格驾驶人的行驶速度和加/减速幅度一般要小于外向型性格的驾驶人。在跟驰时驾驶人对空间和时间距离的选择方面,无论是跟驰距离还是时间间隔,气质类型对非专业驾驶人空间和时间距离选择的影响更加明显,并且各速度段内向型性格驾驶人选择的空间和时间距离一般大于外向型性格驾驶人;相同气质类型的非专业驾驶人选择的时间间隔都要大于专业驾驶人。经研究建立了驾驶人生理心理特性与驾驶行为特性之间的关系,为分析驾驶人行为特性与交通安全之间的关系奠定了良好的基础。     

驾驶风格对驾驶行为的影响

为了揭示驾驶风格对驾驶行为的影响规律,进而提取表征驾驶风格的特征参数,对不同风格驾驶人在感知层和操作层的驾驶行为数据进行了量化分析。首先,基于驾驶行为问卷对18名中国非职业驾驶人进行了驾驶风格问卷调查,并采用主成分分析、K-均值聚类等方法将被试驾驶人分为谨慎型、正常型和激进型3种类型。接着,被试驾驶人在搭载了SmartEye眼动仪的驾驶模拟器上开展了高速公路行车环境下的驾驶试验,同步采集了感知层的视觉特性参数和操作层的驾驶绩效参数,并采用判断抽样的方式将驾驶样本按照驾驶风格和驾驶模式（换道意图和车道保持）进行了划分,共选取了810组有效样本。最后,采用方差分析法分析了不同风格驾驶人在不同驾驶模式下的注视行为、扫视行为、横向控制特性、纵向控制特性方面相关参数的差异显著性,并提取了不同风格间存在显著差异的参数作为表征驾驶风格的特征参数。研究结果表明：驾驶风格越激进,驾驶人对周围环境关注越少,对车辆的横向控制稳定性越差,急加速和急减速行为发生的频次越高;不同风格驾驶人在意图时窗内对后视镜的注视次数（p=0.002）、方向盘转角熵值（p=0.04）、加速踏板开度（p=0.01）、制动踏板开度（p=0.02）这4个参数的差异均较为显著,因此可作为表征驾驶风格的特征参数。     

Traffic safety culture of professional drivers in the State of Qatar

Professional drivers play a significant role within the traffic system of the State of Qatar. With developing infrastructure, the need for professional drivers is growing. However, knowledge is lacking about their perception of traffic safety. Therefore, this study investigates the personal acceptance of risky driving and suggested traffic laws among this specific group of drivers, in order to create understanding about their likelihood to commit certain risky driving behaviors and their resistance to the implementation of certain traffic laws. The aim of this study is to establish which personal attributes of professional drivers in Qatar could influence a high likelihood to commit risky driving behaviors, estimating which specific groups of professional drivers impose the highest risk to violate certain traffic laws. Results indicate that transportation mode, origin and years of driving experience are all personal attributes that have a significant impact on the professional driver's risk to commit risky driving behaviors and their opposition to the implementation of related traffic laws. Distressing results have been found for the high likelihood to violate speed in school zones and the high risk to be distracted by any type of phone use while driving, suggesting the need to put emphasize on these safety hazards during the training programs of professional drivers at professional driving schools in the State of Qatar.     

信控交叉口多绿尾相位切换模式下驾驶行为的对比研究

针对中国城市道路交叉口控制实践中,绿灯信号相位多种切换方式同时存在而未有详细优劣对比的现状开展研究.设计1组驾驶模拟器实验,对绿灯闪烁(GSF)与绿灯倒计时(GSC)2种典型的相位切换模式下的驾驶行为进行对比观测和分析,建立驾驶员在2种切换模式下的"通过/停止"决策行为模型.研究结果表明,驾驶员在不同模式相位切换期间的驾驶行为具有一定复杂性,与驾驶环境(如车速、黄灯启亮时与停车线的距离)和个体特征(如性别、年龄、驾龄)等密切相关.驾驶环境方面,GSC模式对驾驶员有较为强烈的催促作用,驾驶员在GSC模式下的平均通过率提高了10% ~20%,且陷入两难区的概率比GSF模式下提高了7.2%.黄灯通过率和黄灯启亮时车辆与停车线的距离呈负相关关系,且与黄灯启亮时的车速在不同相位切换模式下表现出相反的趋势:在GSC模式下呈现正相关关系,而在GSF模式下呈现负相关关系.驾驶员个体特征方面,男性、20~30岁、3~4年驾龄的驾驶员通过率明显高于其它分组.      

Driver behavior formulation in intersection dilemma zones with phone use distraction via a logit-Bayesian network hybrid approach

Use of cellular phone while driving is one of the top contributing factors that induce traffic crashes, resulting in significant loss of life and property. A dilemma zone is a circumstance near signalized intersections where drivers hesitate when making decisions related to their driving behaviors. Therefore, the dilemma zone has been identified as an area with high crash potential. This article utilizes a logit-based Bayesian network (BN) hybrid approach to investigate drivers' decision patterns in a dilemma zone with phone use, based on experimental data from driving simulations from the National Advanced Driving Simulator (NADS). Using a logit regression model, five variables were found to be significant in predicting drivers' decisions in a dilemma zone with distractive phone tasks: older drivers (50–60 years old), yellow signal length, time to stop line, handheld phone tasks, and driver gender. The identified significant variables were then used to train a BN model to predict drivers' decisions at a dilemma zone and examine probabilistic impacts of these variables on drivers' decisions. The analysis results indicate that the trained BN model was effective in driver decision prediction and variable influence extraction. It was found that older drivers, a short yellow signal, a short time to stop line, nonhandheld phone tasks, and female drivers are factors that tend to result in drivers proceeding through intersections in a dilemma zone with phone use distraction. These research findings provide insight in understanding driver behavior patterns in a dilemma zone with distractive phone tasks.     

Traffic safety culture of drivers in Canada: Implications for new traffic law implementation to enhance traffic safety

This study investigates Canada's traffic safety culture (TSC) as part of a global research project. The TSC survey data collected by an online survey is used to predict drivers' perception of changes in traffic problems in the past 3 years, driver's support, opposition towards enforcement of additional traffic laws, and drivers' perceived threat towards risky driving behaviors. A two-step procedure is followed to build models. The first step includes feature selection using the chi-square test of independence. The second step comprises building classification models using the Random Forest technique. Results suggest that drivers' personal attributes like the number of accident records, driving frequency, geographic region of nationality, and religion are top predictor variables for drivers' perception towards changes in traffic problems. In addition, compared to others most drivers perceive distracted driving as a major traffic problem today. There is strong disapproval of drivers against the following driving behaviors and strong support to implement laws against it: speeding in school zones, talking on a hand-held cell phone, tying text messages or e-mails, drowsy driving, driving without wearing their seatbelt, drive with passengers not wearing seatbelts, running through red lights, and impaired driving. In contrast, following risky driving behaviors is less of a perceived threat: speeding over limit on a freeway, on a residential street, and in an urban area and talking on a hands-free cell phone while driving. In addition, a driver's accident record is a significant indicator for a perceived threat towards risky driving behaviors followed by age and accumulated demerit points. The results from this study can be used to guide educational campaigns to transform the traffic safety culture of target groups and make more informed policy decisions.     

Factors associated with the red-light running behavior characteristics of motorcyclists

Red-Light Running (RLR) is one of the significant safety problems and the riskiest riding behavior at the mixed traffic intersections. This study aims to explore the rider's violation characteristics and factors influencing the occurrences. Observed data were recorded by the video cameras to monitor a rider's crossing behavior. The red-light runners are classified into the Risk-taking, the Opportunistic, and the Traffic-following riders. The results reveal that most violators are the opportunistic riders. These riders first stop and cross the intersection during the late stage of the red signal. Factors associated with each violation characteristic are unique. The opportunistic riders are likely to be on the right-turn maneuver, driving the manual gear motorcycles, not wearing safety helmets, driving at night, and have narrower crossing distances. Mapping the influencing factors on the chain of events provides useful information for countermeasures development. Based on the findings, the integrated measures to modify the environments and raise awareness are proposed to improve motorcycles' safety in the mixed traffic.     

驾驶人驾驶决策机制遵循最小作用量原理

为提升智能汽车的自主决策能力，使其能够学习人的决策智慧以适应复杂多变的道路交通环境，需要揭示驾驶人决策机制。首先通过对自然驾驶数据的分析，发现在车辆行驶过程中能够反映驾驶人决策行为的主要运动特征参数存在极值现象，而产生极值现象的内在动因是驾驶人遵循“趋利避害”的基本决策机制，即驾驶过程中驾驶人力图实现机动性和安全性综合性能最优。受自然界包括物理和生物行为上的众多极值现象遵循最小作用量原理的启发，提出驾驶人决策机制遵循最小作用量原理的假设。随后建立抽象描述驾驶过程的物理模型，并提出最小作用量决策模型（Least Action Decision-making Model，LADM），通过与传统驾驶决策模型（经典跟车模型和换道模型）对比，分析结果显示LADM模型更具通用性。最后开展了实车试验，采集20名驾驶人在自由行驶、跟车行驶和邻车切入3种工况下的试验数据，分析计算并检验了不同驾驶人行车过程的理论最小作用量和实际作用量。试验结果表明：驾驶人在驾驶过程中的实际作用量与最小作用量之间无显著性差异，体现出驾驶人在行车过程中对安全和高效具有共性追求，验证了驾驶人决策机制遵循最小作用量原理。     

The impact of the existence of the volunteer on handling the waiting time at unsignalized intersection in Yogyakarta,Indonesia

This research focuses on the impact of the existence of the volunteer (VO) in managing traffic at unsignalized intersections. VO is a person who manages traffic at unsignalized intersection without the legality of operations but in exchange for financial compensation from the road users. VO is a unique phenomenon that only occurs in Indonesia's cities. The location of this research was qualified at unsignalized intersections in Yogyakarta, Indonesia, which had a high traffic volume and had VO managing the traffic. The selected data were those during peak traffic conditions, each of which was one day at the same time with and without VO. Traffic volume, speed, waiting time, and vehicle conflict were the data collected and analyzed in this research. The results indicated that although the existence of VO was needed psychologically to lessen the chaotic traffic situation at unsignalized intersection, it did not have a significant impact on reducing the waiting time for vehicles from the minor road to cross and merge to the major road. The difference in waiting time was only around an average of two seconds as a result of the existence of VO compared to non-VO. According to the findings, this research proposes to prohibit the untrained VO since the benefits derived in controlling the chaotic traffic is not proportional to the potential risk of getting hit or lawsuits from drivers who are disturbed by their existence.