Using microscopic video data measures for driver behavior analysis during adverse winter weather:opportunities and challenges

This paper presents a driver behavior analysis using microscopic video data measures including vehicle speed, lane-changing ratio, and time to collision. An analytical framework was developed to evaluate the effect of adverse winter weather conditions on highway driving behavior based on automated （computer） and manual methods. The research was conducted through two case studies. The first case study was conducted to evaluate the feasibility of applying an au- tomated approach to extracting driver behavior data based on 15 video recordings obtained in the winter 2013 at three dif- ferent locations on the Don Valley Parkway in Toronto, Canada. A comparison was made between the automated approach and manual approach, and issues in collecting data using the automated approach under winter conditions were identified. The second case study was based on high quality data collected in the winter 2014, at a location on Highway 25 in Montreal, Canada. The results demonstrate the effectiveness of the automated analytical framework in analyzing driver behavior, as well as evaluating the impact of adverse winter weather conditions on driver behavior. This approach could be applied to evaluate winter maintenance strategies and crash risk on highways during adverse winter weather conditions.

Virtual Reality Driving Simulation for Measuring Driver Behavior and Characteristics

This article provides new insights regarding driver behavior, techniques and adaptability. This study has been done because: 1) driving a vehicle is critical and one of the most common daily tasks;2) simulators are used for the purpose of training and researching driver behavior and characteristics;3) the article addresses driver experience by involving new virtual reality technologies. A simulator has been used to assist novice drivers to learn how to drive in a very safe environment, and researching and collecting data for researchers has become easier due to this secure and user-friendly environment. The theoretical framework of this driving simulation has been designed by using the Unity3D game engine (5.4.f3 version) and was programmed with the C# programming language. To make the driving environment more realistic we, in addition, utilized the HTC Vive Virtual reality headset which is powered by Steamvr. We used Unity Game Engine to design our scenarios and maps because by doing this we are able to be more flexible with designing. In this study, we asked 10 people ranging from ages 19 - 37 to participate in this experiment. Four Japanese divers and six non-Japanese drivers engaged in this study, some of which do not have a driver’s license in Japan. A few Japanese drivers have a license and car, while others have a license but no car. In order to analyze the results of this experiment we are used MatlabR2016b to analyze the gathered data. The result of this research indicates that individual’s behavior and characteristics such as controlling the speed, remaining calm and relaxed when driving, driving at appropriate speeds depending on changes in road structures and etc. can affect their driving performance.

Driving Style Recognition Based on Driver Behavior Questionnaire

The ability to classify driver behavior lays the foundation for more advanced driver assistance systems. The present study aims to research driver pattern and classification feature. Driver behavior self-reported investigation was conducted with standardized driver behavior questionnaire (DBQ) by 225 nonprofessional drivers on the internet in Beijing. Questionnaire’s reliability was verified by statistics analysis. Confirmatory factor analysis (CFA) was used to analyze the underlying factor structure. Speed advantage, space occupation, the contend right of way and the contend space advantage were extracted from the ques-tionnaire results to quantify driver characteristics. Based on fuzzy C-means (FCM) algorithm and taking the four factors as pattern features, the number of driver classification distribution was discussed. Then the number of driver classification was determined by statistical indices. The comparison of classification results with the survey finding on whether the driver occurred in traffic accidents within five years shows that the classification result is the same as the actual driving conditions. Finally, correlation between the demographic and types of driving behavior has been analyzed. Female were more likely than male to careful driving, and the older the driver and the less driving experience, the more careful and moderate driving behavior is.

Vehicles, Advanced Features, Driver Behavior, and Safety: A Systematic Review of the Literature

Driver errors contribute to more than 94% of traffic crashes. Automotive companies are striving to enhance their vehicles to eliminate driver errors and reduce the number of crashes. Various advanced features like lane departure warning (LDW), blind spot warning (BSW), over speed warning (OSW), forward collision warning (FCW), lane keep assist (LKA), adaptive cruise control (ACC), cooperative ACC (CACC), and automated emergency braking (AEB) are designed to assist with, or in some cases take over, certain driving maneuvers. They can be broadly categorized into advanced driver assistance system (ADAS) and automated features. Each of these advanced features focuses on addressing a particular task of driving, thereby, aiding the driver, influencing their behavior, and enhancing safety. Many vehicles with these advanced features are penetrating into the market, yet the total reported number of crashes has increased in recent years. This paper presents a systematic review of these advanced features on driver behavior and safety. The review is categorized into 1) survey and mathematical methods to assess driver behavior, 2) field test methods to assess driver behavior, 3) microsimulation methods to assess driver behavior, 4) driving simulator methods to assess driver behavior, and 5) driver understanding and the effectiveness of advanced features. It is followed by conclusions, knowledge gaps, and need for further research.

Lateral control of autonomous vehicles based on learning driver behavior via cloud model

In order to achieve the lateral control of the intelligent vehicle, use the bi-cognitive model based on cloud model and cloud reasoning, solve the decision problem of the qualitative and quantitative of the lateral control of the intelligent vehicle. Obtaining a number of experiment data by driving a vehicle, classify the data according to the concept of data and fix the input and output variables of the cloud controller, design the control rules of the cloud controller of intelligent vehicle, and clouded and fix the parameter of cloud controller： expectation, entropy and hyper entropy. In order to verify the effectiveness of the cloud controller, joint simulation platform based on Matlab/Simulink/CarSim is established. Experimental analysis shows that： driver＇s lateral controller based on cloud model is able to achieve tracking of the desired angle, and achieve good control effect, it also verifies that a series of mental activities such as feeling, cognition, calculation, decision and so on are fuzzy and uncertain.

Generation and Selection of Driver-Behavior-Based Transferable MotionPrimitives

To integrate driver experience and heterogeneous vehicle platform characteristics in a motion-planning algorithm,based on the driver-behavior-based transferable motion primitives(MPs), a general motion-planning framework for offline generation and online selection of MPs is proposed. Optimal control theory is applied to solve the boundary value problems in the process of generating MPs, where the driver behaviors and the vehicle motion characteristics are integrated into the optimization in the form of constraints. Moreover, a layered, unequal-weighted MP selection framework is proposed that utilizes a combination of environmental constraints, nonholonomic vehicle constraints,trajectory smoothness, and collision risk as the single-step extension evaluation index. The library of MPs generated offline demonstrates that the proposed generation method realizes the effective expansion of MP types and achieves diverse generation of MPs with various velocity attributes and platform types. We also present how the MP selection algorithm utilizes a unique MP library to achieve online extension of MP sequences. The results show that the proposed motion-planning framework can not only improve the efficiency and rationality of the algorithm based on driving experience but can also transfer between heterogeneous vehicle platforms and highlight the unique motion characteristics of the platform.

Preface for the special issue on "Driver behavior,highway capacity and transportation resilience"

Transportation plays a vital role both in urban economy and usuallives, which is recognized as one of the major functions of a city, along with dwelling, work, and recreation. As an organic system, urban traffic has attracted increased attention during recent years. However, the majority of studies focused on roadways and vehicular technologies, and limited research has been conducted to address the driver characteristics and their impact. A major possible reason is the scarcity of reliable data. In fact, traditional traffic data obtained from cross-sectional detectors as well as video capture devices are not sufficient to fully capture driver behavior. Only in the recent years, with the availability of transportation-related ＂big data＂ and particu- larly the overwhelming information onboard and from road- side facilities, the impacts of driver behavior may be investigated in more detail. In general, driver behavior in microscopic level may include car-following, lane-changing, and gap acceptance models, which are believed largely to affect roadway capacity. Furthermore, safety concerns especially when drivingin an urban environment due to the interaction of different modes, also increase the complexity of quantifying the quality of service of various transportation facilities. These potential complexities in driver behavior challenge academia to develop supporting models and methods of analysis.

A Cellular Automaton Model for Heterogeneous and Incosistent Driver Behavior in Urban Traffic

In this paper a cellular automaton model is proposed to describe driver behavior at a single-lane urban roundabout. Driver behavior has been considered as heterogeneous and inconsistent. Most traffic papers in the literature just discussed heterogeneous driver behavior, to our best knowledge. Two truncated Gaussian distributions are used to model heterogeneous and inconsistent driver behavior, respectively. The physical meanings of two truncated distributions are indicated. This method may help enhance a better understanding of driver behavior at roundabout traffic, and even possibly provide references for roundabout design and management.

Evaluating Hazard Response Behavior of a Driver Using Physiological Signals and Car-Handling Indicators in a Simulated Driving Environment

Road traffic accidents are a major cause of casualties and costly implications to all the stakeholders. Research focusing on the driver as one of the causal agent of accidents has been studied for centuries and with the advent of modernized driver assistance technologies. This paper sought to evaluate response of a driver using active-driving performance indicators like reaction time and physiological signal response (surface electromyogram), to understand hazard response behavior. Simulation of driving scenes was done using Unity3D engine and VR Head mounted display. The driver was presented with stimulus (collision objects) of different size and distance. From the results, an event scene that the driver considered hazardous was marked with increased electromyography response distinct from non-event scenes. From the results, we noted an increase in pedal misapplication during hazard response. The proposed approach is applicable in a real time driving analysis for on-road risk level classification.

BEHAVIORAL TOXICOLOGICAL RESEARCHES ON AUTOMOBILE DRIVERS' DRIVING ACCIDENTS

125 automobile drivers, including 85 drivers with accidents (Accident Group, AG) and 40 drivers without accidents (Nonaccident Group, NAG), and 51 controls (Control Group. CG )were tested with WHO Neurobehavioral Core Test Battery (WHO NCTB). The results showed that there were obvious negative mood states such as tension-anxiety and fatigue in AG and drivers with accidents had more poor neurobehavioral performances, especially attention, response speed and perceptual-motor speed than drivers without accidents and controls. We also found that automobile drivers' neurobehavioral functions got weakened with the increase of their age and got strengthened with the elevation of their educational level. And the functions were inversely correlative to the accidents they cuased. The results of our study suggest that WHO NCTB can be an index of researches on driving accidents that automobile drivers caused and can be used in occupational selection and training of drivers.

基于多尺度骨架图和局部视觉上下文融合的驾驶员行为识别方法

识别非驾驶行为是提高驾驶安全性的重要手段之一。目前基于骨架序列和图像的融合识别方法具有计算量大和特征融合困难的问题。针对上述问题,本文提出一种基于多尺度骨架图和局部视觉上下文融合的驾驶员行为识别模型(skeleton-image based behavior recognition network,SIBBR-Net)。SIBBR-Net通过基于多尺度图的图卷积网络和基于局部视觉及注意力机制的卷积神经网络,充分提取运动和外观特征,较好地平衡了模型表征能力和计算量间的关系。基于手部运动的特征双向引导学习策略、自适应特征融合模块和静态特征空间上的辅助损失,使运动和外观特征间互相引导更新并实现自适应融合。最终在Drive&Act数据集进行算法测试,SIBBR-Net在动态标签和静态标签条件下的平均正确率分别为61.78%和80.42%,每秒浮点运算次数为25.92G,较最优方法降低了76.96%。

基于自然驾驶数据的驾驶换道风格分析与辨识

为了研究在换道工况下个体驾驶行为及其辨识,基于NGSIM轨迹数据提取532组换道有效数据。分析处理驾驶员换道特征参数,运用K-means算法对驾驶员进行聚类分析,可分为谨慎型、一般型、激进型三类。并对聚类的参数进行统计分析和时频分析验证各类驾驶人的差异性,并提取出有效参数作为辨识模型的输入,基于惯性权重非线性变化和粒子最大速度非线性递减的策略来改进粒子群,建立起基于改进粒子群优化SVM的驾驶人风格辨识模型,对SVM模型的两个关键参数C和g进行优化更新,驾驶人识别准确率可达到96.25%,且与BP神经网络、PSO-SVM、随机森林、KNN算法相比具有更高的识别准确率。

基于双路时空网络的驾驶员行为识别

驾驶员危险驾驶行为是恶性交通事故发生的主要原因之一,因此识别驾驶员行为具有工程应用上的重要意义。目前,主流基于视觉的检测方法是对驾驶员行为的局部时空特征进行研究,针对全局空间特征及长时序相关性特征研究较少,这在一定程度上无法结合场景上下文信息对危险驾驶行为进行识别。为了解决上述问题,提出一种基于双路时空网络的驾驶员行为识别方法,整合不同时空通路的优点以提高行为特征丰富度。首先,使用一种改进的双流卷积神经网络(TSN)对时空信息进行表征学习,同时降低提取特征的稀疏性;其次,构建一种基于Transformer的串行时空网络补充长时序相关性信息;最后,联合双路时空网络进行融合决策,增强模型的鲁棒性。实验结果表明,所提方法在驾驶员疲劳检测数据集YawDD、驾驶员分心检测数据集SF-DDDD和最新驾驶员行为识别数据集SynDD1这3个公开数据集上分别取得99.85%、99.94%和98.77%的识别准确率,特别是在SynDD1上,与使用动作识别的网络MoviNet-A0相比识别准确率提升了1.64个百分点;消融实验结果也验证了该方法对驾驶员行为有较高的识别精度。

城市道路下分心行为对驾驶员驾驶绩效的影响

为探究城市道路条件下常见驾驶分心行为对驾驶绩效的影响,以16名青年志愿者为被试,开展实车驾驶试验,测量车辆纵向和横向行驶参数,对比分析正常驾驶和分心驾驶状态下的驾驶绩效,并探讨驾驶经验的影响。结果表明:3种分心行为均使横向加速度标准差和纵向加速度标准差增加,而使速度均值降低,说明驾驶员通过速度和转向控制补偿分心状态下的驾驶绩效;交谈对速度均值和纵向加速度均值、阅读广告对速度标准差和横向加速度均值具有显著性影响(p<0.05);接听电话时,经验驾驶员速度标准差明显高于新手(p=0.021);接听电话和阅读广告次任务下,经验驾驶员的横向加速度均值均明显较高(p=0.003,p=0.004),说明经验丰富的驾驶员受分心行为的影响程度更低。

人格特征对驾驶人驾驶行为的影响研究

为了解驾驶人不安全驾驶行为背后的人格特征影响因素,回顾自2000年以来国内外人格特征与驾驶行为关系的相关研究,采用系统评价和Meta分析的优先报告项目(PRISMA)标准在国内外知名数据库检索到17688篇文献,剔除重复文献3160篇、不相关文献14263篇、非优秀文献198篇,最终将67篇优秀文献纳入研究,归纳并梳理文献中不同人格特征与不安全驾驶行为的关系。结果表明:大五人格的5个人格维度与驾驶行为的关系研究结果均存在不一致;感觉寻求和冲动性各自均为不安全驾驶行为的优秀预测因子,但相较之下后者的预测能力更强;驾驶愤怒对不安全驾驶行为的直接预测能力不强;无规范感、特质焦虑、利他及自恋对均不安全驾驶行为有不同程度的影响,但相关研究较少;依恋、特质情商、集体主义和异常人格特质等人格特征受关注较少,需要更多检验;中国人样本的一些研究结果与多数研究结果存在不一致,需开展跨文化研究。

交通替代攻击问卷的修订与信效度检验

目的 修订交通替代攻击问卷并检验其在我国驾驶员群体中的信效度.方法 采用交通替代攻击问卷对317 名驾驶员进行施测,共获得 305 份有效问卷.通过分析后得到修订版的中国交通替代攻击问卷.结果 修订后的交通替代攻击问卷共保留 24 个题目,累积方差贡献率 56.278%;该问卷的模型拟合度较好(GFI =0.870,IFI =0.920,TLI =0.911,NFI =0.860,CFI =0.919,RMSEA =0.062,χ2/df =2.148,SRMR = 0.045).问卷总的内部一致性系数为0.894,问卷各维度的内部一致性系数在0.677~0.936.问卷各维度与总分的相关系数在0.588~0.845 之间,且相关显著.问卷各维度与亲社会和攻击性驾驶行为量表之间部分相关显著,说明修订后的问卷具有良好的相容效度.结论 修订后的交通替代攻击问卷信度和效度较好,可以作为测量我国驾驶员交通替代攻击的一个可靠而有效的工具.

Driver State Detection Based on Cardiovascular System and Driver Reaction Information Using a Graphical Model

Traffic accidents are mainly caused by human error. In an aging society, the number of accidents attributed to elderly drivers is increasing. One noteworthy reason for this is operation misapplication. Studies have been conducted on the use of human-machine interfaces (HMIs) to inform the driver when he or she makes an error and encourage appropriate actions. However, the driver state during the erroneous action has not been investigated. The purpose of this study is to clarify the difference in the driver’s state between normal and surprising situations in a misapplication scenario, utilizing multimodal information such as biometric information and driver operation. We found significant changes in the interaction of components between the normal and the surprised driving state. The results could provide basic knowledge for the future development of a driver assistance system and driver state estimation using data acquired from multiple sensors in the vehicle.

基于模型预测的智能汽车辅助驾驶控制研究

辅助驾驶控制是智能汽车的重要功能,可有效协助驾驶员完成驾驶操作,减少驾驶疲劳并提升驾驶安全。智能车辆的辅助驾驶控制需要在不干扰驾驶员的情况下,提供合适的车辆运动控制量并防止可能发生的碰撞。文章提出了一种基于模型预测控制的智能汽车辅助驾驶控制方法。该控制方法在驾驶员正常操作时符合驾驶员的操作特性,在可能发生碰撞时纠正驾驶员的转向操作,能够保证安全性且不干扰驾驶员。为了避免与周围车辆的碰撞,基于周围车辆位置建立约束条件,并引入模型预测辅助驾驶控制设计。通过仿真研究验证了所提出的辅助驾驶控制方法,结果表明,所提出的预测控制方法可以帮助驾驶员避免与附近车辆可能发生的碰撞,从而保证智能汽车的行驶安全。

场车司机的不安全行为研究及建议

本文以场车司机易造成安全事故的高危、频发的不安全行为作为研究重点,明确其行为表现、事故特征,提出合理措施和相关建议,以降低场车事故风险。

基于深度学习的驾驶员分心检测理论研究

文章提出了一种基于改进Yolov8的驾驶员分心检测方法,在Yolov8的基础上设计了一种包含小目标检测层和新型连接结构的加权双向特征金字塔网络(BiFPN)。在骨干网络中,引入了高效多尺度注意力(EMA)机制模块,以加强对关键信息的学习和提取。提出了一种新型的C2f结构,将动态卷积、注意力机制融入到C2f中。改进模型的精度提高了4.6%,同时参数量降低了27.1%,一定程度上减小了模型的规模,提高了计算效率。最后在公开数据集上进行泛化实验,各项评价指标皆优于原模型。因此,文章研究为分心驾驶行为检测领域提供了一种更高效、更精确的解决方案,为道路安全和交通监管提供有力的支持。