An investigation of train driver visual strategies

Train driving is primarily a visual task; train drivers are required to monitor the dynamic scene visually both outside and inside the train cab. Poor performance on this visual task may lead to errors, such as signals passed at danger. It is therefore important to understand the visual strategies that train drivers employ when monitoring and searching the visual scene for key items, such as signals. Prior to this investigation, a pilot study had already been carried out using an eye tracking technique to investigate train drivers’ visual behaviour and to collect data on driver monitoring of the visual environment, Groeger et al. (2003) Pilot study of train drivers’ eye movements, University of Surrey. However, a larger set of data was needed in order to understand more fully train driver visual behaviour and strategies. In light of this need, the Transport Research Laboratory produced a methodology for the assessment of UK train driver visual strategies, on behalf of the Rail Safety and Standards Board and applied this methodology to conduct a large-scale trial. The study collected a wealth of data on train drivers’ visual behaviour with the aim of providing a greater understanding of the strategies adopted. The corneal dark-eye tracking system chosen for these trials tracks human visual search and scanning patterns, and was fitted to 86 drivers whilst driving in-service trains. Data collected include the duration and frequency of glances made towards different elements of the visual scene. In addition, the train drivers were interviewed after driving the routes, to try and understand the thought processes behind the behaviour observed. Statistical analysis of over 600 signal approaches was conducted. This analysis revealed that signal aspect, preceding signal aspect, signal type and signal complexity are important factors, which affect the visual behaviour of train drivers. Train driver interview data revealed that driver expectation also plays a significant role in train driving. The findings of this study have implications for the rail industry in terms of infrastructure design, design of the driving task and driver training. However, train driving is extremely complex and the data from this study only begin to describe and explain train driver visual strategies in the specific context of signal approaches. This study has provided a wealth of data and further analysis of it is needed to investigate the role of other factors and the complex relationships between factors during signal approaches and other driving situations systematically. Finally, there are important aspects of visual behaviour that cannot be examined using these data or this method. Investigation of other aspects of visual behaviour, such as peripheral vision, will require other methods such as simulation.     

The validity of driving simulation for assessing differences between in-vehicle informational interfaces: A comparison with field testing

Data from on-road and simulation studies were compared to assess the validity of measures generated in the simulator. In the on-road study, driver interaction with three manual address entry methods (keypad, touch screen and rotational controller) was assessed in an instrumented vehicle to evaluate relative usability and safety implications. A separate group of participants drove a similar protocol in a medium fidelity, fixed-base driving simulator to assess the extent to which simulator measures mirrored those obtained in the field. Visual attention and task measures mapped very closely between the two environments. In general, however, driving performance measures did not differentiate among devices at the level of demand employed in this study. The findings obtained for visual attention and task engagement suggest that medium fidelity simulation provides a safe and effective means to evaluate the effects of in-vehicle information systems (IVIS) designs on these categories of driver behaviour.

Statement of Relevance: Realistic evaluation of the user interface of IVIS has significant implications for both user acceptance and safety. This study addresses the validity of driving simulation for accurately modelling differences between interface methodologies by comparing results from the field with those from a medium fidelity, fixed-base simulator.     

Optical information for car following: the driving by visual angle (DVA) model

OBJECTIVE: The present study developed and tested a model of car following by human drivers. BACKGROUND: Previous models of car following are based on 3-D parameters such as lead vehicle speed and distance information, which are not directly available to a driver. In the present paper we present the driving by visual angle (DVA) model, which is based on the visual information (visual angle and rate of change of visual angle) available to the driver. METHOD: Two experiments in a driving simulator examined car-following performance in response to speed variations of a lead vehicle defined by a sum of sine wave oscillations and ramp acceleration functions. In addition, the model was applied to six driving events using real world-driving data. RESULTS: The model provided a good fit to car-following performance in the driving simulation studies as well as in real-world driving performance. A comparison with the advanced interactive microscopic simulator for urban and nonurban networks (AIMSUN) model, which is based on 3-D parameters, suggests that the DVA was more predictive of driver behavior in matching lead vehicle speed and distance headway. CONCLUSION: Car-following behavior can be modeled using only visual information to the driver and can produce performance more predictive of driver performance than models based on 3-D (speed or distance) information. APPLICATION: The DVA model has applications to several traffic safety issues, including automated driving systems and traffic flow models.     

Drivers' and non-drivers' performance in a change detection task with static driving scenes: is there a benefit of experience?

The ‘looked-but-failed-to-see’ phenomenon is crucial to driving safety. Previous research utilising change detection tasks related to driving has reported inconsistent effects of driver experience on the ability to detect changes in static driving scenes. Reviewing these conflicting results, we suggest that drivers' increased ability to detect changes will only appear when the task requires a pattern of visual attention distribution typical of actual driving. By adding a distant fixation point on the road image, we developed a modified change blindness paradigm and measured detection performance of drivers and non-drivers. Drivers performed better than non-drivers only in scenes with a fixation point. Furthermore, experience effect interacted with the location of the change and the relevance of the change to driving. These results suggest that learning associated with driving experience reflects increased skill in the efficient distribution of visual attention across both the central focus area and peripheral objects.     

Role of lateral acceleration in curve driving: driver model and experiments on a real vehicle and a driving simulator

Experimental studies show that automobile drivers adjust their speed in curves so that maximum vehicle lateral accelerations decrease at high speeds. This pattern of lateral accelerations is described by a new driver model, assuming drivers control a variable safety margin of perceived lateral acceleration according to their anticipated steering deviations. Compared with a minimum time-to-lane-crossing (H. Godthelp, 1986) speed modulation strategy, this model, based on nonvisual cues, predicts that extreme values of lateral acceleration in curves decrease quadratically with speed, in accordance with experimental data obtained in a vehicle driven on a test track and in a motion-based driving simulator. Variations of model parameters can characterize "normal" or "fast" driving styles on the test track. On the simulator, it was found that the upper limits of lateral acceleration decreased less steeply when the motion cuing system was deactivated, although drivers maintained a consistent driving style. This is interpreted per the model as an underestimation of curvilinear speed due to the lack of inertial stimuli. Actual or potential applications of this research include a method to assess driving simulators as well as to identify driving styles for on-board driver aid systems.     

司机分心驾驶检测研究进展

随着车辆工业和世界经济的快速发展,私家汽车数量不断增加,导致交通事故越来越多,且交通安全问题已经成为全球关注的焦点问题。司机分心驾驶检测的研究主要分为传统计算机视觉（CV）算法和深度学习算法两种。基于传统CV算法的司机分心检测通过尺度不变特征转换（SIFT）、方向梯度直方图（HOG）等特征算子提取图像特征,然后结合支持向量机（SVM）建立模型并对图像进行分类。然而传统CV算法具有对环境的要求高、运用范围较窄、参数多、计算量大的缺点。近年来深度学习在提取数据特征方面表现出速度快、精度高等优异的性能,因此研究人员开始将深度学习引入到司机分心驾驶检测中。基于深度学习的方法可以实现端到端的司机分心驾驶检测网络,而且取得了很高的准确度。介绍了传统CV算法和深度学习算法在司机分心驾驶检测的研究现状,首先,阐释了传统CV算法用于图像领域和司机分心驾驶检测研究的情况;接着,介绍了基于深度学习的司机分心驾驶研究;而后,从准确度、模型参数量等方面对不同司机分心驾驶检测方法进行比较分析;最后,对现有的研究进行了总结并提出了未来司机分心驾驶检测需要解决的三个问题：驾驶过程中司机分心状态以及分心程度划分规范需进一步完善,需要综合考虑人－车－路三者以及如何才能更有效地减少神经网络参数。     

UAV imagery based potential safety hazard evaluation for high-speed railroad using Real-time instance segmentation

Potential safety hazards (PSHs) along the track needs to be inspected and evaluated regularly to ensure a safe environment for high-speed railroad operations. Other than track inspection, evaluating potential safety hazards in the nearby areas often requires inspectors to patrol along the track and visually identify potential threads to the train operation. The current visual inspection approach is very time-consuming and may raise safety concerns for the inspectors, especially in remote areas. Using the unmanned aerial vehicle (UAV) has great potential to complement the visual inspection by providing a better view from the top and ease the safety concerns in many cases. This study develops an automatic PSH detection framework named YOLARC (You Only Look at Railroad Coefficients) using UAV imagery for high-speed railroad monitoring. First, YOLARC is equipped with a new backbone having multiple available receptive fields to strengthen the multi-scale representation capability at a granular level and enrich the semantic information in the feature space. Then, the system integrates the abundant semantic features at different high-level layers by a light weighted feature pyramid network (FPN) with multi-scale pyramidal architecture and a Protonet with residual structure to precisely predict the track areas and PSHs. A hazard level evaluation (HLE) method, which calculates the distance between identified PSH and the track, is also developed and integrated for quantifying the hazard level. Experiments conducted on the UAV imagery of high-speed railroad dataset show the proposed system can quickly and effectively turn UAV images into useful information with a high detection rate and processing speed.     

高速公路场景的车路视觉协同行车安全预警算法

目的 基于视觉的车辆行驶安全性预警分析技术是目前车辆辅助驾驶的一个重要研究方向,对前方多车道快速行驶的车辆进行精准的跟踪定位并建立稳定可靠的安全距离预警模型是当前研究难点。为此,提出面向高速公路场景的车路视觉协同行车安全预警算法。方法 首先提出一种深度卷积神经网络SF_YOLOv4（single feature you look only once v4）对前方车辆进行精准的检测跟踪;然后提出一种安全距离模型对车辆刹车距离进行计算,并根据单目视觉原理计算车辆间距离;最后提出多车道预警模型对自车行驶过程的安全性进行分析,并对司机给予相应安全提示。结果 实验结果表明,提出的SF_YOLOv4算法对车辆检测的准确率为93.55%,检测速度（25帧/s）领先对比算法,有效降低了算法的时间和空间复杂度;提出的安全距离模型计算的不同类型车辆的刹车距离误差小于0.1 m,与交通法建议的距离相比,本文方法计算的安全距离精确度明显提升;提出的多车道安全预警模型与马自达6（ATENZA）自带的前方碰撞系统相比,能对相邻车道车辆进行预警,并提前0.7 s对前方变道车辆发出预警。结论 提出的多车道预警模型充分考虑高速公路上相邻车道中的车辆位置变化发生的碰撞事故;本文方法与传统方法相比,具有较高实用性,其预警效果更加客观,预警范围更广,可以有效提高高速公路上的行车安全。     

高速轴承动力学有限元分析方法

高速轴承是高速列车安全运行的重要部件,轴承运转过程中的动态特性直接影响 轴承的使用寿命,采用最小部件之间创建面面接触的方式,建立高速轨道客车用双列圆锥滚子 轴承的三维虚拟样机模型,利用ANSYS/LS-DYNA 分别研究客车直线匀速行驶和额定速度下以 最小转弯半径行驶两种工况下轴承运转过程的动态接触特性,得到轴承的速度特性、加速特性 以及滚子承载分布状况、接触单元法向作用力时间历程变化曲线以及保持架的振动曲线,研究 结论可以为高速轴承设计过程中模型构建方案的合理确定提供参考。     

Effects of operation type and handle shape of the driver controllers of high-speed train on the drivers' comfort

The design of a high-speed train controller affects the driver's health, operating performance and even safety. Understanding the effects of the design factors on the physical ergonomics of a high-speed train driver controller is essential for optimizing performance and safety. This study experimentally investigated the role of the operation type and handle shape on the physical ergonomics of a driver controller for a high speed train. Two controllers and six handles with pyriform shape, T-shape, sphere shape, cylinder shape and conical frustum shape were used in the experiment. The results indicated that a controller of the sagittal rotation operation type could significantly reduce the workload of the upper limbs compared to a horizontal rotation operation type controller. The handle shape had significant effect on the wrist angles, hand pressures and subjective assessment scores of upper limb fatigue, wrist discomfort and palm discomfort. The handle shape influenced the wrist angles and hand pressures depending on how the participants held the handle. The results demonstrated that the preferred operation type was rotation in the parasagittal plane and that the handle shape should be convenient for operating with a downward-facing palm posture. Among the tested shapes, the pyriform shape and T-shape were considered to be preferable.     

Evaluating the effects of bilingual traffic signs on driver performance and safety

Variable message signs (VMS) can provide immediate and relevant information to road users and bilingual VMS can provide great flexibility in countries where a significant proportion of the population speak an alternative language to the majority. The study reported here evaluates the effect of various bilingual VMS configurations on driver behaviour and safety. The aim of the study was to determine whether or not the visual distraction associated with bilingual VMS signs of different configurations (length, complexity) impacted on driving performance. A driving simulator was used to allow full control over the scenarios, road environment and sign configuration and both longitudinal and lateral driver performance was assessed. Drivers were able to read one- and two-line monolingual signs and two-line bilingual signs without disruption to their driving behaviour. However, drivers significantly reduced their speed in order to read four-line monolingual and four-line bilingual signs, accompanied by an increase in headway to the vehicle in front. This implies that drivers are possibly reading the irrelevant text on the bilingual sign and various methods for reducing this effect are discussed.     

Evaluating the effects of bilingual traffic signs on driver performance and safety

Variable message signs (VMS) can provide immediate and relevant information to road users and bilingual VMS can provide great flexibility in countries where a significant proportion of the population speak an alternative language to the majority. The study reported here evaluates the effect of various bilingual VMS configurations on driver behaviour and safety. The aim of the study was to determine whether or not the visual distraction associated with bilingual VMS signs of different configurations (length, complexity) impacted on driving performance. A driving simulator was used to allow full control over the scenarios, road environment and sign configuration and both longitudinal and lateral driver performance was assessed. Drivers were able to read one- and two-line monolingual signs and two-line bilingual signs without disruption to their driving behaviour. However, drivers significantly reduced their speed in order to read four-line monolingual and four-line bilingual signs, accompanied by an increase in headway to the vehicle in front. This implies that drivers are possibly reading the irrelevant text on the bilingual sign and various methods for reducing this effect are discussed.     

高速磁浮列车多分区追踪算法研究

磁悬浮列车以其高速、节能、环保的特性成为高速轨道交通研究的最新方向。高速磁浮列车多分区双向追踪时间间隔是决定长大干线磁浮交通运能的关键因素。通过对轨道线路数据,车辆运行动态特性,牵引系统牵引制动特性,列车运营方式,列车追踪原理及单个牵引分区内只能运行一辆列车原则的分析,采用VC++语言对列车追踪过程进行了仿真,基于追踪算法确定了双线对向并单线同向多车追踪的最佳间隔时间。仿真结果验证了理论分析及追踪算法的正确性。     

横风下高速列车系统动力学的平衡状态法

基于车辆-轨道耦合动力学和空气动力学提出了一种快速计算横风下高速列车系统动力学行为的平衡状态方法.首先,忽略轨道不平顺并利用流固耦合联合仿真方法计算横风下高速列车的平衡状态;然后,将平衡状态下的气动力加载到车辆-轨道耦合动力学模型并计算高速列车动力学响应.利用建立的平衡状态方法,研究了列车在速度为13.8 m/s的横风下以350 km/h速度运行时的流固耦合动力学行为.比较了平衡状态方法和联合仿真方法两种方法下列车姿态、安全性和舒适性指标的差异,计算结果差别在3.26%以内.研究结果表明:平衡状态方法计算横风下高速列车流固耦合的效率更高.     

基于眼部自商图—梯度图共生矩阵的疲劳驾驶检测

目的 疲劳驾驶是引发车辆交通事故的主要原因之一,针对现有方法在驾驶员面部遮挡情况下对眼睛状态识别效果不佳的问题,提出了一种基于自商图—梯度图共生矩阵的驾驶员眼部疲劳检测方法。方法 利用以残差网络（residual network,ResNet）为前置网络的SSD（single shot multibox detector）人脸检测器来获取视频中的有效人脸区域,并通过人脸关键点检测算法分割出眼睛局部区域图像;建立驾驶员眼部的自商图与梯度图共生矩阵模型,分析共生矩阵的数字统计特征,选取效果较好的特征用以判定人眼的开闭状态;结合眼睛闭合时间百分比（percentage of eyelid closure,PERCLOS）与最长闭眼持续时间（maximum closing duration,MCD）两个疲劳指标来判别驾驶员的疲劳状态。结果 在六自由度汽车性能虚拟仿真实验平台上模拟汽车驾驶,采集并分析驾驶员面部视频,本文方法能够有效识别驾驶员面部遮挡时眼睛的开闭状态,准确率高达99.12%,面部未遮挡时的识别精度为98.73%,算法处理视频的速度约为32帧/s。对比方法1采用方向梯度直方图特征与支持向量机分类器相结合的人脸检测算法,并以眼睛纵横比判定开闭眼状态,在面部遮挡时识别较弱;以卷积神经网络（convolutional neural network,CNN）判别眼睛状态的对比方法2虽然在面部遮挡情况下的准确率高达98.02%,但眨眼检测准确率效果不佳。结论 基于自商图—梯度图共生矩阵的疲劳检测方法能够有效识别面部遮挡时眼睛的开闭情况和驾驶员的疲劳状态,具有较快的检测速度与较高的准确率。     

基于人-车-路-环境综合计算的驾驶员期望车速

微观交通流仿真是智能运输系统研究与开发的重要手段。期望车速是微观交通流仿真研究的一个重要参数,受到驾驶员特性、车辆特性、道路条件、交通干扰、天气和承运任务急缓等多种因素的影响,准确确定期望车速是驾驶员行为研究的难点。从研究驾驶员心理-物理特性的角度出发,利用层次分析法,对驾驶员决策思维的递阶层次进行量化,建立基于人-车-路-环境综合计算的驾驶员期望车速模型。经过实测数据验证,该方法用于驾驶员期望车速模型的研究是可行的。     

Effect of road layout and road environment on driving performance,drivers' physiology and road appreciation

Abstract

Infrastructural changes were implemented on rural 80km/h roads in The Netherlands in an effort to reduce speeding. The road infrastructure changes were designed to produce discomfort for the speeding driver by providing noxious auditory and haptic feedback. On experimental roads, smooth-surface road width was reduced by using blocks of gravel chippings placed along the centre line and at intervals on road edges. It was predicted that these changes would increase mental load while driving, and thereby decrease speeding. In a field experiment 28 subjects drove an instrumented vehicle over experimental and control roads. A decrease in driving speed and swerving behaviour was found on the experimental roads, and this was coupled with a decrease in heart rate variability, consistent with an increase in mental load. Roads in two different road-side environments (woodland vs. moorland) were also tested. There were differences in driver appraisal of the two environments, but no interactions were observed between these appraisals and driving performance on the experimental roads. It is concluded that the infrastructural measures have a useful role to play in road safety through a reduction in driver speeding.     

基于PSO-IGA算法的高性能十字阵列的设计与综合

前碰撞预警系统是安全辅助驾驶领域的一项重要部分,通过计算机处理交通环境信息,当检测到潜在危险时,及时提醒并辅助驾驶员。采用计算机视觉方法,通过目标检测和跟踪算法,获取图像中目标车辆的位置和轨迹信息,并利用相机标定,计算当前车辆和前方车辆在世界坐标系中的距离、速度及轨迹等信息,综合该信息,实现前车碰撞时间预警、前车并线预警以及非机动车预警算法。在前车并线过程中,利用轨迹信息实时检测前车并线意图,及时提示驾驶员注意避让前方车辆。实验表明,本文提出的预警算法具有较高的准确性和鲁棒性,特别在高架或高速道路场景下,并线预警算法能检测到前车的并线意图,及时预警。     

Design and evaluation of a unified chassis control system for rollover prevention and vehicle stability improvement on a virtual test track

This paper describes the development of a unified chassis control (UCC) scheme and the evaluation of the control scheme on a virtual test track (VTT). The UCC scheme aims to prevent vehicle rollover, and to improve vehicle maneuverability and its lateral stability by integrating electronic stability control (ESC) and active front steering (AFS). The rollover prevention is achieved through speed control, and the vehicle stability is improved via yaw rate control. Since the UCC controller always works with the driver, the overall vehicle performance depends not only on how well the controller works but also on its interactions with the human driver. Vehicle behavior and the interactions between the vehicle, the controller, and the human driver are investigated through a full-scale driving simulator on the VTT which consists of a real-time vehicle simulator, a visual animation engine, a visual display, and suitable human–vehicle interfaces. The VTT has been developed and used for the evaluation of the UCC under various realistic conditions in the laboratory making it possible to evaluate the UCC controller in the laboratory without risk of injury prior to field testing, and promises to significantly reduce the cost of development as well as the overall cycle development time.     

基于VR的驾驶仿真及安全教育系统开发

针对全国道路交通事故高发现状及传统驾驶安全教育方式单一、培训效果差的缺点,基于虚拟现实技术（VR）,在引发交通事故人为因素理论基础上,开发驾驶仿真及安全教育系统。系统基于Unity3D引擎,构建了基于道路实景数据的虚拟场景,并联合SUMO实现了道路交通流仿真,通过VR技术仿真驾驶环境及驾驶行为;基于碰撞检测原理,建立了关卡违规触发机制,编码自定义屏幕空间渲染方式模拟驾驶员视觉效果,并构建了基于图像的交通事故现场三维全景,从认知、感知层面培训驾驶员安全驾驶。实用性测试结果表明,系统实现了不同道路场景、气象条件与交通状况下的驾驶模拟及安全培训,增强了使用者的学习兴趣,提高了使用者驾驶安全素养,具有较强的实用性。