基于机器学习的高速公路大型货车追尾风险预测

针对高速公路大型货车追尾事故频发的问题，评估高速公路大型货车追尾风险，并分析交通流特性对大型货车追尾风险的影响，以降低追尾事故的发生率。根据德国HighD开源数据集，以不同冲突风险等级的碰撞时间(TTC)阈值作为大型货车冲突风险的划分标准，提取大型货车的车辆轨迹与交通特征参数等数据，基于随机森林(RF)、支持向量机(SVM)和人工神经网络(ANN)等3种机器学习模型分别建立高速公路大型货车追尾风险实时预测模型；以混淆矩阵、受试者工作特征曲线的曲线下面积(AUC)和洛伦兹(KS)检验等评价指标，对比分析各模型的整体预测能力，并选取预测精度最好的模型分析各个特征参数对追尾风险的影响程度。研究结果表明：RF模型的预测准确率达75%,相对SVM模型高出8%,相对ANN模型高出10%,且RF模型的预测精确度、召回率、AUC值和KS值均优于SVM模型和ANN模型；最小车头间距、车速标准差和加速度标准差3个参数对大型货车追尾风险影响程度最高。     

浅谈高速公路上大型货车违法占道行驶的管理对策

高速公路上大型货车违法占道行驶行为的管理,是道长期困扰交警部门的难题。对此类违法行为的管理,不仅需要交警等多部门通力协作,形成合力,还需要各部门集思广益,多措并举,才能最终实现高速公路的安全与畅通。     

浅谈高速公路上大型货车违法占道行驶的管理对策

高速公路上大型货车违法占道行驶行为的管理,是道长期困扰交警部门的难题。对此类违法行为的管理,不仅需要交警等多部门通力协作,形成合力,还需要各部门集思广益,多措并举,才能最终实现高速公路的安全与畅通。     

不同区域高速公路货车事故特征研究

为深入探究我国不同区域高速公路货车事故特征的差异性,解决货车事故高发问题,应用分类学和社会网络分析等方法,分别选取南方的浙江省和北方的吉林省两地主要货车事故数据,初步分类分析后,构建事故影响因素关系网络图和中心度排序表,对比2组数据,结果表明:浙江、吉林两地货车事故在数量、严重程度、原因以及复杂程度等方面存在较大差异;气象条件、车辆类型和疲劳驾驶是影响事故严重程度的重要因素;不同区域、天气和驾驶经验对司机的不安全驾驶行为有较大影响。     

考虑均值异质性的高原山区公路责任事故严重度风险因素辨识

为明晰高原山区高速公路事故严重度的影响因素及差异，量化事故数据异质性对事故严重度的交互影响效应，基于云南省某高原山区高速公路550起有责交通事故数据，从责任驾驶员、责任车辆、道路特征和责任事故特性4个方面选取潜在特征变量，采用固定参数有序Probit模型和随机参数有序Probit模型，构建考虑均值异质性的高原山区高速公路责任事故的严重度分析模型，并采用边际弹性系数量化分析不同影响因素对事故严重度的影响程度。结果表明，缓和曲线、有其他妨碍安全驾驶的行为、轻型货车、重型货车、多车、下坡以及曲线路段7个因素显著影响责任事故严重度。“驾驶员存在其他妨碍安全驾驶的行为”和“重型货车”的估计系数分别是服从(0.375, 0.201)和(0.531, 0.988)正态分布的随机变量，总体导致死亡及重伤事故的概率分别增大6.15%和9.61%;当驾驶员在曲线路段存在其他妨碍安全驾驶的违法行为时，重伤或死亡事故概率增大6.54%。当重型货车在缓和曲线路段发生碰撞时，重伤或死亡事故概率会降低8.37%,重型货车因未与前车保持安全距离发生重伤或死亡事故概率会降低9.95%,而当重型货车与小客车碰撞时，重伤或...     

班组安全

高速公路上的交通事故，其中70％的事故涉及到路产损失，但是目前此类事故的处理仍然存在许多不规范的地方。有这样一起交通事故：一部运蔬菜的小货车在高速公路行驶中由于操作不当造成车辆侧翻，车辆倒地后刮擦路面，同时漏油污损路面造成路产损失。事故所在地的路政大队给小货车驾驶员出具车辆暂扣凭证及路产损失评估单，路政大队暂扣小货车，要求驾驶员赔偿路产损失3万元。在这起交通事故中，路政大队能否暂扣车辆？路产损失评估是否合法？     

高速公路安全水平仿真分析

为深入研究高速公路安全系统,用系统动力学(SD)方法分析驾驶员、车辆、道路、应急救援等组成要素以及环境、管理等影响因素之间的关系,建立高速公路安全系统因果关系模型和存量流量模型。用层次分析法(AHP)确定各影响因素的权重。以三福高速公路为例,用Vensim软件模拟高速公路运营期间系统安全的变化情况。结果表明,在该高速公路运营期间,系统安全与车辆安全度呈先上升后下降的趋势、道路安全度呈下降趋势、驾驶员安全度与应急救援能力呈上升趋势。通过与实际数据对比,验证了模型的有效性。     

法理与情理

我上路检查车辆,发现某某货车安全设施不全,我告诉驾驶员,这要罚款两百元。驾驶员向我求情,强烈要求我少罚一点,最好是别罚。我说不行,执法要公正。驾驶员下跪,说昨天刚跟老婆离婚,家里都揭不开锅了,全身负债累累，     

满载火药货车爆炸成碎片

4月12日凌晨5时10分,一辆装载数吨炸药的河北籍中型货车行驶至连霍高速民权段双塔镇境内时,发生大爆炸,现场2人死亡,4人重伤。高速公路路面被炸出一个20多平方米的大坑,深近1米,周围100平方米的路面被炸开多条裂缝,附近近千户房屋受损,民权、兰考县城均有震感……     

货车高速路上爆胎 险!

元月8日15：40左右，一部车牌号为闽A-7P137的厢式货车，在沈海高速公路B道323＋780h处因左前轮爆胎，像一匹脱僵的野马冲过中央隔离带后，侧翻在A道快车道上，满载的瓷砖洒落在路面上，现场一片狼藉……     

大货车驾驶员交通心理与交通安全的研究

笔者从交通心理学的角度,对大货车驾驶员在行驶过程中的心理状况、因攻击性驾驶行为、强烈的冒险动机的驱使、对道路期望心理的失衡及交通安全感偏差等不良心理因素而导致恶性交通事故进行分析。通过研究得出以下结论大货车驾驶员是一特殊的群体,应重视对该群体的教育、培训与管理,从而提高驾驶员的职业道德水平、性格品质、交通安全感及驾驶技能,以改善道路安全环境和交通安全状况、降低道路交通事故发生率。     

A study of Chinese truck drivers’ attitudes toward feedback by technology

Prior research indicates that trucks are very likely to be involved in accidents in China due to the poor condition of trucks and unsafe driving behaviors. The objectives of this study were (1) to investigate the safe/unsafe driving behaviors of Chinese truck drivers, and (2) to examine whether Chinese truck drivers recognize a need for feedback and what their attitudes are in terms of receiving feedback from both technology and from individuals to improve driving safety. Eight sessions of focus group discussions were carried out with Chinese truck drivers, middle-level managers in the freight transportation business, and traffic policemen. The results showed that the most important safety issues to Chinese truck drivers are speeding and overloading. Chinese truck drivers would like to receive more feedback. Yet feedback is considered negative in nature. Although some Chinese truck drivers are afraid of not being able to operate the in-vehicle technology, in general, they are more confident with feedback by in-vehicle technology because it is considered objective and scientific. For this reason this kind of feedback is preferred by some over feedback from a human.     

高速公路交通安全现状分析及管理对策

为了分析高速公路交通安全现状,提出高速公路交通安全综合管理对策,采用对比分析法,研判当前高速公路交通安全形势,统计分析高速公路交通事故特征规律,提出基于“6E”的高速公路交通安全系统管理对策。以1994～2013年期间的高速公路交通事故统计数据为分析对象,采用事故起数、事故死亡人数、事故率、致死率、重大事故起数等作为评价指标,对比交通事故总量和不同等级公路事故情况,分析了当前我国高速公路交通安全形势。从交通安全执法管理的角度,对高速公路交通事故发生时间、事故形态、肇事车辆类型、事故原因、肇事驾驶人驾龄等进行了特征分析,在立法、教育、工程、执法、急救、评价6个方面,提出了高速公路交通安全管理系统化对策,为进一步认识当前高速公路交通安全形势,完善高速公路交通安全管理提供技术支持。     

Motor vehicle towing: An analysis of injuries in a high-risk yet understudied industry

Objectives: National fatality rates for commercial tow truck operators exceed those of other first responders who also perform traffic incident management services. The objectives of the current study are to (1) characterize causal factors associated with injuries among commercial tow truck operators engaged in roadside assistance through analysis of coded and free text data obtained from U.S. Occupational Safety and Health Administration (OSHA) investigation files, and (2) utilize supplemental data sources to analyze environmental factors for injuries in which commercial tow truck operators were struck by roadway traffic. Methods: Searches of OSHA’s online IMIS database were performed to identify investigations of incidents in which tow truck operators were injured while performing roadside assistance duties. Freedom of Information Act (FOIA) requests were submitted to obtain full investigation files for each case. Coded and narrative text analyses were performed to identify causal themes across the identified cases. Results: One-hundred and six cases of tow truck operators being killed or severely injured were identified in IMIS; 41 FOIA requests for related investigation documents were fulfilled. Two major event type themes were identified which accounted for 9 in 10 of the cases identified. These were (1) ‘struck-by’ incidents, which were primarily injuries resulting from contact with roadway traffic, rolling vehicles and equipment or other non-motorized objects; and (2) ‘caught-in or -between’ incidents, which were primarily injuries resulting from being pinned beneath and between vehicles and being caught in moving parts. Conclusions: The towing industry should provide initial and refresher safety training on vehicle loading and unloading, defensive techniques when exposed to traffic on roadways, and proper wheel chocking and braking procedures. States should include tow trucks as a first responder vehicle type in their “Move Over” laws and implement public awareness campaigns to protect all first responders, including tow truck operators.     

新驾驶员反应特性分析

新驾驶员是交通事故的频发人群,并与其驾驶反应特性有关。笔者对96名(男75名,女21名)新驾驶员(驾龄≤3年)的反应时进行测试,与一定样本量(83名)的对照组相应指标作比较,其结果表明:对于简单反应时间,由于信息复杂度低,受试组和对照组未能表现出显著差异;当加大信息复杂度时(3光复合刺激),两组间反应时差异较为显著(p<0.01),即新驾驶员的反应能力较对照组差;应当有针对性地对新驾驶员进行反应时训练以增加其驾车安全性。     

驾驶员风险认知能力对交通安全的影响

为辨别不同驾驶员风险认知能力对驾驶安全的影响。通过设置13种存在潜在风险的交通情景,引入惩罚用时机制,利用驾驶模拟器,对熟练驾驶员和非熟练驾驶员在风险认知培训前后的风险认知能力进行比较,找出不同驾驶员对风险认知的差异。试验结果显示,熟练驾驶员的风险认知能力比非熟练驾驶员要高,两者在引起交通风险的因素上是不同的,培训后的风险认知能力较之前有很大提高。培训能够拓宽驾驶员对潜在风险的认知程度,更好地将驾驶员的操作技能、知识体系和外在行为衔接在一起,从而减少交通事故的发生。     

On-Scene Injury Severity Prediction (OSISP) Algorithm for Truck Occupants

Objective: The aim of this study is to develop an on-scene injury severity prediction (OSISP) algorithm for truck occupants using only accident characteristics that are feasible to assess at the scene of the accident. The purpose of developing this algorithm is to use it as a basis for a field triage tool used in traffic accidents involving trucks. In addition, the model can be valuable for recognizing important factors for improving triage protocols used in Sweden and possibly in other countries with similar traffic environments and prehospital procedures.

Methods: The scope is adult truck occupants involved in traffic accidents on Swedish public roads registered in the Swedish Traffic Accident Data Acquisition (STRADA) database for calendar years 2003 to 2013. STRADA contains information reported by the police and medical data on injured road users treated at emergency hospitals. Using data from STRADA, 2 OSISP multivariate logistic regression models for deriving the probability of severe injury (defined here as having an Injury Severity Score [ISS] > 15) were implemented for light and heavy trucks; that is, trucks with weight up to 3,500 kg and ??16,500 kg, respectively. A 10-fold cross-validation procedure was used to estimate the performance of the OSISP algorithm in terms of the area under the receiver operating characteristic curve (AUC).

Results: The rate of belt use was low, especially for heavy truck occupants. The OSISP models developed for light and heavy trucks achieved cross-validation AUC of 0.81 and 0.74, respectively. The AUC values obtained when the models were evaluated on all data without cross-validation were 0.87 for both light and heavy trucks. The difference in the AUC values with and without use of cross-validation indicates overfitting of the model, which may be a consequence of relatively small data sets. Belt use stands out as the most valuable predictor in both types of trucks; accident type and age are important predictors for light trucks.

Conclusions: The OSISP models achieve good discriminating capability for light truck occupants and a reasonable performance for heavy truck occupants. The prediction accuracy may be increased by acquiring more data. Belt use was the strongest predictor of severe injury for both light and heavy truck occupants. There is a need for behavior-based safety programs and/or other means to encourage truck occupants to always wear a seat belt.     

驾驶员注意力分配定量方法研究

驾驶员在不同的路况环境下存在不同的注意力分配模式,为弄清其分配模式特点,利用露天体育场内作为模拟驾驶环境,并利用眼动设备,记录被试者在直道驾驶任务、弯道驾驶任务、复杂路况驾驶任务等3种路况下的眼动数据,采用划分兴趣区域的方式,把视觉信息源划分为原车道兴趣区域FL(Former Lane)和其他兴趣区域OA(Other Areas)两个兴趣区域,对平均百米注视次数、平均百米注视时间、不同兴趣区域注视次数百分比3个眼动指标在两个兴趣区域内的分布情况进行分析和对比,得出不同路况环境下不同的注意力分配模式来揭示驾驶员驾驶过程中的注意力分配特点。研究结果对于驾驶员驾驶培训侧重点的改进、对于预防交通事故都具有重要意义。     

基于速度特征指标的高速公路线形一致性综合评价

为量化评价已建和拟建高速公路交通安全性,以实测运行车速数据和事故率数据为基础,根据公路空间属性的变化,采用不定长法进行路段划分。考虑不同路段车速协调性和连续性的需求,引入客货差极差比、车速降低系数、路段车速离散度等车速特征指标。通过对运行车速特征指标与事故率的相关性分析,采用多元回归的方法建立了高速公路线形设计一致性综合评价模型,同时结合事故多发点改善效果的显著性,确定我国高速公路安全性服务水平,在此基础上提出线形设计一致性评价标准和评价方法。太旧高速公路的实例应用结果表明,该方法可准确有效地评价线形设计一致性。     

高湿天气的时空变化及其对交通安全的影响

利用1951—1998年40多年的气象资料,根据每日08时的空气相对湿度值超过95%的出现频率,分析了京珠高速等沿线高湿天气的时空变化特点以及该时空分布特征给车辆运行和驾驶带来的影响问题。同时指出:高湿天气具有明显的区段性差异和季节性差异;该差异将会使驾驶环境变得复杂,给车辆的行驶和驾驶人员的心理带来很不利的影响,直接影响着交通安全。还根据分析结论提出了减少安全事故发生的具体措施。