Spatial Regression Analysis of Pedestrian Crashes Based on Point-of-Interest Data

Pedestrian safety has recently been considered as one of the most serious issues in the research of traffic safety. This study aims at analyzing the spatial correlation between the frequency of pedestrian crashes and various predictor variables based on open source point-of-interest (POI) data which can provide specific land use features and user characteristics. Spatial regression models were developed at Traffic Analysis Zone (TAZ) level using 10,333 pedestrian crash records within the Fifth Ring of Beijing in 2015. Several spatial econometrics approaches were used to examine the spatial autocorrelation in crash count per TAZ, and the spatial heterogeneity was investigated by a geographically weighted regression model. The results showed that spatial error model performed better than other two spatial models and a traditional ordinary least squares model. Specifically, bus stops, hospitals, pharmacies, restaurants, and office buildings had positive impacts on pedestrian crashes, while hotels were negatively associated with the occurrence of pedestrian crashes. In addition, it was proven that there was a significant sign of localization effects for different POIs. Depending on these findings, lots of recommendations and countermeasures can be proposed to better improve the traffic safety for pedestrians.

Comparative Analysis of Fatal Pedestrian Crashes between Kansas and USA

Due to the increasing trend of population growth and urbanization, pedestrians form one of the largest single road user groups. However, they are the most neglected group among all road users. Pedestrian safety is now a growing concern in the USA. Identifying the factors associated with fatal pedestrian crashes plays a key role in developing efficient and effective strategies to enhance pedestrian safety. This study addresses safety issues by identifying contributory factors associated with fatal pedestrian crashes in Kansas and the USA. For Kansas, the study uses KARS (Kansas Accident Reporting System) database while for the USA FARS (Fatality Analysis Reporting System) database has been used. Different variables considered in this study are human variables (age, and gender), environmental variables (atmospheric condition and light condition), time (time of day, day of week, and crash month), location (intersection vs. mid-block), and roadway variables (speed limit). Different factors that are found to have an association with fatal pedestrian crashes are male pedestrians, older pedestrians, weekend, off peak hours, winter months, dark hours, non-intersection, clear atmospheric conditions, higher speed limit. The findings from Kansas have been compared with that from the USA. This study helps to implement potential countermeasures by identifying the factors that have an association with fatal pedestrian crashes.

Pedestrian Crash Prediction Models and Validation of Effective Factors on Their Safety (Case Study: Tehran Signalized Intersections)

The quantity and severity of traffic accidents have increased with the development of machinery life and traffic growth in cities and roads in the past 50 years. Among the road users, pedestrians are the most vulnerable groups to be exposed to high risks. Vehicle crashes with pedestrian are almost inevitable and cause injury or death to pedestrian. Crash investigation and statistical studies indicate that percentage of pedestrian deaths caused by vehicle accidents are much more than all deaths. A considerable amount of accidents occur at signalized and urban intersections which are the intensive crash places. Therefore in this paper appropriate models that could specify safety indicators have been indicated with existing information by characterized parametric and nonparametric variables for twenty signalized intersections. Categories and correlations of variables also have been investigated. Three models including Regression, Poisson, and Negative binomial with defined variables have been determined. T and chi square tests, calibration and comparison of variables have been done by curve fitting. The role of each parameter was specified in pedestrian crashes. Validating models had the following outcomes: Pedestrian crash prediction models were based on none linear relations at intersections. Predictable variables, developing extended linear models and also pedestrian crash prediction are on the basis of Negative binomial distribution which is used due to more data dispersion. As observed, the Negative binomial regression because of its more R2 correlation factor has more validity among other regression models such as linear regression and Poisson. Calibrated models are put into sensitivity analysis to study the effect of each previously mentioned parameter in overall performance. Hence much better perception of future transportation plans can be achieved by development of safety models at planning levels.

Partial proportional odds model for analyzing pedestrian crashes,threshold heterogeneity by scale and proportional odds factor

Despite low traffic in Wyoming,pedestrian crash severity accounts for a high number of fatalities in the state.Thus this study was conducted to highlights factors contributing to those crashes.The results highlighted that drivers under influence,type of vehicle,location of crashes,estimated speed of vehicles,driving over the recommended speed are some of factors contributing to the severity of crashes.In this study,we used proportional odds model which assumes that the impact of each attribute is consistent or proportional across various threshold values.However,it has been argued that this assumption might be unrealistic,especially at the presence of extreme values.Thus,the assumption was relaxed in this study by shifting the thresholds based on some explanatory attributes,or proportional odds effects.In addition,we accounted for the spread rate,or scale,of the model’s latent distribution of pedestrian crashes.The results highlighted that the partial proportional odds model through proportional odds factor and scale effects result in a significant improvement in model fit compared with the standard proportional odds model.Comparisons were also made across standard normal,simple partial ordinal model,and partial ordinal accounting for scale heterogeneity.In addition,various potential threshold structures such as symmetric and flexible were considered,but similar goodness of fits were observed across all those models.Extensive discussion has been made regarding the formulation of the implemented methodology,and its implications.

Application of machine learning tools in classifying pedestrian crash types: A case study

Crash occurrence is a complex phenomenon,and crashes associated with pedestrians and bicyclists are even more complex.Furthermore,pedestrian-and bicyclist-involved crashes are typically not reported in detail in state or national crash databases.To address this issue,developers created the Pedestrian and Bicycle Crash Analysis Tool(PBCAT).However,it is labour-intensive to manually identify the types of pedestrian and bicycle crash from crash-narrative reports and to classify different crash attributes from the textual content of police reports.Therefore,there is a need for a supporting tool that can assist practitioners in using PBCAT more efficiently and accurately.The objective of this study is to develop a framework for applying machine-learning models to classify crash types from unstructured textual content.In this study,the research team collected pedestrian crash-typing data from two locations in Texas.The XGBoost model was found to be the best classifier.The high prediction power of the XGBoost classifiers indicates that this machine-learning technique was able to classify pedestrian crash types with the highest accuracy rate(up to 77%for training data and 72%for test data).The findings demonstrate that advanced machine-learning models can extract underlying patterns and trends of crash mechanisms.This provides the basis for applying machine-learning techniques in addressing the crash typing issues associated with non-motorist crashes.

车人碰撞中人地接触损伤仿真及防护方法鲁棒性分析

人车事故中行人主要与车辆和地面接触而遭受损伤。此前已有大量研究在车辆接触阶段对多体行人模型进行了验证,但对地面接触阶段的验证较少。本文中评估了用于预测车辆撞击后行人运动和地面接触的4个PC-Crash行人模型。通过车辆和地面接触的HIC伤害与最近研究中的6个尸体实验数据对比显示,其中PC2014行人模型能很好地预测行人头部损伤,车辆和地面接触HIC的平均误差分别为6.07%和5.85%,数据说明PC2014行人模型可以用于再现人车碰撞事故以及开发未来地面接触伤害防护对策。控制制动防护方法鲁棒性研究显示,在3种行人姿态(奔跑、步行和应急)扰动下,HIC平均降低比例分别为63.2%、57.9%和67.8%,说明控制制动防护方法具有很好的抗行人姿态干扰能力。进一步分析不同姿态下的行人损伤发现,3种步态序列之间均有显著性差异,在应急步态下控制制动防护方法有着更好的抗行人姿态干扰能力;而在奔跑步态序列下,往往可以通过控制制动防护方法产生最低的头地碰撞损伤,表明在使用控制制动防护方法对行人进行保护时须考虑到不同姿态产生的影响,以进一步提高控制制动防护方法的地面伤防护效果。

基于PC-Crash的车辆行人事故深入研究

本文依托于中国交通事故深入研究项目(CIDAS),选取CIDAS数据库中的2起典型车辆碰撞行人案例进行事故重建研究。运用多刚体动力学法建立事故车辆的仿真模型,利用PC-Crash深入分析车辆与行人的碰撞过程,得出碰撞车速,研究碰撞后行人的运动响应和各损伤部位的动力学响应,最后将仿真结果和人体损伤的医学鉴定结果进行对比,验证了利用PC-Crash仿真方法重建车辆碰撞行人事故的准确性。

基于Pc-Crash软件的人-车碰撞事故仿真规律研究

以Pc-Crash软件为平台,仿真分析两例实际的人-车碰撞试验。通过对比仿真结果与试验的吻合情况,并结合其他学者所提及的现象,总结了仿真中人-车接触部位、行人终止点位置及抛距、行人被车辆抛出后的运动情况等与实际情况的耦合程度的规律。发现利用Pc-Crash软件实现对人-车碰撞事故仿真时,人-车接触部位与实际情况可以耦合得很好,行人终止点位置一般总体抛距与实际情况差不多,而行人被车辆抛出后的运动情况则与实际不符,针对这些现象提出一些建议。对更好地利用Pc-Crash软件实现对人-车碰撞事故的仿真分析有一定的意义。

基于PC-Crash的车-人碰撞事故仿真与分析

基于多刚体动力学仿真分析软件PC-Crash,建立车-人碰撞事故的再现模型,分别对同一工况下汽车有无ABS刹车防抱死制动系统、不同碰撞初速度以及利用随机函数确定汽车与行人的随机参数,进行大量的数字化仿真试验,分析事故发生后人体头部重要器官与车体前部碰撞点以及损伤程度的分布规律。根据分析结果,从汽车生产厂商保护行人的角度改善车身布置结构,提出有价值的参考意见。

基于PC-Crash的人车事故再现研究

文章对一起人车相撞的事故进行分析,建立了车辆模型、道路模型、行人模型,运用PC-CRASH软件进行了分析计算,结合事故勘查时的痕迹,行人的受伤情况,再现了事故发生时人车的接触部位和行驶行走方向,为更准确地进行人车交通事故鉴定提供了一种方法。

基于Pc-Crash的道路交通碰撞事故再现分析

针对3起不同类型的交通事故,根据现场采集的事故参与方信息,利用Pc-Crash软件建立多刚体动力学模型,对事故过程进行再现分析,从而确定事故参与者在碰撞前的形态.涉及道路弱势群体参与的碰撞类型,事故再现主要集中在车辆损坏信息与人员伤亡信息的交互融合性;涉及到车辆大面积损坏的碰撞类型,事故再现主要集中在碰撞角度、制动反应时间、转向角度等碰撞序列、车辆运行轨迹与事故现场采集信息的一致性.

基于PC Crash的摩托车-行人碰撞事故分析重建

摩托车-行人事故的碰撞姿态鉴定一直是交通事故鉴定的一个难点。由于摩托车的灵活性,当事故发生在没有监控摄像头覆盖的区域时,应用传统的交通事故鉴定方法很难甚至不可能判断出摩托车-行人事故中碰撞瞬间摩托车的行驶状态,而且国内外也鲜有学者及专家对摩托车事故进行分析研究。基于1例摩托车-行人碰撞的交通事故,在应用传统的事故鉴定方法基础之上,利用PC Crash仿真软件对事故进行仿真分析。通过不断仿真调试,明确了人-车碰撞瞬间的关键姿态,最终重建事故发生的整个过程。此外,在PC Crash仿真软件对事故再现的结果中,摩托车碰撞前车身的具体形态可作为判断碰撞前摩托车是否有避让行人这一关键线索。

基于PC-Crash的人—车碰撞事故再现结果误差分析

基于PC-Crash软件研究车辆与行人碰撞交通事故的三维再现结果的误差来源和影响因素。结合实际案例的模拟过程,通过不断调整参数以达到理想的准确度,分析出对模拟结果影响较大的关键参数,为以后的再现研究提供指导和借鉴,同时验证使用PC-Crash再现人车碰撞事故的可行性与准确性。

基于PC-Crash的机动车与行人交通事故鉴定

以天津市近年小客车与行人交通事故的案例为依据,分析了行人碰撞交通事故中人体受损程度及相关部位,对小客车与行人碰撞时车辆行驶速度进行计算;基于PC-Crash再现仿真平台,对车辆与人体碰撞事故形态的仿真再现方法进行分析总结,细化了小客车事故碰撞形态、多刚体损伤模型及车速变化等信息。结果表明:软件仿真结果与交通事故现场勘查结果一致,为PC-Crash软件人车事故仿真分析提供了理论基础。

行人-车辆碰撞中六岁儿童下肢损伤分析及预测

为揭示汽车-行人碰撞事故中车辆前端结构和速度对儿童行人下肢损伤的差异成因及损伤机理,本研究应用符合我国体征且具有详细解剖学结构的六岁儿童行人损伤仿生模型(TUST IBMs 6YO-P),设置了32组涵盖4类常见车型和8种碰撞速度的汽车-行人碰撞仿真,分析下肢运动学和生物力学响应,采用非线性回归建立预测模型对下肢损伤进行评估。结果表明,碰撞速度对儿童下肢长骨骨折和膝关节损伤有直接影响,汽车前保险杠高度影响股骨的损伤程度和膝关节弯曲角度,扰流板离地高度影响胫、腓骨损伤的严重程度。分析膝关节弯曲角度和韧带断裂数据,得出当膝关节弯曲角度分别超出25.9±0.9°、38.6±0.7°、43.2±0.2°时,撞击侧内侧副韧带(MCL)、前交叉韧带(ACL)、后交叉韧带(PCL)发生断裂,基于碰撞速度和前保险杠高度建立的膝关节弯曲角度预测模型,经反求验证该模型可以有效预测膝关节损伤。综合儿童行人下肢长骨骨折评价参数和所建立的小腿弯矩预测模型,发现碰撞速度高于17.94 km/h将会造成儿童小腿骨折。该研究为行人安全法规制定、行人保护装置研发、AEB系统设计和数字化测评提供科学的参考依据。

典型汽车碰撞事故场景中行人运动轨迹预测方法

为提高未来自动驾驶车辆对弱势道路使用群体的感知和决策融合的可靠性,本文提出一种基于目标检测算法(YOLOv5)、多目标跟踪算法(Deep-Sort)和社交长短时记忆神经网络(social-long short-term memory,Social-LSTM)的行人未来运动轨迹预测方法。结合YOLOv5检测和Deep-Sort跟踪算法,有效解决行人检测跟踪过程中目标丢失问题。提取特定行人目标历史轨迹作为预测框架的输入边界条件,并采用Social-LSTM预测行人未来运动轨迹。并对未来运动轨迹进行透视变换和直接线性变换,转换为世界坐标系中的位置信息,预测车辆与行人的可能未来碰撞位置。结果显示目标检测精度达到93.889%,平均精度均值达96.753%,基于高精度的检测模型最终轨迹预测算法结果显示,预测损失随着训练步长的增加呈递减趋势,最终损失值均小于1%,其中平均位移误差降低了18.30%,最终位移误差降低了51.90%,本研究可为智能车辆避撞策略开发提供理论依据和参考。

行人与大型机动车事故严重程度影响因素分析

行人是最易受伤的交通参与者,而大型机动车是应急灵活性差、碰撞事故冲击能量大的交通工具,因此剖析潜在因素对行人与大型机动车事故严重程度的影响非常重要。基于深圳市2015—2016年2 890条行人与大型机动车事故数据,分析其事故严重程度与时空分布特性。从人、车、路、环境、时间、空间、行为7个方面,选取22个潜在影响因素,建立偏比例优势模型得到19个显著因素,结合其边际效应量化分析各因素对该类事故严重程度的差异性影响。结果表明,8个因素不遵守平行线假设,影响最大的前5个因素为驾驶人驾龄、驾驶人年龄、违法行为、碰撞车速、天气状况,其边际效应绝对值的最大值超过14.5%,表明人、车、环境、行为等动态因素对事故的影响更明显;行人年龄、事故时间、驾驶人性别、道路类型、慢性干扰度、光线亮度的影响次之(<13%)。据此差异性影响提出一些改善建议以提升道路交通安全水平。

汽车碰撞行人交通伤害特点分析

旨在研究道路交通事故中行人损伤的发生特点、事故类型以及损伤机制,从而调查交通事故中行人、 车辆以及环境三者之间的关系。事故分析对行人的年龄分布、性别、损伤部位以及严重程度、车辆类型进行了研 究,对典型案例进行了详细分析,并进行了AIS、GCS分级。讨论了行人损伤的事故特点,提出了影响行人损伤的 因素,讨论了车辆交通环境和道路使用者之间的关系。

轿车-行人碰撞事故再现模型参数敏感性研究

为确定对轿车行人事故再现结果有显著影响的参数,在动力学分析的基础上建立了轿车与行人的碰撞模型,筛选出影响事故再现结果的一般敏感性参数;再运用单因素试验方法对一般敏感性参数进行试验,进一步筛选出对再现结果影响较大的敏感性参数;在此基础上采用多因素分析法将敏感性参数按敏感度大小进行排序,参照所得结果对一起真实的车人碰撞事故进行再现仿真,所得结果与实际相吻合。最终表明,对再现结果有显著影响的参数为轿车碰撞前速度、驾驶员反应时间、减速度、轿车转弯偏角和接触位置,针对性地调整这些参数,能提高事故再现的精度。

行人膝关节韧带的建模及验证研究

基于人体模型THUMS中的膝韧带模型,改进了材料参数和单元属性。采用近年较新的生物力学实验数据,逐个验证韧带模型在不同应变率下的生物逼真度。结果表明,膝韧带ACL、PCL、MCL和LCL在不同应变率下的拉力与位移关系曲线与试验结果吻合较好,说明模型具有较好的生物逼真度,能较准确地模拟韧带的损伤和拉伸响应。