基于驾驶模拟实验的道路作业区驾驶员汇合行为及风险研究

发布时间：2018-02-23 13:05

本文关键词： 道路作业区汇合行为驾驶模拟器避免碰撞时间汇合风险模型　出处：《北京交通大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着交通基础设施建设进程的加快,由于改扩建、维修施工作业而产生的道路作业区越来越多。道路作业区的出现不仅极大地降低了道路通行能力,造成交通拥堵,而且使车流被迫产生分流、合流、跟驰等复杂的驾驶行为,带来交通安全隐患。另外,驾驶员作为交通系统中最活跃又最不可控的因素,在作业区这样复杂的驾驶环境中其行为极易产生较大波动,一旦未能及时采取相应驾驶措施,很有可能造成严重的交通事故。因此,道路作业区的驾驶行为研究是作业区交通安全研究的重要组成部分。本文以驾驶员在道路作业区的汇合行为为研究重点,通过驾驶模拟器建立不同汇合标志位置和不同车流密度的模拟场景,采集不同性别、不同职业人员的实验数据,利用基本统计分析、方差分析等方法对不同条件下的驾驶员行为规律进行分析,其中主要分析汇合过程的行为规律。最后,从汇合间距、汇合时距和避免碰撞时间(TTC,Time To Collision)三个角度进行汇合风险分析,并以TTC为交通安全评价指标,考虑驾驶员自身特性变量、场景变量和汇合过程变量,对汇合过程中的风险建立Logistic回归模型。在道路作业区驾驶员行为规律的研究中发现,在汇合过程中会伴随着车辆速度和加速度的明显变化,而汇合过程中速度、加速度会受车流密度和汇合标志位置的影响。不同特性的驾驶员汇合过程中驾驶行为会有所不同,驾驶员的性别和职业对汇合时间、汇合距离有显著性影响,一般非职业驾驶员和女性驾驶员要低于职业驾驶员和男性驾驶员。同时,以汇合间距和汇合时距为指标的汇合风险分析得出,车流密度、汇合标志位置对汇合过程中的最小间距和时距存在显著性影响,车流密度越大,汇合间距越小,汇合时距越小,也就意味着发生碰撞的风险越大。在汇合风险模型的建立中,本文将汇合过程中的最小TTC划分为严重冲突、一般冲突、轻微冲突和无冲突四个等级,通过建立多元有序Logistic回归模型发现,驾驶员性别、职业、车流密度、前车速度方差、主车速度方差、主车加速度方差和车头间距均值对汇合过程中的危险等级存在直接的显著性影响,男性驾驶员、非职业驾驶员、高密度车流条件下汇合风险较大,而随着速度方差、加速度方差的增大汇合风险会增大,随着车头间距均值增加,汇合风险会减小。汇合标志位置虽然对汇合风险没有直接的影响,但其对汇合过程中的车头间距存在影响,因此其也间接地影响着汇合风险。
[Abstract]:With the acceleration of traffic infrastructure construction process, more and more road operation areas are generated by maintenance and construction work due to reconstruction and extension. The emergence of road operation areas not only greatly reduces the capacity of roads, resulting in traffic congestion. In addition, drivers are the most active and uncontrollable factors in the traffic system. In such a complex driving environment as the operation area, its behavior is prone to fluctuate greatly. Once the corresponding driving measures are not taken in time, it is very likely to cause serious traffic accidents. The study of driving behavior in road operation area is an important part of traffic safety research in operation area. This paper focuses on the convergence behavior of driver in road operation area. The simulation scene of different confluence sign position and different traffic density was established by driving simulator, and the experimental data of different sex and different profession were collected, and the basic statistical analysis was used. Variance analysis and other methods are used to analyze the behavior of drivers under different conditions, in which the behavior of convergence process is mainly analyzed. The confluence risk analysis is carried out from three angles of confluence time interval and collision avoidance time to Collision.Using TTC as the evaluation index of traffic safety, drivers' own characteristic variables, scene variables and confluence process variables are considered. The Logistic regression model is established for the risk in the course of confluence. In the study of the behavior law of the driver in the road operation area, it is found that the vehicle speed and acceleration change obviously during the confluence process, and the speed in the confluence process. The acceleration will be affected by the traffic density and the position of the confluence sign. The driving behavior of drivers with different characteristics will be different. The driver's sex and occupation have significant influence on the meeting time and distance. In general, non-professional drivers and female drivers are lower than professional drivers and male drivers. The minimum distance and time interval in the confluence process are significantly affected by the position of the confluence sign. The greater the traffic density, the smaller the confluence distance, and the smaller the confluence time interval, which means the greater the risk of collision. In this paper, the minimum TTC in confluence process is divided into four grades: serious conflict, general conflict, slight conflict and no conflict. By establishing multiple ordered Logistic regression model, it is found that driver sex, occupation, traffic density, speed variance in front car, and so on. The main vehicle speed variance, the acceleration variance and the average distance between the main vehicle and the vehicle have significant effects on the dangerous grade in the confluence process. Male drivers, non-professional drivers, and high-density vehicle flow have a greater risk of convergence, and the risk of convergence is higher than that of the male drivers, the non-professional drivers, and the high-density vehicle flow. However, with the increase of velocity variance and acceleration variance, the convergence risk will increase, and the convergence risk will decrease with the increase of the average distance between the vehicle heads. Although the convergence sign location has no direct effect on the convergence risk, However, it has an effect on the front distance in the course of confluence, so it also affects the convergence risk indirectly.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491.25

【参考文献】