不同风险感知效用驾驶人视觉行为及特性分析

发布时间：2018-04-10 14:55

本文选题：风险感知效用 + 视觉行为及特性　；参考：《昆明理工大学》2017年硕士论文

【摘要】：国内外研究表明,在大多数交通事故中,人的因素是最主要的致因,驾驶人行车过程中对道路交通风险感知的有效与否是交通安全的重要环节。道路交通系统中驾驶人80%的信息是通过视觉系统执行,其风险感知尤其重要,因此研究驾驶人视觉行为及特性对于道路交通人因安全研究具有必要性。本课题以不同风险感知能力的驾驶人视觉行为及特性为切入点,基于昆明理工大学驾驶模拟系统平台,应用iViewXTMHED4头戴眼动仪及BeGaze3.5驾驶人眼动数据分析软件,研究不同风险感知效用驾驶人视觉行为及特性的差异性及其与风险感知效用的关系,建立驾驶人视觉行为及特性与风险感知效用关系的模型。具体内容如下:首先,招募31名符合试验要求的驾驶人作为本次试验的被试者,在五个模拟风险驾驶情境下进行风险感知实验,实时记录其模拟驾驶过程中的驾驶行为数据、眼动数据及车辆运行数据,并基于模拟风险驾驶情境完成主观风险度评价主观问卷测评;应用驾驶人风险感知效用理论,对被试驾驶人的风险感知效用进行计算,采用聚类分析方法将31名被试分为保守型、中间型、激进型和复合型等四种风险感知类型。其次,对不同风险感知效用类型驾驶人的视觉行为及特性进行研究。通过BeGaze3.5眼动数据分析软件提取被试者注视、扫视、眨眼、眼动等32项数据;研究确定选择其中8项指标作为本次实验的实验指标,即被试驾驶人的注视次数与时间;扫视速度、幅度与时间..眨眼时间、频率及次数。视觉特性研究发现,被试驾驶人的8项实验指标中,平均注视次数、每次注视时间、扫视幅度、眨眼频率、平均眨眼次数五项指标差异性显著。结果表明,不同风险感知效用类型驾驶人的视觉行为及特性有明显差异。最后,基于计划行为理论构建驾驶人视觉行为及特性与风险感知效用之间的关系模型。在对数据进行量纲一化和扩增样本容量的基础上,应用计划行为理论构建了驾驶人驾驶态度、主观规范、知觉行为控制、注视特性、眨眼特性、扫视特性等12个变量的风险感知效用类型模型框架:选择结构方程模型构建驾驶人眼动行为与驾驶人风险感知效用类型的关系模型,模型分析表明拟合精度良好,可用于解释驾驶人眼动行为、行为意向及风险感知效用类型之间关系,且视觉特性及行为对驾驶行为意向呈现正效应,而行为意向又对风险感知效用类型呈现正效应。研究结果可为驾驶人因安全评估提供新的思路。
[Abstract]:Studies at home and abroad show that in most traffic accidents, the human factor is the most important cause, and the effective perception of road traffic risk is an important part of traffic safety.In the road traffic system, 80% of the information of the driver is executed through the visual system, and its risk perception is particularly important. Therefore, it is necessary to study the visual behavior and characteristics of the driver for the study of human safety in road traffic.Based on the driving simulation system platform of Kunming University of Science and Technology, using iViewXTMHED4 headset eye movement instrument and BeGaze3.5 driver eye movement data analysis software, this paper takes the visual behavior and characteristics of drivers with different risk perception abilities as the breakthrough point, and based on the driving simulation system platform of Kunming University of Technology.This paper studies the differences of visual behaviors and characteristics of drivers with different risk perception utility and their relationship with risk perceived utility, and establishes a model of the relationship between drivers' visual behavior and risk perceived utility.The specific contents are as follows: firstly, 31 drivers who meet the requirements of the experiment were recruited as the subjects of this experiment. The risk-aware experiments were conducted in five simulated risk driving situations, and the driving behavior data of the simulated driving process were recorded in real time.Eye movement data and vehicle operation data, and based on the simulation of risk driving situation to complete the subjective risk evaluation of subjective risk assessment subjective questionnaire, using the driver risk perception utility theory, the test driver risk perception utility calculation.Using cluster analysis, 31 subjects were divided into four risk perception types: conservative type, intermediate type, radical type and compound type.Secondly, the visual behavior and characteristics of drivers with different risk perception utility types are studied.32 items of data such as gaze, scan, blink and eye movement were extracted by BeGaze3.5 software, and 8 of them were selected as the experimental indexes, namely, the number and time of gaze, the speed of scan, and so on.Range and time.Blink time, frequency and frequency.The study of visual characteristics showed that there were significant differences in the average fixation times, fixation time, scan amplitude, blink frequency and average blink frequency among the 8 experimental indexes of the test drivers.The results show that there are significant differences in visual behavior and characteristics among different risk perception utility types.Finally, based on the theory of planning behavior, the relationship between driver's visual behavior, characteristics and risk perception utility is established.Based on the dimensionalization of the data and the expansion of the sample size, the driving attitude, subjective norms, perceptual behavior control, gaze characteristics, blink characteristics of the driver are constructed by using the theory of planned behavior.The risk perception utility type model framework of 12 variables, such as scan characteristics, is selected to construct the relationship model between the driver's eye movement behavior and the driver's risk perception utility type. The model analysis shows that the fitting accuracy is good.It can be used to explain the relationship among driver's eye movement behavior intention and risk perception utility type and the visual characteristics and behavior have positive effect on driving behavior intention and behavioral intention has positive effect on risk perception utility type.The results of the study can provide a new idea for driver's safety assessment.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491.25

【参考文献】