基于生理信号的驾驶疲劳综合评价指标的试验研究

发布时间：2018-11-14 11:18

【摘要】：蓬勃发展的汽车工业给我们的生活带来了舒适、快捷,但由之而造成的负面影响也是显而易见的,道路交通事故的增多给人们的生活带来了很大的危害。在导致道路交通事故的众多因素中,疲劳驾驶是其中的主要原因之一,疲劳驾驶直接导致道路交通事故几乎达到事故总数的20%。因此,对驾驶疲劳进行研究,寻找预防和缓解驾驶疲劳的方法,对交通事故的预防和控制具有非常重要的意义。本文主要进行了如下研究工作:(1)探讨了驾驶疲劳的影响因素和形成机理,详细总结了国内外基于生理指标测试的驾驶疲劳研究方法和现状。根据目前国内外驾驶疲劳生理测试研究指标心率、血压、呼吸、体温、脑电、肌肉电阻、胃肠电、诱发电位、血氧饱和度、唾液等,并结合本研究的实验室具体条件、实验可操作性、易操作性、实验要求等因素,最终确定心电、脑电、呼吸频率、体温、肌肉电阻五种生理信号作为本次试验的测试指标。(2)构建了驾驶疲劳生理指标测试驾驶模拟平台。设计搭建驾驶模拟平台,进行模拟驾驶中生理指标的测试。该系统主要包括CS310t-1型汽车综合模拟驾驶系统,以模拟真实驾驶操作环境;美国进口MP150多导生理仪,包括用于放大生理信号的放大器、感知人体生物电信号的传感器及用于数据分析的自带计算机软件Acqknowledge。(3)设计了驾驶疲劳生理指标测试试验流程,并进行试验的实施操作。选择20名在校学生作为试验对象,按照试验流程,应用多导生理仪测试被试连续50分钟驾驶任务过程中的心电、脑电、呼吸频率、体温、肌肉电阻五种生理信号,计算机记录并保存所得试验数据。(4)通过生理仪自带软件Acqknowledge进行信号处理,进而对所得试验数据进行分析和探讨。首先介绍了一般线性回归分析和逐步线性回归分析的数学模型,调用R软件,针对两种回归方法进行了分析比较。从分析结果可知,基于一般线性回归分析得到的驾驶疲劳综合指标方程中出现许多变量不显著,而基于逐步线性回归分析法得到的驾驶疲劳综合指标方程避免该缺点,最后利用回归验证和可靠性验证说明了该驾驶疲劳综合指标能够很好的反映驾驶员的疲劳状态。(5)基于主成分回归分析的驾驶疲劳综合指标的确定。首先介绍了主成分回归分析的数学模型,调用R软件,通过主成分回归分析对该模型进行了回归分析,得到了驾驶疲劳综合指标方程,对得到的回归方程进行回归验证和可靠性验证,说明了得到的驾驶疲劳综合指标能够有效地反映驾驶员的驾驶疲劳状态。
[Abstract]:The booming automobile industry has brought comfort and speed to our life, but the negative effects caused by it are also obvious. The increase of road traffic accidents has brought great harm to people's lives. Among the many factors leading to road traffic accidents, fatigue driving is one of the main reasons. Fatigue driving directly leads to almost 20% of the total number of road traffic accidents. Therefore, it is of great significance for the prevention and control of traffic accidents to study driving fatigue and find ways to prevent and alleviate driving fatigue. The main work of this paper is as follows: (1) the influencing factors and formation mechanism of driving fatigue are discussed, and the research methods and present situation of driving fatigue based on physiological index testing are summarized in detail. According to the research indexes of driving fatigue physiological test at home and abroad, such as heart rate, blood pressure, respiration, body temperature, EEG, muscle resistance, gastrointestinal electricity, evoked potential, blood oxygen saturation, saliva and so on. Experimental maneuverability, ease of operation, experimental requirements and other factors, the final determination of ECG, EEG, respiratory frequency, body temperature, Five physiological signals of muscle resistance are used as the test indexes. (2) A driving simulation platform for driving fatigue physiological index testing is constructed. Design and build driving simulation platform to test physiological indexes in simulated driving. The system mainly includes CS310t-1 integrated simulation driving system to simulate the real driving environment. The MP150 multichannel physiological instrument imported by the United States, including an amplifier for amplifying physiological signals, sensors for sensing human bioelectrical signals and a self-contained computer software Acqknowledge. (3) for data analysis, has designed the test process for driving fatigue physiological indicators. And the implementation of the test operation. Twenty students were selected as the subjects. According to the experimental procedure, the electrocardiogram (ECG), electroencephalogram (EEG), respiratory frequency, body temperature and muscle resistance of the subjects during the driving task for 50 minutes were measured. The computer records and saves the experimental data. (4) signal processing is carried out by the physiological instrument software Acqknowledge, and then the experimental data are analyzed and discussed. Firstly, the mathematical models of general linear regression analysis and stepwise linear regression analysis are introduced, and the two regression methods are analyzed and compared by using R software. The results show that many variables are not significant in the comprehensive index equation of driving fatigue based on general linear regression analysis, but the comprehensive index equation of driving fatigue based on stepwise linear regression method avoids this shortcoming. Finally, regression verification and reliability verification are used to show that the driving fatigue comprehensive index can well reflect the driver's fatigue state. (5) the determination of driving fatigue comprehensive index based on principal component regression analysis. Firstly, the mathematical model of principal component regression analysis is introduced, and the comprehensive index equation of driving fatigue is obtained by means of principal component regression analysis. Regression verification and reliability verification of the obtained regression equation show that the obtained comprehensive index of driving fatigue can effectively reflect the driver's driving fatigue state.
【学位授予单位】：河南理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U491.254

【相似文献】