基于生理信息采集的驾驶疲劳研究

发布时间：2018-10-24 08:29

【摘要】：驾驶疲劳已成为交通事故的主要诱因之一,检测驾驶员的疲劳程度、提前预警,对避免交通事故的发生,提高驾驶员的工作效率,保护驾驶员的健康和安全,具有非常重要的意义和应用价值。论文旨在分别揭示模拟驾驶舱的环境温度、主观评测与心电参数指标及驾驶疲劳状态之间的关系,同时拟将环境因素和主观评测作为数据分析和状态评估的考虑参数,提出一种实用性的驾驶疲劳检测方法。论文首先在简单介绍了驾驶疲劳检测方法分类的基础上,重点分析了基于生理信息的驾驶疲劳检测方法的国内外研究现状。随后通过设计驾驶疲劳基本信息调查问卷和问卷信息收集分析,明确了对驾驶疲劳有明显影响的几大因素,据此设计了相应的驾驶疲劳模拟实验方案,同时基于LABVIEW的开发环境编写了相应的数据采集、存储、处理和特征提取软件程序。在以上工作的基础上,先后分别在三种环境温度(15度、25度和35度)下进行了大量的模拟实验,通过后期对采集到的主观疲劳评测数据和客观心电数据的对比分析,提取出受环境温度影响不明显的8项心电指标(HRV三角形指数、TINN、VLF%、LF%、HF%、LF、HF和LF/HF)。后分别利用主、客观数据对比法和基于指标聚类值的ROC曲线分析的方法,对这8项指标的驾驶疲劳判别阈值进行了初步的确定,随后对这两种方法得到的指标阈值的判定准确度进行了进一步的分析,发现HF%、LF/HF、RR triangular index、LF%和HF这五个指标的判定准确度比较高,疲劳阈值具有一定的可参考性。论文的研究为驾驶疲劳的研究提供了新的思路,在考虑环境温度影响的条件下,驾驶疲劳指示性指标判别阈值的初步确定为后续驾驶疲劳检测和预防方法的研究奠定了良好的基础,同时有利于进一步推进驾驶疲劳研究向实车应用发展。
[Abstract]:Driving fatigue has become one of the main inducements of traffic accidents. To detect the fatigue degree of drivers, to warn them in advance, to avoid traffic accidents, to improve the working efficiency of drivers, and to protect the health and safety of drivers, It has very important meaning and application value. The purpose of the paper is to reveal the relationship between the ambient temperature of the simulated cockpit, subjective evaluation and ECG parameters and driving fatigue state respectively. At the same time, the environmental factors and subjective evaluation are considered as the parameters of data analysis and state assessment. A practical driving fatigue detection method is proposed. On the basis of a brief introduction to the classification of driving fatigue detection methods, this paper focuses on the research status of driving fatigue detection methods based on physiological information at home and abroad. Then, by designing the basic information questionnaire of driving fatigue and collecting and analyzing the information of the questionnaire, several factors that have obvious influence on driving fatigue are identified, according to which the corresponding experimental scheme of driving fatigue simulation is designed. At the same time, the software program of data acquisition, storage, processing and feature extraction is written based on LABVIEW development environment. On the basis of the above work, a large number of simulation experiments were carried out at three ambient temperatures (15 degrees, 25 degrees and 35 degrees), respectively. Through the comparison and analysis of subjective fatigue evaluation data and objective ECG data collected in the later period, Eight ECG indexes (HRV triangle index, TINN,VLF%,LF%,HF%,LF,HF and LF/HF) which were not affected by ambient temperature were extracted. After that, by using subjective and objective data comparison method and ROC curve analysis method based on index clustering value, the threshold of driving fatigue discrimination of these eight indexes is preliminarily determined. Then the accuracy of the two methods is further analyzed. It is found that HF%,LF/HF,RR triangular index,LF% and HF have higher accuracy and fatigue threshold has some referential value. The research in this paper provides a new way of thinking for the study of driving fatigue under the condition of considering the influence of environmental temperature. The preliminary determination of the threshold value of driving fatigue indicator index lays a good foundation for the further study of the methods of driving fatigue detection and prevention, and is conducive to the further development of driving fatigue research towards the application of real vehicles.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.6;TP274

【相似文献】