驾驶行为多重分形特征在驾驶疲劳检测中的应用

doi:10.13229/j.cnki.jdxbgxb20200061

摘要/Abstract

摘要：

本文旨在分析驾驶行为多重分形特征对驾驶疲劳检测模型的提升作用。利用UC-win/Road驾驶模拟软件采集行驶速度、加速度、方向盘转角和方向盘角速度等数据，并计算数据的均值、标准差和多重分形特征，比较不同特征的使用是否会对支持向量机（SVM）驾驶疲劳检测模型的精度造成影响。研究表明：在多重分形特征指标中，加速度的奇异强度与驾驶员疲劳状态相关性显著，且受时间窗宽度影响较小；加速度的奇异强度能帮助提高驾驶疲劳检测模型的精度，具有一定的应用价值。

关键词: 道路与铁道工程, 驾驶疲劳, 交通安全, 多重分形特征

Abstract:

Driver fatigue caused by long-term driving is usually accompanied by a decline in driving performance. Therefore， driver fatigue threatens the safety of the drivers and passengers seriously. The development of driver fatigue detection technology can help remind the drivers and take relevant measures against driver fatigue， thereby improving traffic safety to a certain degree. The purpose of this paper is to improve the classification accuracy of the driver fatigue detection model. To achieve this goal， this paper introduces multi-fractal features of the driver behavior data. Six male subjects participated in the driving simulation experiment and UC-win/Road driving simulation software was used to collect driver behavior data such as driving speed， acceleration， steering wheel angle and steering wheel angular velocity. Indicators of the mean， the standard deviation and several different multi-fractal features of each kind of data were calculated. The changes of the drivers' subjective fatigue were also measured with a time interval of 600 s. And the relationship between the indicators and driver fatigue were measured. Additionally， the accuracies of driver fatigue detection models based on support vector machine （SVM） considering different features and different time window were compared. The research shows： The correlation between the singularity strength of acceleration （A₀） and driver fatigue is significant； Compared with the correlation between other indicators and driver fatigue， that between A₀ and driver fatigue is less affected by the time window width； The singularity strength of acceleration can help improve the accuracy of SVM； Compared with the time window of 15 s， a wider time window of 30 s can make the improvement more obvious. Therefore， it has certain application value in driver fatigue detection technology.

Key words: road and railway engineering, driver fatigue, traffic safety, multi-fractal features

中图分类号:

U492.8

张姝玮,郭忠印,杨轸,柳本民. 驾驶行为多重分形特征在驾驶疲劳检测中的应用[J]. 吉林大学学报(工学版), 2021, 51(2): 557-564.

Shu-wei ZHANG,Zhong-yin GUO,Zhen YANG,Ben-min LIU. Application of multi⁃fractal features of driving performance in driver fatigue detection[J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(2): 557-564.

图/表 14

图1

表1

图2

图3

表2

表3

图4

图5

图6

表4

表5

表6

表7

图7

参考文献 21

1	裴玉龙, 马艳丽.疲劳对驾驶员感知判断及操作特性的影响[J]. 吉林大学学报: 工学版, 2009, 39(5) : 1151-1156.
	Pei Yu-long, Ma Yan-li. The effect of fatigue on driver's perception and operating characteristics [J]. Journal of Jilin University (Engineering and Technology Edition), 2009, 39(5): 1151-1156.
2	Phillips R O, Kecklund G, Anund A, et al. Fatigue in transport: a review of exposure, risks, checks and controls[J]. Transport Reviews, 2017, 37(6):742-766.
3	毛科俊, 赵晓华, 房瑞雪, 等. 驾驶疲劳声音对策有效性的驾驶模拟[J]. 吉林大学学报: 工学版, 2010, 40(6) : 1533-1537.
	Mao Ke-jun, Zhao Xiao-hua, Fang Rui-xue, et al. Driving simulation of the effectiveness of driving fatigue sound countermeasures [J]. Journal of Jilin University (Engineering and Technology Edition), 2010, 40(6): 1533-1537.
4	Ting P H, Hwang J R, Doong J L, et al. Driver fatigue and highway driving: a simulator study[J]. Physiology & Behavior, 2008, 94(3): 448-453.
5	毛喆. 基于驾驶员生理特征分析的驾驶疲劳状态识别方法研究[D]. 武汉: 武汉理工大学计算机科学与技术学院, 2006.
	Mao Zhe. Study on the recognition method of driving fatigue state based on the analysis of driver physiological characteristics[D]. Wuhan: School of Computer Science and Technology, Wuhan University of Technology, 2006.
6	黄皓. 基于驾驶操作及车辆状态的疲劳驾驶行为检测研究[D]. 南京: 东南大学交通学院, 2016.
	Huang Hao. Research on fatigue driving behavior detection based on driving operation and vehicle status[D]. Nanjing: College of Transportation, Southeast University, 2016.
7	张晖. 基于驾驶行为的疲劳状态识别研究[D]. 武汉: 武汉理工大学能源与动力工程学院, 2009.
	Zhang Hui. Research on fatigue state recognition based on driving behavior[D]. Wuhan: School of Energy and Power Engineering, Wuhan University of Technology, 2009.
8	吴超仲, 张晖, 毛喆, 等. 基于驾驶操作行为的驾驶员疲劳状态识别模型研究[J]. 中国安全科学学报, 2007, 17(4): 162-165.
	Wu Chao-zhong, Zhang Hui, Mao Zhe, et al. Research on driver fatigue recognition model based on driving operation behavior[J]. Chinese Journal of Safety Science, 2007, 17(4): 162-165.
9	胥川, 裴赛君, 王雪松. 基于无侵入测量指标的个体差异化驾驶疲劳检测[J]. 中国公路学报, 2016, 29(10): 118-125.
	Xu Chuan, Pei Sai-jun, Wang Xue-song. Individual differential driving fatigue detection based on non-intrusive measurement index[J]. China Journal of Highway and Transport, 2016, 29(10): 118-125.
10	金雪. 基于驾驶行为的疲劳驾驶检测方法研究[D]. 北京: 北京工业大学城市交通学院, 2015.
	Jin Xue. Research on fatigue driving detection method based on driving behavior[D]. Beijing: School of Urban Transportation, Beijing University of Technology, 2015.
11	李兆飞, 柴毅, 李华锋. 多重分形去趋势波动分析的振动信号故障诊断[J]. 华中科技大学学报: 自然科学版, 2012, 40(12) : 5-9, 17.
	Li Zhao-fei, Chai Yi, Li Hua-feng. Fault diagnosis of vibration signals based on multi-fractal detrending wave analysis[J]. Journal of Huazhong University of Science and Technology (Natural Science Edition),2012, 40(12): 5-9, 17.
12	胡江, 苏怀智, 马福恒, 等. MF-DFA在大坝安全监测序列分析和整体性态识别中的应用[J]. 水利水电科技进展, 2014, 34(3): 50-55.
	Hu Jiang, Su Huai-zhi, Ma Fu-heng, et al. Application of MF-DFA in dam safety monitoring sequence analysis and overall behavior identification[J]. Advances in Science and Technology of Water Resources and Hydropower, 2014, 34(3): 50-55.
13	Peng C K, Havlin S, Stanley H E, et al. Quantification of scaling exponents and crossover phenomena in nonstationary heartbeat time series[J]. Chaos, 1995, 5(1): 82-87.
14	Kantelhardt J W, Zschiegner S A, Braun P, et al. Multifratal detrended fluctuation analysis of nonstationary time series[J]. Physica A: Statistical Mechanics and its Applications, 2002, 316(1-4):87-114.
15	Larue G S, Rakotonirainy A, Pettitt A N. Driving performance impairment due to hypovigilance on monotonous roads[J]. Accident Analysis and Prevention, 2011, 43(6): 2037-2046.
16	Åkerstedt T, Gillberg M. Subjective and objective sleepiness in the active individual[J]. International Journal of Neuroscience, 1990, 52(1/2): 29-37.
17	Ingre M, Akersted T, Peters B, et al. Subjective sleepiness, simulated driving performance and blink duration: examining individual differences[J]. Journal of Sleep Research, 2006, 15(1): 47-53.
18	Baccarini L M R, Rocha e Silva V V, de Menezes B R, et al. SVM practical industrial application for mechanical faults diagnostic[J].Expert Systems with Applications, 2011, 38(6): 6980-6984.
19	谢宏, 杨硕富, 夏斌, 等. 基于支持向量机的驾驶疲劳脑电特征递增选择算法研究[J]. 生物医学工程学杂志, 2013, 30(6): 1321-1325.
	Xie Hong, Yang Shuo-fu, Xia Bin, et al. Research on incremental selection algorithm of EEG features in driving fatigue based on support vector machine[J]. Journal of Biomedical Engineering, 2013, 30(6): 1321-1325.
20	王振华, 贾银山, 陈兴. 基于SVM的驾驶员疲劳检测研究[J]. 科学技术与工程, 2011(8): 1828-1832.
	Wang Zhen-hua, Jia Yin-shan, Chen Xing. Research on driver fatigue detection based on SVM[J]. Science Technology and Engineering, 2011(8): 1828-1832.
21	王琳虹, 李世武, 高振海, 等. 基于粒子群优化与支持向量机的驾驶员疲劳等级判别[J]. 哈尔滨工业大学学报, 2014, 46(12): 102-107.
	Wang Lin-hong, Li Shi-wu, Gao Zhen-hai, et al. Discrimination of driver fatigue level based on particle swarm optimization and support vector machine[J]. Journal of Harbin Institute of Technology, 2014, 46(12): 102-107.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

驾驶员编号	性别	年龄/岁	驾龄/年
A	男	28	4
B	男	30	7
C	男	33	7
D	男	46	14
E	男	49	21
F	男	55	30

指标	描述	15 s		30 s
指标	描述	均值	标准差	均值	标准差
SM	行驶速度均值/(m·s^-1)	27.769	12.400	27.764	12.441
AM	加速度均值/(m²·s^-1)	0.132	0.213	0.133	0.199
STM	方向盘转角均值	0.011	0.021	0.011	0.021
SVM	方向盘角速度均值	0.000	0.001	0.000	0.001
SSD	行驶速度标准差/(m·s^-1)	0.850	0.719	1.685	1.430
ASD	加速度标准差/(m²·s^-1)	0.079	0.092	0.112	0.095
STSD	方向盘转角标准差	0.002	0.005	0.003	0.005
SVSD	方向盘角速标准差	0.007	0.019	0.009	0.018

时间窗宽度/s	指标	Spearman相关系数
15	SM	0.417**
	AM	0.321**
	STM	0.005
	SVM	-0.093
	SSD	0.398*
	ASD	0.371*
	STSD	0.302*
	SVSD	0.523**
30	SM	0.306
	AM	0.317
	STM	0.019
	SVM	-0.107
	SSD	0.322
	ASD	0.329
	STSD	0.179
	SVSD	0.342

趋势函数	速度		加速度
趋势函数	拟合公式	R²	拟合公式	R²
DFA1	y=0.984x-5.127	0.954	y=0.800x-3.850	0.933
DFA2	y=0.880x-4.944	0.968	y=0.705x-3.687	0.939
DFA3	y=0.767x-4.739	0.984	y=0.585x-3.483	0.950
DFA4	y=0.664x-4.546	0.996	y=0.484x-3.311	0.956
DFA5	y=0.582x-4.390	0.997	y=0.410x-3.183	0.959
DFA6	y=0.532x-4.289	0.993	y=0.359x-3.094	0.959

指标描述	15 s		30 s
指标描述	均值	标准差	均值	标准差
速度序列的奇异强度S₀	1.953	0.182	1.938	0.202
速度序列的广义Hurst指数S₁	0.554	0.079	0.543	0.072
速度序列的标度指数S₂	0.108	0.159	0.086	0.145
加速度序列的奇异强度A₀	1.501	0.393	1.467	0.407
加速度序列的广义Hurst指数A₁	0.395	0.080	0.371	0.071
加速度序列的标度指数A₂	-0.210	0.160	-0.258	0.140