A Method for Identifying Drivers' Risk Perception Based on LightGBM

LI Qing; JING Yunchao; ZHU Tong; ZHU Zishuo; LI Haimei

doi:10.3963/j.jssn.1674-4861.2021.04.003

Volume 39 Issue 4

Aug. 2021

Turn off MathJax

Article Contents

Article Navigation > Journal of Transport Information and Safety > 2021 > 39(4): 16-25

LI Qing, JING Yunchao, ZHU Tong, ZHU Zishuo, LI Haimei. A Method for Identifying Drivers' Risk Perception Based on LightGBM[J]. Journal of Transport Information and Safety, 2021, 39(4): 16-25. doi: 10.3963/j.jssn.1674-4861.2021.04.003

Citation:

LI Qing, JING Yunchao, ZHU Tong, ZHU Zishuo, LI Haimei. A Method for Identifying Drivers' Risk Perception Based on LightGBM[J]. Journal of Transport Information and Safety, 2021, 39(4): 16-25. doi: 10.3963/j.jssn.1674-4861.2021.04.003

Citation:

PDF( 2410 KB)

A Method for Identifying Drivers' Risk Perception Based on LightGBM

doi: 10.3963/j.jssn.1674-4861.2021.04.003

LI Qing^1
,,
JING Yunchao¹,
ZHU Tong^{1
,
,},
ZHU Zishuo¹,
LI Haimei^{2, 3}

1.
College of Transportation, Chang'an University, Xi'an 710064, China
2.
Key Laboratory of Automotive Transportation Safety Enhancement Technology of Ministry of Transport, Chang'an University, Xi'an 710064, China
3.
School of Automobile, Chang'an University, Xi'an 710064, China

Received Date: 2021-03-31

Abstract

Abstract

Accident risk is related to risk perception, and a driving simulation is designed to evaluate the ability of drivers' risk perception considering hazard target' type(explicit/implicit)and amount(single/double). Data of driving behaviors and eye movement characteristics is collected under different risk scenarios. Mantel-Haenszel test is used to analyze differences of hazard target' factors and drivers' characteristics under different risk perception levels. The Spearman correlation test is used to study relationships among driving behaviors, eye movement characteristics, and risk perception ability. The results show that hazard target' factors are negatively correlated with risk perception. Driving age, vehicle speed, longitudinal acceleration, braking depth, braking response time, and response position are significantly related to risk perception. Drivers with poor risk perception would drive at higher speeds with faster acceleration and deeper brake. They need more time to react to emergencies. The evaluation set of risk perception ability is constructed, and the importance of features is ranked using the Random Forest algorithm. The model of risk perception is built based on LightGBM with the effects of different features analyzed. The results show that the identification effect of the model is the best based on LightGBM compared with SVM and AdaBoost. The F1 value reaches 86.07%with an accuracy of 86.14%, which can classify drivers with different risk perception levels.
- traffic safety,
- risk perception,
- hazard target,
- LightGBM algorithm

FullText(HTML)

References(24)

References

[1]	艾倩楠. 基于驾驶员指标的驾驶员风险感知水平评价[J]. 中国安全科学学报, 2018, 28(12): 144-149. https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201812024.htm AI Qiannan. Evaluation of driver risk perception level based on driver indicators[J]. China Safety Science Journal, 2018, 28(12): 144-149. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZAQK201812024.htm
[2]	SIREN A, KJÆR M R. How is the older road users'perception of risk constructed?[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2011, 14(3): 222-228. doi: 10.1016/j.trf.2011.01.002
[3]	HORSWILL M S, MCKENNA F P. Drivers'hazard perception ability: situation awareness on the road[J]. A Cognitive Approach to Situation Awareness: Theory and Application, 2004, 1(9): 155-175. http://www.researchgate.net/profile/Frank_Mckenna2/publication/271767714_Drivers'_hazard_perception_ability_Situation_awareness_on_the_road/links/5665a1c808ae418a786f1af0
[4]	CRUNDALL D, KROLL V. Prediction and perception of hazards in professional drivers: Does hazard perception skill differ between safe and less-safe fire-appliance drivers?[J]. Accident Analysis & amp; Prevention, 2018(121): 335-346. http://www.sciencedirect.com/science?_ob=ShoppingCartURL&_method=add&_eid=1-s2.0-S0001457518301970&originContentFamily=serial&_origin=article&_ts=1526696796&md5=c44700b2eed87a5b83ac6005daee9a50
[5]	张晖, 严新平, 尤志栋, 等. 驾驶人安全意识提升方法与实现[J]. 交通信息与安全, 2011, 29(6): 89-93. doi: 10.3963/j.ISSN1674-4861.2011.06.021 ZHANG Hui, YAN Xinping, YOU Zhidong, et al. Method andrealization of driving safety awareness enhancement[J]. Journal of Transport Information and Safety, 2011, 29(6): 89-93. (in Chinese). doi: 10.3963/j.ISSN1674-4861.2011.06.021
[6]	HORSWILL M S. Hazard perception in driving[J]. Current Directions in Psychological Science, 2016, 25(6): 425-430. doi: 10.1177/0963721416663186
[7]	HORSWILL M S, HILL A, WETTON M. Can a video-based hazard perception test used for driver licensing predict crash involvement?[J]. Accident Analysis & amp; Prevention, 2015(82): 213-219. http://d.wanfangdata.com.cn/periodical/ChlQZXJpb2RpY2FsRW5nTmV3UzIwMjEwMzAyEiA3NGQ4ZGM1YjdkNTFmMjk0MjYwNDkxODVhZWJiYjI0NhoIZzl2MmlyNnk%3D
[8]	BOROWSKY A, SHINAR D, ORON-GILAD T. Age, skill, and hazard perception in driving[J]. Accident Analysis & amp; Prevention, 2010, 42(4): 1240-1249. http://drivingassessment.uiowa.edu/DA2007/PDF/089_BorowskyShinar.pdf
[9]	UNDERWOOD G, CRUNDALL D, CHAPMAN P. Driving simulator validation with hazard perception[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2011, 14(6): 435-446. doi: 10.1016/j.trf.2011.04.008
[10]	王丰元, 陈晓婷. 虚拟试验下驾龄对驾驶人危险感知特性的影响[J]. 现代交通技术, 2018, 15(1): 61-64. doi: 10.3969/j.issn.1672-9889.2018.01.017 WANG Fengyuan, CHEN Xiaoting. Influence of driving age on drivers'danger perception characteristics under virtual experiment[J]. Modern Traffic Technology, 2018, 15(1): 61-64. (in Chinese). doi: 10.3969/j.issn.1672-9889.2018.01.017
[11]	文森, 梁波, 肖尧, 等. 基于反应时间的公路隧道接近段停车视距研究[J]. 交通信息与安全, 2021, 39(2): 43-52. WEN Sen, LIANG Bo, XIAO Yao, et al. Research on stop-up visual distance of highway tunnel approach based on response time[J]. Journal of Transport Information and Safety, 201, 39(2): 43-52. (in Chinese).
[12]	PARK S W, CHOI E S, LIM M H, et al. Association between unsafe driving performance and cognitive-perceptual dysfunction in older drivers[J]. PM & amp; R, 2011, 3(3): 198-203. http://www.sciencedirect.com/science?_ob=ShoppingCartURL&_method=add&_eid=1-s2.0-S1934148210013572&originContentFamily=serial&_origin=article&_ts=1496624436&md5=5c78a9a31976143c935949bdbd1c3dcf
[13]	LIANG B, LIN Y. Using physiological and behavioral measurements in a picture-based road hazard perception experiment to classify risky and safe drivers[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 2018(58): 93-105. http://www.onacademic.com/detail/journal_1000040427726710_1d28.html
[14]	谷志朋, 杨京帅, 楚彭子, 等. 基于危险源位置的驾驶人危险感知研究[J]. 武汉理工大学学报(交通科学与工程版), 2020, 44(5): 789-793. doi: 10.3963/j.issn.2095-3844.2020.05.004 GU Zhipeng, YANG Jingshuai, CHU Pengzi, et al. Study on drivers'danger perception based on danger source location[J]. Journal of Wuhan University of Technology(Traffic Science & amp; Engineering), 2020, 44(5): 789-793. (in Chinese). doi: 10.3963/j.issn.2095-3844.2020.05.004
[15]	ADAMO S H, CAIN M S, MITROFF S R. Self-induced attentional blink: A cause of errors in multiple-target search[J]. Psychological Science, 2013, 24(12): 2569-2574. doi: 10.1177/0956797613497970
[16]	SALL R J, FENG J. Dual-target hazard perception: Could identifying one hazard hinder a driver's capacity to find a second?[J]. Accident Analysis & amp; Prevention, 2019(131): 213-224. http://www.sciencedirect.com/science/article/pii/S0001457518311011
[17]	LI Y, ZHENG Y, WANG J, et al. Crash probability estimation via quantifying driver hazard perception[J]. Accident Analysis & amp; Prevention, 2018(116): 116-125. http://www.onacademic.com/detail/journal_1000039927060210_9f53.html
[18]	TŪSKĖ V, ŠEIBOKAITĖ L, ENDRIULAITIENĖ A, et al. Hazard perception test development for lithuanian drivers[J]. IATSS Research, 2019, 43(2): 108-113. doi: 10.1016/j.iatssr.2018.10.001
[19]	王至恒. 基于车辆运行参数的驾驶危险状态辨识仿真研究[D]. 西安: 长安大学, 2018. WANG Zhiheng. Simulation research on driving hazard state identification based on vehicle operating parameters[D]. Xi'an: Chang'an University, 2018. (in Chinese).
[20]	薛清文, 蒋愚明, 陆键. 基于轨迹数据的危险驾驶行为识别方法[J]. 中国公路学报, 2020, 33(6): 84-94. doi: 10.3969/j.issn.1001-7372.2020.06.008 XUE Qingwen, JIANG Yuming, LU Jian. A method for identification of dangerous driving behavior based on trajectorydata[J]. China Journal of Highway and Transport, 2020, 33(6): 84-94. (in Chinese). doi: 10.3969/j.issn.1001-7372.2020.06.008
[21]	张晖, 肖逸影, 吴超仲, 等. 中国危险品车辆驾驶人驾驶行为影响因素分析[J]. 交通信息与安全, 2020, 38(5): 137-144+153. doi: 10.3963/j.jssn.1674-4861.2020.05.016 ZHANG Hui, XIAO Yiying, WU Chaozhong, et al. Analysis of factors influencing driving behavior of dangerous goods vehicle drivers in china[J]. Journal of Transport Information and Safety, 2020, 38(5): 137-144+153. (in Chinese). doi: 10.3963/j.jssn.1674-4861.2020.05.016
[22]	LONG S, RUOSONG C. Effects of self-assessed ability and driving experience on hazard perception[J]. Journal of Psychological Science, 2016, 39(6): 1346-1352. http://www.psysci.org/CN/article/downloadArticleFile.do?attachType=PDF&id=9791
[23]	HILL A, HORSWILL M S, WHITING J, et al. Computer-based hazard perception test scores are associated with the frequency of heavy braking in everyday driving[J]. Accident Analysis & amp; Prevention, 2019(122): 207-214. http://www.ncbi.nlm.nih.gov/pubmed/30390516
[24]	黄兆国, 过秀成, 贾亮. 基于随机森林的雨天车辆跟驰风险行为研究[J]. 交通信息与安全, 2020, 38(1): 27-34. https://www.cnki.com.cn/Article/CJFDTOTAL-JTJS202001007.htm HUANG Zhaoguo, GUO Xiucheng, JIA Liang. Study on risk behavior of vehicle following in rainy day based on random forest[J]. Journal of Transport Information and Safety, 2020, 38(1): 27-34. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-JTJS202001007.htm

Relative Articles

Supplements(0)

Cited By

Proportional views

Proportional views

通讯作者: 陈斌, bchen63@163.com

1.
沈阳化工大学材料科学与工程学院沈阳 110142

Figures(6) / Tables(9)

Get Citation

PDF

XML

Article Metrics

Article views (842) PDF downloads(50)

A Method for Identifying Drivers' Risk Perception Based on LightGBM

doi: 10.3963/j.jssn.1674-4861.2021.04.003

Abstract

References

Proportional views

Catalog

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Proportional views

Related

A Method for Identifying Drivers' Risk Perception Based on LightGBM

doi: 10.3963/j.jssn.1674-4861.2021.04.003

Abstract

References

Proportional views

Catalog

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Proportional views

Related

Export File

Citation

Format

Content