[1]刘瑞,贺经纬,朱西产,等.基于自然驾驶数据的跟车场景潜在危险估计模型[J].东南大学学报(自然科学版),2019,49(4):788-795.[doi:10.3969/j.issn.1001-0505.2019.04.024]
 Liu Rui,He Jingwei,Zhu Xichan,et al.Potential risk assessment model in car following based on naturalistic driving data[J].Journal of Southeast University (Natural Science Edition),2019,49(4):788-795.[doi:10.3969/j.issn.1001-0505.2019.04.024]
点击复制

基于自然驾驶数据的跟车场景潜在危险估计模型()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
49
期数:
2019年第4期
页码:
788-795
栏目:
交通运输工程
出版日期:
2019-07-20

文章信息/Info

Title:
Potential risk assessment model in car following based on naturalistic driving data
作者:
刘瑞1贺经纬2朱西产1马志雄1
1同济大学汽车学院, 上海 201804; 2上汽大众汽车有限公司前瞻研究与智能驾驶研发部, 上海 201805
Author(s):
Liu Rui1 He Jingwei2 Zhu Xichan1 Ma Zhixiong1
1School of Automotive Studies, Tongji University, Shanghai 201804, China
2Department of Research & Engineering Driver Assistance Systems, SAIC Volkswagen Automotive Co., Ltd., Shanghai 201805, China
关键词:
主动安全 跟车场景 危险估计 潜在危险 驾驶行为
Keywords:
active safety car following risk assessment potential risk driver behavior
分类号:
U461.91
DOI:
10.3969/j.issn.1001-0505.2019.04.024
摘要:
为了改善主动安全系统的性能,提出一种跟车场景潜在危险估计模型.将前车制动时为避免碰撞留给本车制动之前的时间作为跟车场景潜在危险估计指标,即时间裕度.将TTC(time to collision)定义为明显危险估计指标.使用自然驾驶数据来确定时间裕度中的目标车和本车的减速度.为了得到真实可信的驾驶员制动行为特性,讨论了驾驶员制动行为的收敛性.使用核密度估计来描述驾驶员的纵向加速行为,使用相对熵来表征2个不同数据集之间的差异.使用稳定收敛的数据集提取了驾驶员制动行为的特征参数,并根据这些特征参数得到了考虑驾驶员制动极限的时间裕度.最后使用跟车危险工况制动开始时刻的时间裕度构建了跟车场景潜在危险估计模型.分析稳定跟车工况中的明显危险和潜在危险表明,潜在危险估计模型可以描述两车相对速度较小情况的危险等级.真实跟车危险工况验证发现,使用潜在危险估计模型可以更早地发现危险发生的可能.
Abstract:
A potential risk assessment model is proposed to improve the performance of the active safety system. The time before the driver of the host vehicle has to brake to avoid the collision when the target vehicle brakes is defined as a potential risk measure, i.e. time margin. Whereas time to collision(TTC)is defined as an obvious risk assessment measure. The naturalistic driving data was employed to determine the brake deceleration of the host vehicle and target vehicle in the time margin. To obtain the authentic braking behavior characteristics, the convergence of the braking behavior of the driver was discussed. The kernel density estimation was used to describe the accelerating behavior of the driver. And the relative entropy(Kullback-Leibler divergence)was applied to represent the distinctions between two different datasets. The braking behavior characteristic parameters were extracted by using the dataset from which a convergent braking behavior can be obtained. And, the time margin considering the limitation of the driver was obtained based on the braking behavior of the driver. Finally, the time margins at the braking starting time in the dangerous car following scenarios were employed to construct a potential risk assessment model. Analyses of the potential risk and obvious risk in the steady car following scenarios show that the potential risk assessment model can evaluate the risk level when the relative speed between the host vehicle and the target vehicle is small. Verifications based on the dangerous car following scenarios reveal that the potential risk assessment model can help to detect the possibility of the danger earlier.

参考文献/References:

[1] National Highway Traffic Safety Administration. Traffic safety facts 2015: A compilation of motor vehicle crash data from the fatality analysis reporting system and the general estimates system[R]. Washington, DC: National Highway Traffic Safety Administration,2017.
[2] Zhang Y, Antonsson E K, Grote K.A new threat assessment measure for collision avoidance systems[C]// IEEE Intelligent Transportation Systems Conference. Toronto, ON, Canada, 2006: 968-975.DOI:10.1109/itsc.2006.1706870.
[3] Lee K, Peng H. Evaluation of automotive forward collision warning and collision avoidance algorithms[J].Vehicle System Dynamics, 2005, 43(10): 735-751. DOI:10.1080/00423110412331282850.
[4] Brannstrom M, Coelingh E, Sjoberg J. Model-based threat assessment for avoiding arbitrary vehicle collisions[J].IEEE Transactions on Intelligent Transportation Systems, 2010, 11(3): 658-669. DOI:10.1109/tits.2010.2048314.
[5] Eidehall A, Petersson L. Statistical threat assessment for general road scenes using Monte Carlo sampling[J]. IEEE Transactions on Intelligent Transportation Systems, 2008, 9(1): 137-147. DOI:10.1109/tits.2007.909241.
[6] Tan H S, Huang J H. DGPS-based vehicle-to-vehicle cooperative collision warning: Engineering feasibility viewpoints[J]. IEEE Transactions on Intelligent Transportation Systems, 2006, 7(4): 415-428. DOI:10.1109/tits.2006.883938.
[7] Labayrade R, Royere C, Aubert D. A collision mitigation system using laser scanner and stereovision fusion and its assessment[C]//IEEE Proceedings of Intelligent Vehicles Symposium.Las Vegas, NV, USA, 2005: 441-446. DOI:10.1109/ivs.2005.1505143.
[8] Moon S, Yi K. Human driving data-based design of a vehicle adaptive cruise control algorithm[J].Vehicle System Dynamics, 2008, 46(8): 661-690. DOI:10.1080/00423110701576130.
[9] Kiefer R J, LeBlanc D J, Flannagan C A. Developing an inverse time-to-collision crash alert timing approach based on drivers’ last-second braking and steering judgments[J].Accident Analysis & Prevention, 2005, 37(2): 295-303. DOI:10.1016/j.aap.2004.09.003.
[10] 吴斌, 朱西产, 沈剑平, 等. 基于自然驾驶研究的直行追尾危险场景诱导因素分析[J]. 同济大学学报(自然科学版), 2018, 46(9): 1253-1260.DOI:10.11908/j.issn.0253-374x.2018.09.013.
Wu B, Zhu X C, Shen J P, et al. Analysis of causation of rear-end incidents based on naturalistic driving study[J]. Journal of Tongji University(Natural Science), 2018, 46(9): 1253-1260. DOI:10.11908/j.issn.0253-374x.2018.09.013. (in Chinese)
[11] Mulder M, Abbink D A, van Paassen M M, et al. Design of a haptic gas pedal for active car-following support[J].IEEE Transactions on Intelligent Transportation Systems, 2011, 12(1): 268-279. DOI:10.1109/tits.2010.2091407.
[12] Brackstone M, McDonald M. Driver headway: How close is too close on a motorway?[J].Ergonomics, 2007, 50(8): 1183-1195. DOI:10.1080/00140130701318665.
[13] Kondoh T, Yamamura T, Kitazaki S, et al. Identification of visual cues and quantification of drivers’ perception of proximity risk to the lead vehicle in car-following situations[J].Journal of Mechanical Systems for Transportation and Logistics, 2008, 1(2): 170-180. DOI:10.1299/jmtl.1.170.
[14] 李霖, 贺锦鹏, 刘卫国, 等. 基于驾驶员紧急制动行为特征的危险估计算法[J]. 同济大学学报(自然科学版), 2014, 42(1): 109-114. DOI:10.3969/j.issn.0253-374x.2014.01.018.
Li L, He J P, Liu W G, et al. Threat assessment algorithm based on characteristics of driver emergency braking behavior[J]. Journal of Tongji University(Natural Science), 2014, 42(1): 109-114. DOI:10.3969/j.issn.0253-374x.2014.01.018. (in Chinese)
[15] Lu G Q, Cheng B, Lin Q F, et al. Quantitative indicator of homeostatic risk perception in car following[J].Safety Science, 2012, 50(9): 1898-1905. DOI:10.1016/j.ssci.2012.05.007.
[16] Shalev-Shwartz S, Shammah S, Shashua A. On a formal model of safe and scalable self-driving cars[J]. arXiv, 2017:1708.06374.
[17] Sokolovskij E. Experimental investigation of the braking process of automobiles[J].Transport, 2005, 20(3): 91-95. DOI:10.3846/16484142.2005.9638002.
[18] Brach R M, Rudny D F, Sallmann D W. Comparison of tire friction test methodologies used in accident reconstruction[C]//SAE Technical Paper Series. Warrendale, PA, USA: SAE International, 1998. DOI:10.4271/980367.
[19] Bosetti P, da Lio M, Saroldi A. On curve negotiation: from driver support to automation[J].IEEE Transactions on Intelligent Transportation Systems, 2015, 16(4): 2082-2093. DOI:10.1109/tits.2015.2395819.
[20] Butakov V, Ioannou P. Personalized driver/vehicle lane change models for ADAS[J].IEEE Transactions on Vehicular Technology, 2015, 64(10): 4422-4431. DOI:10.1109/tvt.2014.2369522.
[21] Bishop C M. Pattern recognition and machine learning[M]. New York: Springer, 2006: 122-123.
[22] Jones M C, Marron J S, Sheather S J. Progress in data-based bandwidth selection for kernel density estimation[J]. Computational Statistics, 1996, 11(3): 337-381.
[23] Silverman B W. The kernel method for multivariate data[M]//Density Estimation for Statistics and Data Analysis. Boston, MA, USA: Springer US, 1986: 75-94. DOI:10.1007/978-1-4899-3324-9_4.
[24] Botev Z I, Grotowski J F, Kroese D P. Kernel density estimation via diffusion[J].The Annals of Statistics, 2010, 38(5): 2916-2957. DOI:10.1214/10-aos799.
[25] Hall P, Marron J S, Park B U. Smoothed cross-validation[J].Probability Theory and Related Fields, 1992, 92(1): 1-20. DOI:10.1007/bf01205233.
[26] MacKay D J C. Information theory, inference and learning algorithms[M]. Cambridge University Press, 2003: 36-37.
[27] Wang W S, Liu C, Zhao D. How much data are enough? A statistical approach with case study on longitudinal driving behavior[J]. IEEE Transactions on Intelligent Vehicle, 2017, 2(2): 85-98. DOI:10.1109/tiv.2017.2720459.

备注/Memo

备注/Memo:
收稿日期: 2018-12-04.
作者简介: 刘瑞(1989—),男,博士生;朱西产(联系人),男,博士,教授,博士生导师,xczhu@163.com.
基金项目: 国家重点研发计划资助项目(2016YFB0100904-2).
引用本文: 刘瑞,贺经纬,朱西产,等.基于自然驾驶数据的跟车场景潜在危险估计模型[J].东南大学学报(自然科学版),2019,49(4):788-795. DOI:10.3969/j.issn.1001-0505.2019.04.024.
更新日期/Last Update: 2019-07-20