- 无标题文档
查看论文信息

中文题名:

 

考虑驾驶人特性的行车风险短时预测方法研究

    

姓名:

 吴静成    

学号:

 1049722003809    

保密级别:

 公开    

论文语种:

 chi    

学科代码:

 082300    

学科名称:

 工学 - 交通运输工程    

学生类型:

 硕士    

学校:

 武汉理工大学    

院系:

 交通与物流工程学院    

专业:

 交通运输工程    

研究方向:

 交通信息与安全    

第一导师姓名:

 吕能超    

第一导师院系:

 智能交通系统研究中心    

完成日期:

 2023-05-26    

答辩日期:

 2023-05-20    

中文关键词:

 

交通工程 ; 行车风险 ; 驾驶人特性 ; 短时预测 ; LGBM

    

中文摘要:

先进驾驶辅助系统可获取大量自身和周边环境等精细化信息,为行车风险辨识和预警提供数据支撑,同时驾驶人是“人-车-路-环境”交通系统的核心,驾驶人特性直接影响行车安全水平,因此,基于车载感知数据和驾驶人特性及早准确地感知风险是预警系统的关键。本文通过试验平台采集行车过程中所涉及的人-车-路-环境等各方面数据,选取合适模型辨识风险发生前某项或某组合数据的异常变动,结合驾驶人特性尽早且准确预测行车风险,为车载安全系统或驾驶人留足充足反应时间,同时通过对风险辨识及预测结果分析,为安全行车提供指导意见,提高行车安全水平。本文主要内容如下:

(1)自然驾驶试验设计与数据处理。本文设计实车试验数据采集平台,开展92人次实车驾驶试验,采集包含行车环境、驾驶人操纵信息和车辆运行状态信息等数据。针对实车数据存在的数据丢帧、数据异常及存在噪声等问题,进行相应的补帧、修复和降噪处理,总结并计算行车风险研究中常见的替代安全指标(Surrogate Safety Measures, SSM),形成丰富且准确的自然驾驶数据集。

(2)行车风险辨识方法研究。针对车辆运动学特征,在建立假设条件的基础上进行蒙特卡洛仿真模拟,依据仿真结果进行SSM风险阈值标定工作;总结并参考以往风险辨识方法,选取碰撞时间倒数(Time-to-collision inverse, TTCi)和制动减速度(Braking deceleration, BD)两个指标作为风险事件辨识依据,结合视频认定最终事件;统计安全和风险状态下各指标的分布形态,分析行车风险特性。

(3)驾驶人行车风险及避险特性研究。针对驾驶人特性,主要研究不同性别、经验和驾驶风格驾驶人在面对行车风险时的反应特性及风险特性。结合驾驶人操纵信息和TTCi变化规律,确定反应时间和三种典型避险反应策略提取标准,进一步利用Mann-Whitney U和Kruskal-Wallis H两种非参数检验方法分析不同驾驶人及不同反应策略下反应时间和风险特性是否存在显著性差异。结果表明,驾驶人特性对行车风险有显著性影响。

(4)行车风险短时预测方法研究。通过构建包含风险标签和人、车、路及环境在内的车载数据集进行数据聚合和集成,利用可变滑动窗口和下采样方法构建行车风险短时预测建模的样本集;结合4种机器学习方法探索风险演变规律;使用轻度提升机(Light Gradient Boosting Machine, LGBM)和Boruta结合方法用于行车风险短时预测模型中的特征筛选,并结合驾驶人特性进行预测算法改进,预测模型测试在不同数据源下表现情况,并使用SHAP分析方法分析不同特征变量对模型预测结果的影响,以及两两特征之间的相会影响。结果表明,最佳风险预测窗口提前长度及窗口长度分别为1.6 s和1.2 s;改进的LGBM算法(Consider Driver Characteristics-LGBM, CDC-LGBM)具有最好F1值,达到89.03%,该方法能够更好的适用与行车风险短时预测,能够很好的满足未来ADAS系统的风险预测需求;同时使用不同的数据源进行预测效果存在差异,包含三者全部信息的数据预测效果对于分别预测有十分明显的提升。

参考文献:

[1] 交通运输部. 2020年交通运输行业发展统计公报[R]. 北京: 中华人民共和国交通运输部, 2020.

[2] Wang J, H Huang, Li Y, et al. Driving risk assessment based on naturalistic driving study and driver attitude questionnaire analysis[J]. Accident Analysis & Prevention, 2020, 145: 105680.

[3] Wang L, Abdel-Aty M, Lee J. Safety analytics for integrating crash frequency and real-time risk modeling for expressways[J]. Accident Analysis & Prevention, 2017, 104: 58-64.

[4] Wu Y, Abdel-Aty M, Lee J. Crash risk analysis during fog conditions using real-time traffic data[J]. Accident Analysis & Prevention, 2017, 114: 4-11.

[5] 陈亮. 基于驾驶人交通特性的汽车驾驶模拟器系统有效性研究[D].昆明理工大学, 2021.

[6] Xie K, Yang D, Ozbay K, et al. Use of real-world connected vehicle data in identifying high-risk locations based on a new surrogate safety measure[J]. Accident Analysis & Prevention, 2019, 125: 311-319.

[7] Kim Y, Park J, Oh C. A Crash Prediction Method Based on Artificial Intelligence Techniques and Driving Behavior Event Data[J]. Sustainability, 2021, 13.

[8] Guo M, Zhao X, Yao Y, et al. A study of freeway crash risk prediction and interpretation based on risky driving behavior and traffic flow data[J]. Accident Analysis & Prevention, 2021, 160(1): 106328.

[9] Zhao P, Lee C. Assessing rear-end collision risk of cars and heavy vehicles on freeways using a surrogate safety measure[J]. Accident; analysis and prevention, 2018, 113: 149.

[10] Li P, Abdel-Aty M, J Yuan. Real-time crash risk prediction on arterials based on LSTM-CNN[J]. Accident Analysis & Prevention, 2019, 135.

[11] Yu R, Abdel-Aty M. Investigating the different characteristics of weekday and weekend crashes[J]. Journal of Safety Research, 2013, 46: 91-97.

[12] Yang K, Wang X, Yu R. A Bayesian dynamic updating approach for urban expressway real-time crash risk evaluation[J]. Transportation research, 2018, 96: 192-207.

[13] Sun J, Sun J. A dynamic Bayesian network model for real-time crash prediction using traffic speed conditions data[J]. Transportation Research Part C, 2015, 54: 176-186.

[14] Wu Y, Abdel-Aty M, Park J, et al. Effects of Connected-Vehicle Warning Systems on Rear-End Crash Avoidance Behavior Under Fog Conditions[J]. Transportation research, 2018, 95: 481-492.

[15] Qsa B, Tfa B, Sla B. Investigating rear-end collision avoidance behavior under varied foggy weather conditions: A study using advanced driving simulator and survival analysis[J]. Accident Analysis & Prevention, 2020, 139: 105499.

[16] Qsa B, Tfa B, Jwa B, et al. An integrated methodology for real-time driving risk status prediction using naturalistic driving data[J]. Accident Analysis & Prevention, 2021, 156: 106122.

[17] 朱西产, 魏昊舟, 马志雄. 基于自然驾驶数据的跟车场景潜在风险评估[J]. 中国公路学报, 2020, 33(4): 13.

[18] 张香. 基于自然驾驶的驾驶人跟驰行为分析及危险状态识别[D]. 长安大学, 2021.

[19] Bezuglov A, Comert G. Short-term freeway traffic parameter prediction: Application of grey system theory models[J]. Expert Systems with Applications, 2016, 62: 284-292.

[20] Wang J, Luo T, Fu T. Crash prediction based on traffic platoon characteristics using floating car trajectory data and the machine learning approach[J]. Accident Analysis & Prevention, 2019, 133: 105320.1-105320.12.

[21] Yan L, Zhang Y, Yi H, et al. Hazardous Traffic Event Detection Using Markov Blanket and Sequential Minimal Optimization (MB-SMO)[J]. Sensors, 2016, 16(7): 1084.

[22] Chai C, Wong Y D. Fuzzy Cellular Automata Model for Signalized Intersections[J]. Computer-Aided Civil and Infrastructure Engineering, 2016, 30(12): 951-964.

[23] Shi X, Wong Y D, Li Z F, et al. A feature learning approach based on XGBoost for driving assessment and risk prediction[J]. Accident Analysis & Prevention, 2019, 129: 170-179.

[24] 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(06): 84-94.

[25] Abdel-Aty M, Uddin N, Pande A, et al. Predicting Freeway Crashes From Loop Detector Data by Matched Case-Control Logistic Regression[J]. Transportation Research Record Journal of the Transportation Research Board, 2004, 1897: 88-95.

[26] Feng G, Fang Y. Individual driver risk assessment using naturalistic driving data[J]. Accident Analysis & Prevention, 2013, 61: 3-9.

[27] Jiang X, Abdel-Aty M, Hu J, et al. Investigating macro-level hotzone identification and variable importance using big data: A random forest models approach[J]. Neurocomputing, 2016, 18: 53-63.

[28] Da Y, Yw A, Feng S A, et al. Freeway accident detection and classification based on the multi-vehicle trajectory data and deep learning model[J]. Transportation Research Part C: Emerging Technologies, 2021, 130(9):103303.1-103303.22.

[29] 吕能超, 彭凌枫, 吴超仲, 文家强. 区分冲突类型的路段实时碰撞风险预测模型[J]. 中国公路学报, 2022, 35(01): 93-108.

[30] Panagopoulos G . Forecasting Markers of Habitual Driving Behaviors Associated with Crash Risk[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(2): 841-854.

[31] Chen J, Wu Z C, Zhang J. Driving Safety Risk Prediction Using Cost-Sensitive With Nonnegativity-Constrained Autoencoders Based on Imbalanced Naturalistic Driving Data[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(12): 4450-4465.

[32] 廖纪勇, 吴晟, 李武, 等. 基于k-means++聚类的轻型汽车行驶工况构建[J]. 武汉理工大学学报:交通科学与工程版, 2021, (2): 378-383.

[33] Chen, Wang, Chengcheng, et al. A crash prediction method based on bivariate extreme value theory and video-based vehicle trajectory data.[J]. Accident; analysis and prevention, 2018, 123(2): 365-373.

[34] J. Narváez, P. Estrada, P. Vidal-Montal, et al. Impact of previous glucocorticoid therapy on diagnostic accuracy of [18F] FDG PET-CT in giant cell arteritis[J]. Seminars in Arthritis and Rheumatism, 2023, 60(6): 152183.

[35] Ozbay K, Yang H, Bartin B, et al. Derivation and Validation of New Simulation-Based Surrogate Safety Measure[J]. Transportation Research Record, 2018, 2083(1): 105-113.

[36] Mahmud A, Luis A, Shamsul B, et al. Application of proximal surrogate indicators for safety evaluation: A review of recent developments and research needs[J]. IATSS Research, 2017, 41(4):153-163.

[37] Arun A, Haque M M, Bhaskar A, et al. A Bivariate Extreme Value Model for Estimating Crash Frequency by Severity using Traffic Conflicts[J]. Analytic Methods in Accident Research, 2021, 32(11): 100180.

[38] A S M S M, A L F, B M S H, et al. Application of proximal surrogate indicators for safety evaluation: A review of recent developments and research needs[J]. IATSS Research, 2017, 41(4): 153-163.

[39] Mahajan K, Velaga N R. Sleep-deprived car-following: Indicators of rear-end crash potential[J]. Accident Analysis & Prevention, 2021, 156(8): 106123.

[40] Weng J, Xue S, Yang Y, et al. In-depth analysis of drivers' merging behavior and rear-end crash risks in work zone merging areas[J]. Accident Analysis & Prevention, 2015, 77(4): 51-61.

[41] Horst R. Time-to-collision as a cue for decision-making in braking[J]. Vision in Vehicles, 1991.

[42] 潘恒彦,王永岗,李德林,陈俊先,宋杰,杨钰泉.基于交通冲突的长纵坡路段追尾风险评估及预测[J/OL]. 吉林大学学报(工学版): 1-11[2023-03-09]. DOI: 10.13229/j.cnki.jdxbgxb20210912.

[43] 戢晓峰, 谢世坤, 覃文文, 等. 基于轨迹数据的山区危险性弯道路段交通事故风险动态预测[J]. 中国公路学报, 2022, 35(04): 277-285.

[44] Minderhoud M M, Bovy P. Extended time-to-collision measures for road traffic safety assessment[J]. Accident Analysis & Prevention, 2001(33): 89-97

[45] Summala H. Brake Reaction Times and Driver Behavior Analysis[J]. Transportation Human Factors, 2000, 2(3):217-226.

[46] 李霖, 朱西产, 马志雄. 驾驶员在真实交通危险工况中的制动反应时间[J]. 汽车工程, 2014(10):1225-1229.

[47] 郭应时, 王畅, 付锐, 等. 城市道路环境中驾驶人应激响应时间特性[J]. 中国公路学报, 2013, 026(006):135-142.

[48] 李霖, 朱西产, 陈海林. 驾驶员制动和转向避撞极限[J]. 同济大学学报(自然科学版), 2016 ,44(11): 1743-1748.

[49] 秦雅琴, 李秋谷, 谢济铭. 驾驶人风险感知研究综述[J]. 昆明理工大学学报(自然科学版), 2021, 46(05): 125-138.

[50] 孙龙, 常若松 . 驾驶经验和危险类型对年轻驾驶员危险知觉的影响(英文)[J]. 心理科学, 2018, 41(02): 447-452.

[51] 熊坚, 施锦浩, 万华森. 人车路综合风险场模型构建及驾驶风格评估[J]. 交通运输系统工程与信息, 2021, 21(06): 105-114.

[52] 龚珊, 尹相勇, 朱爱华. 基于浮动车的路段行程时间卡尔曼滤波预测算法[C]// 2008第四届中国智能交通年会论文集. 2008.

[53] 于永堂, 郑建国, 张继文, 黄鑫, 等. 基于卡尔曼滤波与指数平滑法融合模型的沉降预测新方法[J]. 岩土工程学报, 2021, 43(S1): 127-131.

[54] 冯延超. 基于支持向量机的车辆换道行为识别与横向轨迹预测研究[D]. 河北工业大学, 2018.

[55] 贺玉龙, 刘磊, 迟佳欣. 高速公路车辆换道行为风险研究[J]. 重庆交通大学学报(自然科学版), 2021, 40(04): 26-33.

[56] Michiel M. Minderhoud, Piet H.L. Bovy, Extended time-to-collision measures for road traffic safety assessment[J]. Accident Analysis & Prevention, 2001, 33(1):89-97.

[57] 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(06): 84-94.

[58] Japan Society of Traffic Engineers (2005),“Traffic Engineering Handbook”.

[59] 吴斌, 朱西产, 沈剑平. 基于自然驾驶数据的驾驶员紧急制动行为特征[J]. 同济大学学报(自然科学版), 2018, 46(11): 1514-1519+1535.

[60] Archer J. Indicators for traffic safety assessment and prediction and their application in micro-simulation modelling: a study of urban and suburban intersections[J]. KTH, 2005.

[61] 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]. Jmtl, 2008, 1(2):170-180.

[62] Lu G, Bo C, Lin Q, et al. Quantitative indicator of homeostatic risk perception in car following[J]. Safety Science, 2012, 50(9): 1898-1905.

[63] 王建强, 吴剑, 李洋. 基于人-车-路协同的行车风险场概念、原理及建模[J]. 中国公路学报, 2016, 29(01): 105-114.

[64] 张旭欣, 王雪松, 马勇, 等. 驾驶行为与驾驶风险国际研究进展[J]. 中国公路学报,2020, 33(06): 1-17.

[65] 朱宇, 赵祥模, 徐志刚, 等. 基于蒙特卡洛模拟的无人车高速公路变道虚拟测试场景自动生成算法[J]. 中国公路学报, 2022, 35(03): 89-100.

[66] Yan Kuang, Xiaobo Qu, Shuaian Wang, A tree-structured crash surrogate measure for freeways[J]. Accident Analysis & Prevention, 2015, 77(3): 137-148.

[67] Shangguan Q, Fu T, Jiang R, et al. Use of Naturalistic Driving Data to Quantify Influencing Factors of Driving Risk based on a New Surrogate Measure of Safety[C]// the 100th Annual Meeting of Transportation Research Board. 2021.

[68] Cunto F, Saccomanno F F. Calibration and validation of simulated vehicle safety performance at signalized intersections[J]. Accident Analysis & Prevention, 2008, 40(3): 1171-1179.

[69] 吕能超, 任泽远, 段至诚, 等. Near-crash事件中驾驶人行为特征分析[J]. 中国安全科学学报, 2017, 27(06): 19-24.

[70] Shandah U, Saccomanno F, Persaud B. Application of traffic microsimulation for evaluating safety performance of urban signalized intersections[J]. Transportation Research Part C, 2015, 60(11): 96-104.

[71] 杨曼, 吴超仲, 张晖, 等. 行车安全事件的驾驶风险影响因素研究[J].交通信息与安全, 2018, 36(05): 34-39.

[72] 熊晓夏, 陈龙, 梁军, 等. 基于驾驶员避撞行为的行车风险判别方法的仿真研究[J]. 汽车工程, 2019, 41(02): 153-160+169.

[73] Naji H, Xue Q, Lyu N, et al. Evaluating the Driving Risk of Near-Crash Events Using a Mixed-Ordered Logit Model[J]. Sustainability, 2018, 10(8): 2868.

[74] 吕能超, 高谨谨, 王维锋, 等. 网联环境下基于精简车头时距特性的驾驶风格分类[J]. 交通信息与安全, 2022, 40(01): 116-125+168.

中图分类号:

 U491    

条码号:

 002000073670    

馆藏号:

 YD10001752    

馆藏位置:

 203    

备注:

 403-西院分馆博硕论文库;203-余家头分馆博硕论文库    

无标题文档

   建议浏览器: 谷歌 火狐 360请用极速模式,双核浏览器请用极速模式