[1]陈敬松,王艳,周栩佳,等.“两客一危”车辆危险驾驶行为前兆因素识别及预测模型*[J].中国安全生产科学技术,2025,21(3):209-217.[doi:10.11731/j.issn.1673-193x.2025.03.027]
CHEN Jingsong,WANG Yan,ZHOU Xujia,et al.Identification and prediction model on precursor factors of dangerous driving behavior for “two types of passenger vehicles and one type of hazardous materials transport vehicle”[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(3):209-217.[doi:10.11731/j.issn.1673-193x.2025.03.027]
点击复制
“两客一危”车辆危险驾驶行为前兆因素识别及预测模型*
CHEN Jingsong,WANG Yan,ZHOU Xujia,et al.Identification and prediction model on precursor factors of dangerous driving behavior for “two types of passenger vehicles and one type of hazardous materials transport vehicle”[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(3):209-217.[doi:10.11731/j.issn.1673-193x.2025.03.027]
《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]
- 卷:
- 21
- 期数:
- 2025年3期
- 页码:
- 209-217
- 栏目:
- 职业安全卫生管理与技术
- 出版日期:
- 2025-03-31
文章信息/Info
- Title:
- Identification and prediction model on precursor factors of dangerous driving behavior for “two types of passenger vehicles and one type of hazardous materials transport vehicle”
- 文章编号:
- 1673-193X(2025)-03-0209-09
- Keywords:
- two types of passenger vehicles and one type of hazardous materials transport vehicle; dangerous driving behavior; precursor factor; behavior prediction
- 分类号:
- X951
- 文献标志码:
- A
- 摘要:
- 为深入分析“两客一危”车辆的危险驾驶行为，基于上海市“两客一危”的危险驾驶行为数据，梳理包括车道偏离、车距过近、分神驾驶、前向碰撞、疲劳驾驶和其他类，共6类危险行为数据集，并基于多元Logistic模型对危险行为的潜在前兆因素进行识别和分析。研究结果表明：前序时间内的累计危险行为次数、持续驾驶时长和驾驶时间（4∶00~6∶00）对这6种危险行为均具有显著影响，前序的车道偏离、分神驾驶、前向碰撞、疲劳驾驶行为分别对6种危险驾驶行为具有不同的影响效果；对比LSTM，Bi-LSTM与Bi-GRU3种模型的预测效果，可知Bi-LSTM在测试集上的预测效果最好，其准确率比LSTM模型高7百分点，比Bi-GRU模型高4百分点，并且Bi-LSTM和Bi-GRU模型在疲劳驾驶和其他类的预测效果最好。研究结果可为“两客一危”车辆危险驾驶行为的预防、管理以及智能监控系统的开发提供参考，有助于提升道路交通安全水平。
- Abstract:
- In order to analyze the dangerous driving behavior of “two types of passenger vehicles and one type of hazardous materials transport vehicle”,the dangerous driving behaviordata of these special vehicles in Shanghai was analyzed.Six types of dangerous behavior datasets were sorted out,including lane deviation,close vehicle distance,distracted driving,forward collision,fatigue driving and other.The potential precursor factors of dangerous behavior were identified and analyzed based on multivariate Logistic model.The results show that the cumulative number of dangerous behavior,continuous driving duration and driving time period (4∶00~6∶00) in the preceding timehave significant influence on these six types of dangerous behavior,and the preceding lane deviation,distracted driving,forward collision and fatigue driving have different influence on the six types of dangerous driving behavior,respectively.By comparing the prediction effect of LSTM,Bi-LSTM and Bi-GRU models,it can be seen that Bi-LSTM has the best prediction effect on the test set,and its accuracy is 7 percentage points higher than that of LSTM model and 4 percentage points higher than that of Bi-GRU model,and Bi-LSTM and Bi-GRU models have the best prediction effect on fatigue driving and other type.The research results can provide reference and effective prediction model for the prevention and management on dangerous driving behavior of “two types of passenger vehicles and one type of hazardous materials transport vehicle” and the development of intelligent monitoring system,which is helpful to improve the level of road traffic safety.
参考文献/References:
［1］中华人民共和国交通运输部.关于加强道路运输车辆动态监管工作的通知［EB/OL］.（2011-04-07）［2025-02-18］.https://xxgk.mot.gov.cn/2020/jigou/ysfws/202006/t20200623_3315851.html.
［2］国家统计局.2024年中国统计年鉴［M］.北京:中国统计出版社，2024.
［3］高丽娜.深度学习在危险驾驶识别中的应用［J］.电子技术（上海）,2024,53(7):404-405.
GAO Li’na.Application of deep learning in dangerous driving recognition［J］.Electronic Technology,2024,53(7):404-405.
［4］张永辉.危险驾驶行为识别方法研究综述［J］.交通科技与管理,2021(2):227-228.
［5］郑志晓,李鹏辉,胡孟夏,等.视觉分神对驾驶员跟驰行为的影响［J］.公路交通科技,2020,37(1):122-130.
ZHENG Zhixiao,LI Penghui,HU Mengxia,et al.Influence of visual distraction on driver’s car-following behavior［J］.Journal of Highway and Transportation Research and Development,2020,37(1):122-130.
［6］朱彤彤,蔡乙男,冯芳,等.车辆安全保障系统的设计［J］.电子制作,2017(21):9-10,17.
［7］MORADI A,NAZARI S SH,RAHMANI K.Sleepiness and the risk of road traffic accidents:a systematic review and meta-analysis of previous studies ［J］.Transportation Research Part F:Traffic Psychology and Behaviour,2019,65:620-629.
［8］陶达,黄玉勇,吴玉卓，等.长时驾驶任务中脑力负荷多维特征变化趋势研究［J］.中国公路学报,2023,36(11):456-464.
TAO Da,HUANG Yuyong,WU Yuzhuo,et al.Changing Mechanism of mental workload multidimensional characteristics during long-term driving Tasks［J］.China Journal of Highway and Transport,2023,36(11):456-464.
［9］赵强.基于深度学习的危险驾驶行为检测技术研究［D］.鞍山：辽宁科技大学,2023.
［10］张志远.基于深度学习的机动车驾驶人危险驾驶行为检测方法研究［D］.杭州：杭州电子科技大学,2023.
［11］董鑫磊.危险驾驶行为预警系统研究［D］.辽宁：辽宁科技大学,2022.
［12］刘永涛,乔洁,魏朗,等.危险驾驶行为辨识算法研究［J］.计算机工程与设计,2014,35(4):1322-1326.
LIU Yongtao,QIAO Jie,WEI Lang,et al.Research on identification algorithm of dangerous driving behavior［J］.Computer Engineering and Design,2014,35(4):1322-1326.
［13］何晖宇,丁瑞,应聃,等.苜蓿叶立交交织区冲突特征及严重程度影响因素分析［J］.交通运输系统工程与信息，2024，24（5）：246-258.
HE Yuhui,DING Rui,YING Dan,et al.Conflict features and impact factors of severity in weaving area of cloverleaf interchange［J］.Journal of Transportation Systems Engineering and Information Technology,2024,24(5):246-258.
［14］林庆峰,成波.基于Logistic回归的驾驶人跟车危险认知模型［J］.中国公路学报,2011,24(6):103-108.
LIN Qingfeng,CHENG Bo.Model of driver risk perception in car-following based on Logistic regression［J］.China Journal of Highway and Transport,2011,24(6):103-108.
［15］柳鹏飞,陆见光,徐磊,等.公路货运危险驾驶行为智能预测技术研究［J］.汽车技术,2024(3):56-62.
LIU Pengfei,LU Mingguang,XU Lei,et al.Research on intelligent prediction technology of dangerous driving behavior in highway freight［J］.Automobile Technology,2024(3):56-62.
［16］杨婧宜,丁渭平,朱洪林.基于Bi-LSTM的多类型特征融合驾驶员疲劳识别方法［J］.汽车工程学报,2023,13(2):159-168.
YANG Jingyi,DING Weiping,ZHU Honglin.Driver fatigue recognition using multi-type features fusion based on Bi-LSTM［J］.Chinese Journal of Automotive Engineering,2023,13(2):159-168.
［17］许新亮.网联环境下基于驾驶意图的车辆轨迹预测算法研究［D］.秦皇岛：燕山大学,2023.
［18］徐文翔,王俊骅,傅挺.基于注意力机制及分层网络的危险驾驶行为预测方法［J］.同济大学学报(自然科学版),2022,50(5):722-730.
XU Wenxiang,WANG Junhua,FU Ting.Aggressive driving behavior prediction method based on attention mechanism and hierarchical network［J］.Journal of Tongji University：Natural Science,2022,50(5):722-730.
［19］上海市公安局.上海市道路交通管理条例［EB/OL］.(2016-12-29)［2025-02-18］.https://gaj.sh.gov.cn/shga/wzXxfbGj/detail?pa=110ef360e4374a41b31df9d314512365ba1ed78d12a678c9e946c2b3dc0c23f8.
［20］中华人民共和国中央人民政府.道路交通安全违法行为记分管理办法［EB/OL］.(2021-12-17)［2025-02-18］.https://www.gov.cn/gongbao/content/2022/content_5679697.htm.
［21］朱凌建，荀子涵，王裕鑫，等.基于CNN-Bi LSTM的短期电力负荷预测［J］.电网技术,2021，11（45):4532-4539.
ZHU Lingjian,XUN Zihan,WANG Yuxin,et al.Short-term power load forecasting based on CNN-BiLSTM ［J］.Power System Technology,2021，11（45):4532-4539.
［22］包智鹏,支永帅,张素民,等.基于BiGRU的多模态驾驶行为及轨迹预测［J］.大连理工大学学报,2021,61(3):246-254.
BAO Zhipeng,ZHI Yongshuai,ZHANG Suming,et al.BiGRU based multi-modal maneuvers and trajectory prediction［J］.Journal of Dalian University of Technology,2021,61(3):246-254.
相似文献/References:
[1]冯永刚,郑亮.危险驾驶行为在分心与事故间的中介效应分析*[J].中国安全生产科学技术,2024,20(8):210.[doi:10.11731/j.issn.1673-193x.2024.08.028]
FENG Yonggang,ZHENG Liang.Analysis on mediating effect of dangerous driving behavior between distraction and accident[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(3):210.[doi:10.11731/j.issn.1673-193x.2024.08.028]
备注/Memo
- 备注/Memo:
-
收稿日期： 2024-07-06
* 基金项目： 道路交通安全管控技术国家工程研究中心开放课题（2024GCZXKFKT19A）；国家自然科学基金项目(52172348，52202419)；上海市科学计划委员会科研计划项目（19DZ1202100）；北二环高速公路改扩建期间精细化管控策略研究项目（BEH-2023-ZX-082）
作者简介： 陈敬松，博士，高级工程师，主要研究方向为智慧交通与运营管理。
通信作者： 周欣，本科，工程师，主要研究方向为智慧交通与运营管理。
