多源数据驱动的隧道场景驾驶员识别方法

doi:10.16097/j.cnki.1009-6744.2024.04.009

交通运输系统工程与信息 ›› 2024, Vol. 24 ›› Issue (4): 81-93.DOI: 10.16097/j.cnki.1009-6744.2024.04.009

• 智能交通系统与信息技术 • 上一篇下一篇

多源数据驱动的隧道场景驾驶员识别方法

金盛^*1,2，周梦涛¹，白聪聪¹

1. 浙江大学，智能交通研究所，杭州310058；2.浙江大学，中原研究院，郑州450000

收稿日期:2024-04-15 修回日期:2024-05-29 接受日期:2024-06-11 出版日期:2024-08-25 发布日期:2024-08-21
作者简介:金盛(1982- )，男，浙江温州人，教授，博士。
基金资助:
国家自然科学基金 (72361137006)；浙江省“尖兵” “领雁”研发攻关计划项目 (2022C01042)；浙江省杰出青年基金 (LR23E080002)。

Driver Identification Method in Tunnel Scenarios Based on Multi-source Data

JIN Sheng^*1,2, ZHOU Mengtao¹, BAI Congcong¹

1. Institute of Intelligent Transportation Systems, Zhejiang University, Hangzhou 310058, China; 2. Zhongyuan Institute, Zhejiang University, Zhengzhou 450000, China

Received:2024-04-15 Revised:2024-05-29 Accepted:2024-06-11 Online:2024-08-25 Published:2024-08-21
Supported by:
NationalNaturalScienceFoundation of China (72361137006)；"Pioneer" and "Leading Goose" R&D Program of Zhejiang (2022C01042)；Natural Science Foundation of Zhejiang Province (LR23E080002)。

摘要/Abstract

摘要： 针对隧道环境中监控图像分辨率低与车辆运动轨迹特征异质性减弱导致的驾驶员识别准确率偏低问题，本文提出一种融合卷积与多头注意力机制的驾驶员识别方法(Multi-scaleCNN with Multi Attention)，通过充分利用驾驶过程人—车—路—环境多源信息的协同耦合关系提升识别精度。首先，设计开展实车驾驶试验，构建针对隧道路段的人—车—路多源驾驶数据库并设计特征集合；其次，搭建驾驶员识别模型框架，该框架通过多尺度卷积神经网络学习驾驶过程中的局部波动，并通过并行的多头自注意力层结构捕捉驾驶时间序列的长期依赖性，实现局部信息与全局信息的有效整合，从而提升隧道场景的驾驶员识别效果。结果显示，与其他先进的算法相比，所提出的模型在驾驶员身份识别任务中的准确率高达99.07%，调和F1分数达到99.03%，充分证明了所提方法的有效性。此外，通过特征贡献度评估方法对隧道场景下驾驶员身份识别任务中的特征重要性进行深入探究发现，相较于车辆历史运动数据，驾驶员心理、生理及视觉特征显示出更高的贡献度。研究结果可为隧道场景多源数据应用提供支持，并对隧道安全监管提供技术支撑。

关键词: 交通工程, 驾驶员识别, 深度学习, 隧道场景, 多头注意力机制, 多尺度卷积

Abstract: In response to the problem of low driver recognition accuracy caused by the low resolution of monitoring images and the weakening of vehicle motion trajectory features in tunnel environments, this paper proposes a driver recognition method (Multi-scale Convolutional Neural Network with Multi Attention) that integrates convolution with multi-head attention mechanism. The collaborative coupling relationship of multi-source information of human-vehicle road-environment during the driving process was utilized to improve recognition accuracy. First, real vehicle driving experiments are designed and conducted to establish a multi-source driving database for tunnel sections and design feature sets. Second, a driver recognition model framework is built. This framework learns local fluctuations in the driving process through a multi-scale convolutional neural network and captures the long-term dependency of driving time series through parallel multi-head self-attention layer structures, effectively integrating local and global information to enhance the driver recognition effect in tunnel scenarios. The results show that compared with other advanced algorithms, the proposed model achieves an accuracy of 99.07% and a macro F1 score of 99.03% in driver identification tasks, fully demonstrating the effectiveness of the proposed method. In addition, through the feature contribution evaluation method, the importance of features in driver identification tasks in tunnel scenarios is explored in depth. It is found that compared with historical vehicle motion data, driver psychological, physiological, and visual features show higher contribution. The research results can provide support for the application of multi-source data in tunnel scenarios and provide technical support for tunnel safety supervision.

Key words: traffic engineering, driver recognition, deep learning, tunnel scenarios, multi-head attention mechanism, multi-scale convolution

中图分类号:

U491.2

金盛, 周梦涛, 白聪聪. 多源数据驱动的隧道场景驾驶员识别方法[J]. 交通运输系统工程与信息, 2024, 24(4): 81-93.

JIN Sheng, ZHOU Mengtao, BAI Congcong. Driver Identification Method in Tunnel Scenarios Based on Multi-source Data[J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 24(4): 81-93.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2024.04.009

http://www.tseit.org.cn/CN/Y2024/V24/I4/81

参考文献

[1]向恬君,白婧荣,唐伯明,等.毗邻互通立交特长隧道驾驶负荷研究[J].交通运输系统工程与信息,2022,22 (1): 301-310. [XIANG T J, BAI J R, TANG B M, et al. Velocity behavior and road alignment safety evaluation in complex scenarios of bridge and tunnel combination [J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(1): 301-310.]

[2]潘福全,泮海涛,王铮,等.照度与纵坡耦合作用下海底隧道出入口段驾驶人视觉特征分析与建模[J].交通运输系统工程与信息,2021, 21(1): 142-148. [PAN F Q, PAN H T, WANG Z, et al. Analysis and modeling of drivers' visual characteristics at entrance and exit of undersea tunnel with coupling of illuminance and longitudinal slope[J]. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(1): 142-148.]

[3]陈志军,胡军楠,冷姚,等.基于轻量化网络和注意力机制的智能车快速目标识别方法[J].交通运输系统工程与信息,2022, 22(6): 105-113. [CHEN Z J, HU J N, LENG Y, et al. Intelligent vehicle target fast recognition based on lightweight network and attention mechanism [J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(6): 105-113.]

[4]梁华刚,黄伟浩,薄颖,等.基于多特征融合的隧道场景车辆再识别[J]. 中国公路学报, 2023, 36(8): 280 291. [LIANG H G, HUANG W H, BO Y, et al. Multifeature-fusion-based vehicle reidentification for tunnel seenes[J]. China Journal of Highway and Transport, 2023, 36(8): 280-291.]

[5] LU J J, XING Y Y, WANG C, et al. Risk factors affecting the severity of traffic accidents at Shanghai river-crossing tunnel[J]. Traffic Injury Prevention, 2016, 17(2): 176 180.

[6] WU L, ZHU T, LIU H X, et al. Comparative research on driver eye movement on freeways and urban roads[J]. Journal of Intelligent Transportation and Urban Planning, 2015, 3(3): 70-78.

[7] HCHOI G, SBAK E, BPAN S. User identification system using 2D resized spectrogram features of ECG[J]. IEEE Access, 2019, 7: 34862-34873.

[8] LU L J, MAO J N, WANG WQ,et al. A study of personal recognition method based on EMG signal[J]. IEEE Transactions on Biomedical Circuits and Systems, 2020, 14: 681-691.

[9] MOHAMMADNAZAR A, ARVIN R, KHATTAK A. Classifying travelers' driving style using basic safety messages generated by connected vehicles: Application of unsupervised machine learning[J]. Transportation Research Part C: Emerging Technologies, 2021, 122: 102917.

[10] VIROJBOONKIATE N, CHANAKITKARNCHOK A, VATEEKUL P, et al. Public transport driver identification system using histogram of acceleration data [J]. Journal of Advanced Transportation, 2019, 2019: 6372597.

[11] KWAK B I, WOO J Y, KIM H K. Know your master: Driver profiling-based anti-theft method[C]// 2016 14th Annual Conference on Privacy, Security and Trust, Online: IEEE Xplore, 2016: 211-218.

[12] KWAK B I, HAN M L, KIM H K. Driver identification based on wavelet transform using driving patterns[J]. IEEE Transactions on Industrial Informatics, 2021, 17 (4): 2400-2410.

[13] CHEN J, WU Z Z, ZHANG J. Driver identification based on hidden feature extraction by using adaptive nonnegativity-constrained autoencoder[J]. Applied Soft Computing, 2019, 74: 1-9.

[14] 胡宏宇, 刘家瑞,高菲,等.基于一维卷积神经网络的驾驶人身份识别方法[J].中国公路学报,2020,33(8): 195-203. [HU H Y, LIU J R, GAO F, et al. Driver identification based on 1-D convolutional neural networks [J]. China Journal of Highway and Transport, 2020, 33 (8): 195-203.]

[15] HU H Y, LIU J R, CHEN G Y, et al. Driver identification through vehicular CAN bus data: An ensemble deep learning approach[J]. Iet Intelligent Transport Systems, 2023, 17(5): 867-877.

[16] AZADANI N M, AZZEDINE B. DriverRep: Driver identification through driving behavior embeddings[J]. Journal of Parallel and Distributed Computing, 2022, 162: 105-117.

多源数据驱动的隧道场景驾驶员识别方法

Driver Identification Method in Tunnel Scenarios Based on Multi-source Data

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	赵红专, 张继康, 潘佳雯, 袁泉, 许恩永, 魏金占, 周旦, 刘承堃. 基于改进视觉算法的自动驾驶风险预判模型[J]. 交通运输系统工程与信息, 2024, 24(5): 79-90.
[2]	胡正华, 周继彪, 毛新华, 张敏捷. 基于图像序列分析的城市道路交通事故预测[J]. 交通运输系统工程与信息, 2024, 24(5): 91-102.
[3]	何晖宇, 丁瑞, 应聃, 张宇豪, 张河山, 徐进. 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.
[4]	唐铁桥, 钟静燃, 袁小婷, 覃梦欣. 基于扩展决策场的机场落客区车辆－旅客行为建模分析[J]. 交通运输系统工程与信息, 2024, 24(5): 309-317.
[5]	黄岩, 李海军, 闫学东, 段克. 雾天网联车辆跟驰模型构建及行为影响分析[J]. 交通运输系统工程与信息, 2024, 24(4): 41-49.
[6]	田鑫, 卢凯, 高志刚. 基于车速引导的交叉口公交优先自适应控制模型[J]. 交通运输系统工程与信息, 2024, 24(4): 69-80.
[7]	魏书樵, 倪颖, 孙剑, 邱红桐. 基于生成对抗模仿学习的路段非机动车行为仿真[J]. 交通运输系统工程与信息, 2024, 24(4): 105-115.
[8]	孟云伟, 全振宇, 王子骁, 刘向洋, 李斌斌, 青光焱, 刘中帅. 亮度和速度协同下公路隧道口路段驾驶舒适度研究[J]. 交通运输系统工程与信息, 2024, 24(4): 116-126.
[9]	程国柱, 李天仪, 汪国鹏. 基于冰雪路面危险驾驶行为谱的行车风险识别方法[J]. 交通运输系统工程与信息, 2024, 24(4): 127-138.
[10]	尚婷, 易爱强, 何军, 刘祥. 自然驾驶状态下城市地下环道驾驶人心理负荷研究[J]. 交通运输系统工程与信息, 2024, 24(4): 274-282.
[11]	张金雷, 杨健, 刘晓冰, 陈瑶, 杨立兴, 高自友. 基于计算机视觉的轨道交通站内火灾检测与定位[J]. 交通运输系统工程与信息, 2024, 24(3): 53-63.
[12]	马书红, 陈西芳, 杨磊, 赵玉哲, 曾玉. 可达性对城市群多模式交通碳排放的空间异质性影响[J]. 交通运输系统工程与信息, 2024, 24(3): 64-74.
[13]	吴场建, 曹奇, 任刚. 考虑路径关系的干线多路径绿波优化模型[J]. 交通运输系统工程与信息, 2024, 24(3): 103-113.
[14]	杨亚璪, 宾涛. 电动汽车集中型充电站选址定容模型[J]. 交通运输系统工程与信息, 2024, 24(3): 204-212.
[15]	胡立伟, 陈琛, 赵雪亭, 刘冰, 侯智, 张瑞杰, 贺雨. 基于行车风险场的快速路短交织区车辆交互风险识别[J]. 交通运输系统工程与信息, 2024, 24(3): 221-231.