基于深度学习的车联边缘网络交通事故风险预测算法研究

赵海涛; 程慧玲; 丁仪; 张晖; 朱洪波

doi:10.11999/JEIT190595

基于深度学习的车联边缘网络交通事故风险预测算法研究

doi: 10.11999/JEIT190595

1.
教育部泛在网络健康服务系统工程研究中心南京 210003
2.
江苏省无线通信重点实验室南京 210003
3.
南京邮电大学通信与信息工程学院南京 210003

基金项目: 国家自然科学基金(61771252)，江苏省自然科学基金(BK20171444)，江苏省高校重点自然科学研究重大项目(18KJA510005)，江苏省“六大人才高峰”B类资助项目(DZXX-041)，江苏省科协青年科技人才托举工程资助培养项目，江苏省研究生科研创新计划项目(KYCX19_0949)

详细信息

作者简介:
赵海涛：男，1983年生，博士，副教授，研究方向为物联网与移动边缘计算

程慧玲：女，1995年生，硕士生，研究方向为移动边缘计算与人工智能

丁仪：女，1995年生，硕士生，研究方向为物联网路由优化和边缘计算

张晖：男，1982年生，博士，副教授，研究方向为未来无线网络

朱洪波：男，1956年生，博士，教授，研究方向为移动通信与宽带无线技术、无线通信与电磁兼容

通讯作者:
赵海涛　zhaoht@njupt.edu.cn

中图分类号: TP399
计量
- 文章访问数: 3529
- HTML全文浏览量: 1358
- PDF下载量: 207
- 被引次数: 0
出版历程
- 收稿日期: 2019-08-06
- 修回日期: 2019-11-05
- 网络出版日期: 2019-11-13
- 刊出日期: 2020-01-21

Research on Traffic Accident Risk Prediction Algorithm of Edge Internet of Vehicles Based on Deep Learning

1.
Ministry of Education Ubiquitous Network Health Service System Engineering Research Center, Nanjing 210003, China
2.
Jiangsu Key Wireless Communication Laboratory, Nanjing 210003, China
3.
College of Communication and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China

Funds: The National Natural Science Foundation of China (61771252), The Natural Science Foundation Project of Jiangsu Province (BK20171444), The Jiangsu Province University Natural Science Research Major Project (18KJA510005), “The Six talents High Peaks” Class B Funding Project of Jiangsu Province (DZXX-041), The Jiangsu Provincial Association for Science and Technology Talents Entrustment Project, Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX19_0949)

摘要

摘要: 针对传统交通事故风险预测算法无法自动判别数据特征，且模型表达能力差等问题。该文提出一种基于深度学习的车联边缘网络交通事故风险预测算法，该算法首先针对车载自组织网络中采集的大量交通数据，采用边缘服务器中建立的卷积神经网络自主提取多维特征，经归一化、去均值等预处理后，再将得到的新变量输入卷积层、采样层进行训练，最后根据全连接层输出的判别值，得到模拟预测交通事故发生的风险性。仿真结果表明，该算法被验证能够预测交通事故发生的风险性，较传统的机器学习算法BP神经网络、逻辑回归具有更低的损失与更高的预测准确度。
- 车联边缘网络 /
- 机器学习 /
- 卷积神经网络 /
- 边缘服务器
Abstract: For the problem that the traditional traffic accident risk prediction algorithm can not automatically discriminate data features, and the model expression ability is poor, a traffic accident risk prediction algorithm based on deep learning is proposed. The algorithm firstly extracts multi-dimensional features by using the convolutional neural network established in the edge server for a large amount of traffic data collected in the edge network of vehicles. After normalization, de-equalization and other pre-processing, the new variables are input into the convolutional layer and the pooling layer for training. Finally, based on the output discrimination value of the fully connected layer, the risk of traffic accidents can be predicted by simulation. The simulation results show that the algorithm is validated to predict the risk of traffic accidents, and has lower loss and higher prediction accuracy than the traditional machine learning BP neural network algorithm and Logical Regression algorithm.
- Edge network of vehicles /
- Machine learning /
- Convolutional neural network /
- Edge server

HTML全文

图 1 输入层具有3个神经元的感知机建模图

下载: 全尺寸图片幻灯片

图 2 含有多层隐含层的卷积神经网络交通事故风险预测建模图

下载: 全尺寸图片幻灯片

图 3 车联边缘网络系统架构图

下载: 全尺寸图片幻灯片

图 4 交通事故风险预测算法流程图

下载: 全尺寸图片幻灯片

图 5 卷积神经网络与BP神经网络、逻辑回归预测损失对比图

下载: 全尺寸图片幻灯片

图 6 卷积神经网络较BP神经网络、逻辑回归预测准确度对比图

下载: 全尺寸图片幻灯片

图 7 不同激活函数对卷积、BP神经网络、逻辑回归算法预测损失的影响对比图

下载: 全尺寸图片幻灯片

图 8 不同激活函数对卷积、BP神经网络、逻辑回归算法预测准确度的影响对比图

下载: 全尺寸图片幻灯片

参考文献(16)

GOTHANE S and SARODE M V. Analyzing factors, construction of dataset, estimating Importance of factor, and generation of association rules for Indian road accident[C]. The 6th IEEE International Conference on Advanced Computing, Bhimavaram, India, 2016: 15–18. doi: 10.1109/IACC.2016.13.

IKRAM N and MAHAJAN S. Road accidents: Overview of its causes, avoidance scheme and a new proposed technique for avoidance[C]. The 3rd International Conference on Computing for Sustainable Global Development, New Delhi, India, 2016: 497–499.

LIN Lei, WANG Qian, and SADEK A W. Data mining and complex network algorithms for traffic accident analysis[J]. Transportation Research Record: Journal of the Transportation Research Board, 2014, 2460(1): 128–136. doi: 10.3141/2460-14

ZONG Fang, CHEN Xiangru, TANG Jinjun, et al. Analyzing traffic crash severity with combination of information entropy and Bayesian network[J]. IEEE Access, 2019, 7: 63288–63302. doi: 10.1109/ACCESS.2019.2916691

GHOSH B, ASIF M T, and DAUWELS J. Bayesian prediction of the duration of non-recurring road incidents[C]. 2016 IEEE Region 10 Conference, Singapore, 2016: 87–90. doi: 10.1109/TENCON.2016.7847964.

秦利燕, 邵春福, 赵亮. 道路交通事故宏观预测模型[J]. 武汉理工大学学报: 交通科学与工程版, 2010, 34(1): 154–157. doi: 10.3963/j.issn.1006-2823.2010.01.038

QIN Liyan, SHAO Chunfu, and ZHAO Liang. Macro prediction model of road traffic accident based on neural network[J]. Journal of Wuhan University of Technology:Transportation Science &Engineering, 2010, 34(1): 154–157. doi: 10.3963/j.issn.1006-2823.2010.01.038

TAO Lu, ZHU Dunyao, YAN Linxin, et al. The traffic accident hotspot prediction: Based on the logistic regression method[C]. 2015 International Conference on Transportation Information and Safety, Wuhan, China, 2015: 107–110. doi: 10.1109/ICTIS.2015.7232194.

OLUTAYO V A and ELUDIRE A A. Traffic accident analysis using decision trees and neural networks[J]. International Journal of Information Technology and Computer Science, 2014, 6(2): 22–28. doi: 10.5815/ijitcs.2014.02.03

YU B, WANG Y T, YAO J B, et al. A comparison of the performance of ANN and SVM for the prediction of traffic accident duration[J]. Neural Network World, 2016, 26(3): 271–287. doi: 10.14311/NNW.2016.26.015

ZHAO Haitao, MAO Tianqi, DUAN Jiaxiu, et al. FMCNN: A factorization machine combined neural network for driving safety prediction in vehicular communication[J]. IEEE Access, 2019, 7: 11698–11706. doi: 10.1109/ACCESS.2019.2891619

ZHANG Jian, LI Zhibin, PU Ziyuan, et al. Comparing prediction performance for crash injury severity among various machine learning and statistical methods[J]. IEEE Access, 2018, 6: 60079–60087. doi: 10.1109/ACCESS.2018.2874979

XU Jinlai, PALANISAMY B, LUDWIG H, et al. Zenith: Utility-aware resource allocation for edge computing[C]. 2017 IEEE International Conference on Edge Computing, Honolulu, USA, 2017: 47–54. doi: 10.1109/IEEE.EDGE.2017.15.

ZHENG Yang, WANG Jianqiang, LI Xiaofei, et al. Driving risk assessment using cluster analysis based on naturalistic driving data[C]. The 17th IEEE International Conference on Intelligent Transportation Systems, Qingdao, China, 2014: 2584–2589. doi: 10.1109/ITSC.2014.6958104.

CHEN Yuanfang, SHU Lei, and WANG Lei. Poster abstract: Traffic flow prediction with big data: A deep learning based time series model[C]. 2017 IEEE Conference on Computer Communications Workshops, Atlanta, USA, 2017: 1010–1011. doi: 10.1109/INFCOMW.2017.8116535.

梁晓萍, 郭振军, 朱昌洪. 基于头脑风暴优化算法的BP神经网络模糊图像复原[J]. 电子与信息学报, 2019, 41(12): 2980–2986. doi: 10.11999/JEIT190261

LIANG Xiaoping, GUO Zhenjun, and ZHU Changhong. BP neural network fuzzy image restoration based on brainstorm optimization algorithm[J]. Journal of Electronics and Information Technology, 2019, 41(12): 2980–2986. doi: 10.11999/JEIT190261

殷礼胜, 唐圣期, 李胜, 等. 基于整合移动平均自回归和遗传粒子群优化小波神经网络组合模型的交通流预测[J]. 电子与信息学报, 2019, 41(9): 2273–2279. doi: 10.11999/JEIT181073

YIN Lisheng, TANG Shengqi, LI Sheng, et al. Traffic flow prediction based on hybrid model of auto-regressive integrated moving average and genetic particle swarm optimization wavelet neural network[J]. Journal of Electronics &Information Technology, 2019, 41(9): 2273–2279. doi: 10.11999/JEIT181073

施引文献

资源附件(0)

访问统计