《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (6): 1748-1755.DOI: 10.11772/j.issn.1001-9081.2021061411
所属专题: 2021年全国开放式分布与并行计算学术年会(DPCS 2021)论文
• 2021年全国开放式分布与并行计算学术年会(DPCS 2021)论文 • 上一篇 下一篇
收稿日期:
2021-08-06
修回日期:
2021-09-17
接受日期:
2021-11-17
发布日期:
2022-01-10
出版日期:
2022-06-10
通讯作者:
孙丽珺
作者简介:
郭嘉宸(1997-),男,山东烟台人,硕士研究生,CCF会员,主要研究方向:深度学习、智能交通基金资助:
Jiachen GUO, Yushen YANG, Yan WANG, Shilong MAO, Lijun SUN()
Received:
2021-08-06
Revised:
2021-09-17
Accepted:
2021-11-17
Online:
2022-01-10
Published:
2022-06-10
Contact:
Lijun SUN
About author:
GUO Jiachen,born in 1997,M. S. candidate. His researchinterests include deep learnSupported by:
摘要:
精细化短时交通流预测是保证智能交通系统(ITS)合理决策的前提。为了建立无人驾驶汽车换道模型、预测车辆轨迹、引导车辆出行,及时为每条车道预测车流量成为亟须解决的问题,然而精细化短时交通流预测面临着以下挑战:一是交通流数据日益多元化,传统预测方法难以满足ITS高精度、短时延的要求;二是为每条车道训练预测模型会造成大量的资源浪费。针对以上问题,提出利用卷积-门控循环单元(Conv-GRU)结合灰色关联度分析法(GRA)建立精细化短时交通流预测模型预测车道流量。考虑到深度学习训练时间长、推理时间相对较短的特点,提出云-雾部署方案;同时,为避免为每条车道训练预测模型,在云-雾部署方案的基础上提出了模型迁移部署方案,该方案仅需训练部分车道的预测模型,然后通过GRA将训练好的预测模型迁移部署到关联车道进行预测。对真实交通流数据集进行大量对比实验的结果表明:与传统深度学习预测方法相比,所提模型拥有更精准的预测性能,与卷积-长短期记忆(Conv-LSTM)网络相比在提高精度的基础上运行时间更短,且能在保证高精度预测的情况下实现模型迁移,比训练每条车道的预测模型节省了约49%的训练时间。
中图分类号:
郭嘉宸, 杨宇燊, 王研, 毛仕龙, 孙丽珺. 精细化短时交通流预测模型及迁移部署方案[J]. 计算机应用, 2022, 42(6): 1748-1755.
Jiachen GUO, Yushen YANG, Yan WANG, Shilong MAO, Lijun SUN. Refined short-term traffic flow prediction model and migration deployment scheme[J]. Journal of Computer Applications, 2022, 42(6): 1748-1755.
符号 | 含义 |
---|---|
一个周期的天数 | |
一个周期内第 | |
一个周期内的车道流量矩阵 | |
经过无量纲化处理后的车道流量矩阵 | |
车道 | |
车道 | |
分辨系数 | |
第 | |
对第 |
表1 主要符号说明
Tab. 1 Summary of main symbols
符号 | 含义 |
---|---|
一个周期的天数 | |
一个周期内第 | |
一个周期内的车道流量矩阵 | |
经过无量纲化处理后的车道流量矩阵 | |
车道 | |
车道 | |
分辨系数 | |
第 | |
对第 |
模型 | 工作日 | 休息日 | ||
---|---|---|---|---|
本文模型 | 13.998 | 22.245 | 12.217 | 16.640 |
Conv-LSTM | 14.602 | 22.834 | 13.329 | 17.833 |
GRU | 16.994 | 24.821 | 13.145 | 17.651 |
LSTM | 14.469 | 22.519 | 12.432 | 17.232 |
表2 不同方法在工作日和休息日的交通流量预测性能
Tab. 2 Performance of different methods in weekday and weekend traffic flow prediction
模型 | 工作日 | 休息日 | ||
---|---|---|---|---|
本文模型 | 13.998 | 22.245 | 12.217 | 16.640 |
Conv-LSTM | 14.602 | 22.834 | 13.329 | 17.833 |
GRU | 16.994 | 24.821 | 13.145 | 17.651 |
LSTM | 14.469 | 22.519 | 12.432 | 17.232 |
迭代 步数 | 运行时间/s | 迭代 步数 | 运行时间/s | ||
---|---|---|---|---|---|
Conv-aLSTM | 本文模型 | Conv-aLSTM | 本文模型 | ||
50 | 161 | 122 | 200 | 645 | 489 |
100 | 320 | 243 | 250 | 784 | 639 |
150 | 471 | 376 |
表3 不同迭代次数下模型运行时间
Tab. 3 Model running time under different iteration times
迭代 步数 | 运行时间/s | 迭代 步数 | 运行时间/s | ||
---|---|---|---|---|---|
Conv-aLSTM | 本文模型 | Conv-aLSTM | 本文模型 | ||
50 | 161 | 122 | 200 | 645 | 489 |
100 | 320 | 243 | 250 | 784 | 639 |
150 | 471 | 376 |
模型 | 推理时间/s | ||
---|---|---|---|
Conv-LSTM | 250 | 13.967 0 | 20.333 5 |
本文模型 | 236 | 13.107 5 | 19.342 5 |
表4 模型推理时间及性能对比
Tab. 4 Model reasoning time and performance comparison
模型 | 推理时间/s | ||
---|---|---|---|
Conv-LSTM | 250 | 13.967 0 | 20.333 5 |
本文模型 | 236 | 13.107 5 | 19.342 5 |
1 | NALLAPERUMA D, NAWARATNE R, BANDARAGODA T, et al. Online incremental machine learning platform for big data-driven smart traffic management[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(12): 4679-4690. 10.1109/tits.2019.2924883 |
2 | VLAHOGIANNI E I, KARLAFTIS M G, GOLIAS J C. Short-term traffic forecasting: where we are and where we’re going[J]. Transportation Research Part C: Emerging Technologies, 2014, 43(Pt 1): 3-19. 10.1016/j.trc.2014.01.005 |
3 | WEIGEND A S, GERSHENFELD N A. Time Series Prediction: Forecasting the Future and Understanding the Past[M]. New York: Routledge, 2018: 2-11. |
4 | KUMAR S V, VANAJAKSHI L. Short-term traffic flow prediction using seasonal ARIMA model with limited input data[J]. European Transport Research Review, 2015, 7: No.21. 10.1007/s12544-015-0170-8 |
5 | KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]// Proceedings of the 25th International Conference on Neural Information Processing Systems. Red Hook, NY: Curran Associates Inc., 2012: 1097-1105. |
6 | 薛春铭,谭国真,丁男,等. 基于博弈论的人类驾驶与无人驾驶协作换道模型[J]. 计算机工程, 2017, 43(12):261-266. 10.3969/j.issn.1000-3428.2017.12.047 |
XUE C M, TAN G Z, DING N, et al. Cooperative lane-changing model of human driving and unmanned driving based on game theory[J]. Computer Engineering, 2017, 43(12): 261-266. 10.3969/j.issn.1000-3428.2017.12.047 | |
7 | 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. 10.7511/dllgxb202103004 |
BAO Z P, ZHI Y S, ZHANG S M, et al. BiGRU based multi-model maneuvers and trajectory prediction[J]. Journal of Dalian University of Technology, 2021, 61(3): 246-254. 10.7511/dllgxb202103004 | |
8 | MU S D, XIONG Z X, TIAN Y X. Intelligent traffic control system based on cloud computing and big data mining[J]. IEEE Transactions on Industrial Informatics, 2019, 15(12): 6583-6592. 10.1109/tii.2019.2929060 |
9 | WILLIAMS B M, HOEL L A. Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process: theoretical basis and empirical results[J]. Journal of Transportation Engineering, 2003, 129(6): 664-672. 10.1061/(asce)0733-947x(2003)129:6(664) |
10 | ZHAO J, SUN S L. High-order Gaussian process dynamical models for traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(7): 2014-2019. 10.1109/tits.2016.2515105 |
11 | VAPNIK V N. An overview of statistical learning theory[J]. IEEE Transactions on Neural Networks, 1999, 10(5): 988-999. 10.1109/72.788640 |
12 | BENGIO Y, COURVILLE A, VINCENT P. Representation learning: a review and new perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8): 1798-1828. 10.1109/tpami.2013.50 |
13 | FENG X X, LING X Y, ZHENG H F, et al. Adaptive multi-kernel SVM with spatial-temporal correlation for short-term traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(6): 2001-2013. 10.1109/tits.2018.2854913 |
14 | LV Y S, DUAN Y J, KANG W W, et al. Traffic flow prediction with big data: a deep learning approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(2): 865-873. 10.1109/tits.2015.2399852 |
15 | YU X J, SUN L J, YAN Y, et al. A short-term traffic flow prediction method based on spatial-temporal correlation using edge computing[J]. Computers and Electrical Engineering, 2021, 93: No.107219. 10.1016/j.compeleceng.2021.107219 |
16 | ZHENG J H, HUANG M F. Traffic flow forecast through time series analysis based on deep learning[J]. IEEE Access, 2020, 8: 82562-82570. 10.1109/access.2020.2990738 |
17 | SHAO H X, SOONG B H. Traffic flow prediction with Long Short-Term Memory Networks (LSTMs)[C]// Proceedings of the 2016 IEEE Region 10 Conference. Piscataway: IEEE, 2016: 2986-2989. 10.1109/tencon.2016.7848593 |
18 | LUO X L, LI D Y, YANG Y, et al. Spatiotemporal traffic flow prediction with KNN and LSTM[J]. Journal of Advanced Transportation, 2019, 2019: No.4145353. 10.1155/2019/4145353 |
19 | GUO S N, LIN Y F, LI S J, et al. Deep spatial-temporal 3D convolutional neural networks for traffic data forecasting[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20(10): 3913-3926. 10.1109/tits.2019.2906365 |
20 | KHAJEH HOSSEINI M, TALEBPOUR A. Traffic prediction using time-space diagram: a convolutional neural network approach[J]. Transportation Research Record, 2019, 2673(7): 425-435. 10.1177/0361198119841291 |
21 | 邓烜堃,万良,丁红卫,等. 基于深度学习的交通流量预测研究[J]. 计算机工程与应用, 2019, 55(2):228-235. 10.1109/icdmw.2019.00040 |
DENG X K, WANG L, DING H W, et al. Traffic flow prediction based on deep neural networks[J]. Computer Engineering and Applications, 2019, 55(2):228-235. 10.1109/icdmw.2019.00040 | |
22 | 徐凌伟,权天祺. 移动协作通信系统中断概率性能智能预测[J]. 聊城大学学报(自然科学版), 2020, 33(6): 33-39, 75. 10.19728/j.issn1672-6634.2020.06.004 |
XU L W, QUAN T Q. OP performance intelligent prediction of mobile cooperative communication system[J]. Journal of Liaocheng University (Natural Science Edition), 2020, 33(6): 33-39, 75. 10.19728/j.issn1672-6634.2020.06.004 | |
23 | 杜圣东,李天瑞,杨燕,等. 一种基于序列到序列时空注意力学习的交通流预测模型[J]. 计算机研究与发展, 2020, 57(8): 1715-1728. 10.7544/issn1000-1239.2020.20200169 |
DU S D, LI T R, YANG Y, et al. A sequence-to-sequence spatial-temporal attention learning model for urban traffic flow prediction[J]. Journal of Computer Research and Development, 2020, 57(8): 1715-1728. 10.7544/issn1000-1239.2020.20200169 | |
24 | 谷远利,陆文琦,李萌,等. 基于组合深度学习的快速路车道级速度预测研究[J]. 交通运输系统工程与信息, 2019, 19(4): 79-86. 10.1016/j.trc.2019.07.003 |
GU Y L, LU W Q, LI M, et al. Lane-level traffic speed prediction for expressways based on a combined deep learning model[J]. Transportation Systems Engineering and Information, 2019, 19(4):79-86. 10.1016/j.trc.2019.07.003 | |
25 | HUANG Y P, QIAN L P, FENG A Q, et al. Short-term traffic prediction by two-level data driven model in 5G-enabled edge computing networks[J]. IEEE Access, 2019, 7: 123981-123991. 10.1109/access.2019.2938236 |
26 | 张琛,徐国丽. 基于云遗传的RBF神经网络的交通流量预测[J]. 计算机工程与应用, 2014, 50(16): 216-220. 10.3778/j.issn.1002-8331.1208-0505 |
ZHANG C, XU G L. Prediction for traffic flow of RBF neural network based on cloud genetic algorithm[J]. Computer Engineering and Applications, 2014, 50(16): 216-220. 10.3778/j.issn.1002-8331.1208-0505 | |
27 | 刘思峰,郭天榜,党耀国,等. 灰色系统理论及其应用[M]. 2版. 北京:科学出版社, 1999:17-29. |
LIU S F, GUO T B, DANG Y G, et al. Grey System Theory and Its Application[M]. 2nd ed. Beijing: Science Press, 1999: 17-29. |
[1] | 杨鑫, 陈雪妮, 吴春江, 周世杰. 结合变种残差模型和Transformer的城市公路短时交通流预测[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2947-2951. |
[2] | 石乾宏, 杨燕, 江永全, 欧阳小草, 范武波, 陈强, 姜涛, 李媛. 面向空气质量预测的多粒度突变拟合网络[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2643-2650. |
[3] | 刘瑞华, 郝子赫, 邹洋杨. 基于多层级精细特征融合的步态识别算法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2250-2257. |
[4] | 吕锡婷, 赵敬华, 荣海迎, 赵嘉乐. 基于Transformer和关系图卷积网络的信息传播预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1760-1766. |
[5] | 夏进, 王正群, 朱世明. 基于时间序列分解的交通流量预测模型[J]. 《计算机应用》唯一官方网站, 2023, 43(4): 1129-1135. |
[6] | 张江峰, 闫涛, 陈斌, 钱宇华, 宋艳涛. 全局时空特征耦合的多景深三维形貌重建[J]. 《计算机应用》唯一官方网站, 2023, 43(3): 894-902. |
[7] | 马栋林, 马司周, 王伟杰. 基于图卷积网络和门控循环单元的多站点气温预测模型[J]. 《计算机应用》唯一官方网站, 2022, 42(1): 287-293. |
[8] | 王智文, 蒋联源, 王宇航, 王日凤, 张灿龙, 黄镇谨, 王鹏涛. 基于尺度自适应局部时空特征的足球比赛视频中的多运动员行为表示[J]. 计算机应用, 2016, 36(8): 2134-2138. |
[9] | 向坚 朱红丽. 基于三维特征提取的人体运动数据分析和检索[J]. 计算机应用, 2008, 28(5): 1344-1346. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||