基于维度加权的残差LSTM短期交通流量预测

doi:10.19678/j.issn.1000-3428.0052120

计算机工程 ›› 2019, Vol. 45 ›› Issue (6): 1-5. doi: 10.19678/j.issn.1000-3428.0052120

所属专题：智能交通专题；

基于维度加权的残差LSTM短期交通流量预测

李月龙^1,a³,唐德华^1,a^2,³,姜桂圆²,肖志涛^1,b³,耿磊^1,b³,张芳^1,b³,吴骏³

^1. 天津工业大学 a.计算机科学与技术学院; b.电子与信息工程学院,天津 300387
^2. 南洋理工大学计算机科学与工程学院,新加坡 639798
^3. 天津市光电检测技术与系统重点实验室,天津 300387

收稿日期:2018-07-16 修回日期:2018-08-31 出版日期:2019-06-15 发布日期:2019-06-17
作者简介:李月龙(1982—),男,副教授、博士,主研方向为智能交通、机器学习、模式识别|唐德华,硕士研究生;姜桂圆,研究员、博士;肖志涛,教授、博士|肖志涛,教授、博士;耿磊、张芳、吴骏,副教授、博士
基金资助:
国家自然科学基金(61771340,61302127,61601325);天津市自然科学基金(18JCYBJC15300)

Short Term Traffic Flow Forecasting Based on Dimension Weighted Residual LSTM

Yuelong LI^1,a³,Dehua TANG^1,a^2,³,Guiyuan JIANG²,Zhitao XIAO^1,b³,Lei GENG^1,b³,Fang ZHANG^1,b³,Jun WU³

^1. a.School of Computer Science and Technology;b.School of Electronics and Information Engineering,Tianjin Polytechnic University,Tianjin 300387,China
^2. School of Computer Science and Engineering,Nanyang Technological University,Singapore 639798,Singapore
^3. Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems,Tianjin 300387,China

Received:2018-07-16 Revised:2018-08-31 Online:2019-06-15 Published:2019-06-17

摘要/Abstract

摘要：

基于神经网络的交通流量预测由于嵌入了部分手工设计的特征,使得提取的网络特征功能单一,存在适应性及鲁棒性差、数据局部特征刻画不准确等问题。为此,提出基于残差长短期记忆网络(LSTM)的交通流量预测方法,利用集成学习思想将空间分布的数据端到端训练到残差LSTM网络中,同时在每个LSTM单元后引入维度加权单元,显式建模特征维度之间的相互依赖关系。实验结果表明,该方法能实现短期交通流量数据的自适应建模分析。

关键词: 智能交通, 短期交通流量预测, 残差连接, 长短期记忆网络, 维度加权

Abstract:

Due to the fact that the current neural network-based traffic flow forecasting method embeds part of the manually designed features,the feature extracted by the network has single function,bad adaptability,poor robustness and inaccurate characterization of the local features.Therefore,this paper proposes a traffic flow forecasting method based on residual Long Short Term Memory(LSTM).The model uses the idea of ensemble learning to train the spatially distributed data into a residual LSTM network.In addition,dimension weighted units are introduced after each LSTM unit to present the interdependencies between different dimensions of modeling features.Experimental results show that the method can realize adaptive modeling analysis of short term traffic flow data.

Key words: intelligent transportation, short term traffic flow forecasting, residual connection, Long Short Term Memory(LSTM) network, dimension weighted

李月龙, 唐德华, 姜桂圆, 肖志涛, 耿磊, 张芳, 吴骏. 基于维度加权的残差LSTM短期交通流量预测[J]. 计算机工程, 2019, 45(6): 1-5.

Yuelong LI, Dehua TANG, Guiyuan JIANG, Zhitao XIAO, Lei GENG, Fang ZHANG, Jun WU. Short Term Traffic Flow Forecasting Based on Dimension Weighted Residual LSTM[J]. Computer Engineering, 2019, 45(6): 1-5.

http://www.ecice06.com/CN/Y2019/V45/I6/1

图/表 4

参考文献 16

[1]	HUANG Wenhao, SONG Guojie, HONG Haikun , et al. Deep architecture for traffic flow prediction:deep belief networks with multitask learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2014,15(5):2191-2201.
[2]	贾显超, 陈旭梅, 弓晋丽 , 等. 基于混沌理论的短期交通流量多步预测[J]. 交通信息与安全, 2013,31(6):27-32.
[3]	曹政才, 韩丁富, 王永吉 . 面向城市交通网络的一种新型动态路径寻优方法[J]. 电子学报, 2012,40(10):2062-2067.
[4]	韩志聪, 樊彦国, 吴会胜 , 等. 基于GA-SVR模型的短期交通流量预测方法研究[J]. 公路交通科技, 2017,34(1):130-136.
[5]	DU Shengdong, LI Tianrui, GONG Xun. Traffic flow forecasting based on hybrid deep learning framework[C]//Proceedings of International Conference on Intelligent Systems and Knowledge Engineering.Washington D.C., USA: IEEE Press, 2018: 1-6.
[6]	JEONG Y S, BYON Y J, CASTRO-NETO M M , et al. Supervised weighting-online learning algorithm for short-term traffic flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2013,14(4):1700-1707. doi: 10.1109/TITS.2013.2267735
[7]	ZHAO Zheng, CHEN Weihai, WU Xingming , et al. LSTM network:a deep learning approach for short-term traffic forecast[J]. IET Intelligent Transport Systems, 2017,11(2):68-75.
[8]	HUANG Wenhao, SONG Guojie, HONG Haikun , et al. Deep architecture for traffic flow prediction:deep belief networks with multitask learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2014,15(5):2191-2201. doi: 10.1109/TITS.2014.2311123
[9]	LÜ Yisheng, DUAN Yanjie, KANG Wenwen , 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]	MA Xiaolei, YU Haiyang, WANG Yunpeng , et al. Large-scale transportation network congestion evolution prediction using deep learning theory[J]. PLoS One, 2015,10(3):1-17.
[11]	PRAKASH A, HASAN S A, LEE K , et al. Neural paraphrase generation with stacked residual LSTM networks[EB/OL].[2018-03-20]. URL
[12]	Live traffic information from highways England[EB/OL].( 2015-12-10)[2018-03-20]. URL
[13]	PENG Zhenrui, GAO Pu, MENG Jianjun . Traffic forecasting using least squares support vector machines[J]. Transportmetrica, 2009,5(3):193-213.
[14]	ZHANG Yanru, HAGHANI A . A gradient boosting method to improve travel time prediction[J]. Transportation Research Part C:Emerging Technologies, 2015,58:308-324.
[15]	LINGRAS P, SHARMA S, ZHON G M . Prediction of recreational travel using genetically designed regression and time-delay neural network models [EB/OL].[2018-03-20]. URL
[16]	DUAN Yanjie, LÜ Yisheng, WANG Feiyue. Travel time prediction with LSTM neural network[C]//Proceedings of the 19th International Conference on Intelligent Transportation Systems.Washington D.C., USA: IEEE Press, 2016: 1053-1058.

模型	预测时间为15 min	预测时间为30 min	预测时间为45 min
SVM	5.675	15.767	36.597
GBRT	5.234	15.154	33.381
ARIMA	7.139	16.431	32.645
DBN	5.618	15.392	29.176
RNN	3.302	14.021	27.559
Stacked LSTM(6层)	4.133	13.174	26.501
Stacked-OW-LSTM(6层)	3.562	12.667	26.219
Stacked LSTM(3层)	2.963	11.718	24.634
Stacked-OW-LSTM(3层)	2.451	10.995	24.317
2D-ODC-LSTM	1.811	8.717	21.842
Res-LSTM(6层)	1.568	6.972	19.406
Res-OW-LSTM(6层)	0.899	5.247	18.628

模型	预测时间为15 min	预测时间为30 min	预测时间为45 min
SVM	5.675	15.767	36.597
GBRT	5.234	15.154	33.381
ARIMA	7.139	16.431	32.645
DBN	5.618	15.392	29.176
RNN	3.302	14.021	27.559
Stacked LSTM(6层)	4.133	13.174	26.501
Stacked-OW-LSTM(6层)	3.562	12.667	26.219
Stacked LSTM(3层)	2.963	11.718	24.634
Stacked-OW-LSTM(3层)	2.451	10.995	24.317
2D-ODC-LSTM	1.811	8.717	21.842
Res-LSTM(6层)	1.568	6.972	19.406
Res-OW-LSTM(6层)	0.899	5.247	18.628

模型	预测时间为15 min	预测时间为30 min	预测时间为45 min
Res-LSTM	1.568	6.972	19.406
Res-OW-LSTM(tanh、r=16)	1.478	6.744	19.693
Res-OW-LSTM(ReLU、r=8)	1.133	6.402	19.531
Res-OW-LSTM(ReLU、r=16)	0.899	5.247	18.628
Res-OW-LSTM(ReLU、r=32)	1.363	6.460	19.566
Res-OW-LSTM(ReLU、r=64)	1.479	5.935	18.853
Res-OW-LSTM(ReLU、r=128)	1.561	6.859	20.301
Res-OW-LSTM(ReLU、r=256)	1.055	6.349	18.890
Res-OW-LSTM(ReLU、r=320)	1.208	6.419	19.631
Res-OW-LSTM(ReLU、r=500)	1.423	6.572	19.738

模型	预测时间为15 min	预测时间为30 min	预测时间为45 min
Res-LSTM	1.568	6.972	19.406
Res-OW-LSTM(tanh、r=16)	1.478	6.744	19.693
Res-OW-LSTM(ReLU、r=8)	1.133	6.402	19.531
Res-OW-LSTM(ReLU、r=16)	0.899	5.247	18.628
Res-OW-LSTM(ReLU、r=32)	1.363	6.460	19.566
Res-OW-LSTM(ReLU、r=64)	1.479	5.935	18.853
Res-OW-LSTM(ReLU、r=128)	1.561	6.859	20.301
Res-OW-LSTM(ReLU、r=256)	1.055	6.349	18.890
Res-OW-LSTM(ReLU、r=320)	1.208	6.419	19.631
Res-OW-LSTM(ReLU、r=500)	1.423	6.572	19.738

选择文件类型/文献管理软件名称

选择包含的内容

基于维度加权的残差LSTM短期交通流量预测

Short Term Traffic Flow Forecasting Based on Dimension Weighted Residual LSTM

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 4

参考文献 16

相关文章 15

编辑推荐

Metrics

本文评价

[1]	席建锐, 唐红梅, 梁春阳, 刘鑫. 基于改进隐函数的点云物体重建[J]. 计算机工程, 2023, 49(7): 214-222.
[2]	赵宏, 陈志文, 郭岚, 安冬. 基于ViT与语义引导的视频内容描述生成[J]. 计算机工程, 2023, 49(5): 247-254.
[3]	曹书鑫, 冯藤藤, 葛凤培, 梁春燕. 基于尺度相关‐双向长短期记忆网络模型的说话人识别[J]. 计算机工程, 2023, 49(4): 289-296.
[4]	邹国建, 赖子良, 李晔. 基于时空注意力网络的动态高速路网交通速度预测[J]. 计算机工程, 2023, 49(2): 303-313.
[5]	葛宇然, 付强. 基于时空联合学习的城市交通流短时预测模型[J]. 计算机工程, 2023, 49(1): 270-278.
[6]	梁登玉, 刘大明. 融合多粒度信息与外部知识的短文本匹配模型[J]. 计算机工程, 2022, 48(8): 129-135,143.
[7]	冯思芸, 施振佺, 曹阳. 基于全局时空特性的城市路网交通速度预测模型[J]. 计算机工程, 2022, 48(5): 112-117.
[8]	彭桐歆, 韩勇, 王程, 张志浩. 面向短时地铁客流量预测的混合深度学习模型[J]. 计算机工程, 2022, 48(5): 297-305.
[9]	贺小伟, 徐靖杰, 王宾, 吴昊, 张博文. 基于GRU-LSTM组合模型的云计算资源负载预测研究[J]. 计算机工程, 2022, 48(5): 11-17,34.
[10]	张毅, 林云汉, 刘双元. 一种面向低采样率的点云数据处理网络[J]. 计算机工程, 2022, 48(11): 240-246.
[11]	史占堂, 马玉鹏, 赵凡, 马博. 基于CNN‐Head Transformer编码器的中文命名实体识别[J]. 计算机工程, 2022, 48(10): 73-80.
[12]	江旭, 钱雪忠, 宋威. 结合残差BiLSTM与句袋注意力的远程监督关系抽取[J]. 计算机工程, 2022, 48(10): 110-115,122.
[13]	谭嵋, 刘士豪, 周婉, 陈国文, 胡学敏. 基于深度时空Q网络的机器人疏散人群算法[J]. 计算机工程, 2021, 47(6): 305-311.
[14]	耿俊杰, 李晓明, 颜金尧. 基于网络流量预测的DASH系统优化[J]. 计算机工程, 2021, 47(5): 292-300.
[15]	包恒彬, 马玉鹏, 杨奉毅, 韩云飞. 基于时空注意力机制的加油站级客流量预测[J]. 计算机工程, 2021, 47(4): 291-297.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于维度加权的残差LSTM短期交通流量预测

Short Term Traffic Flow Forecasting Based on Dimension Weighted Residual LSTM

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 4

参考文献 16

相关文章 15

编辑推荐

Metrics

本文评价