基于时空特征向量的长短期记忆人工神经网络的城市公交旅行时间预测

doi:10.11772/j.issn.1001-9081.2020060467

计算机应用 ›› 2021, Vol. 41 ›› Issue (3): 875-880.DOI: 10.11772/j.issn.1001-9081.2020060467

所属专题：前沿与综合应用

基于时空特征向量的长短期记忆人工神经网络的城市公交旅行时间预测

张欣环¹, 刘宏杰², 施俊庆¹, 毛程远¹, 孟国连¹

1. 浙江师范大学道路与交通工程研究中心, 浙江金华 321004;
2. 西安交通大学电子信息工程学院, 西安 710049

收稿日期:2020-06-18 修回日期:2020-10-12 出版日期:2021-03-10 发布日期:2020-12-17
通讯作者: 刘宏杰
作者简介:张欣环(1977-),女,陕西乾县人,讲师,博士,主要研究方向:城市公共交通系统评价与优化、网联自动车路系统;刘宏杰(1988-),男,陕西渭南人,博士研究生,主要研究方向:智慧交通、人工智能、机器学习、网联自动车路系统;施俊庆(1985-),男,浙江永康人,副教授,博士,主要研究方向:智能交通、交通流理论、网联自动车路系统;毛程远(1982-),男,浙江杭州人,讲师,博士,主要研究方向:城市交通规划与管理、交通流理论;孟国连(1986-),女,山东潍坊人,实验师,博士研究生,主要研究方向:城市交通规划与管理、交通流理论。
基金资助:
浙江省教育厅项目（Y201738488）；浙江省自然科学基金资助项目（LY18G010009，LY18G030021）；教育部留学回国人员科研启动基金资助项目（ZC304012027）。

LSTM and artificial neural network for urban bus travel time prediction based on spatiotemporal eigenvectors

ZHANG Xinhuan¹, LIU Hongjie², SHI Junqing¹, MAO Chengyuan¹, MENG Guolian¹

1. Road and Traffic Engineering Research Center, Zhejiang Normal University, Jinhua Zhejiang 321004, China;
2. School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an Shaanxi 710049, China

Received:2020-06-18 Revised:2020-10-12 Online:2021-03-10 Published:2020-12-17
Supported by:
This work is partially supported by Project of the Zhejiang Provincial Educational Committee (Y201738488), the Zhejiang Provincial Natural Science Foundation (LY18G010009, LY18G030021), the Scientific Research Start Foundation for the Returned Scholars, Ministry of Education of China (ZC304012027).

摘要/Abstract

摘要： 针对“随着预测距离的增加，旅行时间预测的难度加大”的问题，提出了一种基于时空特征向量的长短期记忆（LSTM）和人工神经网络（ANN）的综合预测模型。首先，将24 h切分为288个时间切片，以生成时间特征向量；然后，基于时间切片建立LSTM时间窗口模型，该模型可解决长期预测的窗口移动问题；其次，将公交线路切分为多个空间切片，并使用当前空间切片的共同平均速度作为瞬时速度，同时将每个空间切片的预测时间用作空间特征向量，并将其发送到新型的混合神经网络模型LSTM-A中，该模型结合两种预测模型的优点并解决了公交旅行时间预测问题；最后，基于实验数据集进行了实验和测试：将公交站点间的预测问题划分为线路切片预测子问题，并针对每个相关的子问题引入了实时计算的概念，从而避免了复杂路况带来的预测误差。实验结果表明，所提算法在准确性、适用性方面均优于单个神经网络模型。综上，所提的新型混合神经网络模型LSTM-A能从时间特征的维度实现长距离到站预测、从空间特征的维度实现短距离到站预测，从而有效地解决了城市公交旅行时间预测问题，避免了公交车辆的远程依赖和错误积累。

关键词: 城市交通, 长短期记忆网络, 人工神经网络, 长短期记忆人工神经网络, 旅行时间预测

Abstract: Aiming at the problem that "with the increase of the prediction distance, the prediction of travel time becomes more and more difficult", a comprehensive prediction model of Long Short Term Memory (LSTM) and Artificial Neural Network (ANN) based on spatiotemporal eigenvectors was proposed. Firstly, 24 hours were segmented into 288 time slices to generate time eigenvectors. Secondly, the LSTM time window model was established based on the time slices. This model was able to solve the window movement problem of long-time prediction. Thirdly, the bus line was divided into multiple space slices and the average velocity of the current space slice was used as the instantaneous velocity. At the same time, the predicted time of each space slice would be used as the spatial eigenvector and sent to the new hybrid neural network model named LSTM-A (Long Short Term Memory Artificial neural network). This model combined with the advantages of the two prediction models and solved the problem of bus travel time prediction. Finally, based on the experimental dataset, experiments and tests were carried out:the prediction problem between bus stations was divided into sub-problems of line slice prediction, and the concept of real-time calculation was introduced to each related sub-problem, so as to avoid the prediction error caused by complex road conditions. Experimental results show that the proposed algorithm is superior to single neural network models in both accuracy and applicability. In conclusion, the proposed new hybrid neural network model LSTM-A can realize the long-distance arrival time prediction from the dimension of time feature and the short-distance arrival time prediction from the dimension of spatial feature, thus effectively solving the problem of urban bus travel time prediction and avoiding the remote dependency and error accumulation of buses.

Key words: urban traffic, Long Short-Term Memory (LSTM) network, Artificial Neural Network (ANN), Long Short-Term Memory Artificial neural network (LSTM-A), travel time prediction

中图分类号:

TP18
U491

张欣环, 刘宏杰, 施俊庆, 毛程远, 孟国连. 基于时空特征向量的长短期记忆人工神经网络的城市公交旅行时间预测[J]. 计算机应用, 2021, 41(3): 875-880.

ZHANG Xinhuan, LIU Hongjie, SHI Junqing, MAO Chengyuan, MENG Guolian. LSTM and artificial neural network for urban bus travel time prediction based on spatiotemporal eigenvectors[J]. Journal of Computer Applications, 2021, 41(3): 875-880.

参考文献

[1] XU H,YING J. Bus arrival time prediction with real-time and historic data[J]. Cluster Computing,2017,20(4):3099-3106.
[2] PENG Z,JIANG Y,YANG X,et al. Bus arrival time prediction based on PCA-GA-SVM[J]. Neural Network World,2018,28(1):87-104.
[3] YAO B,CHEN C,CAO Q,et al. Short-term traffic speed prediction for an urban corridor[J]. Computer-Aided Civil and Infrastructure Engineering,2017,32(2):154-169.
[4] CRISTIANINI N,SHAWE-TAYLOR J. An Introduction to Support Vector Machines and Other Kernel-Based Learning Methods[M]. Cambridge:Cambridge University Press,2000:78-82.
[5] ABIDIN A F,KOLBERG M. Towards improved vehicle arrival time prediction in public transportation:integrating SUMO and Kalman filter models[C]//Proceedings of the 17th UKSim-AMSS International Conference on Modelling and Simulation. Piscataway:IEEE,2015:147-152.
[6] 范光鹏, 孙仁诚, 邵峰晶. 基于LSTM和Kalman滤波的公交车到站时间预测[J]. 计算机应用与软件,2018,35(4):91-96.(FAN G P,SUN R C,SHAO F J. Bus arrival time prediction based on LSTM and Kalman filtering[J]. Computer Applications and Software,2018,35(4):91-96.)
[7] HE Z,ZHENG L,CHEN P,et al. Mapping to cells:a simple method to extract traffic dynamics from probe vehicle data[J]. Computer-Aided Civil and Infrastructure Engineering,2017,32(3):252-267.
[8] LIU W,LIU J,JIANG H,et al. WiLocator:WiFi-sensing based real-time bus tracking and arrival time prediction in urban environments[C]//Proceedings of the IEEE 36th International Conference on Distributed Computing Systems. Piscataway:IEEE, 2016:529-538.
[9] FAROOQ M U,SHAKOOR A,SIDDIQUE A B. GPS based public transport arrival time prediction[C]//Proceedings of the 2017 International Conference on Frontiers of Information Technology. Piscataway:IEEE,2017:76-81.
[10] DHIVYABHARATHI B,HIMA E S,VANAJAKSHI L. Stream travel time prediction using particle filtering approach[J]. Transportation Letters,2018,10(2):75-82.
[11] 张孝梅, 陈旭梅, 张溪. 基于非参数回归-粒子滤波模型的公交到站时间预测[J]. 公路交通科技,2020,37(4):118-124. (ZHANG X M,CHEN X M,ZHANG X. Prediction of bus arrival time based on nonparametric regression and particle filtering model[J]. Journal of Highway and Transportation Research and Development,2020,37(4):118-124.)
[12] CHEN C. An arrival time prediction method for bus system[J]. IEEE Internet of Things Journal,2018,5(5):4231-4232.
[13] ZHANG K, JIA N, ZHENG L, et al. A novel generative adversarial network for estimation of trip travel time distribution with trajectory data[J]. Transportation Research Part C:Emerging Technologies,2019,108:223-244.
[14] GAL A,MANDELBAUM A,SCHNITZLER F,et al. Traveling time prediction in scheduled transportation with journey segments[J]. Information Systems,2017,64:266-280.
[15] KUMAR,B A,VANAJAKSHI L,SUBRAMANIAN S C. Bus travel time prediction using a time-space discretization approach[J]. Transportation Research Part C:Emerging Technologies, 2017,79:308-332.
[16] WU R,LUO G,SHAO J,et al. Location prediction on trajectory data:a review[J]. Big Data Mining and Analytics,2018,1(2):108-127.
[17] RAHMAN M M,WIRASINGHE S C,KATTAN L. Analysis of bus travel time distributions for varying horizons and real-time applications[J]. Transportation Research Part C:Emerging Technologies,2018,86:453-466.
[18] MA X,TAO Z,WANG Y,et al. Long short-term memory neural network for traffic speed prediction using remote microwave sensor data[J]. Transportation Research Part C:Emerging Technologies,2015,54:187-197.
[19] ZHENG L,ZHU C,ZHU N,et al. Feature selection-based approach for urban short-term travel speed prediction[J]. IET Intelligent Transport Systems,2018,12(6):474-484.

基于时空特征向量的长短期记忆人工神经网络的城市公交旅行时间预测

LSTM and artificial neural network for urban bus travel time prediction based on spatiotemporal eigenvectors

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]	李康康, 张静. 基于注意力机制的多层次编码和解码的图像描述模型[J]. 计算机应用, 2021, 41(9): 2504-2509.
[2]	党伟超, 李涛, 白尚旺, 高改梅, 刘春霞. 基于自注意力长短期记忆网络的Web软件系统实时剩余寿命预测方法[J]. 计算机应用, 2021, 41(8): 2346-2351.
[3]	张元钧, 张曦煌. 基于图卷积与长短期记忆网络的动态网络表示学习模型[J]. 计算机应用, 2021, 41(7): 1857-1864.
[4]	武国亮, 徐继宁. 基于命名实体识别任务反馈增强的中文突发事件抽取方法[J]. 计算机应用, 2021, 41(7): 1891-1896.
[5]	武光利, 李雷霆, 郭振洲, 王成祥. 基于改进的双向长短期记忆网络的视频摘要生成模型[J]. 计算机应用, 2021, 41(7): 1908-1914.
[6]	徐萌, 王亚锟. 基于双向长短期记忆网络的DA40飞机碳刹车片剩余寿命预测[J]. 计算机应用, 2021, 41(5): 1527-1532.
[7]	李文惠, 曾上游, 王金金. 基于改进注意力机制的图像描述生成算法[J]. 计算机应用, 2021, 41(5): 1262-1267.
[8]	许力, 李建华. 基于句法依存分析的图网络生物医学命名实体识别[J]. 计算机应用, 2021, 41(2): 357-362.
[9]	刘思嘉, 童向荣. 基于强化学习的城市交通路径规划[J]. 计算机应用, 2021, 41(1): 185-190.
[10]	钱斌, 郑楷洪, 陈子鹏, 肖勇, 李森, 叶纯壮, 马千里. 基于残差连接长短期记忆网络的时间序列修复模型[J]. 计算机应用, 2021, 41(1): 243-248.
[11]	肖勇, 郑楷洪, 郑镇境, 钱斌, 李森, 马千里. 基于多尺度跳跃深度长短期记忆网络的短期多变量负荷预测[J]. 计算机应用, 2021, 41(1): 231-236.
[12]	傅魁, 梁少晴, 李冰. 基于改进的深度Q网络结构的商品推荐模型[J]. 计算机应用, 2020, 40(9): 2613-2621.
[13]	林志兴, 王立可. 基于深度特征和Seq2Seq模型的网络态势预测方法[J]. 计算机应用, 2020, 40(8): 2241-2247.
[14]	胡学敏, 童秀迟, 郭琳, 张若晗, 孔力. 基于深度视觉注意神经网络的端到端自动驾驶模型[J]. 计算机应用, 2020, 40(7): 1926-1931.
[15]	曹堉, 王成, 王鑫, 高悦尔. 基于时空节点选择和深度学习的城市道路短时交通流预测[J]. 计算机应用, 2020, 40(5): 1488-1493.