Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow Forecasting
Due to the periodic and dynamic changes of traffic flow and the spatial–temporal coupling interaction of complex road networks, traffic flow forecasting is highly challenging and rarely yields satisfactory prediction results. In this paper, we propose a novel methodology named the Augmented Multi-co...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | ISPRS International Journal of Geo-Information |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2220-9964/11/2/88 |
| _version_ | 1827654604652806144 |
|---|---|
| author | Chi Zhang Hong-Yu Zhou Qiang Qiu Zhichun Jian Daoye Zhu Chengqi Cheng Liesong He Guoping Liu Xiang Wen Runbo Hu |
| author_facet | Chi Zhang Hong-Yu Zhou Qiang Qiu Zhichun Jian Daoye Zhu Chengqi Cheng Liesong He Guoping Liu Xiang Wen Runbo Hu |
| author_sort | Chi Zhang |
| collection | DOAJ |
| description | Due to the periodic and dynamic changes of traffic flow and the spatial–temporal coupling interaction of complex road networks, traffic flow forecasting is highly challenging and rarely yields satisfactory prediction results. In this paper, we propose a novel methodology named the Augmented Multi-component Recurrent Graph Convolutional Network (AM-RGCN) for traffic flow forecasting by addressing the problems above. We first introduce the augmented multi-component module to the traffic forecasting model to tackle the problem of periodic temporal shift emerging in traffic series. Then, we propose an encoder–decoder architecture for spatial–temporal prediction. Specifically, we propose the Temporal Correlation Learner (TCL) which incorporates one-dimensional convolution into LSTM to utilize the intrinsic temporal characteristics of traffic flow. Moreover, we combine TCL with the graph convolutional network to handle the spatial–temporal coupling interaction of the road network. Similarly, the decoder consists of TCL and convolutional neural networks to obtain high-dimensional representations from multi-step predictions based on spatial–temporal sequences. Extensive experiments on two real-world road traffic datasets, PEMSD4 and PEMSD8, demonstrate that our AM-RGCN achieves the best results. |
| first_indexed | 2024-03-09T21:46:53Z |
| format | Article |
| id | doaj.art-dc709ea855264da8a2e88da0b6f973a1 |
| institution | Directory Open Access Journal |
| issn | 2220-9964 |
| language | English |
| last_indexed | 2024-03-09T21:46:53Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | ISPRS International Journal of Geo-Information |
| spelling | doaj.art-dc709ea855264da8a2e88da0b6f973a12023-11-23T20:15:33ZengMDPI AGISPRS International Journal of Geo-Information2220-99642022-01-011128810.3390/ijgi11020088Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow ForecastingChi Zhang0Hong-Yu Zhou1Qiang Qiu2Zhichun Jian3Daoye Zhu4Chengqi Cheng5Liesong He6Guoping Liu7Xiang Wen8Runbo Hu9Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, ChinaDepartment of Computer Science, The University of Hong Kong, Hong Kong 999077, ChinaInstitute of Computing Technology, Chinese Academy of Sciences, Beijing 101408, ChinaShopee Information Technology Co., Ltd., Shenzhen 518063, ChinaAcademy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, ChinaCollege of Engineering, Peking University, Beijing 100871, ChinaXi’an Research Institute of Surveying and Mapping, Xi’an 710000, ChinaDidi Chuxing, Beijing 100085, ChinaDidi Chuxing, Beijing 100085, ChinaDidi Chuxing, Beijing 100085, ChinaDue to the periodic and dynamic changes of traffic flow and the spatial–temporal coupling interaction of complex road networks, traffic flow forecasting is highly challenging and rarely yields satisfactory prediction results. In this paper, we propose a novel methodology named the Augmented Multi-component Recurrent Graph Convolutional Network (AM-RGCN) for traffic flow forecasting by addressing the problems above. We first introduce the augmented multi-component module to the traffic forecasting model to tackle the problem of periodic temporal shift emerging in traffic series. Then, we propose an encoder–decoder architecture for spatial–temporal prediction. Specifically, we propose the Temporal Correlation Learner (TCL) which incorporates one-dimensional convolution into LSTM to utilize the intrinsic temporal characteristics of traffic flow. Moreover, we combine TCL with the graph convolutional network to handle the spatial–temporal coupling interaction of the road network. Similarly, the decoder consists of TCL and convolutional neural networks to obtain high-dimensional representations from multi-step predictions based on spatial–temporal sequences. Extensive experiments on two real-world road traffic datasets, PEMSD4 and PEMSD8, demonstrate that our AM-RGCN achieves the best results.https://www.mdpi.com/2220-9964/11/2/88traffic flow forecastingspatial–temporal predictiongraph convolutional networksaugmented multi-component |
| spellingShingle | Chi Zhang Hong-Yu Zhou Qiang Qiu Zhichun Jian Daoye Zhu Chengqi Cheng Liesong He Guoping Liu Xiang Wen Runbo Hu Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow Forecasting ISPRS International Journal of Geo-Information traffic flow forecasting spatial–temporal prediction graph convolutional networks augmented multi-component |
| title | Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow Forecasting |
| title_full | Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow Forecasting |
| title_fullStr | Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow Forecasting |
| title_full_unstemmed | Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow Forecasting |
| title_short | Augmented Multi-Component Recurrent Graph Convolutional Network for Traffic Flow Forecasting |
| title_sort | augmented multi component recurrent graph convolutional network for traffic flow forecasting |
| topic | traffic flow forecasting spatial–temporal prediction graph convolutional networks augmented multi-component |
| url | https://www.mdpi.com/2220-9964/11/2/88 |
| work_keys_str_mv | AT chizhang augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT hongyuzhou augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT qiangqiu augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT zhichunjian augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT daoyezhu augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT chengqicheng augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT liesonghe augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT guopingliu augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT xiangwen augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting AT runbohu augmentedmulticomponentrecurrentgraphconvolutionalnetworkfortrafficflowforecasting |