Model-enhanced spatial-temporal attention networks for traffic density prediction
Abstract Traffic density is a crucial indicator for evaluating the level of service, as it directly reflects the degree of road congestion and driving comfort. However, accurately predicting real-time traffic density has been a significant challenge in Intelligent Transportation Systems (ITS) due to...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-11-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01669-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571153331257344 |
|---|---|
| author | Qi Guo Qi Tan Yue Peng Long Xiao Miao Liu Benyun Shi |
| author_facet | Qi Guo Qi Tan Yue Peng Long Xiao Miao Liu Benyun Shi |
| author_sort | Qi Guo |
| collection | DOAJ |
| description | Abstract Traffic density is a crucial indicator for evaluating the level of service, as it directly reflects the degree of road congestion and driving comfort. However, accurately predicting real-time traffic density has been a significant challenge in Intelligent Transportation Systems (ITS) due to the nonlinear and spatial-temporal dynamic complexity of traffic density. In this paper, we propose a novel Model-enhanced Spatial-Temporal Attention Network (MSTAN), which constructs a spatial-temporal traffic kernel density model using the Kernel Density Estimation (KDE) method to process the spatiotemporal data and calculate the probabilities of various spatiotemporal events. These probabilities are input into the attention mechanism, enabling the model to recognize the inherent connection between dynamic and distant events. Through this fusion, the network can deeply learn and analyze the spatial-temporal properties of traffic features. Furthermore, this paper utilizes the attention mechanism to dynamically model spatial-temporal dependencies, capturing real-time traffic conditions and density, and constructs a spatial-temporal attention module for learning. To validate the performance of the proposed MSTAN model, experiments are conducted on two public datasets of California highways (PeMS04 and PeMS08). The experimental results demonstrate that the MSTAN model outperforms existing state-of-the-art baseline models in terms of prediction accuracy, thus proving the effectiveness of the model both theoretically and practically. |
| format | Article |
| id | doaj-art-2ca9ddc012654beb8fd5e334336998cc |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-2ca9ddc012654beb8fd5e334336998cc2025-02-02T12:49:16ZengSpringerComplex & Intelligent Systems2199-45362198-60532024-11-0111111510.1007/s40747-024-01669-9Model-enhanced spatial-temporal attention networks for traffic density predictionQi Guo0Qi Tan1Yue Peng2Long Xiao3Miao Liu4Benyun Shi5College of Computer and Information Engineering (College of Artificial Intelligence), Nanjing Tech UniversityCollege of Computer and Information Engineering (College of Artificial Intelligence), Nanjing Tech UniversityCollege of Computer and Information Engineering (College of Artificial Intelligence), Nanjing Tech UniversityCollege of Computer and Information Engineering (College of Artificial Intelligence), Nanjing Tech UniversityCollege of Computer and Information Engineering (College of Artificial Intelligence), Nanjing Tech UniversityCollege of Computer and Information Engineering (College of Artificial Intelligence), Nanjing Tech UniversityAbstract Traffic density is a crucial indicator for evaluating the level of service, as it directly reflects the degree of road congestion and driving comfort. However, accurately predicting real-time traffic density has been a significant challenge in Intelligent Transportation Systems (ITS) due to the nonlinear and spatial-temporal dynamic complexity of traffic density. In this paper, we propose a novel Model-enhanced Spatial-Temporal Attention Network (MSTAN), which constructs a spatial-temporal traffic kernel density model using the Kernel Density Estimation (KDE) method to process the spatiotemporal data and calculate the probabilities of various spatiotemporal events. These probabilities are input into the attention mechanism, enabling the model to recognize the inherent connection between dynamic and distant events. Through this fusion, the network can deeply learn and analyze the spatial-temporal properties of traffic features. Furthermore, this paper utilizes the attention mechanism to dynamically model spatial-temporal dependencies, capturing real-time traffic conditions and density, and constructs a spatial-temporal attention module for learning. To validate the performance of the proposed MSTAN model, experiments are conducted on two public datasets of California highways (PeMS04 and PeMS08). The experimental results demonstrate that the MSTAN model outperforms existing state-of-the-art baseline models in terms of prediction accuracy, thus proving the effectiveness of the model both theoretically and practically.https://doi.org/10.1007/s40747-024-01669-9Spatial-temporal predictionAttention mechanismTraffic density predictionKernel density estimation |
| spellingShingle | Qi Guo Qi Tan Yue Peng Long Xiao Miao Liu Benyun Shi Model-enhanced spatial-temporal attention networks for traffic density prediction Complex & Intelligent Systems Spatial-temporal prediction Attention mechanism Traffic density prediction Kernel density estimation |
| title | Model-enhanced spatial-temporal attention networks for traffic density prediction |
| title_full | Model-enhanced spatial-temporal attention networks for traffic density prediction |
| title_fullStr | Model-enhanced spatial-temporal attention networks for traffic density prediction |
| title_full_unstemmed | Model-enhanced spatial-temporal attention networks for traffic density prediction |
| title_short | Model-enhanced spatial-temporal attention networks for traffic density prediction |
| title_sort | model enhanced spatial temporal attention networks for traffic density prediction |
| topic | Spatial-temporal prediction Attention mechanism Traffic density prediction Kernel density estimation |
| url | https://doi.org/10.1007/s40747-024-01669-9 |
| work_keys_str_mv | AT qiguo modelenhancedspatialtemporalattentionnetworksfortrafficdensityprediction AT qitan modelenhancedspatialtemporalattentionnetworksfortrafficdensityprediction AT yuepeng modelenhancedspatialtemporalattentionnetworksfortrafficdensityprediction AT longxiao modelenhancedspatialtemporalattentionnetworksfortrafficdensityprediction AT miaoliu modelenhancedspatialtemporalattentionnetworksfortrafficdensityprediction AT benyunshi modelenhancedspatialtemporalattentionnetworksfortrafficdensityprediction |