scholarly journals Traffic Speed Prediction: An Attention-Based Method

Sensors ◽  
2019 ◽  
Vol 19 (18) ◽  
pp. 3836 ◽  
Author(s):  
Duanyang Liu ◽  
Longfeng Tang ◽  
Guojiang Shen ◽  
Xiao Han
Keyword(s):  
Intelligent Transportation Systems ◽  
Transportation Systems ◽  
Similar Distribution ◽  
Short Term ◽  
Time Step ◽  
Temporal Clustering ◽  
Spatial Features ◽  
Speed Prediction ◽  
Traffic Speed ◽  
Prediction Approach

Short-term traffic speed prediction has become one of the most important parts of intelligent transportation systems (ITSs). In recent years, deep learning methods have demonstrated their superiority both in accuracy and efficiency. However, most of them only consider the temporal information, overlooking the spatial or some environmental factors, especially the different correlations between the target road and the surrounding roads. This paper proposes a traffic speed prediction approach based on temporal clustering and hierarchical attention (TCHA) to address the above issues. We apply temporal clustering to the target road to distinguish the traffic environment. Traffic data in each cluster have a similar distribution, which can help improve the prediction accuracy. A hierarchical attention-based mechanism is then used to extract the features at each time step. The encoder measures the importance of spatial features, and the decoder measures the temporal ones. The proposed method is evaluated over the data of a certain area in Hangzhou, and experiments have shown that this method can outperform the state of the art for traffic speed prediction.

scholarly journals Space-Time Hybrid Model for Short-Time Travel Speed Prediction

2018 ◽  
Vol 2018 ◽  
pp. 1-9 ◽  
Author(s):  
Qi Fan ◽  
Wei Wang ◽  
Xiaojian Hu ◽  
Xuedong Hua ◽  
Zhuyun Liu
Keyword(s):  
Hybrid Model ◽  
Intelligent Transportation Systems ◽  
Traffic Management ◽  
Arima Model ◽  
Periodic Variation ◽  
Travel Speed ◽  
Transportation Systems ◽  
Speed Prediction ◽  
Traffic Speed ◽  
Short Time

Short-time traffic speed forecasting is a significant issue for developing Intelligent Transportation Systems applications, and accurate speed forecasting results are necessary inputs for Intelligent Traffic Security Information System (ITSIS) and advanced traffic management systems (ATMS). This paper presents a hybrid model for travel speed based on temporal and spatial characteristics analysis and data fusion. This proposed methodology predicts speed by dividing the data into three parts: a periodic trend estimated by Fourier series, a residual part modeled by the ARIMA model, and the possible events affected by upstream or downstream traffic conditions. The aim of this study is to improve the accuracy of the prediction by modeling time and space variation of speed, and the forecast results could simultaneously reflect the periodic variation of traffic speed and emergencies. This information could provide decision-makers with a basis for developing traffic management measures. To achieve the research objective, one year of speed data was collected in Twin Cities Metro, Minnesota. The experimental results demonstrate that the proposed method can be used to explore the periodic characteristics of speed data and show abilities in increasing the accuracy of travel speed prediction.

Hybrid short-term freeway speed prediction methods based on periodic analysis

2015 ◽  
Vol 42 (8) ◽  
pp. 570-582 ◽  
Author(s):  
Yajie Zou ◽  
Xuedong Hua ◽  
Yanru Zhang ◽  
Yinhai Wang
Keyword(s):  
Traffic Data ◽  
Arima Models ◽  
Periodic Patterns ◽  
Short Term ◽  
Time Step ◽  
Hybrid Prediction ◽  
Residual Part ◽  
Periodic Analysis ◽  
Speed Prediction ◽  
Prediction Approach

Short-term traffic speed forecasting is an important issue for developing Intelligent Transportation Systems applications. So far, a number of short-term speed prediction approaches have been developed. Recently, some multivariate approaches have been proposed to consider the spatial and temporal correlation of traffic data. However, as traffic data often demonstrates periodic patterns, the existing methodologies often fail to take into account spatial and temporal information as well as the periodic features of traffic data simultaneously in the multi-step prediction. This paper comprehensively evaluated the multi-step prediction performance of space time (ST) model, vector autoregression (VAR), and autoregressive integrated moving average (ARIMA) models using the 5 minute freeway speed data collected from five loop detectors located on an eastbound segment of Interstate 394 freeway, in Minnesota. To further consider the cyclical characteristics of freeway speed data, hybrid prediction approaches were proposed to decompose speed into two different components: a periodic trend and a residual part. A trigonometric regression function is introduced to capture the periodic component and the residual part is modeled by the ST, VAR, and ARIMA models. The prediction results suggest that for multi-step freeway speed prediction, as the time step increases, the ST model demonstrates advantages over the VAR and ARIMA models. Comparisons among the ST, VAR, ARIMA, and hybrid models demonstrated that modeling the periodicity and the residual part separately can better interpret the underlining structure of the speed data. The proposed hybrid prediction approach can accommodate the periodic trends and provide more accurate prediction results when the forecasting horizon is greater than 30 min.

scholarly journals Traffic Information Enrichment: Creating Long-Term Traffic Speed Prediction Ensemble Model for Better Navigation through Waypoints

2020 ◽  
Vol 11 (1) ◽  
pp. 315
Author(s):  
Milan Simunek ◽  
Zdenek Smutny
Keyword(s):  
Intelligent Transportation Systems ◽  
Large Scale ◽  
Absolute Error ◽  
Transportation Systems ◽  
Traffic Information ◽  
Ensemble Model ◽  
Temporal Data ◽  
Speed Prediction ◽  
Traffic Speed

Traffic speed prediction for a selected road segment from a short-term and long-term perspective is among the fundamental issues of intelligent transportation systems (ITS). During the course of the past two decades, many artefacts (e.g., models) have been designed dealing with traffic speed prediction. However, no satisfactory solution has been found for the issue of a long-term prediction for days and weeks using the vast spatial and temporal data. This article aims to introduce a long-term traffic speed prediction ensemble model using country-scale historic traffic data from 37,002 km of roads, which constitutes 66% of all roads in the Czech Republic. The designed model comprises three submodels and combines parametric and nonparametric approaches in order to acquire a good-quality prediction that can enrich available real-time traffic information. Furthermore, the model is set into a conceptual design which expects its usage for the improvement of navigation through waypoints (e.g., delivery service, goods distribution, police patrol) and the estimated arrival time. The model validation is carried out using the same network of roads, and the model predicts traffic speed in the period of 1 week. According to the performed validation of average speed prediction at a given hour, it can be stated that the designed model achieves good results, with mean absolute error of 4.67 km/h. The achieved results indicate that the designed solution can effectively predict the long-term speed information using large-scale spatial and temporal data, and that this solution is suitable for use in ITS.

scholarly journals Graph Convolutional Network: Traffic Speed Prediction Fused with Traffic Flow Data

Sensors ◽  
2021 ◽  
Vol 21 (19) ◽  
pp. 6402
Author(s):  
Duanyang Liu ◽  
Xinbo Xu ◽  
Wei Xu ◽  
Bingqian Zhu
Keyword(s):  
Traffic Flow ◽  
Intelligent Transportation Systems ◽  
Transportation Systems ◽  
Flow Data ◽  
Convolutional Network ◽  
Convolutional Networks ◽  
Speed Prediction ◽  
Traffic Speed ◽  
Real World Datasets ◽  
Relationship Of

Traffic speed prediction plays an important role in intelligent transportation systems, and many approaches have been proposed over recent decades. In recent years, methods using graph convolutional networks (GCNs) have been more promising, which can extract the spatiality of traffic networks and achieve a better prediction performance than others. However, these methods only use inaccurate historical data of traffic speed to forecast, which decreases the prediction accuracy to a certain degree. Moreover, they ignore the influence of dynamic traffic on spatial relationships and merely consider the static spatial dependency. In this paper, we present a novel graph convolutional network model called FSTGCN to solve these problems, where the model adopts the full convolutional structure and avoids repeated iterations. Specifically, because traffic flow has a mapping relationship with traffic speed and its values are more exact, we fused historical traffic flow data into the forecasting model in order to reduce the prediction error. Meanwhile, we analyzed the covariance relationship of the traffic flow between road segments and designed the dynamic adjacency matrix, which can capture the dynamic spatial correlation of the traffic network. Lastly, we conducted experiments on two real-world datasets and prove that our model can outperform state-of-the-art traffic speed prediction.

Sign in / Sign up

Export Citation Format

Share Document