Dual-attention network with multitask learning for multistep short-term speed prediction on expressways
- PDF / 3,870,094 Bytes
- 22 Pages / 595.276 x 790.866 pts Page_size
- 79 Downloads / 224 Views
(0123456789().,-volV)(0123456789(). ,- volV)
ORIGINAL ARTICLE
Dual-attention network with multitask learning for multistep shortterm speed prediction on expressways Yanyun Tao1 • Guoqi Yue1 • Xiang Wang1 Received: 29 January 2020 / Accepted: 26 October 2020 Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract In this study, a dual-attention network (DAN) with multitask learning is proposed to solve the short-term prediction problems of traffic speed. The proposed DAN includes a road-type attention module (RAM), which performs accurate short-term speed prediction using road-type attention scores, a low-speed attention module (LAM), which is trained on weighted samples and fits low speed, and a decision support module, which outputs either RAM or LAM by estimating the level of the predicted speed. DAN can improve the transfer in the feature and speed prediction task layers by learningassociated and time-dependent tasks. The Shanghai expressway dataset is used to test and compare the proposed method and 15 other techniques. The results show that DAN with a multitask loss function obtains the smallest mean squared error (MSE) and mean absolute percentage error (MAPE) in most cases. LAM efficiently improves the predictive accuracy of low-speed samples, whereas RAM performs better in terms of the overall error reduction. DAN achieves the largest Rsquared of 0.93 with a small reduction in R-squared by 0.12% from the training data to the test data, thereby illustrating its excellent generalization. DAN outperforms the other models by at least 13.5% in terms of the MSE and by 5.07% in terms of the MAPE on different road types. Adding LAM effectively improves the MAPE by at least 21.4% over RAM without increasing the error of the other speed levels. In terms of the MSE, RAM outperforms DAN by 12.6% in the best case. This study proved that the short-term speed prediction based on DAN has the ability to improve the accuracy on low-speed level and the generalization on different road types. Keywords Short-term speed prediction Multitask learning Neural network Dual-attention
1 Introduction Expressways are the main part of an urban road network [1]. Expressway congestion results in financial and time losses and causes environmental pollution and energy waste [2, 3]. Short-term traffic prediction aims to forecast and analyze the trend of a traffic state in a short period based on the existing road traffic state and provides an early warning of possible congestion. This process is a necessary component of intelligent and advanced traffic management systems [4, 5]. A congested segment affects the traffic state of the adjacent road network when congestion spreads. If congestion can be predicted effectively & Xiang Wang [email protected] 1
School of Rail Transportation, Soochow University, Suzhou, China
and accurately, an appropriate traffic management can be implemented to prevent congestion in advance, and the traffic condition of the entire expressway network can be improved. Thus, how to capture th
Data Loading...
