Efficient deep learning based method for multi‐lane speed forecasting: a case study in Beijing. Issue 14 (22nd February 2021)

• Record Type: Journal Article
• Authors: Lu, Wenqi; Yi, Ziwei; Liu, Wan; Gu, Yuanli; Rui, Yikang; Ran, Bin
• Is Part Of: IET intelligent transport systems. Volume 14:Issue 14(2020)
• Page Start: 2073
• Page End: 2082
• Publication Date: 2021-02-22
• Subjects: recurrent neural nets -- traffic engineering computing -- road traffic -- deep learning (artificial intelligence) -- random forests -- real‐time systems; multilane speed forecasting -- Beijing -- multilane traffic flow prediction -- combination method -- lane‐level traffic flow -- traffic speed prediction -- optimal input time window -- RF‐CWRNN method -- training time -- deep learning based traffic prediction methods -- real‐time multilane traffic condition forecasting -- automated vehicle highway system -- connected vehicle highway system -- clockwork recurrent neural network -- random forest -- temporal feature; Intelligent transportation systems -- Periodicals; Electronics in transportation -- Periodicals; 388.31205
• DOI: 10.1049/iet-its.2020.0410
• Languages: English
• ISSNs: 1751-956X

Abstract

Abstract : Real‐time and accurate multi‐lane traffic condition forecasting is of great importance to the connected and automated vehicle highway system. However, the majority of existing deep learning based traffic prediction methods focus on pursuing the precision of the methods while neglect to improve the efficiency of the methods. To achieve the high accuracy and high efficiency of multi‐lane traffic flow prediction simultaneously, this study proposes a novel combination method via the integration of the clockwork recurrent neural network (CWRNN) and random forest (RF) method, which is RF‐CWRNN. To the best of the authors' knowledge, this is the first time that the CWRNN is introduced to capture the temporal feature of lane‐level traffic flow and make traffic speed prediction. Meanwhile, the RF method is employed to measure the temporal relevance of the traffic flow and determine the optimal input time window. To verify the performance of the RF‐CWRNN method, the ground‐truth data of the expressways in Beijing were utilised to carry out experiments. The results indicate that the RF‐CWRNN method is superior to the baseline models in terms of accuracy and robustness. Besides, the proposed method can save plenty of training time compared with the classical long short‐term memory neural network.