A long short-term memory-based framework for crash detection on freeways with traffic data of different temporal resolutions

Feifeng Jiang; Kwok Kit Richard Yuen; Eric Wai Ming Lee

doi:10.1016/j.aap.2020.105520

A long short-term memory-based framework for crash detection on freeways with traffic data of different temporal resolutions

Accid Anal Prev. 2020 Jun:141:105520. doi: 10.1016/j.aap.2020.105520. Epub 2020 Apr 8.

Authors

Feifeng Jiang¹, Kwok Kit Richard Yuen², Eric Wai Ming Lee²

Affiliations

¹ Department of Architecture and Civil Engineering, City University of Hong Kong, Hong Kong, China. Electronic address: feifjiang2-c@my.cityu.edu.hk.
² Department of Architecture and Civil Engineering, City University of Hong Kong, Hong Kong, China.

PMID: 32278148
DOI: 10.1016/j.aap.2020.105520

Abstract

Traffic crash detection is a major component of intelligent transportation systems. It can explore inner relationships between traffic conditions and crash risk, prevent potential crashes, and improve road safety. However, there exist some limitations in current studies on crash detection: (1) The commonly used machine learning methods cannot simulate the evolving transitions of traffic conditions before crash occurrences; (2) Current models collected traffic data of only one temporal resolution, which cannot fully represent traffic trends in different time intervals. Therefore, this study proposes a Long short-term memory (LSTM) based framework considering traffic data of different temporal resolutions (LSTMDTR) for crash detection. LSTM is an effective deep learning method to capture the long-term dependency and dynamic transitions of pre-crash conditions. Three LSTM networks considering traffic data of different temporal resolutions are constructed, which can comprehensively indicate traffic variations in different time intervals. A fully-connected layer is used to combine the outputs of three LSTM networks, and a dropout layer is used to reduce overfitting and improve prediction performance. The LSTMDTR model is implemented on datasets of I880-N and I805-N in California, America. The results indicate that the LSTMDTR model can obtain satisfactory performance on crash detection, with the highest crash accuracy of 70.43 %. LSTMDTR models constructed on one freeway can be transferred to other similar freeways, with 65.12 % of crash accuracy on transferability. Compared with machine learning methods and LSTM models with one or two temporal resolutions, the LSTMDTR model has been validated to perform better on crash detection and transferability. A proper number of neurons in the LSTMDTR model should be determined in real applications considering acceptable detection performance and computation time. The dropout technique can reduce overfitting and improve the generalization ability of the LSTMDTR model, increasing crash accuracy from 64.49 % to 70.43 %.

Keywords: Crash detection; Deep learning methods; Different temporal resolutions; Long short-term memory networks; Traffic condition.