A CNN-LSTM-attention model for near-crash event identification on mountainous roads
To enhance traffic safety on mountainous roads, this study proposes an innovative CNN-LSTM-Attention model designed for the identification of near-crash events, utilizing naturalistic driving data from the challenging terrains in Yunnan, China. A combination of a threshold method complemented by man...
Saved in:
Main Authors: | , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2024
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/178868 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-178868 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1788682024-07-12T15:33:33Z A CNN-LSTM-attention model for near-crash event identification on mountainous roads Zhao, Jing Yang, Wenchen Zhu, Feng School of Civil and Environmental Engineering Engineering Traffic safety Near-crash events To enhance traffic safety on mountainous roads, this study proposes an innovative CNN-LSTM-Attention model designed for the identification of near-crash events, utilizing naturalistic driving data from the challenging terrains in Yunnan, China. A combination of a threshold method complemented by manual verification is used to label and annotate near-crash events within the dataset. The importance of vehicle motion features is evaluated using the random forest algorithm, revealing that specific variables, including x-axis acceleration, y-axis acceleration, y-axis angular velocity, heading angle, and vehicle speed, are particularly crucial for identifying near-crash events. Addressing the limitations of existing models in accurately detecting near-crash scenarios, this study combines the strengths of convolutional neural networks (CNN), long short-term memory (LSTM) networks, and an attention mechanism to enhance model sensitivity to crucial temporal and spatial features in naturalistic driving data. Specifically, the CNN-LSTM-Attention model leverages CNN to extract local features from the driving data, employs LSTM to track temporal dependencies among feature variables, and uses the attention mechanism to dynamically fine-tune the network weights of feature parameters. The efficacy of the proposed model is extensively evaluated against six comparative models: CNN, LSTM, Attention, CNN-LSTM, CNN-Attention, and LSTM-Attention. In comparison to the benchmark models, the CNN-LSTM-Attention model achieves superior overall accuracy at 98.8%. Moreover, it reaches a precision rate of 90.1% in detecting near-crash events, marking an improvement of 31.6%, 14.8%, 63.5%, 8%, 23.5%, and 22.6% compared to the other six comparative models, respectively. Published version This research was jointly funded by the National Key Research and Development Program of China (2022YFC3002601), and the Science and Technology Innovation Program of the Department of Transportation, Yunnan Province, China (2022-107). 2024-07-09T08:14:56Z 2024-07-09T08:14:56Z 2024 Journal Article Zhao, J., Yang, W. & Zhu, F. (2024). A CNN-LSTM-attention model for near-crash event identification on mountainous roads. Applied Sciences, 14(11), 4934-. https://dx.doi.org/10.3390/app14114934 2076-3417 https://hdl.handle.net/10356/178868 10.3390/app14114934 2-s2.0-85195948865 11 14 4934 en Applied Sciences © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering Traffic safety Near-crash events |
spellingShingle |
Engineering Traffic safety Near-crash events Zhao, Jing Yang, Wenchen Zhu, Feng A CNN-LSTM-attention model for near-crash event identification on mountainous roads |
description |
To enhance traffic safety on mountainous roads, this study proposes an innovative CNN-LSTM-Attention model designed for the identification of near-crash events, utilizing naturalistic driving data from the challenging terrains in Yunnan, China. A combination of a threshold method complemented by manual verification is used to label and annotate near-crash events within the dataset. The importance of vehicle motion features is evaluated using the random forest algorithm, revealing that specific variables, including x-axis acceleration, y-axis acceleration, y-axis angular velocity, heading angle, and vehicle speed, are particularly crucial for identifying near-crash events. Addressing the limitations of existing models in accurately detecting near-crash scenarios, this study combines the strengths of convolutional neural networks (CNN), long short-term memory (LSTM) networks, and an attention mechanism to enhance model sensitivity to crucial temporal and spatial features in naturalistic driving data. Specifically, the CNN-LSTM-Attention model leverages CNN to extract local features from the driving data, employs LSTM to track temporal dependencies among feature variables, and uses the attention mechanism to dynamically fine-tune the network weights of feature parameters. The efficacy of the proposed model is extensively evaluated against six comparative models: CNN, LSTM, Attention, CNN-LSTM, CNN-Attention, and LSTM-Attention. In comparison to the benchmark models, the CNN-LSTM-Attention model achieves superior overall accuracy at 98.8%. Moreover, it reaches a precision rate of 90.1% in detecting near-crash events, marking an improvement of 31.6%, 14.8%, 63.5%, 8%, 23.5%, and 22.6% compared to the other six comparative models, respectively. |
author2 |
School of Civil and Environmental Engineering |
author_facet |
School of Civil and Environmental Engineering Zhao, Jing Yang, Wenchen Zhu, Feng |
format |
Article |
author |
Zhao, Jing Yang, Wenchen Zhu, Feng |
author_sort |
Zhao, Jing |
title |
A CNN-LSTM-attention model for near-crash event identification on mountainous roads |
title_short |
A CNN-LSTM-attention model for near-crash event identification on mountainous roads |
title_full |
A CNN-LSTM-attention model for near-crash event identification on mountainous roads |
title_fullStr |
A CNN-LSTM-attention model for near-crash event identification on mountainous roads |
title_full_unstemmed |
A CNN-LSTM-attention model for near-crash event identification on mountainous roads |
title_sort |
cnn-lstm-attention model for near-crash event identification on mountainous roads |
publishDate |
2024 |
url |
https://hdl.handle.net/10356/178868 |
_version_ |
1806059814141296640 |