Intelligent train operation based on deep learning from excellent driver manipulation patterns
Abstract In the application of deep learning to realize intelligent train operation, there are some problems, such as the single learning task. Especially when using the gradient descent approach to optimize the structure, weight and threshold of a deep network, it is easy in this task to fall into...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-09-01
|
| Series: | IET Intelligent Transport Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/itr2.12201 |
| _version_ | 1828147308888326144 |
|---|---|
| author | Kai Xu Yongchao Tu Wenxuan Xu Shixun Wu |
| author_facet | Kai Xu Yongchao Tu Wenxuan Xu Shixun Wu |
| author_sort | Kai Xu |
| collection | DOAJ |
| description | Abstract In the application of deep learning to realize intelligent train operation, there are some problems, such as the single learning task. Especially when using the gradient descent approach to optimize the structure, weight and threshold of a deep network, it is easy in this task to fall into a local optimum. This leads to excessive reliance on manual tuning experience. Aiming at the above issues, this paper proposes a new approach of train manipulation and prediction based on a long short‐term memory (LSTM) deep network. From the perspective of automatic hyper‐parameter optimization, the gradient‐free intelligent search method is principally chosen to optimize the architecture and parameters of a LSTM deep network, so as to improve the manipulation accuracy based on learning from excellent drivers. This method first selects excellent driver data through the Pareto dominance principle and crowding distance calculation; on this basis, a step‐by‐step method is used to optimize the structure, weight and threshold of the LSTM network. Particularly, in the first step, we adopt a genetic algorithm to search for the optimal deep network structure, which overcomes the problem that the structure is difficult to determine. In the second step, we optimize the parameters of the deep network, a process that is divided into two stages of ‘rough learning’ and ‘precise learning’. In the ‘rough learning’ stage, we use the multi‐population chained multi‐agent (MPCMA) algorithm to preliminarily optimize the LSTM network parameters. In the ‘precise learning’ stage, the Adam algorithm is applied to further finely optimize the network parameters. Finally, through simulation experiments, it is verified that the proposed method improves the accuracy of train manipulation and prediction, and shows strong robustness in situations of multiple manipulation sequences and different temporary speed limits. |
| first_indexed | 2024-04-11T20:59:02Z |
| format | Article |
| id | doaj.art-4f7ada817fb149cc9e3d8b739d91bfb7 |
| institution | Directory Open Access Journal |
| issn | 1751-956X 1751-9578 |
| language | English |
| last_indexed | 2024-04-11T20:59:02Z |
| publishDate | 2022-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Intelligent Transport Systems |
| spelling | doaj.art-4f7ada817fb149cc9e3d8b739d91bfb72022-12-22T04:03:34ZengWileyIET Intelligent Transport Systems1751-956X1751-95782022-09-011691177119210.1049/itr2.12201Intelligent train operation based on deep learning from excellent driver manipulation patternsKai Xu0Yongchao Tu1Wenxuan Xu2Shixun Wu3College of Information Science and Engineering Chongqing Jiaotong University Chongqing ChinaCollege of Information Science and Engineering Chongqing Jiaotong University Chongqing ChinaCollege of Electrical Engineering Chongqing University Chongqing ChinaCollege of Information Science and Engineering Chongqing Jiaotong University Chongqing ChinaAbstract In the application of deep learning to realize intelligent train operation, there are some problems, such as the single learning task. Especially when using the gradient descent approach to optimize the structure, weight and threshold of a deep network, it is easy in this task to fall into a local optimum. This leads to excessive reliance on manual tuning experience. Aiming at the above issues, this paper proposes a new approach of train manipulation and prediction based on a long short‐term memory (LSTM) deep network. From the perspective of automatic hyper‐parameter optimization, the gradient‐free intelligent search method is principally chosen to optimize the architecture and parameters of a LSTM deep network, so as to improve the manipulation accuracy based on learning from excellent drivers. This method first selects excellent driver data through the Pareto dominance principle and crowding distance calculation; on this basis, a step‐by‐step method is used to optimize the structure, weight and threshold of the LSTM network. Particularly, in the first step, we adopt a genetic algorithm to search for the optimal deep network structure, which overcomes the problem that the structure is difficult to determine. In the second step, we optimize the parameters of the deep network, a process that is divided into two stages of ‘rough learning’ and ‘precise learning’. In the ‘rough learning’ stage, we use the multi‐population chained multi‐agent (MPCMA) algorithm to preliminarily optimize the LSTM network parameters. In the ‘precise learning’ stage, the Adam algorithm is applied to further finely optimize the network parameters. Finally, through simulation experiments, it is verified that the proposed method improves the accuracy of train manipulation and prediction, and shows strong robustness in situations of multiple manipulation sequences and different temporary speed limits.https://doi.org/10.1049/itr2.12201Optimisation techniquesRail‐traffic system controlInterpolation and function approximation (numerical analysis)Linear algebra (numerical analysis)Combinatorial mathematicsNeural nets |
| spellingShingle | Kai Xu Yongchao Tu Wenxuan Xu Shixun Wu Intelligent train operation based on deep learning from excellent driver manipulation patterns IET Intelligent Transport Systems Optimisation techniques Rail‐traffic system control Interpolation and function approximation (numerical analysis) Linear algebra (numerical analysis) Combinatorial mathematics Neural nets |
| title | Intelligent train operation based on deep learning from excellent driver manipulation patterns |
| title_full | Intelligent train operation based on deep learning from excellent driver manipulation patterns |
| title_fullStr | Intelligent train operation based on deep learning from excellent driver manipulation patterns |
| title_full_unstemmed | Intelligent train operation based on deep learning from excellent driver manipulation patterns |
| title_short | Intelligent train operation based on deep learning from excellent driver manipulation patterns |
| title_sort | intelligent train operation based on deep learning from excellent driver manipulation patterns |
| topic | Optimisation techniques Rail‐traffic system control Interpolation and function approximation (numerical analysis) Linear algebra (numerical analysis) Combinatorial mathematics Neural nets |
| url | https://doi.org/10.1049/itr2.12201 |
| work_keys_str_mv | AT kaixu intelligenttrainoperationbasedondeeplearningfromexcellentdrivermanipulationpatterns AT yongchaotu intelligenttrainoperationbasedondeeplearningfromexcellentdrivermanipulationpatterns AT wenxuanxu intelligenttrainoperationbasedondeeplearningfromexcellentdrivermanipulationpatterns AT shixunwu intelligenttrainoperationbasedondeeplearningfromexcellentdrivermanipulationpatterns |