2019 IEEE Intelligent Transportation Systems Conference (ITSC) 2019
DOI: 10.1109/itsc.2019.8917180
|View full text |Cite
|
Sign up to set email alerts
|

A Deep Reinforcement Learning Approach to High-speed Train Timetable Rescheduling under Disturbances

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
19
0
1

Year Published

2020
2020
2024
2024

Publication Types

Select...
4
3
2

Relationship

0
9

Authors

Journals

citations
Cited by 53 publications
(20 citation statements)
references
References 15 publications
0
19
0
1
Order By: Relevance
“…Currently, the development and implementation of systems for the automation of these processes using artificial intelligence is just getting underway. The work [3] is one of the most recent publications on the implementation of a deep learning approach to reschedule disruptions in high-speed traffic. The approach presented in this article allows us to work only with high-speed traffic.…”
Section: Analysis Of Current Developmentsmentioning
confidence: 99%
See 1 more Smart Citation
“…Currently, the development and implementation of systems for the automation of these processes using artificial intelligence is just getting underway. The work [3] is one of the most recent publications on the implementation of a deep learning approach to reschedule disruptions in high-speed traffic. The approach presented in this article allows us to work only with high-speed traffic.…”
Section: Analysis Of Current Developmentsmentioning
confidence: 99%
“…In [3], an approach is presented that is most closely approximated to the task of dispatch control. However, it, like [5], is solved only for one type of traffic.…”
Section: Artificial Intelligencementioning
confidence: 99%
“…The main outcome reveals that the method generates positive changes in the timetable in 57% of the instances tested. Ning et al [2019] presents an application of DQN to minimise the average total delay for a high-speed timetable rescheduling problem. The model has a centralised dispatcher agent that receives an encoded timetable as the state and acts by defining the departure sequence of the trains in each station.…”
Section: Related Workmentioning
confidence: 99%
“…Although traditional optimization methods are constantly improved in terms of computation efficiency, they are still restricted by the size of problem instances. To overcome this limitation, a few studies [8]- [12] applied machine learning techniques to realize TTR, in particular the technique of Reinforcement Learning.…”
Section: Introductionmentioning
confidence: 99%