Cloud-Edge-End Collaboration for Intelligent Train Regulation Optimization in TACS
Bibliographic record
Abstract
Advancing from large-scale complex railway network construction to refined network operation management is a significant trend in promoting the high-quality development of modern rail transportation services. With the emergence of the next-generation train control system–Train Autonomous Circumambulation System (TACS), the transportation environment manifests obvious intricate correlations with strong couplings, multiple constraints, and rapid evolution. Most of the existing works focus on low-dimensional passenger flow prediction and independent train adjustment optimization, while the potential of network-level situation assessment and generalized experience across multi-tasks are neglected. In this paper, we propose a novel cloud-edge-end collaboration empowered TACS intelligent train regulation optimization scheme with situation awareness at end layer, arithmetic provision at edge layer, and intelligent fusion at cloud layer. Specifically, a Graph Convolutional Network (GCN)-based passenger flow prediction model is introduced to enable accurate assessment of the urban rail transit operation situation at the network level. Moreover, the Deep Reinforcement Learning (DRL)-based train dynamic adjustment algorithm is proposed to ensure efficient matching of passenger and traffic flows. In addition, a Actor-Mimic based multi-task and transfer reinforcement learning method is implemented in TACS to facilitate generalizing the trained experience across multiple tasks and accelerate the ability to adapt to new environments. Extensive simulation results illustrate that the proposed scheme can effectively improve the transportation capacity matching of TACS and enhance the generalization of train dynamic adjustment strategies.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".