Optimizing Airport Ground Movements Using Multi-Agents Reinforcement Learning
Bibliographic record
Abstract
This paper presents an efficient methodology for optimizing aircraft ground trajectories at airports using multi-agent reinforcement learning. Within this context, each aircraft is modeled as an agent navigating in a undirected graph representing the airport environment. The graph is defined as a set of edges and nodes, where edges represent taxiways, while nodes are junctions between these taxiways (or runways). In addition, the paper proposes a new approach to construct a secondary directed graph. This secondary graph simplifies the calculation of an agent (i.e., aircraft) trajectory by integrating geometric constraints, including the avoidance of sharp turns exceeding 45 degrees and the navigation around prohibited taxiways. Agents were trained using the Proximal Policy Optimization (PPO) algorithm to select routes that minimize travel distances while optimizing speed to meet specific arrival time constraints. The proposed methodology was tested and validated at two airports: Montreal Trudeau International Airport (CYUL) and Toronto Pearson Airport (CYYZ). Simulation results showed that, for both airports, agents successfully optimized their trajectories, systematically finding the shortest routes within the graph that met all constraints, and adjusting their speed to ensure on-time arrivals.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 source (direct Gemma or distilled Codex), 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".