.png){kind=icon}
.png){kind=icon}
Return
A scalable reinforcement learning-based approach to dynamic airspace sectorization
.png){kind=icon}
OA
AI
王
A
W
J
DOI：10.1016/j.trc.2026.105824.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Novel reinforcement learning framework enables real-time dynamic airspace sectorization. • Comprehensive multi-objective optimization balances workload minimization with sectorization continuity constraints. • Framework trained on three weeks of historical flight trajectories from UK airspace and validated on independent scenarios. • The approach reduces air traffic controller workload by 50.59% and computation time by 99.98% compared to genetic algorithm benchmarks. • Statistical validation confirms significant, repeatable performance improvements.
Keywords:
Dynamic airspace sectorization
Reinforcement learning
Air traffic controllers’ workload
Tiled Voronoi partitioning
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.9
Papers:
4.7K
Citations:
3.2W
Organization
No organization information available
Cited Papers