.png){kind=icon}
返回
A data-driven trajectory optimization framework for terminal maneuvering area operations
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ast.2022.108010.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The contradiction between air traffic capacity and demand constrains the development of the civil aviation industry. This paper proposes a Data-Driven Trajectory Optimization Framework for Terminal Maneuvering Area operations (DDTO-TMA) to enhance the capacity of Terminal Maneuvering Area (TMA). First, a Wasserstein-distance-based spectral clustering algorithm is applied to divide historical trajectories into several groups based on flight performance. Then, based on clustering results, a set of candidate Four-Dimensional Trajectories (4DTs) is created via flyable trajectories expansion. Finally, a combined algorithm with Mahalanobis-distance-based conflict detection and Pareto frontier determination is designed to identify optimal trajectories from candidate 4DTs in an efficient way. Two validations have been performed for Guangzhou Baiyun International Airport (ZGGG) TMA. The first validation evaluates the performance of the proposed DDTO-TMA. The results show that the DDTO-TMA can generate optimal trajectories that are not only of the interest of most stakeholders but also conflict-free. Besides, with the increase in the size of the historical trajectory set, the performance of DDTO-TMA gradually increases. The second validation estimates the improvement of a TMA operation after adopting DDTO-TMA. The results illustrate that, compared with the baseline, the proposed DDTO-TMA could enhance the arrival capacity by 30% without sacrificing stakeholders' interests. Furthermore, the proposed optimal trajectory generator can be easily transferred to different airports due to its data -driven characteristics.(c) 2022 Elsevier Masson SAS. All rights reserved.
Keyword:
Air traffic management
Trajectory optimization
Data-driven method
Conflict detection
Pareto optimal
期刊
IF:
5.8
论文数:
1.0W
被引数:
3.0W
机构
暂无机构信息
引用论文
.png){kind=icon}
Tinnitus Retraining Therapy (TRT) as a Method for Treatment of Tinnitus and Hyperacusis Patients耳鸣再训练疗法 (TRT) 作为治疗耳鸣和高亢患者的方法
PREAI
分享收藏
分享收藏Deep reinforcement learning based path stretch vector resolution in dense traffic with uncertainties
OAAI
OAAI分享收藏
查看更多内容