A Deep Unsupervised Learning Approach for Airspace Complexity Evaluation

Li, Biyue; Du, Wenbo; Zhang, Yunlong; Tang, Ke; Cao, Xianbin

doi:10.1109/tits.2021.3106779

Cited by 18 publications

(8 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some complexity models which use expert ATCO knowledge are described in [8][9][10][11][12][13][14][15][16][17]. Other models which use complexity indicators are based on mathematical or statistical methods (e.g., [18][19][20][21][22][23][24][25]). On one hand, the latter make the development of the model easier since it is difficult to acquire ATCO input.…”

Section: Air Traffic Complexity Models and Indicatorsmentioning

confidence: 99%

Defining Terminal Airspace Air Traffic Complexity Indicators Based on Air Traffic Controller Tasks

Jurinić,

Juričić,

Antulov-Fantulin

et al. 2024

Aerospace

View full text Add to dashboard Cite

This paper focuses on terminal air traffic complexity indicators. By thorough analysis of previous research, the benefits and limitations of the existing terminal complexity models are identified. According to these findings, a new approach for determining terminal air traffic complexity indicators is proposed which assumes that terminal complexity could be determined based on approach air traffic controller (ATCO) tasks. The comprehensive list of general approach ATCO tasks was defined using a literature review and observation of training exercises, forming the basis for subsequent expert group workshops which enabled the acquisition of ATCOs’ knowledge data. Through these workshops, new approach ATCO tasks were additionally identified, and terminal complexity indicators were defined with airspace and traffic parameters. These new tasks and indicators present a novelty in this field of research since they incorporate ATCOs’ knowledge as the data input and consider various traffic scenarios, all types of traffic, weather conditions, and off-nominal situations.

show abstract

Section: Air Traffic Complexity Models and Indicatorsmentioning

confidence: 99%

Defining Terminal Airspace Air Traffic Complexity Indicators Based on Air Traffic Controller Tasks

Jurinić,

Juričić,

Antulov-Fantulin

et al. 2024

Aerospace

View full text Add to dashboard Cite

show abstract

“…The representative list of 28 complexity factors, such as aircraft density, variability in aircraft speed, and geometric volume of sectors can be referred to [6]. The detailed descriptions of these complexity factors can also be found in our previous work [7]. Furthermore, the airspace complexity is discretized into three levels: high, normal, and low.…”

Section: A Problem Formulationmentioning

confidence: 99%

“…1. More detailed descriptions of the airspace sectors can also be found in our previous work [7]. The study period is from 8:00 to 24:00 on July 28, 2010.…”

Section: A Datasetmentioning

confidence: 99%

“…A pioneering work by Chatterji used artificial neural networks to establish a nonlinear mapping between complexity factors and airspace complexity [5]. Subsequently, a series of works have been applied to calculate the real-world airspace complexity through supervised learning [6], unsupervised learning [7], transfer learning [8], etc. However, the above methods are insufficient to model the temporal dependencies implicit in the variation of airspace complexity.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Spatiotemporal Hybrid Model for Airspace Complexity Prediction

et al. 2023

IEEE Intell. Transport. Syst. Mag.

View full text Add to dashboard Cite

Airspace complexity is a key indicator that reflects the safety of airspace operations in air traffic management systems. Furthermore, to achieve efficient air traffic control, it is necessary to accurately predict the airspace complexity. In this article, we propose a novel spatial-temporal hybrid deep learning model for airspace complexity prediction to efficiently capture spatial correlations as well as temporal dependencies pertaining to the airspace complexity data. Specifically, we apply convolutional networks to discover the short-term temporal patterns and skip long short-term memory networks to model the long-term temporal patterns of airspace complexity data. Furthermore, it is observed that the graph attention network in our proposed model, which emphasizes capturing the spatial correlations of the airspace sectors, can significantly improve the prediction accuracy. Extensive experiments are conducted on the real data of six airspace sectors in Southwest China. Experimental results show that our spatial-temporal deep learning approach is superior to state-of-the-art methods.

show abstract

“…It has the advantages of accurately simulating the real‐world ATC operational scenes and flight status, allowing for a detailed and in‐depth evaluation of the airspace planning scheme from the perspectives of ATC and airspace users (flights). Furthermore, it provides unrivaled advantages in terms of human factor analysis with respect to both ATC controllers and pilots, as well as their interactions [4]. However, due to factors such as the difficulty of fully systematic analysis of airspace operation, the personal preferences and fuzzy subjective judgments of human experts, and conflicting conclusions made by different criteria, the evaluation of airspace planning scheme based on ATC‐Flight real‐time simulation continues to face challenges.…”

Section: Introductionmentioning

confidence: 99%

Plithogenic multi‐criteria decision making approach on airspace planning scheme evaluation based on ATC‐flight real‐time simulation

Liang

Han

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

Airspace planning scheme evaluation based on ATC-Flight real-time simulation which incorporates air traffic control simulator and flight simulator has become increasingly important. It has irreplaceable advantages of truly simulating and in-depth evaluating airspace operation, especially for analyzing human factors. As a multi-criteria decision making (MCDM) problem, it poses challenges which this study aims to overcome from three aspects. Firstly, the complexity of airspace and the characteristics of ATC-Flight realtime simulation are analyzed to build a multi-criteria evaluation system. The criteria include air traffic flow, airspace operational performance, flight procedure quality, cost, controller workload and pilot workload. Secondly, a novel weighting method based on plithogenic set is presented to derive robust subjective criteria weights from a group of human decision makers. The subjective weights are then combined with objective weights using game theory to determine cumulative criteria weights. Thirdly, technique for order preference by similarity to ideal solution (TOPSIS) and grey correlation analysis are integrated to perform comprehensive evaluation and ranking. The first series of ATC-Flight real-time simulations in China are conducted to illustrate a case study. The results demonstrated that the proposed MCDM approach could improve the accuracy and discriminability of evaluation results, and outperformed other existing well-known MCDM algorithms.

show abstract

A Deep Unsupervised Learning Approach for Airspace Complexity Evaluation

Cited by 18 publications

References 52 publications

Defining Terminal Airspace Air Traffic Complexity Indicators Based on Air Traffic Controller Tasks

Defining Terminal Airspace Air Traffic Complexity Indicators Based on Air Traffic Controller Tasks

A Spatiotemporal Hybrid Model for Airspace Complexity Prediction

Plithogenic multi‐criteria decision making approach on airspace planning scheme evaluation based on ATC‐flight real‐time simulation

Contact Info

Product

Resources

About