Learning Air Traffic as Images: A Deep Convolutional Neural Network for Airspace Operation Complexity Evaluation

Xie, Hua; Zhang, Minghua; Ge, Jiaming; Dong, Xianjun; Chen, Haiyan

doi:10.1155/2021/6457246

Cited by 23 publications

(11 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to the evaluation of the reliability of the models, it is also important to evaluate the performance of the algorithms in terms of resource utilization. ere is very little work evaluating the computational complexity of deep learning models although there is some proposal [84] that considers some factors, such as the number of layers, the size of the input matrix, and other factors depending on the specific algorithm. In CNN, the number and size of convolution kernels and the number of output channels of each layer are considered.…”

Section: Discussion As Seen In Figuresmentioning

confidence: 99%

Hybrid Deep Learning Models for Sentiment Analysis

2021

View full text Add to dashboard Cite

Sentiment analysis on public opinion expressed in social networks, such as Twitter or Facebook, has been developed into a wide range of applications, but there are still many challenges to be addressed. Hybrid techniques have shown to be potential models for reducing sentiment errors on increasingly complex training data. This paper aims to test the reliability of several hybrid techniques on various datasets of different domains. Our research questions are aimed at determining whether it is possible to produce hybrid models that outperform single models with different domains and types of datasets. Hybrid deep sentiment analysis learning models that combine long short-term memory (LSTM) networks, convolutional neural networks (CNN), and support vector machines (SVM) are built and tested on eight textual tweets and review datasets of different domains. The hybrid models are compared against three single models, SVM, LSTM, and CNN. Both reliability and computation time were considered in the evaluation of each technique. The hybrid models increased the accuracy for sentiment analysis compared with single models on all types of datasets, especially the combination of deep learning models with SVM. The reliability of the latter was significantly higher.

show abstract

Section: Discussion As Seen In Figuresmentioning

confidence: 99%

Hybrid Deep Learning Models for Sentiment Analysis

2021

View full text Add to dashboard Cite

show abstract

“…Radišić et al [21] put forward novel complexity indicators under trajectory-based operation, and Andraši [22] estimated the air traffic complexity using neural networks. Xie et al [23] used images instead of complexity factors to evaluate the operational complexity by deep convolutional neural networks. ty between the two clustering results is analysed to investigate whether there is a correlation between complexity and performance or not.…”

Section: Overall Frameworkmentioning

confidence: 99%

Establishing the Correlation Between Complexity and Performance for Arrival Operations

Zhang

Xiang

Zhou

et al. 2022

PROMTT

View full text Add to dashboard Cite

Air traffic complexity indicators play an essential role in measuring operational performance and controller workload. However, current studies mainly depend on the manual scoring method to scale performance or workload. This paper focuses on arrival operations and presents a data-driven strategy to establish the correlation between complexity and performance to avoid the subjectivity of the currently used manual scoring method. Firstly, we present twenty-six indicators for describing air traffic complexity and two indicators for arrival operational performance. Secondly, the clustering method distinguishes peak and off-peak situations for arrival operation. Moreover, clustering results are compared to investigate the correlation between complexity and performance initially. Thirdly, the classification method is adopted to determine such correlation further. In addition, we also identify the affecting factors which could influence operational performance. Finally, trajectories of arrival aircraft landing at Guangzhou Baiyun International Airport (ZGGG) are used for case validation. The results indicate that there is a strong correlation between complexity and performance. The accuracy and precision of classification are approximately 90%. Furthermore, the number of aircraft significantly impacts the arrival operational performance within TMA.

show abstract

“…The conclusions drawn from these projects are important for the ATC service, as an increasing airspace complexity directly leads to an increase in controllers' workload. [9]. Thus, many benefits can be subtracted from the establishment of a specific complexity measure through allowing a more direct 1226 (2022) 012018 IOP Publishing doi:10.1088/1757-899X/1226/1/012018 2 assessment of the controllers' workload.…”

Section: Introduction and Objectivesmentioning

confidence: 99%

“…Thus, many benefits can be subtracted from the establishment of a specific complexity measure through allowing a more direct 1226 (2022) 012018 IOP Publishing doi:10.1088/1757-899X/1226/1/012018 2 assessment of the controllers' workload. Due to this relationship between the sector operation complexity (SOC) and controllers' workload, SOC will have an extraordinary role in air traffic management, e.g., airspace reconfiguration, air traffic flow management, and allocation of the ATC service resources [9]. These reasons place the study of complexity as a topic of great interest [10].…”

Section: Introduction and Objectivesmentioning

confidence: 99%

Dynamic sector characterisation model with the application of machine learning techniques

Moreno

Comendador

Jurado

et al. 2022

IOP Conf. Ser.: Mater. Sci. Eng.

View full text Add to dashboard Cite

The ATC service has the objective of controlling airspace operations safely and efficiently. This control is becoming more and more difficult due to the increasing complexity of airspace. With the objective of collaborating and facilitating the provision of the control service, FLUJOS project aims to develop a methodology to characterise ATC sectors according to their complexity. This paper shows the first results obtained in this project. A methodology is proposed that first performs a statistical analysis of the data present in the flight plans of individual aircraft. The statistical analysis will be used to estimate the impact of air traffic flows. With this, the complexity of ATC sectors will finally be determined. In addition, a machine learning tool will be added to develop a dynamic methodology. After evaluating the methodology with data from Spanish airspace in 2019, it can be said that the results obtained are logical from an operational point of view, and that they allow a fairly accurate classification of the sectors based on their complexity. However, the proposed methodology is still a preliminary version, so more work will have to be done to add variables to achieve the objective of obtaining an even more accurate and realistic classification.

show abstract

Learning Air Traffic as Images: A Deep Convolutional Neural Network for Airspace Operation Complexity Evaluation

Cited by 23 publications

References 31 publications

Hybrid Deep Learning Models for Sentiment Analysis

Hybrid Deep Learning Models for Sentiment Analysis

Establishing the Correlation Between Complexity and Performance for Arrival Operations

Dynamic sector characterisation model with the application of machine learning techniques

Contact Info

Product

Resources

About