Anomaly Detection on ADS-B Flight Data Using Machine Learning Techniques

Osman Taşdelen*, Levent Çarkacioglu, Behçet Uğur Töreyin

*Bu çalışma için yazışmadan sorumlu yazar

Araştırma sonucu: Kitap/Rapor/Konferans Bildirisinde BölümKonferans katkısıbilirkişi

3 Atıf (Scopus)

Özet

With the rapid increase in the number of flights all over the world, the management and control of flight operations has become difficult in recent years. Moreover, the expectations for the aviation sector indicate that this increase will continue in the upcoming years. Therefore, safer and systematic monitoring systems by eliminating the requirement of human-dependent tracking during the air travel of an aircraft and automating the detection of abnormal situations has become a major problem in aviation sector. With the recent advances in artificial intelligence, a safer and systematic tracking system for controlling the airspace by eliminating the need for human-dependent tracking during the flight of aircraft in the air has become possible. In this study, we aimed to create a system that detects and predicts movements to indicate abnormal, dangerous situations in the airspace by monitoring radar flight data using machine learning and deep learning techniques. We applied two different methods, i.e., Proximity Based kNN and Auto Encoder We used real-life historical radar flight data set which consists of Flight Radar 24 data were converted from ADS-B messages for learning. We created simulation data and used this data for testing and validation for our trained model. Within the scope of this project, we also developed a system to monitor air traffic through radar tracks with our model and present the abnormal situations to the user through a visual interface for decision support. In this visualization, we present the abnormal situations if one of the algorithms labeled as anomaly. Results for both methods have shown that our findings were similar to the real-life predictions.

Orijinal dilİngilizce
Ana bilgisayar yayını başlığıComputational Collective Intelligence - 13th International Conference, ICCCI 2021, Proceedings
EditörlerNgoc Thanh Nguyen, Ngoc Thanh Nguyen, Lazaros Iliadis, Ilias Maglogiannis, Bogdan Trawiński
YayınlayanSpringer Science and Business Media Deutschland GmbH
Sayfalar771-783
Sayfa sayısı13
ISBN (Basılı)9783030880804
DOI'lar
Yayın durumuYayınlandı - 2021
Etkinlik13th International Conference on Computational Collective Intelligence, ICCCI 2021 - Virtual, Online
Süre: 29 Eyl 20211 Eki 2021

Yayın serisi

AdıLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Hacim12876 LNAI
ISSN (Basılı)0302-9743
ISSN (Elektronik)1611-3349

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???13th International Conference on Computational Collective Intelligence, ICCCI 2021
ŞehirVirtual, Online
Periyot29/09/211/10/21

Bibliyografik not

Publisher Copyright:
© 2021, Springer Nature Switzerland AG.

Parmak izi

Anomaly Detection on ADS-B Flight Data Using Machine Learning Techniques' araştırma başlıklarına git. Birlikte benzersiz bir parmak izi oluştururlar.

Alıntı Yap