Artificial Intelligence Perfecting Autonomous Distress Locator: Defusing False Positives

• Daniel Valério Leite
• Li Weigang
• Pastor Willy Gonzáles Taco

University of Brasilia (UnB), Post-Graduate Program in Transport (PPGT) · Post-Graduate Program in Informatics (PPGI), Computational Model Laboratory and Artificial Intelligence · for Transport (TransLab/UnB) · Federal Highway Police University (UniPRF)

Evento

XXII SITRAER – Brasília, DF, Brasil, 22 a 24 de outubro de 2025

Volume

Anais do XXII Simpósio de Transporte Aéreo – SITRAER 2025 · ISBN 978-65-272-2024-4 · DOI 10.29327/9786527220244

Editora

Even3 Publicações

Editores

Fabiana Serra de Arruda, Camila Lucena, Giovanna Miceli Ronzani Borille, Evandro José da Silva

Abstract

Emergency Locator Transmitters (ELTs) are devices equipped in aircraft to provide coordinates in emergency situations, although during collision with the ground the ELT is damaged preventing transmission of the signal with the coordinates, as well as when the aircraft is submerged in rivers or seas the warning signal is not transmitted. Some regions of Brazil, especially the Amazon rainforest, have communication problems with pilots to resolve questions about aircraft safety. The lack of communication linked to the activation of the ELT can trigger search and rescue missions. Thus, to respond to this challenge, we propose the application of an intelligent system, called the in-flight emergency locator transmitter (ELT-DT), which is autonomously activated during because it detects the anomalous behavior of the aircraft and already activates the transmission before it is damaged with the ground. The problem is that ELT-DT can generate false alerts, because aircraft may present some angulations considered abnormal, but this may be due to transient turbulence. Then, the method will take place in data modeling with characteristics of angles of attack that indicate that the aircraft is in a state of pre-stall, stall - when the aircraft loses support in flight -, as well as when it returns to the normal state. Previously said, it will be possible to perform the method with stall data combined with artificial intelligence techniques. The article proposes the use of Artificial Intelligence (AI) to mitigate false alerts that may trigger unnecessary searches. Based on a proof of concept (PoC) to evaluate the capacity of Artificial Intelligence (AI) to solve and manage the ELT-DT. Soon, tests will be done with AI in search of the best performance AI. The results show that AI can mitigate false alerts from ELT-DT and prevent the triggering of rescuers unnecessarily generating large expenses to public coffers.

Keywords: ELT-DT; Artificial intelligence; Autonomous distress locator; Stall; Alert