Pregledni rad
The Proposal of a Concept of Artificial Situational Awareness in ATC
Tomislav Rdišić
; Fakultet prometnih znanosti Sveučilišta u Zagrebu
Petar Andraši
; Fakultet prometnih znanosti Sveučilišta u Zagrebu
Doris Novak
; Fakultet prometnih znanosti Sveučilišta u Zagrebu
Tea Rogošić
; Fakultet prometnih znanosti Sveučilišta u Zagrebu
Sažetak
Automation is one of the most promising solutions to the airspace capacity problem. However, we believe that in order to safely implement advanced automation concepts in air traffic control, it is necessary for AI and humans to share situational awareness. One of the main objectives of this concept proposal is to explore the effects and possi-bilities of distributed human-machine situational awareness in en-route air traffic control operations. Instead of automating isolated individual tasks, such as conflict detection or coordination, we propose to create a basis for automation by developing an intelligent situation-aware system. The sharing of the same situational awareness be-tween the members of the air traffic controller team and AI enables the automated system to reach the same conclu-sions as air traffic controllers when faced with the same problem and to be able to explain the reasons for these conclusions. Machine learning can be used to predict, estimate and filter at the level of individual probabilistic events, an area in which it has shown great ability so far, whereas the reasoning engine can be used to represent knowledge and draw conclusions based on all the available data and explain the reasons for these conclusions. In this way, the artificial situational awareness system will pave the way for future advanced automation based on machine learning. Here, we will explore which technologies and concepts are useful in building the artificial situational awareness system and propose the methodology for testing the AI situational awareness.
Ključne riječi
Air traffic control, artificial intelligence, situational awareness, concept proposal
Hrčak ID:
244897
URI
Datum izdavanja:
20.7.2020.
Posjeta: 727 *