Unearthing strategies from visio-physiological data of traffic conflict situations in terminal maneuvering area

Abstract

Airspace will becoming increasingly crowded and complex, human Air Traffic Control Officers (ATCOs) need to transit from the role of an operator to that of a supervisor of automation. Current Artificial Intelligence in Air Traffic Management only encompasses providing information and advisories, but not executing solutions automatically. Hence, there is to understand the strategies used by the human ATCOs to provide a template for the Artificial Intelligence to smoothen the process of automation acceptance. This paper will examine the maneuver patterns used by human ATCOs in resolving waypoint conflicts in two cases, the converging conflict and overtaking conflict, and in sequencing conflicting for landing. The focus of this paper will be on the Terminal Maneuvering Area control sector of the Air Traffic Management framework. Adoption rates of maneuver patterns identified will also be computed and compared to uncover preferred methods of resolution maneuvers. The identified maneuver patterns will also be validated with eye tracking data to uncover unique signatures to the maneuver patterns for recommendations for the suitability of integration into the automation system. The performances of each experiment will be studied and correlated to the number of commands executed or number of fixations. An understanding of maneuver patterns used to resolve conflicts in the Terminal Maneuvering Area will help drive future automation acceptance to alleviate the human ATCOs’ workload. However, the overall results have concluded that the maneuver patterns and eye tracking signatures is not suited to for integration with an automation framework.