Machine learning methods for planning conflict-free trajectories
View/
Keywords
Human-centric artificial intelligence ; Machine learning ; Imitation learning ; Supervised learning ; Air traffic management ; Air traffic control ; Trajectory predictionAbstract
Safe and efficient transportation, in terms of cost, time and distance covered, in the aviation domain is provided through the Air Traffic Management (ATM) system, which includes all airborne and ground-based operations required to ensure safe and efficient traffic flow. Every year the volume of air traffic increases pushing the ATM system to its limits, requiring it to handle
greater complexity and density of traffic.
Different initiatives worldwide, such as NextGen in the US and SESAR in Europe, have been investigating the implementation of automation to enhance the efficiency and cost-effectiveness of the Αir Τraffic Μanagement (ATM) system. Towards this goal Artificial Intelligence and Machine Learning (AI/ML) methods are considered for providing accurate predictions of flight trajectories and addressing complexity issues while ensuring safety.
During airborne operations safety between aircraft is provided by the Air Traffic Control (ATC) service. According to International Civil Aviation Organisation (ICAO) Annex 11, ATC is “a service provided for the purpose of: (a) preventing collisions: (1) between aircraft, and (2) on the maneuvering area between aircraft and obstructions; and (b) expediting and maintaining an orderly flow of air traffic”. This includes imposing certain separation minima between aircraft, detecting conflicts that breach separation minima (loss of separation) between flights and their resolution by appropriate actions.
The provision of safe ATC services determines traffic volume, which must not exceed the airspace’s capacities declared. However, capacities should be utilized to the maximum extent due to increased demand and the need for the optimal utilization of resources, without compromising the efficiency and safety of flights. This trade-off introduces challenging issues in the aviation industry, where AI/ML can provide solutions.
The objective of this Ph.D. study is to explore and present state of the art AI/ML algorithms towards planning conflicts-free trajectories in computationally efficient ways, following a methodology combining data-driven and agent-based approaches.
In the context of this study the conflicts-free trajectory planning task is defined to incorporate trajectory prediction and conflicts detection and resolution. While trajectory prediction concerns predicting the spatiotemporal evolution of the aircraft state along a trajectory (also called, trajectory evolution), conflicts detection and resolution concerns the detection of conflicts that breach separation minima (loss of separation) between flights and their resolution by appropriate actions. Therefore, the objective of the conflicts-free trajectory planning task is to predict the evolution of trajectories, regulating flights to avoid loss of separation. While trajectory planning may take place at the pre-tactical phase of operations, methods developed in this study are expected
to have a large impact in the tactical phase of operations. Aiming to model stakeholders’ decisions to planning conflicts-free trajectories, the major emphasis of this study is to imitate flights’ trajectories and air traffic controller’s behavior according to demonstrations provided by historical data.
Department
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών ΣυστημάτωνNumber of pages
141Language
EnglishCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα