Abstract

The Air Traffic Management (ATM) community strives to reduce the environmental impact per flight. Continuous Descent Operation (CDO) has been identified by the ATM community as one of the operational improvements that could reduce aviation’s environmental impact, both in terms of aircraft noise and gaseous emissions. In the current ATM system, CDOs are only feasible in low-density traffic. The ultimate goal is an ATM system that facilitates CDOs in high-density traffic. This research described in this thesis focused on two features of such an ATM system: decision support that enables the air traffic controller to accurately set up traffic for CDO, and delegation of the spacing task to the flight crew during the CDO. Two enablers are real-time availability of meteorological data, and accurate trajectory prediction. In this thesis new methods were developed and validated to infer wind, air pressure, and air temperature profiles from aircraft surveillance data. Trajectory prediction was defined as a machine learning problem, enabling predictions based on historic aircraft trajectory and meteorological data without the need for explicit modeling of the aircraft performance and procedures. A decision support tool was developed further and tested using a human-in-the-loop experiment. The tool enabled the subjects to set up traffic for CDO in high-density traffic at an acceptable work load level and high level of situation awareness. Monte-Carlo simulations were carried out to assess the runway capacity that can be achieved when delegating the spacing task to the flight crew. These simulations showed the feasibility of CDOs in high-density traffic.