Nikoleris, Tasos | Engineering Papers

NASA NTRS2018

Supervised Learning Applied to Air Traffic Trajectory Classification

Given the recent increase of interest in introducing new vehicle types and missions into the National Airspace System, a transition towards a more autonomous air traffic control system is required in order to enable and handle increased density and complexity. This paper presents an exploratory effort of the needed autonomous capabilities by exploring supervised learning techniques in the context of aircraft trajectories. In particular, it focuses on the application of machine learning algorithms and neural network models to a runway recognition trajectory-classification study. It investigates the applicability and effectiveness of various classifiers using datasets containing trajectory records for a month of air traffic. A feature importance and sensitivity analysis are conducted to challenge the chosen time-based datasets and the ten selected features. The study demonstrates that classification accuracy levels of 90% and above can be reached in less than 40 seconds of training for most machine learning classifiers when one track data point, described by the ten selected features at a particular time step, per trajectory is used as input. It also shows that neural network models can achieve similar accuracy levels but at higher training time costs.

Bosson, Christabelle↗

NASA NTRS2018

Supervised Learning Applied to Air Traffic Trajectory Classification

Given the recent increase of interest in introducing new vehicle types and missions into the National Airspace System, a transition towards a more autonomous air traffic control system is required in order to enable and handle increased density and complexity. This paper presents an exploratory effort of the needed autonomous capabilities by exploring supervised learning techniques in the context of aircraft trajectories. In particular, it focuses on the application of machine learning algorithms and neural network models to a runway recognition trajectory-classification study. It investigates the applicability and effectiveness of various classifiers using datasets containing trajectory records for a month of air traffic. A feature importance and sensitivity analysis are conducted to challenge the chosen time-based datasets and the ten selected features. The study demonstrates that classification accuracy levels of 90% and above can be reached in less than 40 seconds of training for most machine learning classifiers when one track data point, described by the ten selected features at a particular time step, per trajectory is used as input. It also shows that neural network models can achieve similar accuracy levels but at higher training time costs.

Bosson, Christabelle S.↗

NASA NTRS2016

Comparison of Fuel Consumption of Descent Trajectories Under Arrival Metering

This paper compares fuel consumption of descent trajectories from cruise altitude to meter fix when the required time of arrival is later than the nominal time of arrival at the meter fix. The required delay, which is the difference between the nominal and the required times of arrival, is achieved by either slowing down the aircraft in the cruise and descent phases or flying a longer route at a constant altitude. Performance models of ten different Boeing and Airbus aircraft, obtained from the Base of Aircraft Data, are employed for generating the results. It is demonstrated that the most fuel-efficient speed control strategy for absorbing delay is first reducing descent speed as much as possible and then reducing cruise speed. This is a common finding for all ten aircraft considered. For some aircraft, flying at a fixed flight path angle and constant Mach-calibrated-airspeed results in lower fuel consumption compared to standard descent at idle-thrust and constant Mach-calibrated- airspeed. Finally, for the cases examined, it is shown that executing a path stretch maneuver at cruise altitude and descent at a reduced speed is more fuel efficient than inserting an intermediate-altitude cruise segment.

Nikoleris, Tasos↗

NASA NTRS2016

Performance of an Automated System for Control of Traffic in Terminal Airspace

This paper examines the performance of a system that performs automated conflict resolution and arrival scheduling for aircraft in the terminal airspace around major airports. Such a system has the potential to perform separation assurance and arrival sequencing tasks that are currently handled manually by human controllers. The performance of the system is tested against several simulated traffic scenarios that are characterized by the rate at which air traffic is metered into the terminal airspace. For each traffic scenario, the levels of performance that are examined include: number of conflicts predicted to occur, types of resolution maneuver used to resolve predicted conflicts, and the amount of delay for all flights. The simulation results indicate that the percentage of arrivals that required a maneuver that changes the flight's horizontal route ranged between 11% and 15% in all traffic scenarios. That finding has certain implications if this automated system were to be implemented simply as a decision support tool. It is also found that arrival delay due to purely wake vortex separation requirements on final approach constituted only between 29% and 35% of total arrival delay, while the remaining major portion of it is mainly due to delay back propagation effects.

air traffic control↗