This study proposes an autonomous aircraft taxi agent that can be used to recommend the pilot the optimal speed profile to achieve optimal fuel burn and to arrive on time at the target position on the taxiway while considering potential interactions with surrounding traffic. The problem is modeled as a control decision problem that is solved by training the agent under a Deep Reinforcement Learning (DRL) mechanism using the Proximal Policy Optimization (PPO) algorithm. The reward function is designed to consider the fuel burn, taxi time, and delay time. Accordingly, the trained agent will learn to taxi the aircraft between any pair of locations on the airport surface in a timely manner while maintaining safety and efficiency. As a result, in more than 97.8% of the evaluated sessions, the controlled aircraft reached the target position with a time difference falling within the range of −20 to 5 s. Moreover, compared to actual fuel burn, the proposed autonomous taxi agent demonstrated a reduction of 29.5%, equivalent to reducing 13.9 kg of fuel consumption per aircraft. This benefit in fuel burn reduction can complement the emission reductions achieved by solving other sub-problems, such as pushback control and taxi-route assignments, to achieve much higher performance.

Aircraft crew are aware of the delay they have experienced at departure. However, uncertainties ahead, and in particular holdings at arrival, can have an impact on the final performance of their operations. When optimizing a trajectory, the expected cost at the arrival gate should be considered. Consequently, taking into account potential congestion and extra delay at the arrival airspace is paramount to avoid making sub-optimal decisions during the early stages of a flight. This paper presents a framework to optimize trajectories in the execution phase of the flight considering expected delays at arrival. A flight from Athens (LGAV) to London Heathrow (EGLL) is used as an illustrative example, systematically exploring a range of departure delays and expected holdings at arrival.

This contribution discusses the imaging of large antenna array navigation systems by means of near-field measurements. Examples for such navigation systems are the instrument landing system (ILS) and the Doppler VHF omnidirectional radio range (DVOR). In this context, imaging means the determination of individual array antenna amplitudes and phases with near-field measurements and a corresponding prediction of the far-field radiation characteristics according to required specifications of flight inspections. Near-field measurement results of a large ILS antenna array are presented, and fundamentals are explained with numerical simulations. In particular, a simple simulation scheme allows the investigation of basic measurement requirements and errors that are essential for the later application. This work is part of the Navant-NG II (navaid antenna characterization – next generation) project dealing with near-field inspection techniques using a unmanned aerial vehicle. From the academic point of view, this contribution summarizes the fundamental issues and probable approaches for solutions of such imaging techniques, which are a mathematical ill-conditioned problem that exceeds the current state-of-the-art near-field inspection of single antennas.

Slot allocation in a single airport aims to maximize the utilization of airport-declared capacity under operational and regulation constraints, while that in a multi-airport system (MAS) has to take airspace capacity into account. This is due to the fact that the conflict of using the limited capacity of certain departure/arrival fixes in the terminal airspace could induce unnecessary flight delays. The uncertainty of flying times between the airport and congested fixes makes it even more complicated for slot allocation in a MAS. Traffic flow may exceed capacity when the flying times of flights change. In this paper, the authors propose an uncertainty slot allocation model for a MAS (USAM). The objective of the model is to minimize the total displacement of slot requests in the MAS while considering all of the capacity constraints, as well as the uncertainty of flying time. The constraints of departure/arrival fixes are formulated as chance constraints, and then the Lyapunov theorem is applied for reformulation. The USAM is applied in the MAS of the Guangdong-Hong Kong-Macao Greater Bay Area (GBA). Specifically, the impact of the uncertainty of flying times from five airports to airspace fix YIN is investigated. Results show that the total displacement would increase if the uncertainty of flying time was considered. The optimized schedule using the USAM, however, is more robust and can satisfy capacity constraints under various scenarios.

This paper presents a machine learning-based approach for predicting the taxi-out time, with the departure process decomposed into two components: the time taken to travel from the gate to the departure queue, and the time spent in the departure queue. Gradient-Boosted Decision Tree (GBDT) models are trained to predict the two components using different feature sets, and a comparison of both model shows that they can provide better prediction accuracy compared with conventional methods, with a Root Mean Squared Error (RMSE) of 1.79 minutes and 0.92 minutes when predicting the taxiing and queuing times respectively, and 78% and 96% of predictions falling within a ±2 minute error margin. Predictions from the GBDT model are analysed and interpreted using SHAP (SHapley Additive exPlanations) values, a well-recognised technique for providing interpretability to many different black-box models, and allowing feature importance to be evaluated at global (model) and local (individual prediction) levels. In particular, the most important feature groups for the taxiing and queuing models are respectively the route features and runway queuing features. The model explainability provides a pathway towards the certification of machine learning techniques in Air Traffic Controller (ATCO) decision support tools.

Air traffic flow management balances demand and capacity strategically by applying various initiatives such as ground delay programs and controlled enroute delays. The delay assigned to each flight is determined by the estimated time of arrival and the maximum allowed airborne delay (buffer) set to absorb uncertainties and minimize arrival runway throughput and capacity loss. Current operations often apply a constant buffer regardless of the projected traffic. This research uses high-fidelity traffic simulations to investigate the effect of a dynamically selected buffer optimizing the daily flow control. Performance is measured by three metrics: ground delay, airborne delay and capacity loss, which assures that runway pressure is maintained. Simulations over 162 days of traffic demonstrate the potential for considerable savings using the proposed method. Furthermore, an initial feasibility investigation of machine learning applied to the dynamic buffer selection problem is performed, and is concluded that, despite a certain loss of optimality and estimation accuracy challenges, such techniques can aid real-life implementation.

System Wide Information Management (SWIM) provides a digital data-sharing infrastructure that includes the standardization of the data format and exchange protocol internationally so that the required information can be shared among related civil aviation stakeholders efficiently and safely. In our project, we study the availability of the Aeronautical Mobile Airport Communications System (AeroMACS) as a wireless medium to achieve Aircraft Access to SWIM (AAtS) around the airport. We carried out flight experiments with an AeroMACS prototype and SWIM test bed over Sendai City to confirm the possibility of expanding AeroMACS coverage. In this paper, we focus on tracking, antenna placement and handover evaluations. First, the results of tracking tests show that our system can track the aircraft by using only ADS-B position reports and keep the communication link between the base station and the aircraft. Second, the results of antenna placement tests show that both the main and sub antennas for large aircraft should be mounted at the bottom of the aircraft body to obtain higher combined gain. On the other hand, the main and sub antennas for small aircraft should be mounted at the bottom and top of the aircraft body respectively, to obtain diversity gain during the aircraft turn. Third, the results of handover tests show that our system needs parameter optimization regarding handover in the air. Finally, we demonstrate SWIM-based information sharing over extended AeroMACS coverage.

A “detect-and-avoid” capability providing a “remain well-clear” function will be needed for unmanned aircraft to fly safely in uncontrolled airspace. It could also be used in controlled airspace provided that the action chosen by the pilot, based on the system suggestive guidance, is compatible with the current air traffic control clearance or an amended clearance. The study reported in this paper looks at the potential operational consequences of the interaction between a pilot and the air traffic controller in controlled airspace when a clearance amendment must be requested by the pilot using the recently standardized detect-and-avoid system, Airborne Collision Avoidance Systems Xu (ACAS Xu). Numerical simulations on a potential encounter are performed. A model of pilot behavior when faced with a remain well-clear alert (including pilot-controller communication delays) is used. An analysis of the qualitative results outlines several areas of concern: the possible confusion caused by multiple changes in remain well-clear guidance, the undesirable effects when both aircraft are equipped with ACAS Xu, and the potential interaction with short-term conflict alerts displayed at the air traffic controller’s workstation.

Strategic 4D trajectory planning is a promising technology for next-generation air traffic management and systems. Some approaches attempt to satisfy the capacity constraint to reduce traffic congestion, while others aim to reduce potential conflicts between trajectories. This paper investigates two approaches to organizing the real traffic in the French airspace at the strategic level. The first approach minimizes interaction between trajectories, while the second reduces traffic congestion so that the controller maintains the traffic without much effort. The associated optimization problems are formulated and resolved using an approximative approach based on simulated annealing. The departure time perturbation was introduced to study the robustness of the two proposed methods. An evaluation of the robustness is performed using Monte Carlo simulation. According to the results, the strategic deconfliction method completely solved all interactions between trajectories, and the strategic decongestion method reduced traffic congestion by 99.94%. Furthermore, a comparative study shows that the method reducing congestion is more robust against the departure time perturbation than the method minimizing interaction between trajectories. These findings encourage appropriate use of the proposed methods in the strategic 4D trajectory planning framework.