Search

The air traffic control paradigm is shifting from sector-based operations to cross-border flow-centric approaches to overcome sectors’ geographical limits. Under the flow-centric paradigm, prediction of the traffic flow at major flow intersections, defined as flow coordination points in this paper, may assist controllers in coordinating intersecting traffic flows which is the main challenge for implementing flow-centric concepts. This paper proposes to predict the flow at coordination points through a transformer neural network model. Firstly, the flow coordination points, i.e., the major flow intersections, are identified by hierarchical clustering of flight trajectory intersections whose location and connectivity characterize daily traffic flow patterns as a graph. The number of coordination points is optimized through graph analysis of the daily flow pattern evolution. Secondly, air traffic flow features in the airspace during a period are described as a “paragraph” whose “sentences” consist of the time and callsign sequences of flights transiting through the identified coordination points. Finally, a transformer neural network model is adopted to learn the sequential flow features and predict the future number of flights passing the coordination points. The proposed method is applied to French airspace based on one-month ADS-B data (from Dec 1, 2019, to Dec 31, 2019), including 158,856 flights. Results show that the proposed prediction model can approximate the actual flow values with a coefficient of determination (R2) between 0.909 to 0.99 and a mean absolute percentage error (MAPE) varying from 27.4% to 11.7% with respect to a 15-minute to 2-hour prediction window. The sustainability of the prediction accuracy under an increasing prediction window demonstrates the potential of the proposed model for longer-term flow prediction. Civil Aviation Authority of Singapore (CAAS) National Research Foundation (NRF) Published version This research is supported by the National Research Foundation, Singapore, and the Civil Aviation Authority of Singapore, under the Aviation Transformation Programme. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not reflect the views of National Research Foundation, Singapore and the Civil Aviation Authority of Singapore.

Improving short-term air traffic flow prediction can help forecast demand and maximize existing capacity by tactical air traffic flow management. Most existing studies in flow prediction lacks consideration of the dynamic, structural, and interrelated nature of air traffic flows in the airspace. Therefore, this paper proposes to predict sector entry flows based on graph convolutional networks, which consider the dynamic spatial-temporal features of air traffic from a graph perspective. First, we specify a sector entry flow based on its upstream and downstream sectors. Then, each entry flow is denoted as a node in a graph. The weighted edges between the nodes are learned from a Word2vec model based on air traffic flows among the nodes. With the weighted graph constructed and the temporal flows on the nodes extracted from the flight trajectories, an Attention-based Spatial-Temporal Graph Convolutional Network (ASTGCN) module is adopted to capture spatial-temporal features of recent, daily-periodic, and weekly-periodic flows in the graph. Finally, The outputs from the ASTGCN module based on the three features are fused to generate the final prediction results. The proposed method is applied on 164 sectors of French airspace for one-month ADS-B data )from Dec 1, 2019, to Dec 31, 2019) which includes 158,856 flights. Results show that, the proposed method outperforms the well established Long short-term memory (LSTM) model, and demonstrates better capability in predicting rapid changes in traffic flow and has relatively smaller decrease in prediction accuracy as the prediction time-window increases. Civil Aviation Authority of Singapore (CAAS) National Research Foundation (NRF) Submitted/Accepted version This research is supported by the National Research Foundation, Singapore, and the Civil Aviation Authority of Singapore, under the Aviation Transformation Programme.

Inefficiencies in flight operations, like deviations and non-optimal flight speed or altitude, are directly linked to flight emission inefficiencies. Quantifying these emission inefficiencies and studying potential mitigation strategies is certainly beneficial for the sustainability of the aviation industry. In this paper, we analyze emission inefficiencies in Dutch and French airspaces using flight data from 2019. The emission inefficiency analysis quantifies the excess carbon emissions for each flight by comparing its emissions with a set of optimal alternative trajectories. We find that around 19% of excess emissions existed in 2019 within the airspace of interest. We also study the potential reduction of emissions by replacing short-range flights with electric aircraft. We propose a simple electric aircraft energy model and relate that to emissions in electric generations in different countries. We find that besides the significant increase in air traffic demand caused by the limited capacities of electric flights, the emissions caused by electricity generation cannot be neglected. Significant reductions can only be achieved when emissions caused by electricity generation are low, as is the case currently in France. However, more emissions can be indirectly generated if the electricity used to power the future electric aircraft is itself produced from high emission sources, as is the case currently in the Netherlands. The paper also provides further insights and recommendations on the data sources, research approach, and future research for aviation sustainability.

Finding a shortest path with various constraints and cost functions is a common problem in operations research. This paper deals with the special case of the problem of computing multiple shortest paths with capacity constraints on a subgraph that evolves over time. Multiple shortest paths is understood as paths for each vehicle considered. The static special case is modeled as a Mixed Integer Linear Programming (MILP) problem, so that it can be solved directly by a standard commercial solver. The time-dependent nature of the problem is then modeled thanks to a sliding-window approach. This study is motivated by the problem of minimizing the environmental impact of air transport at the level of a complete air network, considering thereby several aircraft. Both CO2 and non-CO2 effects are taken into account to calculate the impact. The proposed methodology takes into account a network point of view where airspace (subgraph) capacities evolve as well as the traffic itself over time. Encouraging numerical results on the CO2 /contrail-safe aircraft trajectories application are obtained and show that the environmental impact can be reduced while maintaining safety by guaranteeing the respect of airspace capacities.

International audience; The Extended ATC Planning (EAP) function aims at bridging the gap between Air Traffic Flow & Capacity Management (ATFCM) and Air Traffic Control (ATC) by predicting air traffic congested areas tens of minutes before their formation, and by suggesting real-time and fine-tuning measures to alleviate airspace "complexity". Current Air Traffic Flow Prediction methods focus on crude aircraft count in a sector, and hence are unable to distinguish between low and high complexity situations for a similar aircraft count. Complexity indicators, on the other hand, aggregate air traffic measurements and workload to describe the perceived complexity. However, the evaluation of workload is a long-debated issue and an inherently ill-posed problem. In this work, we present an intrinsic complexity metric, independent of any traffic control system, and address the prediction task of the EAP function using a novel Encoder-Decoder Long Short-Term Memory (LSTM) neural network. The complexity is measured by the eigenvalues of a linear dynamical system fitted to the aircraft's speed vectors. The Encoder-Decoder LSTM network uses a sequence of (discrete) aircraft states to predict the complexity of the air traffic in all areas of an airspace, in a time horizon of 40 minutes. Simulated traffic corresponding to one day of traffic over the French airspace is used to train and validate the model. Our experiments show that the proposed model achieves a Mean Absolute Error of 0.08 in predicting the normalized complexity value 40 minutes in the future.

International audience; As the number of vehicles equipped with radars sensors is rapidly increasing, the risk of harmful interference is rising to undesirable levels, especially since the radar waveform parameters are not regulated. Interference mitigation techniques are becoming very important for radars to operate properly in this complex environment. This article investigates the potential of V2X (Vehicle-to-Everything) technology as a side communication channel between vehicles, to coordinate usage of the radar bandwidth and thus avoid interference. Vehicles equipped with V2X technology can receive information about the upcoming radar waveform parameters of the vehicles in their vicinity, and adjust their own radar parameters accordingly, then inform them back by means of a V2X message. Simulation results reveal that V2X technology can help to significantly reduce interference levels and might thus be a promising companion communication channel for radar interference avoidance techniques.

We present an automated emergency trajectory generator to compute the best emergency trajectory for a given landing site. A combination of the optimized version of the rapidly exploring random tree algorithm and Dubins paths is used to compute a path connecting the aircraft position with the landing site, which avoids the obstacles in the way. Then, this path is used as input for a trajectory prediction (TP) algorithm, which computes a four-dimensional trajectory by taking into account the current aircraft performance and weather. The set of vertical profiles considered in the TP has been designed in order to cover the widest possible range of emergency situations. Moreover, the aircraft intents considered in these profiles are chosen by taking into account the operational requirements of the air traffic operation system and the input of the flight crew. Among these profiles, two have been tested during the study, to verify the result of the proposed algorithm and its computing time, which is one of the main success criteria. This concept is expected to be part of an advanced flight management system on-board function to help the pilot take efficient and effective decisions in emergency situations and adverse conditions.

International audience; In this paper, an efficient algorithm to generate a short and safe trajectory for an aircraft in a situation of emergency is proposed. The algorithm is to be run on a Flight Management System, hence the computation performance, the size of the stored data and the quality of the solution are taken as primary stakes. The algorithm is based on a front propagation algorithm, the Fast Marching method. Fronts are propagated on a single glide slope, allowing the algorithm to provide a flyable trajectory in a very short time (a few seconds). The algorithm is tested in a mountainous environment, and both shortness and safeness are obtained in response to a critical emergency.

International audience; This paper investigates multi-objective optimization of coordinated patrolling flight of multiple unmanned aerial vehicles in the vicinity of terrain, while respecting their performance parameters. A new efficient modified A-star (A*) algorithm with a novel defined criterion known as individual revisit time cell value is introduced and extended to the whole area of the 3D mountainous environment. As a contribution to solving trade-offs in the optimization problem, revisit time is conjugated with other contrary costs effective in flight planning through Pareto analysis. By introducing the revisit time and applying a specific setup to mitigate computational complexity, the proposed algorithm efficiently revisits the desired zones, which are more important to be revisited during the patrolling mission. The results of the introduced modified A* algorithm are compared in various scenarios with two different algorithms including a complete and optimal algorithm known as Dijkstra, and an evolutionary algorithm known as the genetic algorithm. Simulation results demonstrate that the proposed algorithm generates faster and more efficient trajectories in complex multiagent scenarios due to the introduced cell selection method and dynamic-based simplifications applied in this research.

This document provides a description of initial validation test plan for the trajectory generation system.

National audience

This work present a framework based on datadriven techniques for quantifying and chaos theory for propagating the uncertainty present in the aircraft trajectory prediction process when computing the expected trajectory from a given flight plan. The developed framework employs data assimilation models to capture real-time information from the air traffic system and introduces a novel methodology in order to account for the uncertainty of the weather conditions. The comparison of the resulting set of probabilistic trajectories and the actually flown ones proves how the former could be a key enabler to support envisioned trajectory-based operation concepts and modern airline operations planning. Objectius de Desenvolupament Sostenible::9 - Indústria, Innovació i Infraestructura Objectius de Desenvolupament Sostenible::12 - Producció i Consum Responsables

Accurate prediction of aircraft arrival times is one of the fundamental elements for air traffic controllers to manage an optimal arrival and departure sequencing on the runway, reduce flight delays, and achieve a good collaboration with airports and airlines. In this work, we analyze the feature engineering problem to predict Aircraft Landing Time (LDT) in Extended-TMA with machine learning models. Two main contributions are highlighted in this work. First, the impact of different features in LDT prediction is investigated. Second, a machine learning prediction model is presented to predict LDT. Our case of study is the E-TMA of Singapore Changi Airport (WSSS) with a radius of $100$NM. Firstly, data analysis is conducted to check the availability of different resource data, as well as cleaning the raw trajectory data. Then, feature construction and extraction are discussed in details, machine learning prediction models are proposed to address the LDT prediction. The experimental results show that 4 sets of features play a significant impact on LDT prediction for primary runway-in-use, they are: (1) Control intent: traffic demand, current traffic density, and adjacent flow; (2) Weather: surface wind; (3) Trajectory: the position of aircraft; (4) Seasonality: parts of a day and a week. Moreover, comparing three Machine Learning algorithms, in our study case, Extra-Trees is the best prediction algorithm compared with other machine learning models in terms of Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). It is also found that Machine learning models perform much better than the current operational system. In summary, two main conclusions are drawn from the present work. First, predicting the aircraft LDT is strongly correlated with the TMA density at the flight operation time. Second, feature selection with domain knowledge and expert opinions is very important, and with good features, the model is less sensitive to the choice of machine learning algorithm. Civil Aviation Authority of Singapore (CAAS) Accepted version This research is supported by the Civil Aviation Authority of Singapore under the Aviation Transformation Program.

International audience; This paper aims at presenting a novel approach to airline sentiment analysis processing using Twitter data. By transforming trained sentiment classifiers into regressors, the daily sentiment distribution obtained can be represented as atrimodal Gaussian Mixture leading to a simple but efficient classification algorithm. These classes can be considered as daily sentiment scores. This classification applied to passenger generated tweets and airline generated tweets for five major US airlines highlights major difference in experience between passengers and airlines. This methodology also confirms the existing gap between flight performance and passenger experience and the necessity of considering and implementing passenger centric metrics