Current thesis
  • Analyse statistique de trajectoires avion avec application à la sécurité des vols - Courjault-Radé Vincent
    Ecole doctorale: EDMITT
    Start date: November 2014
    Supervisors: Ludovic d'Estampe(Enac) et Stéphane Puechmorel (Enac)
    Cette thèse a pour but l'étude systématique des moyens statistiques applicables à l'analyse de données de vols qui peuvent être aussi bien des enregistrement radar que des fichiers obtenus à partir des QARs. L'accent sera mis sur deux points spécifiques :
    - La taille des échantillons recueillis, qui pour une trajectoire peut atteindre plusieurs milliers de points, chacun de dimension variant de 3 (plots radar) à 300 (sortie QAR).
    - Le caractère fonctionnel des données de vol, qui doivent être considérées comme des applications d'un intervalle temporel déterminé dans l'espace des mesures.
    L'accent sera mis en particulier sur les méthodes issues de la statistiques sur données fonctionnelles et sur la réduction de dimension.
  • Cooperative aircraft trajectories planning - Breil Romaric
    Ecole doctorale: EDAA
    Start date: January 2014
    Supervisors: Daniel Delahaye (ENAC), Eric Feron (Georgia Tech)
    This thesis aims to develop distributed algorithms allowing to do strategic planning of air traffic, giving both performance and resilience, this second property, inherent to those methods, allowing them to reorganize themselves after a disturbance. More specifically, we will seek to emerge a globally organized traffic structure that statistically reduces interactions between trajectories, only using local rules.
  • - Nicolas Péteilh
    Ecole doctorale: EDAA
    Start date: January 2014
    Supervisors: Christian Bès (UPS), Marcel Mongeau (ENAC)
  • Optimizing Airport Dynamic Strategic Planning under Uncertainty - Elena Mihaela CAPITANUL
    Ecole doctorale: EDMITT
    Start date: october 2013
    Supervisors: Felix MORA CAMINO (ENAC), Alexandru DONCESCU (LAAS)
    Airports of the 21st century are expansive and expensive infrastructures with a significant economic, social and environmental impact at national, regional and international level. If in the past was witnessed an exponential growth of air transportation on markets like US and Europe, today these markets have matured up, privatization with very long term concessions becoming a common practice for all airport categories. In this context, airports must be capable to generate sufficient revenues to finance their operations and investments while being profitable and maintaining a satisfactory quality of service for both their primary clients: airlines and passengers, and also maintaining its role of economic driver supporting in a sustainable manner its local community. The main objective of this thesis is to develop a model to assess the financial risk when engaging in such a challenge. This assessment will be based on the analysis of airport operational costs, long term demand predictions and airport technological advances and will make use of mathematical techniques such as fuzzy modelling, fuzzy logic and dynamic programming. Cases studies considering regional airports in Eastern Europe will be developed.
  • 3D Departure and Arrival Route Design in Terminal Maneuvering Area - Jun ZHOU
    Ecole doctorale: EDMITT
    Start date: october 2013
    Supervisors: Sonia Cafieri (ENAC), Mohammed Sbihi (ENAC)
    The Terminal Maneuvering Area or Terminal Control Area (TMA) is a block of airspace class around a set of airports (in a big city). It is designed to handle aircraft arriving to and departing from airports and perhaps one of the most complex types of airspace. The arrival phase of flights, typically starts when the aircraft leave their cruise level in En-Route phase and ends when aircraft reach the Final Approach Fix. Aircraft arriving to airports in terminal areas are organized in arrival streams. To build such streams, the individual paths of each aircraft have to be gradually merged. The purpose of this thesis is to optimally design departure and arrival routes. Two planification levels will be addressed.
    • The first one aims at designing such routes at stategic level. For a given airport we propose to designdeparture-arrival route for any runway configurations.
    • The second one will address route design at tactical level. For a given configuration, we consider that some arrival or departure routes are blocked by some weather events (thunderstorm). Routes have to be dynamically redesigned in order to address the demand as efficiently as possible.
    Both designs have to take into account operationnal constraints, for instance noise restrictions, aircraft feature (for instance, a heavy aircraft can't land on a short runway), vertical and horizontal flow separation.
    The aim of this thesis is, on the one hand, to develop mathematical models for the addressed problem, and on the other hand to devise efficient numerical methods and corresponding algorithms for the associated (global) optimization problems.
  • Problèmes de localisation de stations ADS-B pour la surveillance aérienne - Ning Wang
    Ecole doctorale: EDAA
    Start date: october 2013
    Supervisors: Alexandre Gondran (ENAC), Marcel Mongeau (ENAC)
    Automatic Dependent Surveillance-Broadcast (ADS-B) is a new cooperative surveillance technology, which is based on satellite positioning. ADS-B OUT service provides air traffic controllers with real-time position information through ADS-B ground stations, which is more accurate than the current radar-based systems. ADS-B IN service allows aircraft to receive traffic and flight information from nearby aircraft or from ground, therefore enhances the pilot situation awareness. Moreover, ADS-B is cheaper than radar in terms of deployment price. For the above reasons, ADSB is gaining acceptance around the world and has become a key technology of the next-generation surveillance system.
    The purpose of this thesis is first to define the ADS-B system location problems, then to model these problems under the form of optimization problems, and finally to develop original methodologies to solve the problems with a special interest to applications in a large country like China.
  • Dynamic Optimization of airspace sectors grouping - Tambet Treimuth
    Ecole doctorale: EDAA
    Start date: January 2013
    Supervisors: Daniel Delahaye (ENAC) , Sandra Ulrich NGUEVEU (LASS)
    The purpose of this thesis is to build optimization algorithms to schedule on a day the scheme of merging and splitting control sectors on a given airspace. One of the main objectives of this plainification is to provide sectors groups with a minimum of coordination and which are more balanced in terms of loadwork control so that each team of controllers can work in the same way. It is also important to optimize the switching times between sector configurations depending on traffic fluctuations. These algorihtmes will allow more to take into account the distances between two successive configurations in order to avoid sudden changes in a given airspace.
  • Dynamic optimization of airspace sectorization based on the 4D trajectory complexity - Marina Sergeeva
    Ecole doctorale: EDMITT
    Start date: December 2012
    supervisors : Daniel DELAHAYE (ENAC) , Catherine MANCEL (ENAC)
    We considered a 4D trajectories optimized set for users constituting a traffic demand over a given period. This demand leads to a congestion of airspace characterized by connections between 4D aircraft trajectories. The purpose of this thesis is to build segmentation algorithms for airspace to construct 3D sectors adapted to the distribution of current traffic. One of the main objectives of this segmentation is to provide areas with a minimum of coordination and more equilibrated in terms of control workload so that each team of controllers working in the same way. These algorithms will allow a gradual deformation of the segments shapes based on changes of traffic flow directions.
  • Optimization of trajectories in random meteorological environment - Cécile Ichard
    Ecole doctorale: EDMITT
    Start date: October 2012
    Supervisors:Marcel Mongeau (ENAC) , Christophe BAEHR (meteo France)
    La trajectoire des avions est un processus pour lequel on dispose d'un prédicteur dont nombre de paramètres ne sont pas connus. Si la plupart de ces paramètres dépendent des aéronefs, ceux qui relèvent des conditions météorologiques leur sont communs. Un certain nombre d'études ont de fait essayé de caractériser les vents (en valeurs directes ou en termes d'erreur par rapport à une prévision) à partir de l'observation radar d'un trafic de plusieurs avions.
    Les services de prévisions météorologiques principaux fournissent depuis plusieurs années des prévisions ensemblistes permettant d'avoir plusieurs scénarii d'évolution. Il s'agit d'un ensemble de prévisions dont les conditions initiales diffèrent pour rendre compte de la méconnaissance de l'état de l'atmosphère en se concentrant sur les zones ayant le plus fort impact sur les prévisions. Cette technique permet ainsi d'explorer l'incertitude sur la condition initiale. Les produits d'ensemble ont été développés au départ pour des échéances de prévisions assez longues avec des résolutions spatiales modestes. Mais depuis, l'accroissement des performances des calculateurs aidant et avec la réduction de la granularité des prévisions, il est possible d'envisager une application de ces techniques d'ensembles aux données aéronautiques.
    Le travail de cette thèse consiste à étudier l'utilisation des prévisions ensemblistes pour l'aéronautique.
  • Optimisation of ocean trafic - Olga Rodionova
    Ecole doctorale: EDMITT
    Start date: November 2011
    Supervisors : Marcel Mongeau (ENAC) and Mohamed Sbihi (ENAC)
    This thesis is devoted to study the air traffic in North Atlantic oceanic airspace. Radar-based surveillance is not applicable in this highly congested airspace. For conflict-free flight progress, the Organized Track System is established in North Atlantic, and flights are prescribed to follow pre-defined oceanic tracks. Large longitudinal separation norms are imposed, preventing aircraft from performing re-routings between the tracks. This leads to an increase in aircraft cruising time and congestion level in continental airspace at input and output. Implementing new technologies and airborne-based control procedures will enable significant decrease in the present separation standards and improvement of the traffic situation in North Atlantic. Making the direct data communication possible between the aircraft will permit to utilize meteorological measurements from preceding aircraft for following flight predictions, increasing the accuracy of such predictions. The study estimates the advantages obtained due to new technologies: trajectory and flight time prediction improvement, congestion level decrease, etc.
  • Optimal control applied on the aircraft conflict avoidance problem - Loïc Cellier
    Ecole doctorale: EDMITT
    Start date: October 2011
    Supervisors: Sonia Cafieri (ENAC) , Frédéric Messine (IRIT)
    The purpose of this Ph.D. thesis is to study models and solution techniques based on optimal control approaches to address the aircraft conflict avoidance problem (in tactical phases and en-route flight configurations). To ensure the minimum safety distances between aircraft, this research focuses primarily on a velocity regulation strategy to perform the separation which preserves the predefined trajectories-directions.