- author = "Stephane Querry",
- title = "Stochastic optimization by evolutionary methods applied to autonomous aircraft flight control",
- school = "Laboratoire des sciences de l'ingenieur, de l'informatique et de l'imagerie (Strasbourg)",
- year = "2014",
- address = "France",
- month = nov,
- keywords = "genetic algorithms, genetic programming, GPU",
-
URL = "
http://www.polyvionics.com/Thesis.pdf",
-
URL = "
http://theses.unistra.fr/ori-oai-search/notice.html?id=2014STRAD031&printable=true",
-
URL = "http://www.theses.fr/2014STRAD031",
-
URL = "http://scdtheses.u-strasbg.fr/ori-oai-search/notice/view/2014STRAD031",
-
URL = "https://publication-theses.unistra.fr/restreint/theses_doctorat/2014/Querry_Stephane_2014_ED269.pdf",
- size = "198 pages",
-
abstract = "The object of this PhD has consisted in elaborating
evolutionary computing algorithms to find interesting
solutions to important problems in several domains of
automation science, applied to aircrafts mission
conduction and to understand what could be the
advantages of using such approaches, compared to the
state-of-the-art, in terms of efficiency, robustness,
and effort of implementation.
New algorithms have been developed, in Identification, Path planning, Navigation and Control and have been tested on simulation and on real world platforms (AR.Drone 3.0 UAV (Parrot), Oktokopter UAV, Twin Otter and military fighter F-16 (NASA LaRC)), to assess the performances improvements, given by the new proposed approaches.
Interestingly enough, all these different techniques interoperate: GANIAC (Genetic Automatic Nonlinear Identification of Aerodynamic Coefficients) is used in GAFPLAN (Genetic Autonomous Flight Planning and Navigation) that is finally used by GAFCON (Genetic Autonomous Flight Control) for a complete sequence that should make it possible for UAVs to fly in an autonomous way, including learning how to fly with GANIAC.
Because these approaches have been designed to be embedded and used in flight, they pave the way to imagining throwing a UAV in the air that would 1) learn how to fly before crashing, then receive a mission by radio. Thanks to the mathematic identification of the platform, the UAV would then be able to find autonomously good trajectories to fly the mission, then good strategies to avoid obstacles or fulfil some flight constraints and finally, autonomously determine the right sequences of commands to actually fly the mission.
Most of these new approaches provide very interesting results; and research work (on control by evolutionary algorithms, identification by genetic programming and relative navigation) should be engaged to plan potential applications in different real world technologies.",
- resume = "Le but de ce doctorat est de determiner dans quelle mesure les algorithmes issus de l'intelligence artificielle, principalement les Algorithmes Evolutionnaires et la Programmation Genetique, pourraient aider les algorithmes de l'automatique classique afin de permettre aux engins autonomes de disposer de capacites bien superieures, et ce dans les domaines de l'identification, de la planification de trajectoire, du pilotage et de la navigation.De nouveaux algorithmes ont ete developpes, dans les domaines de l'identification, de la planification de trajectoire, de la navigation et du controle, et ont ete testes sur des systemes de simulation et des aeronefs du monde reel (Oktokopter du ST2I, Bebop.Drone de la societe Parrot, Twin Otter et F-16 de la NASA) de maniere a evaluer les apports de ces nouvelles approches par rapport a l'etat de l'art.La plupart de ces nouvelles approches ont permis d'obtenir de tres bons resultats compares a l'etat de l'art, notamment dans le domaine de l'identification et de la commande, et un approfondissement des travaux devraient etre engage afin de developper le potentiel applicatifs de certains algorithmes.",
-
notes = "anglais 2014STRAD031
Supervisor Pierre Collet http://www.idref.fr/162130066 (ICube - BFO)
",
