Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Archives des thèses ...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.

Etude et application des méthodes d'apprentissage pour la navigation en environnement inconnu

Authors: Pastor, Philippe;

Etude et application des méthodes d'apprentissage pour la navigation en environnement inconnu

Abstract

L’objet de cette thèse est l’étude des méthodes d'apprentissage par renforcement en vue de son application à la navigation d’un robot mobile autonome. Après une présentation des méthodes d’apprentissage développées depuis les débuts de la Cybernétique jusqu’à aujourd'hui en Intelligence Artificielle, nous présentons les fondements mathématiques de l'apprentissage par renforcement que sont la théorie des automates d’apprentissage et la Programmation Dynamique en temps réel. Les chapitres suivants sont consacrés au problème de la navigation d’un robot mobile autonome évoluant dans un environnement qui lui est inconnu. Pour répondre à ce problème, nous proposons d’utiliser différents algorithmes d’apprentissage par renforcement issus, soit des automates d'apprentissage, soit du Q-learning. Les performances de ces algorithmes sont ensuite comparées à partir d'expérimentations menées sur un système non-holonome. Enfin, le dernier chapitre propose une extension originale de ce type d’apprentissage dans le but de construire une carte représentant la topologie de l'environnement dans lequel le robot évolue. This thesis studies the application of reinforcement learning methods to the autonomous mobile robots navigation problem. After a presentation of the learning methods which have been developped from the begin- ning of cybernetics to the current state-of-the-art in artificial intelligence, we present the mathematical foundations of reinforcement learning, i.e. learning automata theory and real time dynamic programming. The following chapters are devoted to the problem of navigation for an autonomous mobile robot in an unknown environment. To solve this problem, several algorithms based on learning automata or on Q-learning are proposed. The performance of these algorithms are compared on the basis of simulation results for a robot with nonholonomic constraints. The last chapter presents an original extension of this kind of learning for the construction of a topological map of the robot’s environment.

Keywords

Robotique mobile autonome, Programmation dynamique en temps réel, Apprentissage par renforcement, 000, Real-time dynamic programming, Reinforcement learning, Q-learning, Map learning, Automates d'apprentissage, Apprentissage de cartes, Navigation, Learning automata, Autonomous mobile robots

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green