Dissertationen zum Thema „Algorithmes inspirés de la robotique“
Geben Sie eine Quelle nach APA, MLA, Chicago, Harvard und anderen Zitierweisen an
Machen Sie sich mit Top-50 Dissertationen für die Forschung zum Thema "Algorithmes inspirés de la robotique" bekannt.
Neben jedem Werk im Literaturverzeichnis ist die Option "Zur Bibliographie hinzufügen" verfügbar. Nutzen Sie sie, wird Ihre bibliographische Angabe des gewählten Werkes nach der nötigen Zitierweise (APA, MLA, Harvard, Chicago, Vancouver usw.) automatisch gestaltet.
Sie können auch den vollen Text der wissenschaftlichen Publikation im PDF-Format herunterladen und eine Online-Annotation der Arbeit lesen, wenn die relevanten Parameter in den Metadaten verfügbar sind.
Sehen Sie die Dissertationen für verschiedene Spezialgebieten durch und erstellen Sie Ihre Bibliographie auf korrekte Weise.
Meyer, Cédric. „Théorie de vision dynamique et applications à la robotique mobile“. Paris 6, 2013. http://www.theses.fr/2013PA066137.
Der volle Inhalt der QuelleThe recognition of objects is a key challenge in increasing the autonomy of robots and their performances. Although many techniques of object recognition have been developed in the field of frame-based vision, none of them hold the comparison against human perception in terms of performance, weight and power consumption. Neuromorphic engineering models biological components into artificial chips. It recently provides an event-based camera inspired from a biological retina. The sparse, asynchronous, scene-driven visual data generated by these sensors allow the development of computationally efficient bio-inspired artificial vision. I focus in this thesis on studying how event-based acquisition and its accurate temporal precision can change object recognition by adding precise timing in the process. This thesis first introduces a frame-based object detection and recognition algorithm used for semantic mapping. It then studies quantitatively what are the advantages of using event-based acquisition using mutual information. It then inquires into low level event-based spatiotemporal features in the context of dynamic scene to introduce an implementation of a real-time multi-kernel feature tracking using Gabor filters or any kernel. Finally, a fully asynchronous time-oriented architecture of object recognition mimicking V1 visual cortex is presented. It extends the state of the art HMAX model in a pure temporal implementation of object recognition
Sarrabezolles, Louise. „Formalization and Simulation of Bio-Inspired On-chip Dynamic Attractors for Low Power Computer Vision“. Thesis, Institut polytechnique de Paris, 2020. http://www.theses.fr/2020IPPAE001.
Der volle Inhalt der QuelleHigh performance in real-time is one of today's challenge of embedded Computer Vision. The study of the human vision models inspires several solutions. The bio-inspired method used by the BVS-Tech component caught the attention of the community with its hardware characteristics and its application in different industrial projects. However, the lack of mathematical formalization and simulation of the method does not allow to analyse it on academical databases.This thesis focused on the understanding, the formalization and the reproduction of the method. The mathematical model and the simulator produced allowed to validate the concept on artificial images, to analyse its behaviour and performances in a specific application. The promising results shows, however, the strong influence of several parameters. Two extensions of the method have recently been proposed: the automatic adaptation of the feature inputs and the prediction integration
Kachurka, Viachaslau. „Contribution à la perception et l’attention visuelle artificielle bio-inspirée pour acquisition et conceptualisation de la connaissance en robotique autonome“. Thesis, Paris Est, 2017. http://www.theses.fr/2017PESC1072/document.
Der volle Inhalt der QuelleDealing with the field of "Bio-inspired Perception", the present thesis focuses more particularly on Artificial Visual Attention and Visual Saliency. A concept of Artificial Visual Attention, inspired from the human mechanisms, providing a model of such artificial bio-inspired attention, was developed, implemented and tested in the context of autonomous robotics. Although there are several models of visual saliency, in terms of contrast and cognition, there is no hybrid model integrating both mechanisms of attention: the visual aspect and the cognitive aspect.To carryout such a model, we have explored existing approaches in the field of visual attention, as well as several approaches and paradigms in related fields (such as object recognition, artificial learning, classification, etc.).A functional architecture of a hybrid visual attention system, combining principles and mechanisms derived from human visual attention with computational and algorithmic methods, was implemented, explained and detailed.Another major contribution of this doctoral work is the theoretical modeling, development and practical application of the aforementioned Bio-inspired Visual Attention model, providing a basis for the autonomy of assistance-robotic systems.The carried out studies and experimental validation of the proposed models confirmed the relevance of the proposed approach in increasing the autonomy of robotic systems within a real environment
Brilhault, Adrien. „Vision artificielle pour les non-voyants : une approche bio-inspirée pour la reconnaissance de formes“. Toulouse 3, 2014. http://thesesups.ups-tlse.fr/2630/.
Der volle Inhalt der QuelleMore than 315 million people worldwide suffer from visual impairments, with several studies suggesting that this number will double by 2030 due to the ageing of the population. Given the variety of its causes, the volume of people affected, and its consequences on quality of life, visual impairment constitutes a major current health issue. To compensate for the loss of sight, the two main holistic approaches consist of sensorial substitution and neuroprosthetics. Sensorial substitution devices provide visual information through different sensory modalities (i. E. Audition or touch). Neuroprostheses reproduce images acquired by a video camera at the surface of the visual structure implanted (retina, LGN, or visual cortex), respecting their spatial configuration: each electrode corresponds to a given pixel. Unfortunately, current implants still suffer from very low resolution: each image is transmitted via a matrix of only a few dozen electrodes, rendering these systems unsuitable for everyday use. Sensory substitution devices are subject to the same limitations: the amount of information needed to process a natural visual scene is far too important in relation to the output interface resolutions (both auditive and tactile, or through microstimulation). Thus the current holistic systems at present are unable to provide sufficient aid in navigating complex visual environments, and are rarely implemented outside the context of laboratory research. To overcome these obstacles, we propose the use of artificial vision in order to pre-process visual scenes and provide the user with relevant information. We have validated this approach through the development of a novel assistive device for the blind called 'Navig'. Through shape recognition and spatialized sounds synthesis, this system is able to restore visuomotor loops, allowing users to locate and grab objects of interest. With navigation being one of the most challenging tasks for the visually impaired, we also developed guidance features relying on satellite positioning as well as an adapted geographic information system. Given that GPS accuracy in urban areas remains too low to safely guide blind pedestrians, we developed a new positioning method combining GNSS, inertial sensors and the visual detection of geolocalized landmarks. The use of artificial vision succeeded in reducing the average positioning error, and as a result provides accurate navigational markers to guide visually impaired users. To enhance the performance of the visual module, a key component of the system, we further developed a novel bio-inspired multi-resolution algorithm for pattern recognition based on the Spikenet library. It uses latency-based coding of visual information, oriented edge representations and several other mechanisms which essentially mimick the activations of the primary visual cortex. Compared to the original monoscale algorithm, our new architecture captures a far broader spectrum of spatial frequencies. Low-resolution processing allows for improved tolerance to image degradations and deformations, while higher and more discriminative frequencies maintain optimal selectivity. Through our cascaded scheme, combining detections at different resolutions, we significantly reduced processing time. Indeed, a first pass is used to filter objects of interest, and only a few candidates are then tested at a higher resolution
Craye, Céline. „Intrinsic motivation mecanisms for incremental learning of visual saliency“. Electronic Thesis or Diss., Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLY006.
Der volle Inhalt der QuelleConceiving autonomous perceptual systems, such as robots able to accomplish a set of tasks in a safe way, without any human assistance, is one of the biggest challenge of the century. To this end, the developmental robotics suggests to conceive robots able to learn by interacting directly with their environment, just like children would. This thesis is exploring such possibility while restricting the problem to the one of localizing objects of interest (or salient objects) within the robot’s environment.For that, we present in this work a mechanism able to learn visual saliency directly on a robot, then to use the learned model so as to localize salient objects within their environment. The advantage of this method is the creation of models dedicated to the robot’s environment and tasks it should be asked to accomplish, while remaining flexible to any change or novelty in the environment.Furthermore, we have developed exploration strategies based on intrinsic motivations, widely used in developmental robotics, to enable efficient learning of good quality. In particular, we adapted the IAC algorithm to visual saliency leanring, and proposed an extension, RL-IAC to allow an efficient exploration on mobile robots.In order to verify and analyze the performance of our algorithms, we have carried out various experiments on several robotics platforms, including a foveated system and a mobile robot, as well as publicly available datasets
Melnyk, Artem. „Perfectionnement des algorithmes de contrôle-commande des robots manipulateur électriques en interaction physique avec leur environnement par une approche bio-inspirée“. Thesis, Cergy-Pontoise, 2014. http://www.theses.fr/2014CERG0745/document.
Der volle Inhalt der QuelleAutomated production lines integrate robots which are isolated from workers, so there is no physical interaction between a human and robot. In the near future, a humanoid robot will become a part of the human environment as a companion to help or work with humans. The aspects of coexistence always presuppose physical and social interaction between a robot and a human. In humanoid robotics, further progress depends on knowledge of cognitive mechanisms of interpersonal interaction as robots physically and socially interact with humans. An illustrative example of interpersonal interaction is an act of a handshake that plays a substantial social role. The particularity of this form of interpersonal interaction is that it is based on physical and social couplings which lead to synchronization of motion and efforts. Studying a handshake for robots is interesting as it can expand their behavioral properties for interaction with a human being in more natural way. The first chapter of this thesis presents the state of the art in the fields of social sciences, medicine and humanoid robotics that study the phenomenon of a handshake. The second chapter is dedicated to the physical nature of the phenomenon between humans via quantitative measurements. A new wearable system to measure a handshake was built in Donetsk National Technical University (Ukraine). It consists of a set of several sensors attached to the glove for recording angular velocities and gravitational acceleration of the hand and forces in certain points of hand contact during interaction. The measurement campaigns have shown that there is a phenomenon of mutual synchrony that is preceded by the phase of physical contact which initiates this synchrony. Considering the rhythmic nature of this phenomenon, the controller based on the models of rhythmic neuron of Rowat-Selverston, with learning the frequency during interaction was proposed and studied in the third chapter. Chapter four deals with the experiences of physical human-robot interaction. The experimentations with robot arm Katana show that it is possible for a robot to learn to synchronize its rhythm with rhythms imposed by a human during handshake with the proposed model of a bio-inspired controller. A general conclusion and perspectives summarize and finish this work
Denarie, Laurent. „Robotics-inspired methods to enhance protein design“. Phd thesis, Toulouse, INPT, 2017. http://oatao.univ-toulouse.fr/18677/1/Denarie.pdf.
Der volle Inhalt der QuelleCraye, Céline. „Intrinsic motivation mecanisms for incremental learning of visual saliency“. Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLY006/document.
Der volle Inhalt der QuelleConceiving autonomous perceptual systems, such as robots able to accomplish a set of tasks in a safe way, without any human assistance, is one of the biggest challenge of the century. To this end, the developmental robotics suggests to conceive robots able to learn by interacting directly with their environment, just like children would. This thesis is exploring such possibility while restricting the problem to the one of localizing objects of interest (or salient objects) within the robot’s environment.For that, we present in this work a mechanism able to learn visual saliency directly on a robot, then to use the learned model so as to localize salient objects within their environment. The advantage of this method is the creation of models dedicated to the robot’s environment and tasks it should be asked to accomplish, while remaining flexible to any change or novelty in the environment.Furthermore, we have developed exploration strategies based on intrinsic motivations, widely used in developmental robotics, to enable efficient learning of good quality. In particular, we adapted the IAC algorithm to visual saliency leanring, and proposed an extension, RL-IAC to allow an efficient exploration on mobile robots.In order to verify and analyze the performance of our algorithms, we have carried out various experiments on several robotics platforms, including a foveated system and a mobile robot, as well as publicly available datasets
Cambier, Nicolas. „Bio-inspired collective exploration and cultural organisation“. Thesis, Compiègne, 2019. http://www.theses.fr/2019COMP2511.
Der volle Inhalt der QuelleAutomatically-controlled artificial systems have recently been used in numerous settings including environmental monitoring and explorations, with great success. In such cases, the use of multiple robots could increase efficiency, although we should ensure that their communication and organisation strategies are robust, exible, and scalable. These qualities can be ensured through decentralisation, redundancy (many/all robots perform the same task), local interaction, and simplistic rules, as is the case in swarm robotics. One of the key components of swarm robotics is local interaction or communication. The later has, so far, only been used for relatively simple tasks such as signalling a robot's preference or state. However, communication has more potential because the emergence of meaning, as it exists in human language, could allow robots swarms to tackle novel situations in ways that may not be a priori obvious to the experimenter. This is a necessary feature for having swarms that are fully autonomous, especially in unknown environments. In this thesis, we propose a framework for the emergence of meaningful communications in swarm robotics using language games as a communication protocol and probabilistic aggregation as a case study. Probabilistic aggregation can be a prerequisite to many other swarm behaviours but, unfortunately, it is extremely sensitive to experimental conditions, and thus requires specific parameter tuning for any setting such as population size or density.With our framework, we show that the concurrent execution of the naming game and of probabilistic aggregation leads, in certain conditions, to a new clustering and labelling behaviour that is controllable via the parameters of the aggregation controller. Pushing this interplay forward, we demonstrate that the social dynamics of the naming game can select efficient aggregation parameters through environmental pressure. This creates resilient controllers as the aggregation behaviour is dynamically evolved online according to the current environmental setting
Tomic, Marija. „Manipulation robotique à deux mains inspirée des aptitudes humaines“. Thesis, Ecole centrale de Nantes, 2018. http://www.theses.fr/2018ECDN0017/document.
Der volle Inhalt der QuelleThe number of humanoid robots has increased in recent years to be able to collaborate with humans or replace them in tedious tasks. The objective of this thesis is to transfer to humanoid robots, skills or human competences, in particular for movements involving coordination between the two arms. In the first part of the thesis, a process of conversion from a human movement to a robot movement, with the aim of imitation is proposed. Since humans have much more freedom than a humanoid robot, identical movements cannot be produced, the characteristics (body lengths) canal so be different. Our conversion process takes into account the recording of marker locations attached to human bodies and joints to improve the imitation processes. The second part of the thesis aims at analyzing the strategies used by humans to generate movement. Human movements are assumed to be optimal and our goal is to find criteria minimized during manipulations. We hypothesize that this criterion is a combination of classical criteria used in robotics and we look for the weights of each criterion that best represents human movement. In this way, an optimal kinematic control approach can then be used to generate movements of the humanoid robot
Milia, Valentin. „Couplage de modèles de chimie quantique et d'algorithmes haute performance pour l'exploration globale du paysage énergétique de systèmes atomiques et moléculaires“. Electronic Thesis or Diss., Université de Toulouse (2023-....), 2024. http://www.theses.fr/2024TLSEP095.
Der volle Inhalt der QuelleThe primary aim of this thesis is to develop efficient methods for characterizing molecular conformations at a quantum level. Various methods devoted to the computation of molecular potential energy are reviewed, as well as the most popular potential energy surfaces (PES) global exploration schemes. In this context, a key contribution of this thesis is the coupling of the robotics-inspired Iterative Global exploration and LOcal Optimization (IGLOO) method, implemented in the MoMA software, with the quantum Density-Functional based Tight-Binding (DFTB) potential, implemented in the deMonNano software. The IGLOO algorithm integrates the motion planning Rapidly-exploring Random Trees (RRT) algorithm with local optimization and structural filtering. A proof of concept has been done through the identification of low-energy conformations of the alanine dipeptide.The IGLOO/DFTB coupling has been applied to the mapping of the PES of three close-sized molecules of the phthalate family (dibutyl phthalate DBP, benzyl butyl phthalate BBP and di-2-ethylhexyl phthalate DEHP), providing detailed insights into their different conformational landscapes. Various geometrical descriptors have been used to analyze their structure-energy relationships. Coulomb interactions, steric hindrance, and dispersive interactions have been found to drive the geometric properties and a strong correlation has been evidenced between the two dihedral angles describing the side-chains orientation of the phthalate molecules. The results demonstrate the method's capability to identify low-energy minima without prior knowledge of the PES.Furthermore, an innovative algorithm for the large-scale generation of molecular structures, including a conformational variety, is presented. It combines molecular graph generation with atom or fragment addition techniques. It is applied to provide an extensive database of 3D structures of hydrogenated amorphous carbon (a-CH) molecules. The analysis of the database generated in this study provides a comprehensive understanding of the relationship between the geometrical and electronic descriptors of a-C:H structures. These properties are compared with those of compact Polycyclic Aromatic Hydrocarbons and linear chains, representing limit cases.Finally, a review is given on methods aiming at identifying saddle points and transition paths between low-energy conformations on the PES. A first step toward the identification of transition paths between low-energy conformations using a motion planning algorithm, known as Transition-based Rapidly-exploring Random Trees (T-RRT), is presented. A similarity measure, designated as the Symmetrized Segment-Path Distance (SSPD), is used to compare the generated trajectories. Subsequently, a clustering technique, namely the Hierarchical Clustering Analysis (HCA), is employed to group similar trajectories in order to identify the common pathways, thereby providing valuable insights into the dynamics of conformational changes. The methodology has been successfully applied to the identification of low-energy paths between two minima of the alanine dipeptide PES.Overall, the research presents significant advancements in the exploration of complex molecular PES at a quantum level including (i) the IGLOO/DFTB coupling (ii) a novel algorithm for 3D structure generation of large-scale molecules and (iii) an original scheme allowing for the identification of multiple transition paths. Correlations between the structural, energetic and electronic properties have been evidenced for the polluting phthalate molecules and astrophysically relevant hydrogenated amorphous carbon (a-CH) molecules. These contributions pave the way for future research, aiming to extend these methods to larger and more complex systems
Douib, Ameur. „Algorithmes bio-inspirés pour la traduction automatique statistique“. Thesis, Université de Lorraine, 2019. http://www.theses.fr/2019LORR0005/document.
Der volle Inhalt der QuelleDifferent components of statistical machine translation systems are considered as optimization problems. Indeed, the learning of the translation model, the decoding and the optimization of the weights of the log-linear function are three important optimization problems. Knowing how to define the right algorithms to solve them is one of the most important tasks in order to build an efficient translation system. Several optimization algorithms are proposed to deal with decoder optimization problems. They are combined to solve, on the one hand, the decoding problem that produces a translation in the target language for each source sentence, on the other hand, to solve the problem of optimizing the weights of the combined scores in the log-linear function to fix the translation evaluation function during the decoding. The reference system in statistical translation is based on a beam-search algorithm for the decoding, and a line search algorithm for optimizing the weights associated to the scores. We propose a new statistical translation system with a decoder entirely based on genetic algorithms. Genetic algorithms are bio-inspired optimization algorithms that simulate the natural process of evolution of species. They allow to handle a set of solutions through several iterations to converge towards optimal solutions. This work allows us to study the efficiency of the genetic algorithms for machine translation. The originality of our work is the proposition of two algorithms: a genetic algorithm, called GAMaT, as a decoder for a phrase-based machine translation system, and a second genetic algorithm, called GAWO, for optimizing the weights of the log-linear function in order to use it as a fitness function for GAMaT. We propose also, a neuronal approach to define a new fitness function for GAMaT. This approach consists in using a neural network to learn a function that combines several scores, which evaluate different aspects of a translation hypothesis, previously combined in the log-linear function, and that predicts the BLEU score of this translation hypothesis. This work allowed us to propose a new machine translation system with a decoder entirely based on genetic algorithms
Douib, Ameur. „Algorithmes bio-inspirés pour la traduction automatique statistique“. Electronic Thesis or Diss., Université de Lorraine, 2019. http://www.theses.fr/2019LORR0005.
Der volle Inhalt der QuelleDifferent components of statistical machine translation systems are considered as optimization problems. Indeed, the learning of the translation model, the decoding and the optimization of the weights of the log-linear function are three important optimization problems. Knowing how to define the right algorithms to solve them is one of the most important tasks in order to build an efficient translation system. Several optimization algorithms are proposed to deal with decoder optimization problems. They are combined to solve, on the one hand, the decoding problem that produces a translation in the target language for each source sentence, on the other hand, to solve the problem of optimizing the weights of the combined scores in the log-linear function to fix the translation evaluation function during the decoding. The reference system in statistical translation is based on a beam-search algorithm for the decoding, and a line search algorithm for optimizing the weights associated to the scores. We propose a new statistical translation system with a decoder entirely based on genetic algorithms. Genetic algorithms are bio-inspired optimization algorithms that simulate the natural process of evolution of species. They allow to handle a set of solutions through several iterations to converge towards optimal solutions. This work allows us to study the efficiency of the genetic algorithms for machine translation. The originality of our work is the proposition of two algorithms: a genetic algorithm, called GAMaT, as a decoder for a phrase-based machine translation system, and a second genetic algorithm, called GAWO, for optimizing the weights of the log-linear function in order to use it as a fitness function for GAMaT. We propose also, a neuronal approach to define a new fitness function for GAMaT. This approach consists in using a neural network to learn a function that combines several scores, which evaluate different aspects of a translation hypothesis, previously combined in the log-linear function, and that predicts the BLEU score of this translation hypothesis. This work allowed us to propose a new machine translation system with a decoder entirely based on genetic algorithms
Germain, Pascal. „Algorithmes d'apprentissage automatique inspirés de la théorie PAC-Bayes“. Thesis, Université Laval, 2009. http://www.theses.ulaval.ca/2009/26191/26191.pdf.
Der volle Inhalt der QuelleAt first, this master thesis presents a general PAC-Bayes theorem, from which we can easily obtain some well-known PAC-Bayes bounds. Those bounds allow us to compute a guarantee on the risk of a classifier from its achievements on the training set. We analyze the behavior of two PAC-Bayes bounds and we determine peculiar characteristics of classifiers favoured by those bounds. Then, we present a specialization of those bounds to the linear classifiers family. Secondly, we conceive three new machine learning algorithms based on the minimization, by conjugate gradient descent, of various mathematical expressions of the PAC-Bayes bounds. The last algorithm uses a part of the training set to capture a priori knowledges. One can use those algorithms to construct majority vote classifiers as well as linear classifiers implicitly represented by the kernel trick. Finally, an elaborated empirical study compares the three algorithms and shows that some versions of those algorithms are competitive with both AdaBoost and SVM.
Inscrit au Tableau d'honneur de la Faculté des études supérieures
Soussen, Charles. „Algorithmes d'approximation parcimonieuse inspirés d'Orthogonal Least Squares pour les problèmes inverses“. Habilitation à diriger des recherches, Université de Lorraine, 2013. http://tel.archives-ouvertes.fr/tel-00924578.
Der volle Inhalt der QuelleFijany, Amir. „Algorithmes et architectures parallèles en robotique“. Paris 11, 1988. http://www.theses.fr/1988PA112258.
Der volle Inhalt der QuelleFijany, Amir. „Algorithmes et architectures parallèles en robotique“. Grenoble 2 : ANRT, 1988. http://catalogue.bnf.fr/ark:/12148/cb37613611m.
Der volle Inhalt der QuelleWang, Yun. „Stéréovision robotique algorithmes de mise en correspondance /“. Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37610700h.
Der volle Inhalt der QuelleJuston, Raphael. „De l'oeil élémentaire à l'oeil composé artificiel : application à la stabilisation visuelle en vol stationnaire“. Thesis, Aix-Marseille, 2013. http://www.theses.fr/2013AIXM4118.
Der volle Inhalt der QuelleThe biorobotics team from the Institute of Movement Sciences (Marseille, France) takes its inspiration from biological studies on flying insects which are able to navigate into unknown 3D environments with a high maneuverability. These studies led us to build minimalist optical sensors to make aerial robots autonomous for achieving complex tasks such as automatic landing and take-off, obstacle avoidance and very accurate hovering flight depicted in this doctoral thesis. This work presents several bio-inspired visual sensors implemented with different visual processing algorithms. All these sensors are able to locate visual objects (contrasting edges and bars) with unusual properties for optical sensing devices: a blur obtained by defocusing optics related with active retinal micro-movements to improve the sensor resolution. We showed that the resolution in locating contrasting objects can be improved up to 160 fold better than the static resolution defined by the pixel pitch, which means that these bio-inspired optical sensors are endowed with hyperacuity.The thesis presents a miniature artificial compound eye CurvACE (of 1.75g for 2.2cm3) with a panoramic field of view (180x60°). This thesis describes thoroughly the characterization and the implementation of the CurvACE sensor onboard an aerial robot named HyperRob. This artificial compound eye acts as a position sensing device able to measure its position relative to a complex textured scene by fusing the position measurements obtained by 40 pixels. The tethered flying robot HyperRob (a 150-g bi-rotor with a 23-cm wingspan) stabilizes its roll and its position thanks to its hyperacute artificial compound eye
Larabi, Marie-Sainte Souad. „Biologically inspired algortithms for exploratory projection pursuit“. Toulouse 1, 2011. http://www.theses.fr/2011TOU10021.
Der volle Inhalt der QuelleMafrica, Stefano. „Capteurs visuels bio-inspirés pour des applications robotiques et automobiles“. Thesis, Aix-Marseille, 2016. http://www.theses.fr/2016AIXM4029/document.
Der volle Inhalt der QuelleThanks to the advances in the fields of robotics and intelligent transportation systems (ITS), the autonomous vehicles of the future are gradually becoming a reality. As autonomous vehicles will have to behave safely in presence of other vehicles, pedestrians and other fixed and moving objects, one of the most important things they need to do is to effectively perceive both their motion and the environment around them. In this thesis, we first investigated how bio-inspired visual sensors, giving 1-D optic flow using a few pixels based on the findings on the fly’s visual system, could be used to improve automatic parking maneuvers. We subsequently tested a novel bio-inspired silicon retina, showing that the novel pixel, called M2APix, can auto-adapt in a 7-decade range and respond appropriately to step changes up to ±3 decades, while keeping sensitivity to contrasts as low as 2%. We lastly developed and tested a novel optic flow sensor based on this auto-adaptive retina and a new robust method for computing the optic flow, which is robust to the light levels, textures and vibrations that can be found while operating on the road. We also constructed a car-like robot, called BioCarBot, which estimates its velocity and steering angle by means of an extended Kalman filter (EKF) using only the optic flow measurements delivered by two downward-facing sensors of this kind
Wang, Yun. „Stéréovision robotique : algorythmes de mise en correspondance“. Paris 12, 1987. http://www.theses.fr/1987PA120006.
Der volle Inhalt der QuelleLimozin-Long, Pascale. „Vision stéréoscopique appliquée à la robotique“. Nice, 1986. http://www.theses.fr/1986NICE4052.
Der volle Inhalt der QuelleLanneau, Sylvain. „Localisation et estimation basées modèle d’un objet ellipsoidal avec le sens électrique artificiel“. Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2017. http://www.theses.fr/2017IMTA0030/document.
Der volle Inhalt der QuelleThe aim of this thesis is to contribute to the underwater perception for robotics applications using an electric field. We propose new methods for the inspection, the localization and the shape estimation of an ellipsoidal object using a sensor inspired by the weakly electric fish. Firstly, we show that the object can be detected and its material and position relative to the sensor axis discriminated, using simple threshold detections on the measured currents. Then, we propose the successive implementations of three reactive control laws allowing the sensor to head for the object and revolve around it by following its boundaries. After that, we use the MUSIC algorithm in order to localize the object’s center. Finally, the geometrical parameters of the object and its orientation are estimated thanks to an optimization algorithm based on the least squares method and the inversion of the analytical model of the polarization tensor of an ellipsoidal object. We show that these algorithms can be experimentally implemented. For the localization and the shape estimation algorithms, some additional techniques involving sensor movements are proposed in order to significantly reduce the imprecisions due to the gap between the model and the actual currents’ measurements
Scornavacca, Céline. „Supertree methods for phylogenomics“. Montpellier 2, 2009. http://www.theses.fr/2009MON20136.
Der volle Inhalt der QuelleEvolutionary relationships between species can be summarized in the so-called “species tree. Species trees are mainly estimated using gene trees, ie evolutionary trees constructed by analyzing a gene family. However, for both methodological and biological reasons, a gene tree may differ from the species tree. To estimate species tree, biologists then analyze several data sets at a time. This thesis focuses on the “supertree” approach to combine data sets. This approach consists first in constructing trees (commonly called source trees) from primary data, then assembling them into a larger tree, called supertree. When using supertree construction in a divide-and-conquer approach in the attempt to reconstruct the Tree of Life, conservative supertree methods have to be preferred in order to obtain reliable supertrees. In this context, a supertree method should display only information that is displayed or induced by source trees (induction property, PI) and that does not conflict with source trees or a combination thereof (non contradiction property, PC). In this thesis we introduce two combinatorial properties that formalize these ideas. We proposed algorithms that modify the output of any supertree methods such that it verifies these properties. Since no existing supertree method satisfies both PI and PC, we have developed two methods, PhySIC and PhySIC_IST, which directly build supertrees satisfying these properties. Since duplication events often result in the presence of several copies of the same genes in the species genomes, gene trees are usually multi-labeled, ie , a single species can label more than one leaf. Since no supertree method exists to combine multi-labeled trees, until now these gene trees were simply discarded in supertree analyses. In this thesis, we propose several algorithms to extract a maximum amount of speciation signal from multi-labeled trees and put it under the form of single-labeled trees which can be handled by supertree methods
Godzik, Nicolas. „Approches évolutionnaires pour la robotique modulaire et anticipatoire“. Paris 11, 2005. http://www.theses.fr/2005PA112200.
Der volle Inhalt der QuelleIn this thesis, we propose some original approaches to autonomous robotics tha are somehow intermediate between Brooks'subsumption architecture and the pure black-box Evolutionary Robotics approach. Both those approaches badly scale up with the complexity of the task of the robot. The "symbolic controllers" approach uses some high level behaviors as output of the controller (e. G. Move forward, turn left,. . . ). The following step is that of the "supervisors", where the controller simply chooses between different possible behaviors from a library of behaviors that can themselves be the result of previous evolutions (e. G. Explore, go to the light,. . . ). This approach is then validate on a homing experiment with battery recharge. Finally, we studied the coupling between evolution and life-time learning for the robot. The "auto-teaching" architecture, proposed by Nolfi and Parisi, has been thoroughly tested, and some clear defects of its generalization properties have been shown. The AAA architecture (Action, Anticipation Adaptation) was designed to remedy those defects: it uses the errors made by the anticipatory part of the controller compared to its actual sensations to on-line modify the weights of the controller. Even though some long-term weaknesses still remain, this architecture showed some small adaptation capabilities, and some prmosing results were obtained after tuning some parameters of the model
Zidi, Abir. „Méthodes par sous-espaces et algorithmes d’optimisation bio-inspirés pour le débruitage de signaux multidimensionnels et applications“. Thesis, Ecole centrale de Marseille, 2017. http://www.theses.fr/2017ECDM0003/document.
Der volle Inhalt der QuelleThis thesis is devoted to study matrix and tensor ranks of multidimensional signalsand to the development of methods for estimating these ranks in the frameworkof the wavelet transform. For this study, we used the wavelet packet decompositionand the multilinear algebra. A bio-inspired stochastic optimization methodhas been adapted, with the ultimate objective of suppressing noise in multidimensionalimages. In order to ensure this, we have estimated the different values ofthe dimensions of the tensor subspace for all the modes of the coefficients of thewavelet packets.We have applied the proposed denoising methods to various multidimensionalimages: RGB images, multispectral images extracted from hyperspectralimages of metal parts, plant fluorescence images, and multispectral RX images.Finally, a comparative study was carried out with three main types of algorithms: onthe one hand, the Perona-Malik method based on diffusion; Second, the truncationof HOSVD and MWF, and thirdly, a method based on wavelet packet decompositionand MWF, where the dimensions of the signal subspace are estimated by a statisticalcriterion rather than by an optimization method. The results are promising in termsof denoising in grund truth. Ultimately, we achieve an advantageous time savingduring the acquisition of hyperspectral images
Bernard, Mathieu. „Audition active et intégration sensorimotrice pour un robot autonome bioinspiré“. Phd thesis, Université Pierre et Marie Curie - Paris VI, 2014. http://tel.archives-ouvertes.fr/tel-01023986.
Der volle Inhalt der QuelleMezgebe, Tsegay Tesfay. „Algorithmes inspirés des sociales pour concevoir des nouveaux systèmes de pilotage dans le contexte de l'usine du futur“. Electronic Thesis or Diss., Université de Lorraine, 2020. http://www.theses.fr/2020LORR0006.
Der volle Inhalt der QuelleThe use of traditionally centralized control system does not able to meet the rapidly changing customer expectations, high product varieties, and shorter product life-cycles. In particular, the emergence of Cyber Physical System (CPS) which can be seen as interacting networks of physical and computational components has provided the foundation for many new factories’ infrastructures and improved the quality of products and processes. This Cyber Physical System has dramatically impacted the centrally predictive control system in responding to perturbation(s) in the current dynamic market characteristics. Urgent change for example is one of the common perturbations and has significant perturbing ability to a central predictive control system. Accordingly, it is now accepted that using agent-based control system improves the reactivity to treat these perturbation(s) until they are no longer limiting factors. In this study, the use of human-inspired interaction approach (by means of negotiation and consensus-based decision-making algorithms) is explored to design and propose a new control system. It has taken advantages from Industry 4.0 assets and encompassed heterogeneous and intelligent entities (mainly the product entities and resource entities) and discrete event systems. Each entity could have different capability (evolution, learning, etc.) and the whole physical and control system may lead emerging behaviors to dynamically adapt the perturbation(s). Hence, every intelligent entity decides when to broadcast its current state to neighbor entities and the controlling decision depends on the behavior of this state. The negotiation and consensus-based decision-making algorithms were initially formulated and modeled by networking all the contributing and heterogeneous entities considering two real industrial shop floors. Then after, simulation and application implementation tests on the basis of full-sized academic platform called TRACILOGIS platform have been conducted to verify and validate these decision-making algorithms. This has been done with expectations that the applicability of these algorithms will be more adaptable to set best priority-based product sequencing and rescheduling than another decision-making approach called pure reactive control approach. Accordingly, the experimental results have shown that the negotiation and consensus among the decisional entities have significantly minimized the impact of perturbation(s) on a production process launched on the TRACILOGIS platform. Meanwhile, using these two decision-making algorithms has conveyed better global performance (e.g., minimized makespan) over the pure reactive control approach
Orseau, Laurent. „Imitation algorithmique : Apprentissage Incrémental En-ligne de Séquences“. Rennes, INSA, 2007. http://www.theses.fr/2007ISAR0014.
Der volle Inhalt der QuelleIn continual learning, an agent is continually interacting with its environment. At each time step, it receives inputs, uses a small amount of computations (online) and gives outputs. There is no real definition of a goal to learn, the agent must acquire more and more knowledge, incrementally, and re-use it in more complex tasks. In this framework, we are interested in learning complex sequences, involving recurrence, variables and conditions. But the agent cannot use a large number of trials and error, because of its interaction with the environment. How then can learning be possible from a small number of examples?Traditional methods that are able to solve such complex tasks do not fit in the continual learning framework, because difficulties become harder. To simplify the task, an imitation protocol is used, allowing the agent to learn by seeing a teacher doing, but this respects the continual learning constraints and keeps a high autonomy. Imitation is usually used in a robotic framework, so we extend it to learn more complex sequences~: this is Algorithmic Imitation. A learning system, CSAAL, is then developed and tested on experiments showing that it is indeed able to learn complex sequences within few examples. An extension of this system, H-CSAAL, allows to re-use hierarchically recurrent functions, increasing both the autonomy of the agent and its generalization capacities
Montero, Elisabeth. „Calibration strategies for bio-inspired population-based algorithms that solve combinatorial optimization problems“. Nice, 2011. http://www.theses.fr/2011NICE4040.
Der volle Inhalt der QuelleSelection of proper values for parameters of bio-inspired algorithms is crucial to get good performance. To find these values is a complex process because parameter values depend on the problem that is being solved. Moreover, it is a high time consuming task and parameters are interrelated values. The first contribution of this thesis is the comparative study of features and performance of some automated parameter tuning methods. Advantages and disadvantages of these methods were identified. The main disadvantage of tuning methods is that they are high consuming time process. These methods search for parameters values focusing on statistically promising areas of the parameters search space. We also present an analysis of the potential of using these techniques to support the design process of evolutionary algorithms. In most cases, tuning methods are not able to discard elements that do not contribute to their operation. The main contribution of this thesis is related to the ability of varying parameter values during the execution of the algorithm. These kinds of approaches are known as parameter control methods. Two methods are proposed in this thesis for improving the performance of evolutionary algorithms by controlling its operators rates : Ac and Sac. Ac works in a population level rewarding operators that perfom good transformations during the iteration and punishing the others. Sac works in an individual level rewarding or punishing operators according to the performance of their applications in each individual. These both control methods can be incorporated to the evolutionary algorithm without introducing a meaningful cost in terms of time and without changing the original structure of the evolutionary algorithm. Variations of parameter values are based on current and past information about the operators performance. The incorporation of these techniques allows the algorithm to decide when and how much use each operator during the search process. Moreover, we propose the (C,n)-strategy for the CLONALG artificial immune system. In this case we control the amount of selected cells and clones produced in each iteration. The application of our control method allows us to reduce the amount of parameters to tune for the algorithm. Our control technique is able to manage the relationship between intensification and diversification stages of the search by varying the values of these parameters. From this work, we can conclude that parameter tuning techniques are able to find parameter values that show a good performance on the tuned algorithm. Most of these techniques require a previous analysis of the parameter search spaces and, in general, they require a considerable amount time. However, they can be able to make decisions about the design of the evolutionary algorithm. About the incorporation of parameter control strategies, we can conclude that it is possible to improve the efficiency of the algorithm, reducing the tuning effort by decreasing the number of parameters of tune. It is possible to vary parameter values during the search without changing the original design of the algorithm. Finally, the proposed techniques add a minimum amount of easily tunable new parameters. The development of tuning techniques able to work with any kind of parameters and capable to determine proper stopping criteria in order to do not waste resources is still an open research area. In this thesis we have shown that it is possible to perform on-line parameter variation in population-based algorithms. One interesting point of research is the automatic identification of the role of different components of the algorithm in order to automatize the design process of parameter control strategies
Bouzid, Zohir. „Modèles et algorithmes pour les systèmes émergents“. Paris 6, 2013. http://www.theses.fr/2013PA066589.
Der volle Inhalt der QuelleNetworks of autonomous robots are mobile entities that communicate only through their movements and the observation of their respective positions. They are anonymous, without memory and without a global coordinate system nor a common notion of distance. We focus on the algorithmic study of the problems of gathering and convergence of robots when some of them may be subject to failures. Our first contribution is of geometric nature. We provide a protocol to compute the Weber point of a large class of rotational symmetric configurations. Based on this primitive, we present an algorithm that solves thegathering problem in presence of any number of crash failures. Then, we address the convergence problem when robots may incur byzantine failures which are harder to handle than crash failures. We provide several tight bounds relating the degree of synchronicity of the system to its resiliency. Finally, we study robots that are endowed with memory and we show that this model is stronger than the message passing model. The different node types in an uniformized manner. Our experimental results show that this model is able to take in account the correlations between labels of different node types
Côté, Marianne. „Développement d'algorithmes d'estimation de la pose d'objets saisis par un préhenseur robotique“. Master's thesis, Université Laval, 2016. http://hdl.handle.net/20.500.11794/26802.
Der volle Inhalt der QuelleEscobar-Zuniga, María-José. „Modèles bio-inspirés pour l'estimation et l'analyse de mouvement : reconnaissance d'actions et intégration du mouvement“. Nice, 2009. http://www.theses.fr/2009NICE4050.
Der volle Inhalt der QuelleThis thesis addresses the study of the motion perception in mammals and how bio-inspired systems can be applied to real applications. The first part of this thesis relates how the visual information is processed in the mammal's brains and how motion estimation is usually modeled. Based on this analysis of the state of the art, we propose a feedforward V1-MT core architecture. This feedforward V1-MT core architecture will be a basis to study two different kinds of applications. The first application is human action recognition, which is still a challenging problem in the computer vision community. We show how our bio-inspired method can be successfully applied to this real application. Interestingly, we show how several computational properties inspired from motion processing in mammals, allow us to reach high quality results, which will be compared to latest reference results. The second application of the bio-inspired architecture proposed in this thesis, is to consider the problem of motion integration for the solution of the aperture problem. We investigate the role of delayed V1 surround suppression, and how the 2D information extracted through this mechanism can be integrated to propose a solution for the aperture problem. Finally, we highlight a variety of important issues in the determination of motion estimation and additionally we present many potential avenues for future research efforts
Jalel, Sawssen. „Optimisation de la navigation robotique“. Phd thesis, Toulouse, INPT, 2016. http://oatao.univ-toulouse.fr/17515/1/Jalel.pdf.
Der volle Inhalt der QuelleTournassoud, Pierre. „Planification de trajectoires en robotique : complexité et approche pratique“. Paris 11, 1988. http://www.theses.fr/1988PA112070.
Der volle Inhalt der QuelleFuret, Matthieu. „Analyse cinéto-statique de mécanismes de tenségrité : Application à la modélisation de cous d'oiseaux et de manipulateurs bio-inspirés“. Thesis, Ecole centrale de Nantes, 2020. https://tel.archives-ouvertes.fr/tel-03286016.
Der volle Inhalt der QuelleA tensegrity structure is an assembly of elements in compression (bars) and elements in tension (cables, springs) held together in equilibrium. Tensegrity has been known in architecture and art for more than a century and is adapted to the modeling of living organisms. More recently, tensegrity mechanisms have been studied for their promising properties in robotics such as low inertia, natural flexibility and deployment capability. A tensegrity mechanism is obtained when one or more elements are actuated, in our case by cables. This thesis is part of the AVINECK project, in which biologists and roboticians are involved with a dual goal: to have a better understanding of nature and bird necks, and also to draw inspiration from nature in order to design innovative bio-inspired manipulators. First, a study was conducted on the kinematics of bird vertebrae. A modeling of the contacts between articular surfaces allowed to reconstruct in 3D the movement of any bird vertebra with respect to another, based on 3D scans of real vertebrae. Then, several simple mechanisms were compared in order to determine the best candidate for modeling the kinematics of bird necks. Secondly, two 1-degree of freedom (DDL) tensegrity mechanisms potentially interesting for bird neck modeling and manipulator realization were modeled. Starting from an optimal design approach, two optimal mechanisms according to a given criterion were dimensioned and then compared in order to determine the most interesting mechanism from a manipulator design point of view. The 1 DDL mechanism chosen was an antagonistically actuated anti-parallelogram, called the X mechanism. A prototype of a 2 DDL manipulator was then designed and built. A study of the geometrical model of such a manipulator revealed interesting phenomena such as the cuspidal behaviour of the robot under certain conditions. A kinetostatic study of the manipulator was then carried out. As the behavior of the manipulator strongly depends on the actuation strategy, these were listed and some of them were compared. The results obtained were verified and highlighted on a prototype. Finally, the modeling of plane manipulators was extended to N DDL manipulators. A complete dynamic model was obtained, and a design process was carried out to obtain the dimensions and specifications of a 10 DDL bio-inspired planar manipulator
Prébet, Rémi. „Connexité dans les ensembles algébriques réels : algorithmes et applications“. Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS664.
Der volle Inhalt der QuelleThis PhD thesis focuses on the design and the analysis of computer algebra algorithms for solving polynomial systems. More precisely, we address the problems of counting the number of connected components of sets of real solutions to systems of polynomial equations with real coefficients and of answering connectivity queries over such real solution sets. These problems are central in real algebraic geometry and find applications in robotics. The chosen framework is the one of roadmaps, introduced by Canny in 1988: it consists in computing a curve, included in the solution set under consideration, which has a connected intersection with all its connected components. We design an algorithm which, under some regularity assumptions which are satisfied generically, computes such roadmaps in time subquadratic w.r.t. the output size. This latter quantity is nearly optimal. This extends to non compact situations the best complexity results known for such a computational problem. We also show that the cost to compute the number of connected components of a real algebraic curve (lying in a space of arbitrary dimension) is nearly the same as the one of computing the topology of its projection on a generic plane. Last, by combining these results with algorithms of real algebraic geometry, we design an algorithm to decide the cuspidality of robots
Doncieux, Stéphane. „Robotique évolutionniste: conception orientée vers le comportement“. Habilitation à diriger des recherches, Université Pierre et Marie Curie - Paris VI, 2010. http://tel.archives-ouvertes.fr/tel-00547778.
Der volle Inhalt der QuelleTlapale, Olivier Émilien. „Modelling the dynamics of contextual motion integration in the primate“. Nice, 2011. https://tel.archives-ouvertes.fr/tel-00850265.
Der volle Inhalt der QuelleThis thesis addresses the study of motion integration in the primate. Based on anatomical and functional knowledge of two cortical areas involved in motion perception, namely VI and MT, we explain various perceptual and oculo-motor responses found in the literature. First, we build a recurrent model of motion integration where a minimal number of cortical interactions are assumed. Proposing a simple readout mechanism, we are able to reproduce not only motion perception but also the dynamics of smooth pursuit eye movements on various line figures and gratings viewed through different apertures. Second, following perceptual studies concerning motion integration and physiological studies of receptive fields, we construct another dynamical model where motion information is gated by form cues. To this end, we postulate that the visual cortex takes advantage of luminance smoothness in order to gate motion diffusion. Such an elementary diffusion mechanism allows to solve various contextual problems where extrinsic junctions should be eliminated, without relying on complex junction detectors or depth computation. Finally, we rewrite the initial dynamical model into the neural field formalism in order to mathematically analyse its properties. We incorporate the multiplicative feedback term into the formalism, and prove the existence and uniqueness of the solution. To generalise the comparison against visual performance, we propose a new evaluation methodology based on human visual performance and design a database of image sequences taken from biology and psychophysics literature. Offering proper evaluation methodology is essential to continue progress in modelling the neural mechanisms involved in motion processing. To conclude, we investigate the performances of our neural field model by comparison against state of the art computer vision approaches and sequences. We find that, despite its original objective, this model gives results comparable to recent computer vision approaches of motion estimation
Djennas, Meriem. „Les apports des outils de l'intelligence artificielle à l'amélioration du processus de prévision des taux de change : le cas de la couronne norvégienne“. Amiens, 2013. http://www.theses.fr/2013AMIE0052.
Der volle Inhalt der QuelleThe main objective of our research covers the quantitative modeling of foreign exchange rate using an artifïcial intelligence approach, notably the genetic algorithms and neural networks, applied to the Norwegian foreign exchange market. In the context of modeling to explain the movement of exchange rates, the thesis proposes a reflection on the artifïcial intelligence means applied to the chartist-fundamentalist approach of the exchange rate. The results of optimization and simulation show that despite their complexity, the series of exchange rates can be modeled so that the estimated values of the exchange rate approach, wherever possible, the real values of foreign exchange rate. The artifïcial simulation by a neuro-genetic model gives the best result compared with a STAR model and a standard neural model. The second part of the thesis has shown that using a genetic algorithm as a means of optimization has allowed measuring the impact of explanatory variables on the movement of the Norwegian exchange rate
Pauplin, Olivier. „Application de techniques d'évolution artificielle à la stéréovision en robotique mobile autonome“. Paris 5, 2007. http://www.theses.fr/2007PA05S015.
Der volle Inhalt der QuelleThe "Fly Algorithm" is an evolutionary algorithm used for stereo reconstruction It gives a rough description of the visible surfaces of the objects in a 3D scene, from a couple of stereo To fmprove the efficiency and precision of the algorithm so as to apply its output to autonomous mobile robotics, several ways of improvement are examined and^ evaluated (fitness function, genetic operators, self-adaptivity etc. ). Three categories of applications are then explored. Firstly, a way to build a map of the environment of a mobile robot from he successive frames processed by the algorithm. A second type of application relies on the consideration of a speed given to each considered 3D points, and aims to determine the speeds of the objects of the scene. Finally, and application to automated driving (stop and obstacle avoidance) is presented on an electrical vehicle (Cycab)
Magro, Gérard. „Génération de trajectoires sans collision en robotique : principes et algorithmes pour la modélisation et la navigation“. Toulouse, ENSAE, 1989. http://www.theses.fr/1989ESAE0012.
Der volle Inhalt der QuelleBelkhiri, Ayman. „Modélisation dynamique de la locomotion compliante : Application au vol battant bio-inspiré de l'insecte“. Phd thesis, Ecole des Mines de Nantes, 2013. http://tel.archives-ouvertes.fr/tel-00874497.
Der volle Inhalt der QuelleTeichteil, Florent. „Approche symbolique et heuristique de la planification en environnement incertain : optimisation d'une stratégie de déplacement et de prise d'information“. Toulouse, ENSAE, 2005. http://www.theses.fr/2005ESAE0028.
Der volle Inhalt der QuellePeignier, Sergio. „Subspace clustering on static datasets and dynamic data streams using bio-inspired algorithms“. Thesis, Lyon, 2017. http://www.theses.fr/2017LYSEI071/document.
Der volle Inhalt der QuelleAn important task that has been investigated in the context of high dimensional data is subspace clustering. This data mining task is recognized as more general and complicated than standard clustering, since it aims to detect groups of similar objects called clusters, and at the same time to find the subspaces where these similarities appear. Furthermore, subspace clustering approaches as well as traditional clustering ones have recently been extended to deal with data streams by updating clustering models in an incremental way. The different algorithms that have been proposed in the literature, rely on very different algorithmic foundations. Among these approaches, evolutionary algorithms have been under-explored, even if these techniques have proven to be valuable addressing other NP-hard problems. The aim of this thesis was to take advantage of new knowledge from evolutionary biology in order to conceive evolutionary subspace clustering algorithms for static datasets and dynamic data streams. Chameleoclust, the first algorithm developed in this work, takes advantage of the large degree of freedom provided by bio-like features such as a variable genome length, the existence of functional and non-functional elements and mutation operators including chromosomal rearrangements. KymeroClust, our second algorithm, is a k-medians based approach that relies on the duplication and the divergence of genes, a cornerstone evolutionary mechanism. SubMorphoStream, the last one, tackles the subspace clustering task over dynamic data streams. It relies on two important mechanisms that favor fast adaptation of bacteria to changing environments, namely gene amplification and foreign genetic material uptake. All these algorithms were compared to the main state-of-the-art techniques, obtaining competitive results. Results suggest that these algorithms are useful complementary tools in the analyst toolbox. In addition, two applications called EvoWave and EvoMove have been developed to assess the capacity of these algorithms to address real world problems. EvoWave is an application that handles the analysis of Wi-Fi signals to detect different contexts. EvoMove, the second one, is a musical companion that produces sounds based on the clustering of dancer moves captured using motion sensors
Mkaouar, Hachicha Raoudha. „Contribution à la modélisation et résolution du problème d’affectation sous contraintes de compétences et préférences“. Paris 8, 2012. http://octaviana.fr/document/168003317#?c=0&m=0&s=0&cv=0.
Der volle Inhalt der QuelleThe consideration of the human characteristics is essential to enhance the enterprise performance. In this research work, we are particularly interested to ensure a better matching between task requirements and human resources competences taking into account their preferences. To do this, we propose a new approach that consists of four steps to solve the assignment problem under competences and preferences constraints. The first step concerns the identification of the acquired and required competences levels using the 2-tuple linguistic representation model. The second focus on the ranking of the resulting competences levels using the TOPSIS technique. The third step manages the integration of the resulting ranking as well as the preferences’ information into the problem formulation. Finally, a new algorithm based on the hybridization of the bee algorithm and the immune system is developed to solve the proposed problem. Finally, the approach is illustrated through an example depicted from a real case study in a telecommunications sector company
Fernández, Ramírez Arnoldo Apolonio. „Contribution à la manipulation d'objets par ume main robotique avec repositionnement des doigts“. Poitiers, 2007. http://www.theses.fr/2007POIT2311.
Der volle Inhalt der QuelleMouret, Jean-Baptiste. „Pressions sélectives multiples pour l'évolution de réseaux de neurones destinés à la robotique“. Paris 6, 2008. http://www.theses.fr/2008PA066640.
Der volle Inhalt der QuelleEvolutionary algoritms have been successfully used to generate controllers for many robots. However, they struggle to design complex artifacts when the fitness is unable to explicitly guide the process. In this thesis, we draw the hypothesis that these problems originate from the use of a single selection pressure, whereas living organisms are subject to many ones. We investigate here th euse of multiobjective evolutionary algorithms to create such multiple gradients in order to evolve neuro-controllers. We first describe how hypotheses about potential intermediate steps can be used by defining a multiobejctive optimization problem in which each objective corresponds to a sub-task. In the lack of any selection pressure, it is also possible to add an objective which encourages an efficient exploration of the neighborhood of current candidate solutions. We consider several possibilities to instantiate this concept for the evolution of neural networks and we conclude that maintaining the diversity of the behaviors, instead of the diversity of the genotype or the ohenotype, is an efficient way to override the deceptiveness of a fitness function. Last, we show that exaptations can be favored by applying a selection pressure on some modules of the generated neural-networks, possibly linked to genotypic modules. We tested these methods on the evolution of neural networks to compute a Boolean function and to control a light-seeking robot. They seem to be applicable to a wide range of evolutionary robotics problems, from complex locomotion to behavior control
Khoudour, Louahdi. „Analyse spatio-temporelle de séquences d'images lignes : application au comptage des passagers dans les systèmes de transport“. Lille 1, 1997. https://pepite-depot.univ-lille.fr/LIBRE/Th_Num/1997/50376-1997-5.pdf.
Der volle Inhalt der Quelle