Our senses are our mind's window to the outside world and determine how we perceive our environment.Sensory systems are complex multi-level systems that have to solve a multitude of tasks that allow us to understand our surroundings.However, questions on various levels and scales remain to be answered ranging from low-level neural responses to behavioral functions on the highest level.Modeling can connect different scales and contribute towards tackling these questions by giving insights into perceptual processes and interactions between processing stages.In this thesis, numerical simulations of spiking neural networks are used to deal with two essential functions that sensory systems have to solve: pattern recognition and prediction.The focus of this thesis lies on the question as to how neural network connectivity can be used in order to achieve these crucial functions.The guiding ideas of the models presented here are grounded in the probabilistic interpretation of neural signals, Hebbian learning principles and connectionist ideas.The main results are divided into four parts.The first part deals with the problem of pattern recognition in a multi-layer network inspired by the early mammalian olfactory system with biophysically detailed neural components.Learning based on Hebbian-Bayesian principles is used to organize the connectivity between and within areas and is demonstrated in behaviorally relevant tasks.Besides recognition of artificial odor patterns, phenomena like concentration invariance, noise robustness, pattern completion and pattern rivalry are investigated.It is demonstrated that learned recurrent cortical connections play a crucial role in achieving pattern recognition and completion.The second part is concerned with the prediction of moving stimuli in the visual system.The problem of motion-extrapolation is studied using different recurrent connectivity patterns.The main result shows that connectivity patterns taking the tuning properties of cells into account can be advantageous for solving the motion-extrapolation problem.The third part focuses on the predictive or anticipatory response to an approaching stimulus.Inspired by experimental observations, particle filtering and spiking neural network frameworks are used to address the question as to how stimulus information is transported within a motion sensitive network.In particular, the question if speed information is required to build up a trajectory dependent anticipatory response is studied by comparing different network connectivities.Our results suggest that in order to achieve a dependency of the anticipatory response to the trajectory length, a connectivity that uses both position and speed information seems necessary.The fourth part combines the self-organization ideas from the first part with motion perception as studied in the second and third parts.There, the learning principles used in the olfactory system model are applied to the problem of motion anticipation in visual perception.Similarly to the third part, different connectivities are studied with respect to their contribution to anticipate an approaching stimulus.The contribution of this thesis lies in the development and simulation of large-scale computational models of spiking neural networks solving prediction and pattern recognition tasks in biophysically plausible frameworks. / <p>QC 20150504</p>

In this thesis a Proactive Risk-Aware Robotic Sensor Network (RSN) is proposed for the application of Critical Infrastructure Protection (CIP). Each robotic member of the RSN is granted a perception of risk by means of a Risk Management Framework (RMF). A fuzzy-risk model is used to extract distress-based risk features and potential intrusion-based risk features for CIP. Detected high-risk events invoke a fuzzy-auction Multi-Robot Task Allocation (MRTA) algorithm to create a response group for each detected risk. Through Evolutionary Multi-Objective (EMO) optimization, a Pareto set of optimal robot configurations for a response group will be generated using the Non-Dominating Sorting Genetic Algorithm II (NSGA-II). The optimization objectives are to maximize sensor coverage of essential spatial regions and minimize the amount of energy exerted by the response group. A set of non-dominated solutions are produced from EMO optimization for a decision maker to select a single response. The RSN response group will re-organize based on the specifications of the selected response.

<p>Sorted bedforms are spatial extensive (100 m-km) features present on many inner continental shelves with subtle bathymetric relief (cm-m) and localized, abrupt variations in grain size (fine sand to coarse sand/gravel). Sorted bedforms provide nursery habitat for fish, are a control on benthic biodiversity, function as sediment reservoirs, and influence nearshore waves and currents. Research suggests these bedforms are a consequence of a sediment sorting feedback as opposed to the more common flow-bathymetry interaction. This dissertation addresses three topics related to sorted bedforms: 1) Modeling the long-term evolution of bedform patterns, 2) Refinement of morphological and sediment transport relations used in the sorted bedform model with `machine learning'; 3) Development of a new sorted bedform model using these new `data-driven' components.</p><p> Chapter 1 focuses on modeling the long term evolution of sorted bedforms. A range of sorted bedform model behaviors is possible in the long term, from pattern persistence to spatial-temporal intermittency. Vertical sorting (a result of pattern maturation processes) causes the burial of coarse material until a critical state of seabed coarseness is reached. This critical state causes a local cessation of the sorting feedback, leading to a self-organized spatially intermittent pattern, a hallmark of observed sorted bedforms. Various patterns emerge when numerical experiments include erosion, deposition, and storm events. </p><p> Modeling of sorted bedforms relies on the parameterization of processes that lack deterministic descriptions. When large datasets exist, machine learning (optimization tools from computer science) can be used to develop parameterizations directly from data. Using genetic programming (a machine learning technique) and large multisetting datasets I develop smooth, physically meaningful predictors for ripple morphology (wavelength, height, and steepness; Chapter 2) and near bed suspended sediment reference concentration under unbroken waves (Chapter 3). The new predictors perform better than existing empirical formulations. </p><p> In Chapter 3, the new components derived from machine learning are integrated into the sorted bedform model to create a `hybrid' model: a novel way to incorporate observational data into a numerical model. Results suggest that the new hybrid model is able to capture dynamics absent from previous models, specifically, the two observed end-member pattern modes of sorted bedforms (i.e., coarse material on updrift bedform flanks or coarse material in bedform troughs). However, caveats exist when data driven components do not have parity with traditional theoretical components of morphodynamic models, and I address the challenges of integrating these disparate pieces and the future of this type of `hybrid' modeling.</p> / Dissertation

In this work, a general dynamic model was proposed to simulate the dynamic motion of microtubules driven by dynein motors, which is of importance to the design of potential nano-bio machines composed of dynein motors and microtubules. The model was developed based on Newton's law of motion. By incorporating a DPD technique, the general model was applied to simulate the unidirectional motion of microtubule. The functions of dyneins and their coordination with each other, which plays an important role in the motion of microtubules, were studied. By taking into account the bending energy of microtubules, we extended the general model to study possible mechanisms responsible for the microtubule-microtubule and microtubule-wall interactions, which are essential to the design of optimal track patterns for potential nanomachine systems. This study helps to evaluate the influence of bending and rotation on microtubule joining processes, involving bumping force, bending moment and torque generation. Finally, a phenomenal modeling study based on the Monte Carlo method, was conducted to investigate the self-organization of microtubules driven by dynein motors and identify out key parameters that control the self-organized movement of microtubules, giving crucial information for nano device design. This modeling study helps to clarify several important issues regarding the interaction between dynein motors and microtubules as a power transfer medium, which provides important information for the development of potential nanobio-machines using dynein as a biological motor.

Form-finding in Architecture looks at processes in nature to discover a more correct way in which to organize building. It is a study into the capability of discovering optimum form, dynamic adaptability, and exposes a set of unique relationships not relevant to Architecture previously. The beauty of these objects does not have to be designed. It is an emergent property of natural form. However, the wonder lies not in aesthetics, but in the manner in which natural forms come into being seemingly without a plan, at a multitude of scales, and in a vast array of materials. Alone, pattern in nature opens a vast array of potentialities for the study into new methods of architectural design. It is important to note that this inquiry will not be into the aesthetics of self-organized pattern, but the mathematical and procedural processes of formation itself. This study forms a set of principles, methodologies and tools for structuring a full-scale form-finding inquiry through the self-organization of pattern in nature. Following this inquiry one should be able to apply the organizational principles of patterning in nature, specifically breakdown patterns, to inform the programmatic design and layout of shopping malls. The rules set forth outline the formation of breakdown patterns, and the ordering of shopping malls. Through the use of parametric modeling software and computer programming language, sets of digital models efficiently explore of the vast number of potential pattern organizations by mimicking their formation in digital space. Through computational scripting, digital models also reveal formation changes due to the adaptation to site, circulatory loads, and spatial distribution, while still maintaining the laws of pattern formation.

The magnetic and structural properties of nano magnet arrays and ferromagnetic thin films are investigated. Circular x-rays are used and extensive use is made in this Thesis of the X-ray Magnetic Circular Dichroism (XMCD) technique. By means of the XMCD magneto-optic sum rules the values of the orbital and spin moments are determined. In the case of the nano magnet arrays studied, the XMCD technique is used in a spatially resolved mode using Photo Electron Emission Microscopy (PEEM) after circular light excitation. The Extended X-ray Absorption Fine Structure (EXAFS) is studied in both the Co K- and L-edges. In situ Co L-edge X-ray XMCD spectroscopy measurements are presented, in combination with spectro-microscopy results, on Co/Pt and Co/Au based nano-dot arrays, of typical dot lateral size 250×100 nm2, on self organized Si0.5Ge0.5. The Co is only a few atomic layers thick. The dot arrays display a high degree of lateral order and the individual dots, in several cases, exhibit a stable magnetic moment at 300 K. It is found possible to characterize the spin reorientation of these dot arrays. For both systems the in- versus out-of-plane orbital moment anisotropy, is not always related with an out-of-plane magnetization and the occurrence of a spin reorientation. By performing Co K-edge EXAFS measurements the local atomic structure around the Co atoms is characterized. The feasibility of a high precision quantitative structural analysis of L-EXAFS is studied on the system Au/Co/Au/W(110). The spin reorientation transition is studied as a function of the Co thickness and Au cap thickness. The L-edge EXAFS indicates that this reorientation is correlated to a lattice expansion in the perpendicular direction. High precision angle dependent XMCD work is performed on a high temperature exchange bias system. Pinned or frozen magnetic moments are studied within an exchange biased NiFe ferromagnet at the NiFe/FeMn, ferromagnet/antiferromagnet interface by XMCD and complemented by x-ray resonant reflectivity experiments, at the Ni, Fe and Mn L-edges. The Mn L-edge XMCD MnSb and of (Ga, Mn)As layers modified by high temperature annealing is studied. For MnSb an enhanced value is obtained versus theoretical calculations. This result can be explained by means of the enhanced surface to volume ratio for the samples studied. For (Ga, Mn)As differences are found in the local environment of the Mn atoms upon annealing.

Le microenvironnement apporte une multitude d'informations aux cellules régulant ainsi leurs fonctions et leur organisation. L'objectif de cette thèse a été de contrôler différents paramètres du microenvironnement cellulaire in vitro afin de moduler l'autorenouvellement et la destinée des cellules souches embryonnaires (CSE).Cette thèse comporte 3 parties. Premièrement, des films de polyélectrolytes multicouches à base de poly(L-lysine) et de hyaluronane ont été utilisés comme substrats modulables. Les propriétés mécaniques ainsi que la chimie des films régulent la proportion de sous-populations de CSE qui reflètent soit le stade masse cellulaire interne, soit le stade épiblaste.Dans un deuxième temps, un équilibre entre l'expression des marqueurs embryonnaires de l'axe proximodistal a été mis en évidence dans la culture de CSE. L'emploi de micro-patrons adhésifs permettant de contrôler la géométrie des colonies a révélé l'importance des contraintes topologiques sur la distribution des cellules exprimant le marqueur proximal Brachyury. Enfin, l'action combinée de BMP2 et de wnt3a mimant l'environnement biochimique du stade tardif de la ligne primitive a permis d'isoler une population pure et très précoce de progéniteurs cardiaques SSEA1+ multipotents. / The microenvironment provides stem cells with numerous pieces of information. Biochemical and mechanical cues synergize to regulate cell function and organization. The aim of this PhD thesis was to control specific microenvironmental parameters to modulate embryonic stem cell (ESC) self-renewal and fate.First, poly(L-lysine) and hyaluronan based polyelectrolyte multilayer films were used as tunable substrates. Both mechanical and chemical properties of the films influenced the balance between ESC subpopulations reflecting different embryonic stages (inner cell mass versus epiblast)Second, a dynamic equilibrium was found between the expression of embryonic proximal and distal markers within ESC culture. The uses of micropatterned substrates to control colony shape uncovered a key role for geometrical constraints in the distribution of Brachyury expression.Last, BMP2 was used together with secreted wnt3a to mimic the late streak stage of the embryo and to trigger the differentiation of pluripotent cells towards the cardiogenic mesoderm. Responsive cells could be sorted out based on SSEA1 expression. This purified population represents the earliest ESC derived multipotent cardiac progenitor population identified to date.

La dynamique et la rhéologie d'une suspension 2D confinée de vésicules (un modèle de RBCs (globules rouges) ) sont étudiées numériquement en utilisant une méthode de Boltzmann sur réseau frontière immergée. Nous analysons d'abord les situations dans lesquelles les vésicules effectuent le mouvement de chenille de char. Des paires de vésicules se placent dans un état d'équilibre avec une distance relative constante et régulée par le confinement. La distance d'équilibre augmente avec l'intervalle entre les parois suivant une relation linéaire. Cependant, aucune distance d'équilibre stable entre deux vésicules en mouvement de tumbling n’est observée. La présence ou l'absence d'une distance d'équilibre entre deux vésicules dicte l'organisation spatio-temporelle de la suspension. L’organisation de la suspension s’accompagne d’assez amples oscillations de la viscosité normalisée variant en fonction de la concentration, tandis que la viscosité effective ne varie pas.Les interactions dans la direction verticale par rapport au plan de cisaillement sont analysées par des simulations en 3D de capsules et des expériences. Nous montrons que dans une suspension confinée de sang, les RBCs s’organisent spontanément en une structure cristalline sous le seul effet de l'interaction hydrodynamique. Il est en outre démontré que lorsque les RBCs sont remplacés par des particules rigides, l'ordre disparait pour laisser place au désordre. Différents ordres cristallins peuvent apparaître selon la concentration et le confinement. La distance intercellulaire de la structure cristalline est une fonction linéaire du confinement. L’ordre apparaît comme une interaction subtile entre la force de portance qui pousse les RBCs des murs vers le centre et l'interaction hydrodynamique dans la verticale du plan d'écoulement de cisaillement. Cette étude introduit un nouveau paradigme dans le domaine des suspensions non-colloïdales diluées où la prévalence des désordres était mise à jour la règle.La répartition des RBCs au niveau d’une bifurcation est abordée dans nos simulations sur ordinateur ainsi que dans des expériences in vitro. Ces études révèlent que la répartition de RBCs dépend fortement du contraste de viscosité entre la viscosité de l'hémoglobine du RBC et le fluide suspendant, tant que l'hématocrite est inférieure à 20%. Pour des dilutions importantes, nos résultats montrent un nouveau phénomène : la branche de faible débit peut recevoir une concentration plus élevé que la branche de haut débit, en opposition à l'effet Zweifach-Fung. Ce phénomène est observé sous confinement modéré et est le résultat d'une structuration particulière de la suspension cellulaire. Nos résultats suggèrent que les différentes propriétés des RBCs doivent être prises en compte et soigneusement analysées afin d'avoir une bonne compréhension de la distribution de RBCs dans la microcirculation et donc de la livraison de l'oxygène dans la microcirculation en général.Enfin, nous réalisons des simulations numériques d'une grande quantité de RBCs, circulant dans un réseau qui est structuré selon un motif en nid d'abeilles. Nos résultats montrent que tant que l'hématocrite est inférieure à 20%, les RBCs dont la membrane est plus rigide présentent un déplacement latéral plus important dans le réseau. En plus, nous découvrons une différence par rapport à la circulation de RBCs dans un tube droit où le débit pour des globules rigides est plus petit. Au contraire, un débit plus important est observé pour les RBCs plus rigides dans le réseau. Enfin, nous présentons la manifestation d'une diffusion longitudinale plus rapide d’une suspension dense de RBCs de faible déformabilité dans le réseau. Nos résultats fournissent des informations intéressantes sur la livraison de RBCs dans le réseau, ce qui pourrait être important non seulement sur la compréhension de la perfusion du sang et le transit de RBC dans la microcirculation, mais aussi sur des applications pratiques. / Dynamics and rheology of a 2D confined suspension of vesicles (a model for RBCs) is studied numerically by using an immersed boundary lattice Boltzmann method (IB-LBM). We pay a special attention to the link between the spatiotemporal organization of the suspension and rheology. We first analyze situations in which vesicles perform tank-treading. The pair of vesicles settles into an equilibrium state with constant relative distance, which is regulated by the confinement. The equilibrium distance increases with the gap between walls following a linear relationship. However, no stable equilibrium distance between two tumbling vesicles is observed. The presence or the lack thereof of an equilibrium distance between two vesicles dictates the spatiotemporal organization of the suspension (order or disorder). Ordering of the suspension is accompanied with quite ample oscillation of normalized viscosity as a function of concentration, while the effective viscosity exhibits plateau. The oscillations amplitude of normalized viscosity is suppressed when disordered pattern prevails.Beside the interactions in the shear plane discussed in 2D framework, the interactions in the vertical direction to the shear plane are also analyzed by 3D simulations of capsules (a model for RBCs) and experiments. We show that in a confined blood suspension RBCs spontaneously organize in a crystalline-like structure under the sole effect of hydrodynamic interaction. It is further shown that when RBCs are substituted by rigid particles order disappears in favor of disorder. Various crystalline orders take place depending on concentration and confinement. The intercellular distance of the crystalline structure is a linear function of confinement. Order appears as a subtle interplay between the lift force that pushes RBCs away from walls towards the center and hydrodynamics interaction in the vertical of shear flow plane. This study introduces a new paradigm in the field of dilute non-colloidal suspensions where the prevalence of disorder was up-to date the rule.The partition of RBCs at the level of bifurcations is addressed in our computer simulations and in vitro experiments, which reveal that the hematocrit partition depends strongly on the viscosity contrast between the viscosities of the RBC hemoglobin and the suspending fluid, as long as hematocrit is less than 20% (which is the normal range in microcirculation). In the extreme hemodilution, our results exhibit a new phenomenon: the low flow rate branch may receive higher hematocrit than the high flow rate branch in opposition to the known Zweifach-Fung effect. This phenomenon is observed under moderate confinement and is the result of a peculiar structuring of the cell suspension. Our findings suggest that the various RBCs properties must be taken into consideration and carefully analyzed in order to have a firm understanding of RBC distribution in microcirculation and thus oxygen delivery in the microcirculation in general.Finally, we carry out numerical simulations of a large number of RBCs flowing in a network that is structured in a honeycomb pattern. Our results reveal that as long as the hematocrit is less than 20% the RBCs with higher membrane rigidity show a larger lateral displacement in the network. Furthermore, we discover a deviation of RBC flux in network from that in straight tube where the more rigid RBCs get the smaller flux. Oppositely, the larger RBC flux is observed for the more rigid RBCs in the network. Finally, we report on the manifestation of a faster longitudinal diffusion of crowded RBCs with smaller deformability in the network. Our results provide interesting information on the RBC delivery in the network, which should be significant not only in the understanding of the blood perfusion and the RBC transit in the microcirculation but also in practical applications such as cell sorting and chemical analysis.

Actuellement, la musicologie de l’improvisation s’appuie pour l’essentiel sur une lecture intentionnelle du projet artistique ; notre thèse propose une approche différente, en regardant l’improvisation comme un processus complexe, comportant une dimension auto-organisée, à l’image des comportements collectifs observés dans les sociétés animales. Ceux-ci résultent de composantes aléatoires, de nombreuses interactions, de logiques d’amplification et de processus non linéaires. Les outils et les concepts mis au point à l’interface physique-biologie pour comprendre ces dynamiques naturelles nous permettent de construire une démarche de modélisation propre à la musicologie, qui décrit les actes de l’improvisateur à partir d’intuitions musicales pour analyser l’interaction entre ce qui participe de savoir pré-construits et de l’intention, d’une part, et ce qui peut être compris comme un réglage de l’aléatoire, d’autre part. Cette démarche de modélisation est mise en oeuvre sur deux transcriptions du pianiste américain Brad Mehldau. Pour la première (Am Zauberberg), la démarche itérative de modélisation est exposée en détail, partant du modèle le plus pauvre jusqu’à la nécessité d’incorporer le geste de la main. Pour la seconde (Bard), cette démarche est étendue à la conception harmonique, chaque voix d’accompagnement étant conçue comme mue d’un mouvement propre au sein de contraintes d’espace donnée par les autres voix. Nous concluons en ouvrant des perspectives de possibles expérimentations inspirées par ces modèles, du côté de l’enseignement de l’improvisation ou de celui du musicien cherchant à incorporer à son jeu une dimension de lâcher-prise qui est au cœur du processus d’improvisation. / Currently, the musicology of improvisation essentially highlights the intentional part of an artistic project ; our thesis starts with a quite distinct approach, looking at improvisation as a complex process, with a self-organization dimension inspired of the way biologists analyse collective behaviors in animal societies. These behaviors are todays perceived as the result of combined statistical processes at the individual scale, with numerous inter-individual interactions, amplifications, and non linear loops. Such an analysis of observed natural phenomena led biologists and physicists to introduce and set-up concepts and tools that we use here to propose a modeling approach adapted to Musicology. We start with musical intuitions to propose a description of the perceptions and actions of the improviser that puts forward the deep interaction between what is made of pre-builded knowledge and intention, on one side, and of « tuning random behaviors », on the other. This modeling approach is carried out on two transcriptions of the american pianist Brad Mehldau. With the first piece (Am Zauberberg), we fully describe the iterative process leading from the poorest model to the need of incorporating hand-gestures. With the second piece (Bard), we dress the question of improvising harmony, each voice being conceived as animated of its own displacement rules, but spatially constrained by the others. We conclude by opening up prospects of experiments inspired by these models, some concerning the teaching of improvisation, others aiming at better understanding the process by which a musician seeks to incorporate to his play more « letting go », at the heart of what improvisation is about.

O presente trabalho propõe um novo modelo de rede neural artificial voltado a aplicações robóticas, em especial a tarefas de natureza espaço-temporal e de horizonte infinito. Este modelo apresenta três características que o tornam único e que foram tomadas como guia para a sua concepção: auto-organização, representação temporal e aprendizado construtivo. O algoritmo de aprendizagem auto-organizada incorpora todos os mecanismos que são básicos para a auto-organização: competição global, cooperação local e auto-amplificação seletiva. A rede neural é suprida com propriedades dinâmicas através de uma memória de curto prazo. A memória de curto prazo é inserida na estrutura da rede por meio de integradores e diferenciadores, os quais são implementados na camada de entrada da rede. Nesta abordagem existe uma evidente separação de papéis: a rede é responsável pela não-linearidade e a memória é responsável pelo tempo. A construção automática da arquitetura da rede neural é realizada de acordo com uma unidade de habituação. A unidade de habituação regula o crescimento e a poda de neurônios. O procedimento de inclusão, adaptação e remoção de conexões sinápticas é realizado conforme o método de aprendizado hebbiano competitivo. Em muitos problemas práticos, como os existentes na área da robótica, a auto-organização, a representação temporal e o aprendizado construtivo são fatores imprescindíveis para o sucesso da tarefa. A grande dificuldade e, ao mesmo tempo, a principal contribuição deste trabalho consiste em integrar tais tecnologias em uma arquitetura de rede neural artificial de maneira eficiente. Estudos de caso foram elaborados para validar e, principalmente, determinar as potencialidades e as limitações do modelo neural proposto. Os cenários abrangeram tarefas simples de classificação de padrões e segmentação temporal. Os resultados preliminares obtidos demonstraram a eficiência do modelo neural proposto frente às arquiteturas conexionistas existentes e foram considerados bastante satisfatórios com relação aos parâmetros avaliados. No texto são apresentados, também, alguns aspectos teóricos das ciências cognitivas, os fundamentos de redes neurais artificiais, o detalhamento de uma ferramenta de simulação robótica, conclusões, limitações e possíveis trabalhos futuros. / The present work proposes a new artificial neural network model suitable for robotic applications, in special to spatiotemporal tasks and infinite horizon tasks. This model has three characteristics which make it unique and are taken as means to guide its conception: self-organization, temporal representation and constructive learning. The algorithm of self-organizing learning incorporates all the mechanisms that are basic to the self-organization: global competition, local cooperation and selective self-amplification. The neural network is supplied with dynamic properties through a short-term memory. The short-term memory is added in the network structure by means of integrators and differentiators, which are implemented in the input layer of the network. In this approach exists an evident separation of roles: the network is responsible for the non-linearity and the memory is responsible for the time. The automatic construction of the neural network architecture is carried out taking into account habituation units. The habituation unit regulates the growing and the pruning of neurons. The procedure of inclusion, adaptation and removal of synaptic connections is carried out in accordance with competitive hebbian learning technique. In many practical problems, as the ones in the robotic area, self-organization, temporal representation and constructive learning are essential factors to the success of the task. The great difficulty and, at the same time, the main contribution of this work consists in the integration of these technologies in a neural network architecture in an efficient way. Some case studies have been elaborated to validate and, mainly, to determine the potentialities and the limitations of the proposed neural model. The experiments comprised simple tasks of pattern classification and temporal segmentation. Preliminary results have shown the good efficiency of the neural model compared to existing connectionist architectures and they have been considered sufficiently satisfactory with regard to the evaluated parameters. This text also presents some theoretical aspects of the cognitive science area, the fundamentals of artificial neural networks, the details of a robotic simulation tool, the conclusions, limitations and possible future works.