Dissertations / Theses on the topic 'NATO Intelligence'

Approved for public release; distribution is unlimited<br>Despite NATOs long history of close military cooperation and coordination, the Alliance cannot fully support major multinational operations. Tested by conflicts in the 1990s through to the recent war in Libya, NATO capabilities have consistently fallen short. During each conflict, intelligence played a key role in securing victory, but NATOs operational successes were made possible only through substantial U.S. intelligence support. The lack of an independent NATO intelligence capability presents a problem for the United States, which needs a NATO capable of conducting operations in the event of a simultaneous conflict in another region. Given the rise of austerity measures in Europe, it is unlikely that European governments will provide additional funding to establish a comprehensive NATO intelligence capability. Therefore, the U.S. should establish an enhanced intelligence-sharing relationship with NATO to offset the paucity of NATO intelligence support and operational resources. Intelligence sharing can increase international cooperation and allow for greater burden sharing among allies. This is most aptly demonstrated by the United Kingdom-U.S.A. (UKUSA) series of signals intelligence agreements developed during World War II, which arguably provided the foundation for the special relationship between the U.S. and the U.K. today.

Approved for public release; distribution is unlimited<br>This thesis examines how to optimize intelligence sharing in a coalition by a thorough literature review and site visits to intelligence sharing organizations in order to establish best practices for multinational intelligence sharing. The newly established NATO SOF Headquarters (NSHQ) in Mons, Belgium was treated as a test case to validate their intelligence sharing procedures and structures in reference to the authors' identified best practices: mutual gains and benefits; trust; direct control; and accessibility and interoperability. Intelligence support to SOF is a decisive factor, when in conventional operations it often is not; therefore intelligence support to SOF is special - NATO SOF is no exception. The level of intelligence support to SOF normally only exists at the national level, due to bureaucratic obstacles, a need to protect sensitive sources and capabilities, and lack of trust. The NSHQ is experimenting with several innovative methods to enhance trust and streamline intelligence capability amongst NATO SOF forces. There are structural and organizational lessons learned from the establishment of the NSHQ that can be applied to future operations and coalitions.

Thesis (M.A. in Security Studies (Combating Terrorism: Policy and Strategy))--Naval Postgraduate School, March 2008.<br>Thesis Advisor(s): Abenheim, Donald. "March 2008." Description based on title screen as viewed on April 28, 2008. Includes bibliographical references (p. 71-77). Also available in print.

This thesis analyzes intelligence support to NATO Special Operations and the training that facilitates this capability and recommends a model to increase competence in the field of intelligence. The training provided by the NSTEP has led NATO SOF intelligence personnel to fill critical positions that have afforded operational elements to execute missions. The successes of ISAF SOF have been attributed to the skilled analysts that have developed clear pictures of the enemy threat and the analysis of raw information. These skill sets are critical to ensure that the commanders guidance is met and the gaps answered for the operational elements. The steps taken by the NSHQ to meet the need for skilled intelligence professionals have been pivotal to the growth of the NSTEP schoolhouses course curriculum. Additionally, increased accesses to ISR assets and interactions with other intelligence agencies have levied added responsibilities on NATO SOF intelligence officers to become more knowledgeable about multi-disciplinary intelligence operations. This requirement forces NATO SOF intelligence officers to become experts in multiple fields. This access enables these officers to seek out additional training to meet these needs. To ensure these requirements are met, does NSHQ need to develop a curriculum that provides management training for intelligence officers to meet the need of overseeing multisource intelligence elements

O presente trabalho de investigação tem como temática a possível integração do sistema de armas P-3C CUP+ ORION, da Força Aérea Portuguesa, no programa multinacional de vigilância terrestre da NATO, o Alliance Ground Surveillance (AGS). O estudo realizado envolve a verificação da existência de requisitos operacionais definidos para os meios aéreos empregues no âmbito do programa AGS, a análise das capacidades Intelligence Surveillance and Reconnaissance (ISR) da plataforma P-3C, a fim de verificar o cumprimento dos referidos requisitos e validar a sua possível integração no sistema NATO AGS como um ISR National Asset. De modo a cumprir o objetivo inicialmente proposto, a análise efetuada permitiu interligar o conceito de capacidade no âmbito do programa AGS, com o conceito de capacidade ISR da plataforma aérea P-3C. Ambos os conceitos são analisados através do conteúdo obtido pelas diversas fontes bibliográficas, entrevistas realizadas e conversas exploratórias, com a finalidade de se retirarem conclusões que permitiram dar resposta às questões de partida e às questões derivadas. Considerando o enquadramento da cooperação internacional a nível da vigilância, concluiu-se que o sistema de armas em apreço pode integrar o programa da NATO, uma vez que possuí sensores de recolha de informação tecnologicamente evoluídos. Por outro lado, foi possível identificar limitações ao nível da transmissão de informação para outros sistemas, o que poderá trazer alguns problemas ou mesmo comprometer a possível integração no NATO AGS. É recomendada uma análise, avaliação e proposta da: participação da aeronave P-3C no programa NATO AGS; melhoria das capacidades ISR e comunicações da plataforma.

Dissertação de Mestrado em Estratégia<br>No ano em que a NATO comemora o seu 70.º aniversário, a ordem internacional demonstra-se cada vez mais volátil. O fim da Guerra Fria fazia prever o "fim da história" de Fukuyama, uma ideia que rapidamente se dirimiu com o início do novo século, designadamente com o pináculo da ameaça terrorista, a 11 de setembro de 2001. Em 2014, o ambiente estratégico internacional foi novamente alterado com a anexação da Crimeia pela Rússia, num momento em que o país constituía um promissor parceiro da Aliança. Atualmente, ameaças híbridas e conflitos assimétricos representam um desafio à segurança do mundo. A partir desta alteração paradigmática, e tendo presente que caso os esforços diplomáticos não sejam bem-sucedidos a NATO tem o poder de conduzir operações de gestão de crises a fim de reestabelecer a paz, os decisores da Aliança enfrentam um momento na História no qual todos os cenários devem ser considerados. Assim, de modo a reforçar o processo de tomada de decisão da NATO, as Informações Estratégicas desempenham um papel fundamental. Em conjunto, os especialistas das Informações Aliados, os Comandos Estratégicos da NATO e a Divisão Conjunta de Informações e Segurança contribuem com informação atempada, relevante e precisa, sobre a qual os Representantes Permanentes se baseiam na formulação de respostas aos desafios e ameaças que possam surgir. Neste contexto, esta Dissertação visa clarificar o quadro de condução das Informações Estratégicas no seio da estrutura da NATO, assim como os desafios que enfrenta para alcançar máximo potencial. Pretende-se ainda elucidar como o processo de tomada de decisão da NATO é reforçado por este instrumento estratégico, apesar dos constrangimentos que avoca.<br>In the year that NATO celebrates its 70th anniversary, international world order is increasingly volatile. If with the end of the Cold War societies anticipated Fukuyama’s “end of History”, the beginning of the new century quickly dismantled such an idea, namely with an increase in the terrorist threat, reflected by the 9/11 attacks. In 2014, the strategic international environment changed, once again, with the annexation of Crimea by Russia, in an unexpected moment, when the country demonstrated itself as a promising Alliance partner. Today, hybrid threats and asymmetric conflicts pose a challenge to world’s security. From this paradigm change, and being aware that if diplomatic efforts fail, NATO has the military power to undertake crisis-management operations in order to reestablish a peaceful environment, NATO decisionmakers face a moment in History when each scenario needs to be considered. Thus, in order to bolster NATO’s decision-making process, Strategic Intelligence plays a key role within the Alliance. Together, allies’ intelligence experts, NATO strategic commands and the Joint Intelligence and Security Division, contribute with timely, relevant and accurate information in which the Permanent Representatives draw upon to formulate their responses, tackling the challenges and threats that may rise. With this milieu, this Dissertation aims to clarify the framework in which Strategic Intelligence is conducted within NATO structure and the challenges it faces to achieve full potential. Furthermore, it is intended to explain how NATO’s decision-making paradigm, with all its constraints, it is strengthened by this strategic tool.<br>N/A

Les technologies de séquençage ont révolutionné la biologie moléculaire en permettant l’analyse directe des séquences d’ADN et de protéines, ce qui est crucial pour : i) approfondir nos connaissances des systèmes biologiques, ii) le diagnostic précoce de maladies, ou iii) le stockage des données biomoléculaires. Dans cette thèse, nous avons exploré pleinement le potentiel des nanopores solides (SSNs) pour la détection de molécules uniques (protéines), en particulier les membranes nanoporeuses 2-D de type MoS2 . Le principe général de détection d’une molécule unique par des SSNs est que, lorsqu’une molécule biologique (chargée) dans une solution ionique traverse le nanopore, celle-ci bloque le flux des ions durant son passage. La mesure ultra-rapide des fluctuations de courant fournit ainsi des informations structurales sur la molécule à partir de l’analyse de l’amplitude des chutes de courant et de leurs durées, interprétées comme des évènements de translocation. Tout d’abord, à l’aide de simulations de dynamique moléculaire classique "tous atomes" en solvant explicite et non biaisées, nous avons analysé la dynamique de translocation des vingt acides aminés qui composent les protéines et identifié leurs empreintes de courant ionique à partir de techniques d’apprentissage machine. Plus précisément, des méthodes d’apprentissage non supervisées ont été utilisées pour regrouper les niveaux de blocage du courant ionique extraits des séries temporelles enregistrées pendant les simulations. Cela nous a permis de discriminer avec précision les acides aminés chargés positivement et négativement des acides aminés neutres. Ces résultats prometteurs ouvrent la voie au séquençage "à gros grains" des protéines basé sur la charge des acides aminés, un nouveau concept introduit pour la première fois dans ce travail. Deuxièmement, en utilisant une procédure de dynamique moléculaire similaire, nous avons réalisé des simulations de translocation pour douze séquences de peptides distinctes faites d’un acide aminé chargé positivement, d’un chargé négativement et de quatre neutres (composition identique). L’objectif était d’encoder des informations dans ces séquences de peptides et d’explorer le potentiel d’utilisation des nanopores solides de type MoS2 pour des applications de stockage de données moléculaires. En utilisant des techniques d’apprentissage machine supervisées cette fois-ci, en particulier des modèles de classification, nous avons identifié les séquences de peptides spécifiques capables de représenter de manière fiable des données binaires. Les principales caractéristiques de translocation, comme la chute du courant ionique et le temps de résidence dans le pore, ont été extraites et analysées pour discriminer les paires de séquences idéales représentant les bits 0 et 1. Cette approche innovante montre la faisabilité de réaliser des systèmes de stockage de données moléculaires denses, robustes et stables à l’aide de séquences de peptides composés de 3 acides aminés différents, ouvrant la voie aux solutions évolutives et durables pour l’encodage d’informations au niveau moléculaire. Enfin, nous avons exploré l’utilisation de méthodes d’apprentissage profond pour prédire la dynamique des acides aminés des peptides et les fluctuations du courant ionique enregistrées pendant leur passage à travers des nanopores de type MoS2 . En utilisant des réseaux de neurones Long Short-Term Memory (LSTM), qui excellent dans la capture de dépendances temporelles complexes dans les données de séries temporelles, nous avons prédit des courants ioniques et en avons déduit les positions des acides aminés à partir des séries temporelles du courant ionique extraites des simulations de dynamique moléculaire. Ces résultats préliminaires concernant l’intégration des techniques d’apprentissage profond pour étudier des séries temporelles enregistrées par nanopores montrent leur potentiel significatif pour analyser des observations expérimentales<br>Sequencing technology has revolutionized molecular biology by enabling the direct analysis of DNA and protein sequences, which is crucial for: i) a better understanding of biological systems, ii) early disease diagnosis, or iii) biomolecular data storage. In the present thesis, we fully explored the potential of solid-state nanopores (SSNs) for single-biological molecule detection, particularly single-layer MoS2 nanoporous membranes, as versatile technologies for protein sequencing applications. The detection principle using SSNs relies on measuring the relatively small variations of ionic current as charged biomolecules immersed in an electrolyte traverse the nanopore, in response to an external voltage applied across the membrane. The passage of a biomolecule through the pore yields information about its structure and chemical properties.First, through extensive unbiased all-atom classical molecular dynamics (MD) simulations in explicit solvent combined with machine learning techniques, we analyzed translocation dynamics of the twenty proteinogenic amino acids and identified their ionic current fingerprints. Specifically, unsupervised learning methods were employed to cluster blockade levels of ionic current extracted from structural break of time series recorded during MD. It enables the accurate discrimination of positively, negatively charged, and neutral amino acids. These findings pave the way to coarse-grained sequencing of proteins based on the amino acid charge, a new concept introduce for the first time in the present work to the best of our knowledge.Second, using similar MD procedure, we performed translocation simulations of twelve distinct peptide sequences made of 1 positively, one negatively and four neutral amino acids (identical composition). The goal was to encode information into peptide sequences and investigate the potential of using MoS2 SSNs for molecular data storage applications. By leveraging machine learning techniques, in particular classification models, we identified specific peptide sequences capable of reliably representing binary data. Key translocation features, such as ionic current drop and dwell time, were extracted and analyzed to discriminate the ideal pairs of sequences representing bits 0 and 1. This innovative approach highlights the feasibility of achieving high-density, robust, and stable molecular data storage systems using peptide sequences, paving the way for scalable and sustainable solutions in molecular-level information encoding.Finally, we explored the use of deep learning methods to forecast the dynamics of peptide residues and ionic current fluctuations during their passage through MoS2 nanopores. By employing Long Short-Term Memory (LSTM) neural networks, which excel at capturing complex temporal dependencies in time-series data, we predicted ionic currents and infer residue positions from the ionic current time series extracted from MD simulations. These preliminary results about the integration of deep learning techniques to study nanopore recorded time series highlights their significant potential in bridging experimental observations and computational predictions. It provides a robust framework for improving the resolution and accuracy of nanopore-based protein sequencing.This research project demonstrates that MoS2 nanopores are promising platforms for both protein sequencing and molecular data storage applications. By combining MD simulations with artificial intelligence techniques, we established a robust framework for the identification of biological molecules (proteins) and for the encoding of information at the molecular level. The findings presented here contribute to the advancement of next-generation nanotechnology applications, including proteomics, diagnostics, and sustainable molecular data storage solutions

L'objectif principal de ces travaux de thèse est de développer des modèles membranaires biomimétiques pour la reconstitution et l'étude d'interactions protéine/membrane. Dans ce but, deux approches sont adoptées : l'une mettant en œuvre une plateforme basée sur des nanoparticules de silice/Au recouvertes de lipides et l'autre comprenant la formation de bicouches lipidiques découplées d'un support solide d'or. Dans la première approche, nous avons synthétisé des particules de silice de taille nanométrique contenant des grains d'or inclus dans la matrice silicique. Ces nanoparticules sont ensuite recouvertes par différents phospholipides. Les propriétés plasmoniques acquises grâce aux grains d'or sont caractérisées puis utilisées pour suivre l'interaction avec les lipides et/ou les protéines. Le suivi de ces interactions est également visualisé par analyse de la mobilité électrophorétique des particules. La deuxième stratégie développée, consiste à assembler un système membranaire sur une surface solide d'or. Dans un premier temps, une couche de calmoduline est liée à la surface de manière stable. Dans un deuxième temps, une bicouche est formée au-dessus de la couche de calmoduline par deux méthodes. La première méthode consiste à ancrer la bicouche directement sur la couche de protéine par un mécanisme faisant intervenir des lipides chélateurs. Alors que dans la deuxième méthode les lipides sont liés à la surface et découplés grâce à l'utilisation d'une surface d'or modifiée par de la cystéamine et à des lipides fonctionnalisés. L'ancrage est assuré par des groupements succinimidyl et le découplage par des polymères de polyéthylène glycol porté sur un même lipide. Dans les deux stratégies, un réservoir sub-membranaire est créé entre la bicouche étanche et le support. Le suivi des constructions moléculaires est réalisé par résonance plasmonique de surface et analyse du retour de fluorescence. De plus le système est implémenté par des électrodes afin d'étudier l'effet d'application de potentiel sur la bicouche. Après caractérisation, le modèle membranaire est validé par la reconstitution de la translocation de la toxine CyaA de Bordetella pertussis. Cette protéine dispose en effet d'un mécanisme d'internalisation singulier qui permet d'explorer tout le potentiel de notre modèle membranaire<br>The main objective of this work is to develop biomimetic membrane models for the reconstitution and study of protein/membrane interaction. Two devices were designed: one operate a nanometric platform composed of phospholipids coated lipid silica/Au nanoparticles, while the other including tethered lipid bilayer reconstitution on a gold surface. The first approach needs the synthesis of nanometer sized gold/silica particles and that are subsequently coated with different phospholipids. The plasmonic properties provided by gold seeds are characterized and they are of utility to follow the interaction between lipids and/or proteins at the surface. Following of these interactions was also realized with electrophoretic mobility analysis. The second biomimetic device involves a membrane assembly on a gold surface. In a first time, a calmodulin layer is bound on the surface. In a second time, a lipid bilayer is assembled above the calmodulin layer by two approaches. In the first approach the lipid bilayer is anchored on the protein layer with chelators lipid and His-Tag bearing by the proteins. While, in the second approach, lipids are bound on the surface and tethered with the use of a cysteamin modified gold surface and functionalized lipids. The anchorage is realized by succinimidyl group and the tethering by polyethylene glycol group wearing by one kind of lipid. A sub-membrane reservoir is created under the lipid bilayer. The biomimetic model formation was followed by plasmonic resonance and fluorescence recovery after photobleaching. After their characterization the tethered model is validated by reconstitution of a particular mechanism: the CyaA toxin from Bordetella pertussis translocation

Beaucoup d’applications en ingénierie demandent l’utilisation de matériaux intelligents qui peuvent se déformer en réponse à un stimulus extérieur. C’est dans ce contexte, que s’est posé ce projet de recherche. Bénéficiant d’un environnement pluridisciplinaire, grâce à l’association de deux axes de l'Institut Pascal : l’axe MMS (Mécanique, Matériaux et Structures) et l’axe PHOTON (Axe Photonique, Ondes, Nanomatériaux), cette thèse s’intègre parfaitement dans l’action transversale "Matériaux et Modélisations multi-échelles" du laboratoire. La première partie de ce travail s'appuie sur un système expérimental mis au point par une équipe américaine [Chang_10] qui permet la mesure sans contact du pH d'une solution en exploitant les caractéristiques photoniques du système. Ce système est composé d'un réseau d'hydrogel fixé sur un substrat rigide. Un modèle numérique est développé dans le but de simuler le fonctionnement de l'ensemble et d'optimiser le réseau d'hydrogel en vue d'applications dans le domaine médical. La seconde partie de ce travail concerne le développement d'une théorie sur le comportement mécanique de polymères sensibles à la lumière. L'objectif est d'établir une relation liant la déformation du matériau à l’intensité lumineuse. Les résultats obtenus sont comparés avec les résultats expérimentaux issus de la littérature. L'influence des interactions entre les molécules d'azobenzènes sur la déformation du matériau est étudiée<br>Many engineering applications involve stimuli-responsive materials that can change their shape under the action of an external stimulus. It is in this context that this project takes place. Thanks to a multidisciplinary environment with the association of two lines of research of the Institut Pascal: the Mechanical area (Mechanic, Materials and structure) and the Photonic area (Nanostructures and Nanophotonics), this PhD perfectly fits with the “Materials and multi-scale Modeling” transversal action of the laboratory. The first part of this work relies on an experimental system developed by an American team [Chang_10] which allows to measure the pH of a solution without contact, making use of its photonic characteristics. This system is composed of a hydrogel network fixed on a rigid substract. A numerical model is developed in order to simulate its behavior and optimize the hydrogel network with a view to applications in the medical domain. The second part of this PhD is related to the development of a theory on the mechanical behavior of photo-sensitive polymers. The aim is to establish a link between the material deformation and the light intensity. The obtained results are compared to experimental ones from literature. The interaction influence of the azobenzenes molecules on the material strain is studied

A new acid-sensitive drug-delivery nanocarrier has been developed for tumour targeting. The self-assembling co-polymer stearoyl–PEG–poly-sulfadimethoxine methacrylate (stearoyl–PEG–polySDM) was prepared to obtain liposomes with responsive behaviour in the physiopathologic pH range. Stearoyl–PEG–polySDM was synthesised using a multi-step procedure that includes pH-sensitive sulfadimethoxine methacrylate polymerisation by AGET-ATRP at the amino terminal side of stearoyl–PEG–NH2.<br>Il progetto di ricerca sviluppato ha visto lo sviluppo di un nuovo sistema nanovettoriale intelligente per uso farmaceutico con composizione e architettura tali da consentire una risposta adeguata qualora vengano a contatto con condizioni fisiopatologiche alterate. E’ noto da studi riportati in letteratura, che tessuti affetti da particolari patologie manifestano alterazioni molto definite. Nello specifico, i tessuti tumorali in genere manifestano variazioni di pH, temperatura, potenziale ossidoriduttivo, corredo enzimatico rispetto ai tessuti sani. È quindi possibile disegnare nano sistemi combinando chimicamente o fisicamente materiali di diversa natura che rispondano con alterazioni morfologiche o superficiali solo quando vengono a contatto con il microambiente specifico per cui sono stati disegnati. Ciò consentirà di ottenere sistemi sito-specifici.

Il problema del riconoscimento ottico di caratteri all’interno delle immagini è ormai studiato da decenni e, nel corso degli anni, sono stati proposti molti algoritmi che mirano a risolvere il problema a vari livelli di “generalità” (cioè con la presenza o meno di vincoli relativi alle caratteristiche dell’immagine da analizzare). Tuttavia, ad oggi, molti tra i software (che si occupano del riconoscimento di caratteri) sviluppati per il robot Nao di Aldebaran Robotics si appoggiano a librerie software di Optical Character Recognition (OCR) già pronte all'uso come, ad esempio, Tesseract. In questa tesi verrà illustrato un approccio alternativo al problema, che mostra come sia stato possibile creare da zero un software che non utilizza librerie di OCR preesistenti ma che, combinando l'utilizzo di alcune Reti Neurali Convoluzionali (addestrate ad hoc) con alcune funzioni basilari per la manipolazione delle immagini di una libreria di Computer Vision largamente utilizzata (quale è OpenCV), è comunque capace di garantire buoni risultati in termini di accuratezza per quanto riguarda la localizzazione del testo all'interno dell'immagine, la successiva segmentazione dello stesso nelle sue componenti più elementari (cioè righe di testo, parole e caratteri), il riconoscimento dei singoli caratteri e, infine, la riaggregazione delle lettere riconosciute in parole.

Le lancement d'une fusée fait appel à de nombreux équipements pyrotechniques chargés de réaliser des fonctions clés comme l'allumage des moteurs, la séparation et la neutralisation des étages, la découpe de la coiffe et la libération de la charge utile, et enfin, l'autodestruction du lanceur en cas de problème. Traditionnellement, les systèmes pyrotechniques utilisés sont basés sur des ordres détoniques transmis et distribués grâce à des lignes et des relais pyrotechniques intégrant donc des explosifs. Ces chaines et systèmes pyrotechniques basés sur des ordres détoniques sont certes robustes et fiables mais présentent l'inconvénient d'être lourds, volumineux, délicats et couteux dans leur mise en œuvre en raison du caractère dangereux inhérent à tout explosif et système les intégrant. Dans ce contexte, Dassault-Aviation, le CNES et le LAAS ont proposé une nouvelle solution technologique de systèmes pyrotechniques basée sur des ordres non plus détoniques mais électriques et codées intégrant des micro systèmes de mise à feu pyrotechniques codés sécurisés et intelligents. Ces derniers, appelé dans le projet µIFI pour Micro Initiateurs à Fonctions Intégrés intègrent dans quelques cm3, outre une adresse numérique permettant de les identifier pour communiquer, un micro initiateur électro-pyrotechnique, des sécurités électriques, une barrière de sécurité mécanique déplaçable par un micromoteur, un stockage local d'énergie et une électronique numérique. Ce travail de thèse a consisté à concevoir, réaliser et caractériser un Micro Initiateur à Fonction Intégré. Après avoir évalué les différentes technologies existantes, nous avons conçu puis fabriqué les différents éléments constitutifs du µIFI permettant de répondre au cahier des charges du spatial. Après fabrication, chacun de ces éléments ont été caractérisés pour valider leur fonctionnalité et leur conformité au cahier des charges. Un important travail s'est porté essentiellement sur deux points. Le premier concerne la conception et la mise en œuvre d'une technologie MEMS innovante pour réaliser une puce d'initiation optimisée à base de nanothermite Al/CuO déposée sur une plateforme en SU-8 qui a permis d'atteindre des performances d'initiation electro-pyrotechnique jusqu'ici jamais atteinte. Le deuxième point concerne le travail important et délicat d'interfaçage mécanique et électrique des différents éléments constitutifs du µIFI : les puces d'initiation MEMS, un micromoteur brushless permettant de sécuriser mécaniquement l'initiateur dans un volume de moins de 0,5mm3, une électronique numérique, une capacité de stockage de l'énergie et les connectiques associées. Ce fut un défi technologique important de par la complexité du système à intégrer dans moins de 3 cm3 et assurant un fonctionnement avec une fiabilité exemplaire. Ce travail de thèse a permis des avancées technologiques certaine et constitue la première étape d'un processus d'innovation très important dans le domaine de la pyrotechnie spatiale.

La tesi è stata incentrata sul gioco «Indovina chi?» per l’identificazione da parte del robot Nao di un personaggio tramite la sua descrizione. In particolare la descrizione avviene tramite domande e risposte L’obiettivo della tesi è la progettazione di un sistema in grado di capire ed elaborare dei dati comunicati usando un sottoinsieme del linguaggio naturale, estrapolarne le informazioni chiave e ottenere un riscontro con informazioni date in precedenza. Si è quindi programmato il robot Nao in modo che sia in grado di giocare una partita di «Indovina chi?» contro un umano comunicando tramite il linguaggio naturale. Sono state implementate regole di estrazione e categorizzazione per la comprensione del testo utilizzando Cogito, una tecnologia brevettata dall'azienda Expert System. In questo modo il robot è in grado di capire le risposte e rispondere alle domande formulate dall'umano mediante il linguaggio naturale. Per il riconoscimento vocale è stata utilizzata l'API di Google e PyAudio per l'utilizzo del microfono. Il programma è stato implementato in Python e i dati dei personaggi sono memorizzati in un database che viene interrogato e modificato dal robot. L'algoritmo del gioco si basa su calcoli probabilistici di vittoria del robot e sulla scelta delle domande da proporre in base alle risposte precedentemente ricevute dall'umano. Le regole semantiche realizzate danno la possibilità al giocatore di formulare frasi utilizzando il linguaggio naturale, inoltre il robot è in grado di distinguere le informazioni che riguardano il personaggio da indovinare senza farsi ingannare. La percentuale di vittoria del robot ottenuta giocando 20 partite è stata del 50%. Il data base è stato sviluppato in modo da poter realizzare un identikit completo di una persona, oltre a quello dei personaggi del gioco. È quindi possibile ampliare il progetto per altri scopi, oltre a quello del gioco, nel campo dell'identificazione.

Les nano-oscillateurs à transfert de spin sont des composants radiofréquences magnétiques non-linéaires, nanométrique, de faible consommation en énergie et accordables en fréquence. Ce sont aussi potentiellement des candidats prometteurs pour l’élaboration de larges réseaux d’oscillateurs couplés. Ces derniers peuvent être utilisés dans les architectures neuromorphiques qui nécessitent des assemblées très denses d’unités de calcul complexes imitant les neurones biologiques et comportant des connexions ajustables entre elles. L’approche neuromorphique permet de pallier aux limitations des ordinateurs actuels et de diminuer leur consommation en énergie. En effet pour résoudre des tâches cognitives telles que la reconnaissance vocale, le cerveau fonctionne bien plus efficacement en terme d’énergie qu’un ordinateur classique. Au vu du grand nombre de neurone dans le cerveau (100 milliards) une puce neuro-inspirée requière des oscillateurs de très petite taille tels que les nano-oscillateurs à transfert de spin. Récemment, une première démonstration de calcul neuromorphique avec un unique nano-oscillateur à transfert de spin a été établie. Cependant, pour aller au-delà, il faut démontrer le calcul neuromorphique avec plusieurs nano-oscillateurs et pouvoir réaliser l’apprentissage. Une difficulté majeure dans l’apprentissage des réseaux de nano-oscillateurs est qu’il faut ajuster le couplage entre eux. Dans cette thèse, en exploitant l'accordabilité en fréquence des nano-oscillateurs magnétiques, nous avons démontré expérimentalement l'apprentissage des nano-oscillateurs couplés pour classifier des voyelles prononcées avec un taux de reconnaissance de 88%. Afin de réaliser cette tache de classification, nous nous sommes inspirés de la synchronisation des taux d’activation des neurones biologiques et nous avons exploité la synchronisation des nano-oscillateurs magnétiques à des stimuli micro-ondes extérieurs. Les taux de reconnaissances observés sont dus aux fortes accordabilités et couplage intermédiaire des nano-oscillateurs utilisés. Enfin, afin de réaliser des taches plus difficiles nécessitant de larges réseaux de neurones, nous avons démontré numériquement qu’un réseau d’une centaine de nano-oscillateurs magnétiques peut être conçu avec les contraintes standards de nano-fabrication<br>Spin-torque nano-oscillators are non-linear, nano-scale, low power consumption, tunable magnetic microwave oscillators which are promising candidates for building large networks of coupled oscillators. Those can be used as building blocks for neuromorphic hardware which requires high-density networks of neuron-like complex processing units coupled by tunable connections. The neuromorphic approach allows to overcome the limitation of nowadays computers and to reduce their energy consumption. Indeed, in order to perform cognitive tasks as voice recognition or image recognition, the brain is much more efficient in terms of energy consumption. Due to the large number of required neurons (100 billions), a neuromorphic chip requires very small oscillators such as spin-torque nano-oscillators to emulate neurons. Recently a first demonstration of neuromorphic computing with a single spin-torque nano-oscillator was established, allowing spoken digit recognition with state of the art performance. However, to realize more complex cognitive tasks, it is still necessary to demonstrate a very important property of neural networks: learning an iterative process through which a neural network can be trained using an initial fraction of the inputs and then adjusting internal parameters to improve its recognition or classification performance. One difficulty is that training networks of coupled nano-oscillators requires tuning the coupling between them. Here, through the high frequency tunability of spin-torque nano-oscillators, we demonstrate experimentally the learning ability of coupled nano-oscillators to classify spoken vowels with a recognition rate of 88%. To realize this classification task, we took inspiration from the synchronization of rhythmic activity of biological neurons and we leveraged the synchronization of spin-torque nano-oscillators to external microwave stimuli. The high experimental recognition rates stem from the weak-coupling regime and the high tunability of spin-torque nano-oscillators. Finally, in order to realize more difficult cognitive tasks requiring large neural networks, we show numerically that arrays of hundreds of spin-torque nano-oscillators can be designed with the constraints of standard nano-fabrication techniques

Hlavným cieľom práce bolo vytvorenie softvérového riešenia založeného na neurónových sieťach, pomocou ktorého bolo možné detegovať človeka a následne ho nasledovať. Tento výsledok bol dosiahnutý splnením jednotlivých bodov zadania tejto práce. V prvej časti práce je popísaný použitý hardvér, softvérové knižnice a rozhrania pre programovanie aplikácií (API), ako aj robotická platforma dodaná skupinou robotiky a umelej inteligencie ústavu automatizácie a meracej techniky Vysokého Učenia Technického v Brne, na ktorej bol výsledný robot postavený. Následne bola spracovaná rešerš viacerých typov neurónových sietí na detekciu osôb. Podrobne boli popísané štyri detektory. Niektoré z nich boli neskôr testované na klasickom počítači alebo na počítači NVIDIA Jetson Nano. V ďalšom kroku bolo vytvorené softvérové riešenie tvorené piatimi programmi, pomocou ktorého bolo dosiahnuté ciele ako rozpoznanie osoby pomocou neurónovej siete ped-100, určenie reálnej vzdialenosti vzhľadom k robotu pomocou monokulárnej kamery a riadenie roboty k úspešnému dosiahnutiu cieľa. Výstupom tejto práce je robotická platforma umožnujúca detekciu a nasledovanie osoby využiteľné v praxi.

A partite dallo studio dell’attuale sistema di riconoscimento facciale presente sul robot NAO, questo lavoro di tesi si propone di ricercare una soluzione alternativa, basata sul riconoscimento delle caratteristiche facciali che possa essere utilizzata per estendere l’attuale sistema di riconoscimento in uso dal robot. Le soluzioni proposte per il rilevamento e successivo riconoscimento dei volti, nel robot NAO, si basano su delle librerie software pronte all’uso. In particolare il modulo di visione AlFacedetect consente di rilevare un volto ed in seguito riconoscerlo, mentre La fase di apprendimento avviene grazie al modulo Learn Face presente nel software Choregraphe, sviluppato dalla Aldebaran Robotics. Al fine di ricercare soluzioni alternative ai problemi esposti, sarà illustrato il software sviluppato e gli strumenti utilizzati in ciascuna delle fasi sopra indicate, nessuna delle quali utilizza i moduli built-in presenti nel NAO. La fase iniziale è consistita nella realizzazione dei datasets, uno per ogni caratteristica facciale considerata, dai quali si sono ricavati i diversi training set necessari per addestrare le reti neurali convoluzionali a classificare le caratteristiche facciali prese in esame. Per la fase di detection dei volti si sono utilizzati algoritmi noti in letteratura e opportunamente implementati nella libreria di Computer Vision OpenCV, adattati dinamicamente al compito da svolgere, ovvero alla caratteristica facciale da individuare. In questo modo si sono potute sfruttare appieno le potenzialità e l’efficienza delle reti neurali convoluzionali nello svolgere compiti di classificazione, riuscendo a garantire, nella fase di riconoscimento un’elevata precisione, in termini di accuratezza. La fase finale è consistita nel porting di quanto fatto sul robot NAO, in modo da effettuare la predizione di una determinata caratteristica facciale a partire da un’immagine acquisita attraverso la sua videocamera.

Dans cette thèse, nous abordons la navigation autonome d'un robot mobile en milieu domiciliaire congestionné. Cette problématique relève du domaine de la navigation parmi les obstacles amovibles. Nous proposons une architecture robotique permettant la navigation parmi des obstacles fixes, amovibles et interactifs. L'objectif du robot est de rejoindre une position, tout en évitant les obstacles fixes, déplacer les obstacles amovibles s'ils gênent le passage ou bien demander à des obstacles interactifs (humain, robots, etc.) de céder le passage.Dans notre première contribution, nous proposons une architecture robotique hiérarchique baptisée VICA (VIcarious Cognitive Architecture), dont le niveau décisionnel est couplé à une architecture cognitive. Nous nous sommes inspiré des travaux sur la simplixité de Alain Berthoz qui décrivent comment le vivant prépare l'action et anticipe les réactions. L'architecture robotique se compose d'un planificateur global permettant la navigation dans un environnement inconnu et d'un planificateur local dédié à la gestion des obstacles.La seconde met en œuvre un planificateur global dont le but est de rapprocher autant que possible le robot de son objectif, en utilisant l’algorithme H* que nous avons développé.La troisième propose un planificateur local pour la gestion des obstacles. La solution proposée consiste à utiliser la simulation multi-agents dans le but d'anticiper le comportement des obstacles.L'implémentation de cette solution est réalisée dans l'architecture VICA développée sous ROS (Robot Operating System). En parallèle, nous avons développé un robot expérimental pour valider nos résultats<br>In this thesis, we address the autonomous navigation of a mobile robot in a congested indoor environment. This problem is related to navigation among movable obstacles (NAMO). We propose a robotic architecture allowing navigation among: fixed, removable and interactive obstacles. The objective of the robot is to reach a position, while avoiding fixed obstacles, to move removable obstacles if they obstruct the path or to ask interactive obstacles (human, robots, etc.) to give way.In our first contribution, we propose a hierarchical robotic architecture named VICA (VIcarious Cognitive Architecture), whose decisional level is coupled to a cognitive architecture. We are inspired by Alain Berthoz's work on simplexity, which describes how living organisms prepare actions and anticipate reactions. The robotic architecture is composed of a global planner allowing navigation in an unknown environment and a local planner dedicated to obstacle management.The second one implements a global planner whose goal is to bring the robot as close as possible to its goal, using the H* algorithm we have developed.The third one proposes a local planner for obstacle management. The proposed solution consists in using multi-agent simulation in order to anticipate the behavior of obstacles.The implementation of this solution is realized in the VICA architecture developed under ROS (Robot Operating System). In parallel, we have developed an experimental robot to validate our results

The main problem being investigated is why there were such divergent views on the appropriate curriculum for ANC education-in-exile from within the ANC, and in the light of this contestation, what happened in reality to curriculum practice at the institutions. The arguments for Academic, Political and Polytechnic Education are contextualized in the curriculum debates of the times, that is, the 20th century international policy discourse, the African curriculum debates and Apartheid Education in South Africa. This study examines how Academic Education, despite the sharp debates, was institutionalised at the SOMAFCO High School. It also analyses the arguments for and various notions of Political and Polytechnic Education as well as what happened to these in practice at the school. The SOMAFCO Primary School went through three phases of curriculum development. The school opened in 1980 under a ‘caretaker’ staff and without a structured curriculum. During the second phase 1980-1982 a progressive curriculum was developed by Barbara and Terry Bell. After the Bells resigned in 1982, a conventional academic curriculum was implemented by Dennis September, the new principal.

博士<br>國立臺北科技大學<br>機電科技研究所<br>99<br>Indoor air quality (IAQ) has recently received extensive attention from various researchers who have removed indoor pollutants using various methods. Indoor contaminants included formaldehyde from wood or other construction materials and ammonia from toilets.Those substances that might affect human health. Titanium oxide (TiO2) nanopowder was spread on stainless steel filters by using electrophoretic deposition (EPD) to form photocatalytic filters. The morphology of the filter- coated TiO2 nanoparticles was observed via scanning electronic microscope (SEM) and film thickness tester. Referring to the experimental data, electrophoretic deposition (EPD) on photocatalytic filters was used. The filters were then placed in an enclosed circulative chamber to study the decomposition of formaldehyde under different heating ventilation air conditioning (HVAC)-related air conditions. At constant temperature (18-28 ℃), relative humidity (40-60 %) and air velocity (0.4-1.4 m/s), the TiO2 photocatalytic degradation rates were 33-42 %. Therefore, in a sealed system, when one or all of the air temperatures, 1RH, and air velocity are increased, the formaldehyde photodegradation rate also increases. This study also analyzed airflow using fluid simulation software (Fluent) to examine simulated flow field distribution under various filter intervals (1, 2.3 and 3.6 cm) in an air-conditioning circulation chamber. Referring to the simulated experimental data, this work placed electrophoretic deposition (EPD) on photocatalytic filters in an enclosed circulative chamber to decompose formaldehyde, allowing photocatalytic filters to exert the greatest decomposition effect. The results show that in the testing environment, six photocatalytic filters at the interval of 2.3 cm achieved photocatalytic oxidation (PCO) efficiency of 29.7 %, higher than the 1 cm interval at 26.4 %. This result reveals that using simulation software for photocatalyst filter intervals decreases filter design flaws, reducing initial installation cost, and raising photocatalyst utilization efficiency to achieve higher photocatalytic efficiency. In the final part of this study, the intelligent nano photocatalyst air cleaner for degrading the ammonia gas was developed, to dispel unpleasant odors step systemtically according to the concentration of ammonia. The results showed the decomposition efficiency of TiO2 photocatalyst was 41.2 % and the power consumption of photocatalyst reactor for degradation of ammonia with intelligent control and normal control were 7.016 and 11.45 kWh, respectively. The intelligent control combines the UV lamps and fan and save energy about 38.72 % more than the normal control. This method can reduce any cost increase and energy consumption caused by various design flaws of reactors.

碩士<br>國立臺灣大學<br>機械工程學研究所<br>94<br>The research proposed a positioning system with nano resolution. Utilize ultrasonic motor as the actuator, and develop orthogonal signal decoding method for LDGI (Laser Diffraction Grating Interferometer) as displacement feedback with high accuracy. Based on FCMAC (Fuzzy Cerebellar Model Articulation Controller) control algorithm for controlling and predicting the variation of ultrasonic motor friction driving, establishing PC-Based positioning integral system in with BCB (Borland C++ Builder) software and National Instrument DAQ (Data Acquisition) Device, to achieve the demand of intelligent nano positioning control. In recent years, due to the developing of nano technologies, traditionally piezoelectric device is regarded as actuator in precision positional devices, and the stroke is limited. Therefore it’s generally combined with another long-stroke stage to accomplish high-resolution and long-stroke purpose. However, accompanied assembling and volumetric error will be generated, and the controlling system will become much complicated. In this research the selecting actuator is ultrasonic motor, with resonant state to drive in long-stroke and high-speed, and bending mode to provide acting as piezoelectric actuator to achieve high-resolution. Therefore the long-stroke and high-resolution demand of nano positioning system is achievable with ultrasonic motor. In positioning system developing, introduce the principle of ultrasonic motor and simulating at first. For the several controlling modes to achieving long-stroke and high-resolution positioning, design experiments to observe and realize the phenomenon while driving. For the driving status of friction variation because of friction driving property, design FCMAC controlling algorithm for each modes respectively, to make each friction variation status could get a suitable response, prediction, and control. For displacement feedback, utilize LDGI and developing method to eliminate error of orthogonal signal, and developed positioning system base on class concepts for extension requirements. Furthermore, DVD pick-up head and triggering device design are utilizing for measurement applications.

博士<br>國立雲林科技大學<br>工業工程博士班<br>95<br>Nano-particles are advanced materials with 1~100 nanometer (nm) grain size. They have been widely applied in photo-catalysis, carbon nanotube, nano-ceramics, fabric fiber, and compound material industries. Using the wet-type mechanical milling process to make nano-particles is a popular method because of its simplicity and the applicability to all classes of materials. As the grain size of nano-particles decrease, the ratio of surface area to volume and hardness of nano-particles increase. The smaller grain size the nano-particle is, the higher performance of nano-particle has. Therefore, the required qualities of the nano-particle are that the mean of nano-particle grain size and the variance of grain size must be kept small. These two output quality criteria result in a two-objective (Min-Min) optimization problem. However, the effects of process parameters on the output quality criteria remain unknown in the nano-particles milling process. In addition, these two quality criteria maybe conflict. If the process is optimized with respect to only one output criterion, it is possible to obtain a result that is against the other output criterion. Therefore, it is necessary to find the optimal milling parameters to simultaneously achieve the multi-objective desired product qualities in the nano-particle milling process. Furthermore, the mechanical milling process is applied in a mass production to make the nano-particles, another urgent problem accompanies with it. That is, some factors may cause the milling process to shift away from the optimal process condition and result in defective nano-particles. Hence, engineers of the nano-particles milling process have to regulate the process parameters back to the optimal condition immediately. This process regulation scheme is widely known as engineering process control (EPC). The regulation scheme is based on adjusting the process using information about its current deviation from the optimal values. However, the know-how rules used to regulate the process parameters are still unknown in the nano-particles milling process. In addition, it is necessary to develop a process parameter regulation system for the nano-particle milling process that can guide engineers to regulate the process parameter and to set the process parameter values. Therefore, developing a process parameter regulation system which can simultaneously meet multi-quality criteria for the nano-particles milling process is another critical issue. In the dissertation, several computational intelligence techniques are integrated and applied to develop the optimization system and to develop the process parameter regulation system for a nano-particles milling process. Two optimization systems are developed for a nano-particle milling process in this dissertation. An integrated method, which combines the parameter design of Taguchi method, response surface method (RSM) and genetic algorithm (GA) is proposed to investigate the effects of process parameters and to find the optimal process parameters. The other optimization method, called integrated multi-population intelligent genetic algorithm (MPIGA), which integrates the concept of vector evaluated genetic algorithm (VEGA), RSM and generalized Pareto-based scale-independent fitness function (GPSIFF) approach, is proposed to find the Pareto-optimal process parameters. Based on the results of the optimal process parameters, a process parameter regulation system for a combined quality criterion is developed by integrating the fuzzy set theory, adaptive resonance theory (ART-2) neural networks, variable precision rough set (VPRS), decision tree C5.0 algorithm and fuzzy rule-based control techniques. Based on the results of the Pareto-optimal process parameters, the other process parameter regulation system which can simultaneously meet multi-quality criteria is also developed by integrating the fuzzy set theory, ART-2 neural networks, VPRS, decision tree C5.0 algorithm, Pareto-optimal solution and fuzzy rule-based control techniques. Both the two process parameter regulation system can be applied to guide engineers to regulate the process parameters and to set the process parameter values when the nano-particles milling process shifts away from the optimal process condition or produces faulty nano-particles. Finally, performances of the developed process parameter regulation system are verified in the dissertation. The results of the optimization system using GA and RSM approach show that it can find optimal process parameters for a nano-particle milling process. The results of optimization system using integrated MPIGA approach show that it can find the Pareto-optimal process parameters for decision makers to determine the optimal parameters to assure a stable process and product qualities in the nano-particle milling process. Because the two output quality criteria of nano-particle milling process are conflicting, it is suggested to apply the multi-objective evolutionary algorithms, MPIGA, to capture a set of Pareto-optimal solutions in the process optimization of the nano-particle milling process. In addition, the verification results for the proposed process parameter regulation systems show that it can improve the nano-particle milling process from current shift condition to the low shift condition for all the 27 test data. These two proposed regulation system indeed can be applied to guide engineers to set the process parameters when the nano-particle milling process is shifted. The results demonstrate that the extracted rules found by VPRS approach and C5.0 algorithm can be integrated together to build the reasoning rule-base to increase the completeness of reasoning rule-base in developing process parameter regulation system for a nano-particle milling process. The verification results also show that the proposed process parameter regulation system with combined quality criterion has 65.7 % improvement in shortening the shift value and the average number of adjustments is 8.7 times. The regulation system with multi-quality criteria has 85.9 % improvement in shortening the shift value and the average number of adjustments is 4.4 times. Thus, the performance of the regulation system with multi-quality criteria is more efficient than the other regulation system with combined quality criterion in shortening the shift value of a nano-particle milling process. When the process parameter regulation system with combined quality criterion is used to set parameters, the system average in the mean of grain size is 127.9 nm and in the variance of grain size is 1.3nm. But, when the process parameter regulation system with multi-quality criteria is used to regulate parameters, the system average in the mean of grain size is 117.9 nm and in the variance of grain size is 1.7nm. Therefore, the regulation system with multi-quality criteria has better improvement quality than the regulation system with combined quality criterion in the mean of grain size.

Thesis (M.A.)--University of Wisconsin--Madison, 1999.<br>Typescript. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 180-191).