Electromagnetic microactuators with permanent magnets have many potential applications such as micro-energy scavengers, microswitches, micromirrors and microfluidics. However, many electromagnetic microactuator designs utilize either external permanent magnet or external coil, which do not allow tight integration to other MEMS components and further miniaturization. Furthermore, all of the available permanent magnet microfabrication technologies have some drawbacks and improvements are required. Thus the integrated monolithic electromagnetic microactuator is investigated in this project. The three main components of the electromagnetic actuator have been investigated separately. A novel microfabrication technology called ???Template printing???for the fabrication of polymer bonded permanent magnet has been investigated and developed. It is based on ???Screen printing??? which has its drawbacks on alignment accuracy and poor line definition. This is eliminated in ???Template printing??? by photolithography of the photoresist template. The shape and location of the permanent magnet is defined by the template. A new approach based on the filling of dry magnetic powder and vacuum impregnation has been developed to form the polymer bonded permanent magnet. This allows the use of short pot-life matrix material and the elimination of homogenous mixing. A monolithic electromagnetic microactuator has been fabricated successfully. It consists of a 2-layer planar copper microcoil, surface micromachined polyimide beam and Strontium ferrite/EPOFIX permanent magnet (diameter of 460 ??m and 30 ??m thickness). Large deflection in excess of 100 ??m at 35 mA driving current and magnetic force of 0.39 ??N/mA have been achieved. It compares favourably with other much larger electromagnetic actuators that have been reported. ???Template printing??? has the potential of being a low temperature batch process for the microfabrication of thick polymer bonded permanent magnets with high magnetic properties and low residual stress. The fabrication consistency and the quality of template printed magnet can be improved in future studies.

Gac (Momordica Cochinchinensis Spreng or Muricia Cochinchinensis Spreng) is an indigenous fruit of Vietnam, commonly used as a colourant in traditional dishes. Recent studies showed that this fruit had a high carotenoid content, especially lycopene and β-carotene. These carotenoids are currently in high demand as they are natural antioxidants to prevent and treat cancers. Over the last ten years some new products from Gac fruit have been found on the market but the cultivation and consumption of the fruit is still limited. Gac is harvested seasonally and the post harvest processing has not been well developed. To make the best use of the valuable substances in Gac, this project investigated some processing methods to make Gac powder and developed the product quality analyses, potentially applicable to local small-scale manufacture. In order to evaluate the efficiency of the Gac powder process and some of the most important qualities of the Gac powder products, the analytical methods to determine lycopene and β-carotene were developed and validated in this project. The new HPLC-based method included a liquid-liquid extraction, followed by a solid phase extraction before the analysis. Both newly- developed HPLC and SP methods exhibited a high extraction yield, good recovery and reproducibility. Compared with the HPLC, the SP method gives higher mean values and standard deviations, but it is quicker, simpler, less expensive and still reliable for industrial applications due to the use of less toxic solvents. The production of powder from Gac fruit included two main stages: seed removal and aril dehydration. Derived from the same drying operation, the enzymatic-treated Gac powder had a lower carotenoid content compared to the heat-treated powder. The Gac powder produced by the freeze-drying method had the highest carotenoid level and brightest colour. Finally the orange-red Gac powder with high carotenoid content was added to some foods as a colorant and nutrient supplement to illustrate its applications. The colours of foods were significantly improved and the carotenoid contents were increased after processing, indicating a potential market for the Gac powder. / Master of Science (Hons)

Image analysis, and its application to sensory input (computer vision) is a fairly mature field, so it is surprising that its techniques are not extensively used in robotic applications. The reason for this is that, traditionally, robots have been used in controlled environments where sophisticated computer vision was not necessary, for example in car manufacturing. As the field of robotics has moved toward providing general purpose robots that must function in the real world, it has become necessary that the robots be provided with robust sensors capable of understanding the complex world around them. However, when researchers apply techniques previously studied in image analysis literature to the field of robotics, several difficult problems emerge. In this thesis we examine four reasons why it is difficult to apply work in image analysis directly to real-time, general purpose computer vision applications. These are: improvement in the computational complexity of image analysis algorithms, robustness to dynamic and unpredictable visual conditions, independence from domain specific knowledge in object recognition and the development of debugging facilities. This thesis examines each of these areas making several innovative contributions in each area. We argue that, although each area is distinct, improvement must be made in all four areas before vision will be utilised as the primary sensory input for mobile, autonomous robotic applications. In the first area, the computational complexity of image analysis algorithms, we note the dependence of a large number of high-level processing routines on a small number of low-level algorithms. Therefore, improvement to a small set of highly utilised algorithms will yield benefits in a large number of applications. In this thesis we examine the common tasks of image segmentation, edge and straight line detection and vectorisation. In the second area, robustness to dynamic and unpredictable conditions, we examine how vision systems can be made more tolerant to changes of illumination in the visual scene. We examine the classical image segmentation task and present a method for illumination independence that builds on our work from the first area. The third area is the reliance on domain-specific knowledge in object recognition. Many current systems depend on a large amount of hard-coded domainspecific knowledge to understand the world around them. This makes the system hard to modify, even for slight changes in the environment, and very difficult to apply in a different context entirely. We present an XML-based language, the XML Object Definition (XOD) language, as a solution to this problem. The language is largely descriptive instead of imperative so, instead of describing how to locate objects within each image, the developer simply describes the properties of the objects. The final area is the development of support tools. Vision system programming is extremely difficult because large amounts of data are handled at a very fast rate. If the system is running on an embedded device (such as a robot) then locating defects in the code is a time consuming and frustrating task. Many development-support applications are available for specific applications. We present a general purpose development-support tool for embedded, real-time vision systems. The primary case study for this research is that of Robotic soccer, in the international RoboCup Four-Legged league. We utilise all of the research of this thesis to provide the first illumination-independent object recognition system for RoboCup. Furthermore we illustrate the flexibility of our system by applying it to several other tasks and to marked changes in the visual environment for RoboCup itself.

Extensive brackishwater aquaculture, which is a dominant land-based aquaculture system in Indonesia, has experienced variable success in most farming locations in the country due to poor understanding of spatial assessment of environmental factors and rudimentary site selection criteria. Despite tremendous potential, the application of GIS and remote sensing in spatial assessment has tended to focus on Boolean (Crisp) logic that is often unable to effectively handle the complexity and spatial variability of key environmental factors for the development of aquaculture. This study explored the possibility of integrating fuzzy logic techniques into GIS and remote sensing technology to generate more robust mapping protocols in aquaculture, compensating for the disadvantages of the Crisp method. Two models were developed in two different provinces in Indonesia to spatially assess soil and hydrological constraints on extensive brackishwater aquaculture. The soil assessment focussed on acid sulfate soils (ASS) and sandy-textured sediments in Aceh, and the hydrological study focused on investigating important wave parameters that influence the suitability of coastal areas for siting extensive pond units in South Sulawesi. The study showed that fuzzy-based classification methods, integrated into the image analysis, was highly effective in identifying existing and potential pond areas for extensive brackishwater aquaculture compared to the best result of the commonly used Crisp method. By addition of one or more key environmental variables of ASS into the fuzzy-classified existing and potential ponds areas, a very robust predictive tool to identify potential ponds areas affected by ASS in Kembang Tanjung, Aceh was developed. A more detailed assessment of ASS developed in this study also successfully highlighted the severity of sandy-soils and identified them as another key soil variable that has and will severely impact on pond productivity. The second model developed by the study enables fuzzy logic to be integrated into GIS to predict the possible areas impacted by moderate to high energy wave conditions and possible ways of minimising their direct and indirect impacts. The models developed in this study were shown to work well in both study sites and can be applied elsewhere. The mapping outputs are easy to interpret even by stakeholders with no prior training in map reading. Overall, the models have the potential to reduce planning errors and to improve decision making in aquaculture provided that quality data sources are used.

In this document, I explore the problem of scheduling pipelined applications onto large-scale distributed platforms, in order to optimize several criteria. A particular attention is given to throughput maximization (i.e., the number of data sets that can be processed every time unit), latency minimization (i.e., the time required to process one data set entirely), and failure probability minimization. First, I accurately define the models and the scheduling problems, and exhibit surprising results, such as the difficulty to compute the optimal throughput and/or latency that can be obtained given a mapping. In particular, I detail the importance of the communication models, which induce quite different levels of difficulty. Second, I give an overview of complexity results for various cases, both for mono-criterion and for bi-criteria optimization problems. I illustrate the impact of the models on the problem complexity. Finally, I show some extensions of this work to different applicative contexts and to dynamic platforms.

Le cadre général de ce cette thèse est l'analyse non linéaire dans les espaces de Banach réels associée à la géométrie de ces espaces. Ce travail est composé de deux parties. Dans la première partie on s'intéresse principalement aux applications uniformément continues entre espaces de Banach de dimension infinie et à des résultats d'approximation et d'extension de telles applications. La seconde partie aborde la structure asymptotique des espaces de Banach de dimension infinie puis certaines propriétés de régularité des polynômes entre ces espaces en liaison avec cette structure.

Cette these est dedie a l'etude dynamique des applications rationnelles des espaces projectifs. Elle se decompose en 5 chapitres. Le premier traite de problemes locaux et presente la classification analytique et formelle d'une classe de germes super-attractifs. Dans la deuxieme partie est demontre que le courant de Green d'une application rationnelle quelconque n'a de singularites importantes qu'aux points d'indetermination de l'application. Dans la troisieme partie le cas particulier des applications sur les espaces multi-projectifs en dimension deux est traite. Dans la quatrieme partie, on demontre un theoreme optimal pour la convergence des preimages d'une courbe vers le courant de Green, ce dans le cas des applications birationnelles du plan. Enfin la derniere partie est dedie a l'etude de quelques exemples.

Les prothèses métalliques recouvertes ou non de biomatériaux sont largement utilisées en chirurgie orthopédique. Le devenir de ces implants a été étudié et en particulier la dégradation qui peut induire le transfert d'éléments métalliques dans les tissus adjacents. L'hydroxyapatite synthétique, utilisée en tant que matériau de comblement, se résorbe plus vite lorsqu'elle est dopée en zinc que l'HAP pure ou dopée au manganèse. Les causes de cette amélioration sont d'origine biologique et physico­chimique. Les analyses par activation neutronique, spectrométrie d'émission à source de plasma (ICP/AES), diffraction de rayons X et spectrométrie infrarouge ont été effectuée et ont montré que ces hydroxyapatites avaient des compositions et une structure semblables à celles des apatites pures. La localisation du zinc dans la structure cristalline de l'HAP a été déterminée par EXAFS (Extended X­ray Fine Structure). L'association de ces résultats montre que l'addition du zinc dans l'apatite entraîne une déformation locale de la structure ce qui pourrait permettre un relargage éventuel de cet élément dans l'organisme et favoriserait ainsi la résorption de l'hydroxyapatite. Malgré les progrès réalisés, tout implant métallique mis en place dans un corps humain est le siège d'une corrosion ou d'échange avec le tissu vivant pouvant entraîner à terme un descellement de prothèse. Les matériaux de fixation utilisés ne permettent pas d'arrêter la migration des éléments métalliques vers les tissus, ce qui a pour conséquence une contamination des tissus adjacents. Les tissus post­mortem et per­opératoires prélevés autour d'implants sont étudiés par la méthode d'analyse PIXE (Particle Induced X­ray Emission) afin de localiser et d'évaluer qualitativement et quantitativement la contamination à l'aide d'une méthode d'analyse que nous avons développée. Parallèlement, la microscopie électronique à transmission (TEM) a montré la forme sous laquelle ces éléments ont été relargués.

Nous donnons une condition nécessaire et suffisante pour que les coefficients de Taylor à l'origine de séries en plusieurs variables $q_i({mathbf z})=z_iexp(G_i({mathbf z})/F({mathbf z}))$ soient entiers, avec ${mathbf z}=(z_1,dots,z_d)$ et où $F({mathbf z})$ et $G_i({mathbf z})+log(z_i)F({mathbf z})$, $i=1,dots,d$, sont des solutions particulières de certains $A$-systèmes d'équations différentielles linéaires. Ce critère est basé sur les propriétés analytiques de l'application de Landau (classiquement associée aux suites de quotients de factorielles de formes linéaires). Pour démontrer ce critère, nous généralisons entre autres une version en plusieurs variables d'un théorème de Dwork concernant les congruences formelles entre séries formelles, démontrée par Krattenthaler et Rivoal dans og Multivariate $p$-adic formal congruences and integrality of Taylor coefficients of mirror maps fg [arXiv:0804.3049v3, math.NT]. Ce critère en plusieurs variables implique l'intégralité des coefficients de Taylor de nouvelles applications miroir d'une seule variable dans og Tables of Calabi--Yau equations fg [arXiv:math/0507430v2, math.AG] de Almkvist, van Enckevort, van Straten et Zudilin. Dans le cas particulier d'une variable, nous affinons notre critère et démontrons l'intégralité des coefficients de Taylor de racines d'applications miroir. Cela nous permet de démontrer une conjecture de Zhou énoncée dans og Integrality properties of variations of Mahler measures fg [arXiv:1006.2428v1 math.AG].

Une série de résultats fructueux ont été obtenus au cours de notre étude sur la fabrication de supraconducteur YBCO monodomaine à partir d'un procédé par fusion de poudre (PMP). Primo, grâce à une balance de Faraday à haute température, nous avons révélé les transitions de phase dynamiques pendant la croissance de monodomaine YBCO par le procédé PMP, ce qui est un travail de recherche original dans ce domaine. Ces résultats sur les transitions de phase ont contribué à une meilleure compréhension de la fabrication des monodomaine PMP-YBCO. Secundo, nous avons étudié la dynamique de croissance des monodomaines PMP-YBCO. L'effet de la diffusion de l'oxygène sur le taux de croissance des monodomaines YBCO est démontré. De la relation entre les taux de croissance et l'étude de la surfusion, nous déduisons que la croissance de cristaux d'YBCO dépend de deux types de diffusion. L'une a trait au transport des ions yttrium, baryum et cuivre vers l'interface de croissance de la phase 123 à travers la phase liquide fondue par diffusion en solution et interfaciale comme dans un système liquide fondu-solide. L'autre concerne la diffusion de l'oxygène qui repose sur la diffusion interfaciale à partir de l'air comme dans un système vapeur-solide. Enfin, nous avons été les premiers à fabriquer par lot avec succès des monodomaines YBCO par procédé PMP de taille Ф30mm × 15mm. Un champ magnétique important de 739 mT a été piégé par un monodomaine YBCO à parois minces de Φ15mm. Une mesure magnétique record de la densité de courant critique Jcm ~ 1,2 × 105A/cm² (77K) a été atteinte. Les résultats obtenus nous donnent confiance pour fabriquer des échantillons YBCO monodomaine de haute qualité.