Thesis (MScEng)--Stellenbosch University, 2011. / ENGLISH ABSTRACT: Although a numerical model of the mirror support truss of the Southern African Large Telescope (SALT) has already been developed during the design thereof, this thesis focuses on the development of the methods and techniques that would result in a more accurate numerical model of the actual structure that could be used as a basis for a numerical control system. This control system will compensate for de ections in the structure by adjusting the positioning of the individual mirror segments of the primary mirror. The two main components from which the support truss is constructed are the steel nodes, and the struts that connect to them. For this project a smaller, simpler laboratory model was designed and built to have geometrical properties similar to that of the support truss. The methods and techniques that were investigated were carried out on this model. By using numerical design optimisation techniques, improved numerical models of the different strut types were obtained. This was done by performing tests on the struts so that the actual responses of the struts could be obtained. Numerical models of the struts were then created and set up so that they could be optimised using structural optimisation software. Once accurate strut models had been obtained, these strut models were used to construct a numerical model of the assembled structure. No additional optimisation was performed on the assembled structure and tests were done on the physical structure to obtain its responses. These served as validation criteria for the numerical models of the struts. Because of unforeseen deformations of the structure, not all of the measured structural responses could be used. The remaining results showed, however, that the predictive accuracy of the top node displacement of the assembled structure improved to below 1.5%, from over 60%. From these results it was concluded that the accuracy of the entire structure's numerical model could be signi ficantly improved by optimising the individual strut types. / AFRIKAANSE OPSOMMING: Alhoewel daar reeds 'n numeriese model van die spieëlondersteuningsraamwerk van SALT ontwikkel is gedurende die ontwerp daarvan, fokus hierdie tesis op die ontwikkeling van metodes en tegnieke om 'n numeriese model van steeds hoër gehalte van hierdie spesi eke struktuur te verkry wat kan gebruik word as 'n basis vir 'n numeriese beheerstelsel. Hierdie beheerstelsel sal kan kompenseer vir die ondersteuningsraamwerk se vervormings deur om die individuele spieëlsegmente van die primêre spieël se posisionering te verstel. Hierdie stuktuur bestaan uit hoofsaaklik twee komponente, naamlik staalnodusse en die stutte wat aan hulle koppel. Vir hierdie projek is 'n kleiner, eenvoudiger laboratorium-model ontwerp en gebou om geometriese eienskappe soortgelyk aan die van die ondersteuningstruktuur te hê. Die metodes en tegnieke wat ondersoek is, is op hierdie model uitgevoer. Verbeterde numeriese modelle van die verskillende stut-tipes is ontwikkel deur middel van numerieseoptimeringstegnieke. Dit is gedoen deur toetse op die stutte uit te voer sodat hul werklike gedrag bepaal kon word. Numeriese modelle van die stutte is toe geskep en opgestel sodat hulle geoptimeer kon word om dieselfde gedrag as wat gemeet is, te toon. Hierdie geoptimeerde modelle is toe gebruik om numeriese modelle van die toets-struktuur te skep. Geen verdere optimering is op die numeriese model uitgevoer nie en toetse is op die struktuur gedoen om sy werklike gedrag te meet. Data wat deur die toetse verkry is het as validasie kriteria gedien om die akkuraatheid van die numeriese modelle van die stut-tipes te bepaal. Weens die struktuur se onvoorsiene vervorming kon alle gemete struktuurdata nie gebruik word nie. Die oorblywende data het egter getoon dat die akkuraatheid van die finale numeriese modelle van die struktuur verbeter het en dat dit die translasie van die top-node met 'n speling van 1.5% akkuraatheid kon voorspel, teenoor die oorsponlike speling van meer as 60%. Daar is bevind dat die akkuraatheid van die numeriese model van die hele struktuur noemenswaardig verbeter kan word deur die numeriese modelle van die stut-tipes te optimeer.

We develop a model to explain two-way migration of high-skilled individuals between countries that are similar in their economic characteristics. High-skilled migration results from the combination of workers whose abilities are private knowledge, and a production technology that gives incentives to firms for hiring workers of similar ability. In the presence of migration cost, high-skilled workers self-select into the group of migrants. The laissez-faire equilibrium features too much migration, explained by a negative migration externality. We also show that for sufficiently low levels of migration cost the optimal level of migration, while smaller than in the laissez-faire equilibrium, is strictly positive. Finally, we extend our model into different directions to capture stylized facts in the data and show that our baseline results also hold in these more complex modelling environments.

This thesis comprises four essays in the economics of networks and market design. The common thread in all these essays is the presence of complementarities or externalities. Chapter 2 presents a unified model of networks and matching markets. We build on a contribution by Pycia (2012). We show that strong pairwise alignment of agents’ preferences is a necessary and sufficient condition for the existence of strongly stable networks and strongly stable allocations in multilateral matching markets with finite contracts. Strongly stable networks are not necessarily efficient. Although we use a demanding stability concept, strong pairwise alignment allows for complementarities and externalities. In Chapter 3, we generalise the gross substitutes and complements condition introduced by Sun and Yang (2006). Our new condition guarantees the existence of competitive equilibrium in economies with indivisible goods. Competitive equilibrium can be found using an extension of the double-track adjustment process (Sun and Yang, 2009). In this chapter, we also study contract networks (Ostrovsky, 2008). We show that chain-stable contract allocations can exist even in cyclical contractual networks, such as electricity markets, as long as they are appropriately segmented. In Chapter 4, we run a series of experiments to compare the performance of four auctions – first-price, Vickrey, Vickrey-Nearest Rule (Day and Cramton, 2008), and Reference Rule (Erdil and Klemperer, 2010). In our setting, there are two items and three bidders. Two local bidders want an item each, but the global bidder wants both items. We introduce various exposure and package-bidding treatments. We find that the first-price auction always revenue-dominates all the other auctions without any loss in efficiency, strengthening the results of Marszalec (2011). Exposure affects global bidders only in the first-price auction. In other auctions, global bidders often do not take into account the effect of their own bids on their payments. We find no evidence of threshold effects. Finally, in Chapter 5, we develop a new model of online social network formation. In this model, agents belong to many overlapping social groups. We derive analytical solutions for the macroscopic properties of the network, such as the degree distribution. We study the dynamics of homophily – the tendency of individuals to associate with those similar to themselves. We calibrate our model to Facebook data from ten American colleges.

All people experience pain and for some people, acute pain may over time develop into long-term disabling problems. Already at an early stage, it is possible to identify people at risk for long-term problems and psychologically oriented interventions have been shown to successfully prevent future disability. However, not all people are helped by treatment and there is room for improvement. Moreover, subgroups of people suffering from pain, with different profiles of psychological factors have been identified, indicating that people with pain problems differ. The first aim of this dissertation was to improve the understanding of how people differ. The second aim was to use these individual differences and to match people to psychological treatment based on their psychological profile. The third aim was to explore what happens during treatment that might be important for treatment outcome. The findings show that people who belonged to subgroups with elevated levels of psychological factors had less favorable outcomes over time, despite treatment, than people with no elevations. Moreover, people with elevations in several psychological factors had even less favorable outcomes. Psychological treatments aimed at preventing future disability performed well, but using profiles to match people to treatment did not improve outcomes further; people who were matched to a treatment and people who were unmatched had similar outcomes. However, the profiles used for matching were unstable over time and there is need to improve the identification of psychological variables used for treatment matching. Finally, a number of psychological factors were shown to be valuable targets for treatment; if the treatments successfully produced change in people’s thoughts and emotions related to pain the treatment outcomes were better. The findings were summarized in a flow chart showing the recommended clinical approach to people seekinghealth care for acute pain problems.

International Telemetering Conference Proceedings / October 23-26, 2000 / Town & Country Hotel and Conference Center, San Diego, California / Map matching method plays an important role in vehicle location and navigation systems. It employs the information in a digital map to compensate the positioning error. This paper presents a fuzzy-logic-based probabilistic map-matching algorithm used in GPS/INS integrated navigation systems, in which the reliability degree of map matching resolution is given explicitly as the decision basis in selecting matching road segment by utilizing the fuzzy comprehensive judgement. The results of experimental simulations have shown that the system performance gained significant enhancement by introducing this algorithm.

International Telemetering Conference Proceedings / October 25-28, 1999 / Riviera Hotel and Convention Center, Las Vegas, Nevada / In this paper, a devising scheme of Autonomous Vehicle Location and Navigation System is introduced firstly. Then, several key technologies used in the devising scheme are presented, which includes a data fusion method based on extended decentralized kalman filter technology, a map-matching method used to compensate the positioning error, and a digital map data processing method used to realize route planning algorithm. By this time, a sample machine based on the devising scheme introduced in this paper has already been worked out successfully. The availability and the advantages of these technologies have been demonstrated.

Thesis (PhD) -- Stellenbosch University, 2001. / Some digitised pages may appear illegible due to the condition of the original hard copy. / ENGLISH ABSTRACT: A new technique for the design of multimode devices in overmoded waveguide is presented. The technique applies the principle of growth-based design and uses a conceptual functional block representation of the design structure to provide necessary flexibility to the design algorithms. Two growth based design strategies are proposed and evaluated. The first uses a generalized synthesis-oriented scanning technique, and the second uses an evolutionary strategy. The techniques provide reliable solutions to a variety of multimode design problems. In order to facilitate sufficiently fast numerical analysis, novel enhancements of the mode matching technique are developed and the use of surrogate models is investigated. In addition, to allow physical evaluation of the finished devices, original techniques of measuring multimode devices are formulated and utilised. Two practical problems are used to evaluate the performance of the design procedures. The first is the design of overmoded waveguide chokes for microwave heating facilities, and the second is the design of multimode horns for antenna and spatial power combining applications. Various examples of each type of problem are presented with measurements of manufactured solutions. / AFRIKAANSE OPSOMMING: ’n Nuwe tegniek vir die ontwerp van multimodusstelsels binne multimodus golfleier word voorgestel. Die tegniek maak gebruik van die beginsel van groei-georienteerde ontwerp en ontgin ’n konsepsuele funksionele module-voorstelling van die ontwerpstruktuur om die nodige buigsaamheid aan die ontwerpsalgoritmes te verleen. Twee groei-georienteerde ontwerpstrategiee word aangebied en geevalueer. Die eerste is gebasseer op ’n veralgemeende sintese-georienteerde skandeertegniek, en die tweede maak gebruik van ’n evolusie-strategie. Die tegniek verskaf betroubare oplossings vir ’n verskeidenheid van multimodusontwerpsprobleme. Ten einde ’n numeriese analise-tegniek daar te stel wat vinnig genoeg is, word oorspronklike verbeterings van die modal-pas metode ontwikkel en surrogaatmodelle is ook ondersoek. Verder, vir fisiese evaluasie, word oorspronklike meettegnieke vir multimodusstelsels geformuleer en gebruik. Twee praktiese probleme word gebruik om die ontwerpprosedures te evalueer. Die eerste is die ontwerp van multimodus golfleierdrywingsdempers vir mikrogolfverhitting, en die tweede is die ontwerp van multimodus horings vir antenna- en ruimtelike drywingskombineerdertoepasings. Verskeie voorbeelde van elke tipe probleem word gegee met metings van gei'mplementeerde oplossings.

There has been an increasing interest in Semantic Web services (SWS) as a proposed solution to facilitate automatic discovery, composition and deployment of existing syntactic Web services. Successful implementation and wider adoption of SWS by research and industry are, however, profoundly based on the existence of effective and easy to use methods for service semantic description. Unfortunately, Web service semantic annotation is currently performed by manual means. Manual annotation is a difficult, error-prone and time-consuming task and few approaches exist aiming to semi-automate that task. Existing approaches are difficult to use since they require ontology building. Moreover, these approaches employ ineffective matching methods and suffer from the Low Percentage Problem. The latter problem happens when a small number of service elements - in comparison to the total number of elements – are annotated in a given service. This research addresses the Web services annotation problem by developing a semi-automatic annotation approach that allows SWS developers to effectively and easily annotate their syntactic services. The proposed approach does not require application ontologies to model service semantics. Instead, a standard query template is used: This template is filled with data and semantics extracted from WSDL files in order to produce query instances. The input of the annotation approach is the WSDL file of a candidate service and a set of ontologies. The output is an annotated WSDL file. The proposed approach is composed of five phases: (1) Concept extraction; (2) concept filtering and query filling; (3) query execution; (4) results assessment; and (5) SAWSDL annotation. The query execution engine makes use of name-based and structural matching techniques. The name-based matching is carried out by CN-Match which is a novel matching method and tool that is developed and evaluated in this research. The proposed annotation approach is evaluated using a set of existing Web services and ontologies. Precision (P), Recall (R), F-Measure (F) and Percentage of annotated elements are used as evaluation metrics. The evaluation reveals that the proposed approach is effective since - in relation to manual results - accurate and almost complete annotation results are obtained. In addition, high percentage of annotated elements is achieved using the proposed approach because it makes use of effective ontology extension mechanisms.

Adolescence is characterized by systematic and dramatic physical and behavioural changes, the most noticeable physical growth is the rapid increase in stature marked by peak height velocity (PHV). Anecdotally, many people are aware that as youth pass through their adolescent growth spurt there is a perceived period of physical awkwardness; however, there is no scientific agreement as to whether a period of awkwardness associated with the adolescent growth actually exists. Previous research has focused on the development of general motor performance or gross motor coordination. Increases in strength during adolescence may mask the effect of a stage of adolescent awkwardness on general motor performance tasks. To detect adolescent awkwardness it is necessary to measure either performance of skills that specifically do not depend on strength, or body awareness. The purpose of this study was to investigate whether female adolescents’ awareness of their body size and movement was influenced by biological maturation, and whether adolescent awkwardness could be detected in performance of sport specific skills independent of strength. An endpoint matching task was used to measure awareness of foot position in space by measuring the distance (mm) between a reference and matching endpoint position (endpoint matching error (EME)) with eyes either open or closed. The Johnson wall volley and a ball juggling task were used to measure soccer specific skill and coordination. Thirty six female youth soccer players, aged 10-14 years, were recruited. Age at PHV was predicted from measures of age, height, leg length and weight. Three groups were identified: pre-PHV (n=6), PHV (n=5) and post –PHV (n=25). Mean group differences were assessed using ANOVA. It was found that when the endpoint matching task was performed with eyes open there was no significant difference in mean EME between groups (p > 0.05). With eyes closed the PHV group performed worse than the pre-PHV and post-PHV groups. The Post-PHV group significantly (p < 0.05) decreased their EME (22.2 ± 13.9) compared to the PHV group (32.8 ± 17.6) but no difference was found with the Pre-PHV group (27.5 ± 15.7). On the wall volley test the Post-PHV group performed significantly (p < 0.05) better compared to the PHV group but not the Pre-PHV group. The Post-PHV group performed significantly (p < 0.05) better on the ball juggling task compared to both the Pre-PHV and the PHV groups. In conclusion, the results suggest that in the year around PHV body awareness, as measured by EME, may plateau or decrease and that soccer skill performance plateaus. This plateau or decrease in body awareness and soccer skills involving coordination is likely temporary; participants in this study who were > 6 months past PHV had significantly better body awareness and soccer skill performance than those participants who were within 6 months of PHV. The results from the endpoint matching task also indicate that a measure of body size and movement awareness has the potential to be used to measure changes in body awareness during the adolescent growth spurt. The results of this study suggest that teenage clumsiness could exists. However, a definitive study with larger maturity groups followed over time is required to confirm this statement.

This dissertation addresses the problem of inferring scene depth information from a collection of calibrated images taken from different viewpoints via stereo matching. Although it has been heavily investigated for decades, depth from stereo remains a long-standing challenge and popular research topic for several reasons. First of all, in order to be of practical use for many real-time applications such as autonomous driving, accurate depth estimation in real-time is of great importance and one of the core challenges in stereo. Second, for applications such as 3D reconstruction and view synthesis, high-quality depth estimation is crucial to achieve photo realistic results. However, due to the matching ambiguities, accurate dense depth estimates are difficult to achieve. Last but not least, most stereo algorithms rely on identification of corresponding points among images and only work effectively when scenes are Lambertian. For non-Lambertian surfaces, the "brightness constancy" assumption is no longer valid. This dissertation contributes three novel stereo algorithms that are motivated by the specific requirements and limitations imposed by different applications. In addressing high speed depth estimation from images, we present a stereo algorithm that achieves high quality results while maintaining real-time performance. We introduce an adaptive aggregation step in a dynamic-programming framework. Matching costs are aggregated in the vertical direction using a computationally expensive weighting scheme based on color and distance proximity. We utilize the vector processing capability and parallelism in commodity graphics hardware to speed up this process over two orders of magnitude. In addressing high accuracy depth estimation, we present a stereo model that makes use of constraints from points with known depths - the Ground Control Points (GCPs) as referred to in stereo literature. Our formulation explicitly models the influences of GCPs in a Markov Random Field. A novel regularization prior is naturally integrated into a global inference framework in a principled way using the Bayes rule. Our probabilistic framework allows GCPs to be obtained from various modalities and provides a natural way to integrate information from various sensors. In addressing non-Lambertian reflectance, we introduce a new invariant for stereo correspondence which allows completely arbitrary scene reflectance (bidirectional reflectance distribution functions - BRDFs). This invariant can be used to formulate a rank constraint on stereo matching when the scene is observed by several lighting configurations in which only the lighting intensity varies.