<p dir="ltr">The increasing prevalence of decentralized data in real-world applications, characterized by natural distributions and clustered patterns, has amplified the need for distributed machine learning techniques. This thesis delves into the challenges and opportunities of developing innovative distributed solutions, focusing on real-world applications in social network scenarios, such as student modeling and natural language processing.</p><p><br></p><p dir="ltr">This thesis first investigates the limitations of traditional centralized machine learning, which often generalizes poorly to underrepresented patterns due to biases in data availability. Motivated by Federated Learning (FL), we propose a personalized distributed machine learning framework to model student behavior and generate fair predictions based on distinct demographic attributes, addressing biases inherent in education data. Next, we tackle communication constraints--a fundamental challenge in distributed systems, particularly when applying large language models for natural language tasks. We focus on federated multilingual translation, developing an efficient communication method that selectively exchanges only crucial parameters during FL rounds, significantly reducing overhead while maintaining performance.</p><p><br></p><p dir="ltr">Building upon these applications, we explore broader challenges in FL -- model initialization and reliability -- which remain under-explored. Unlike centralized approaches, FL often suffers from limited performance gains and imbalanced predictions across distributed participants when initialized with pre-trained models, raising fairness concerns. To address this, we propose a meta-learning-based pre-training strategy that delivers robust and fair initialization across heterogeneous FL tasks. Finally, we examine model calibration in FL, an essential aspect for confidence alignment and decision reliability. We introduce a similarity-based calibration method that seamlessly adapts to diverse FL algorithms, enhancing both performance and model calibration.</p><p><br></p><p dir="ltr">In conclusion, this thesis highlights the importance of fairness, efficiency, and reliability in distributed machine learning, offering novel solutions to key challenges and advancing the applicability of distributed machine learning across various domains.</p>

Motion planning for robot operations is concerned with path planning and trajectory generation. In multi-robot systems, i.e. with multiple robots operating simultaneously in a shared workspace, the motion planning also needs to coordinate the robots' motions to avoid collisions between them. The multi-robot coordination decides the cycle-time for the planned paths and trajectories since it determines to which extend the operations can take place simultaneously without colliding. To obtain the quickest cycle-time, there needs to bean optimal balance between, on the one hand short paths and fast trajectories, and on the other hand possibly longer paths and slower trajectories to allow that the operations take place simultaneously in the shared workspace. Due to the inter-dependencies, it becomes necessary to consider the path planning, trajectory generation and multi-robot coordination together as one optimisation problem in order to find this optimal balance.This thesis focusses on optimising the motion planning for multi-robot material handling systems of sheet metal parts. A methodology to model the relevant aspects of this motion planning problem together as one multi-disciplinary optimisation problem for Simulation based Optimisation (SBO) is proposed. The identified relevant aspects include path planning,trajectory generation, multi-robot coordination, collision-avoidance, motion smoothness, end-effectors' holding force, cycle-time, robot wear, energy efficiency, part deformations, induced stresses in the part, and end-effectors' design. The cycle-time is not always the (only) objective since it is sometimes equally/more important to minimise robot wear, energy consumption, and/or part deformations. Different scenarios for these other objectives are therefore also investigated. Specialised single- and multi-objective algorithms are proposed for optimising the motion planning of these multi-robot systems. This thesis also investigates how to optimise the velocity and acceleration profiles of the coordinated trajectories for multi-robot material handling of sheet metal parts. Another modelling methodology is proposed that is based on a novel mathematical model that parametrises the velocity and acceleration profiles of the trajectories, while including the relevant aspects of the motion planning problem excluding the path planning since the paths are now predefined.This enables generating optimised trajectories that have tailored velocity and acceleration profiles for the specific material handling operations in order to minimise the cycle-time,energy consumption, or deformations of the handled parts.The proposed methodologies are evaluated in different scenarios. This is done for real world industrial case studies that consider the multi-robot material handling of a multi-stage tandem sheet metal press line, which is used in the automotive industry to produce the cars' body panels. The optimisation results show that significant improvements can be obtained compared to the current industrial practice.

The greening of air transport is the driver for developing technologies to reduce the environmental impact of aviation with the aim of halving the amount of carbon dioxide (COଶ) emitted by air transport, cutting specific emissions of nitrogen oxides (NO୶) by 80% and halving perceived noise by the year 2020. Fuel Cells (FC) play an important role in the new power generation field as inherently clean, efficient and reliable source of power especially when comparing with the traditional fossil-fuel based technologies. The project investigates the feasibility of using an electric hybrid system consisting of a fuel cell and battery to power a small model aircraft (PiperCub J3). In order to meet the desired power requirements at different phases of flight efficiently, a simulation model of the complete system was first developed, consisting of a Proton Exchange Membrane hybrid fuel cell system, 6DoF aircraft model and neural network based controller. The system was then integrated in one simulation environment to run in real-time and finally was also tested in hardware-in-the-loop with real-time control. The control strategy developed is based on a neural network model identification technique; specifically Model Reference Control (MRC), since neural network is well suited to nonlinear systems. To meet the power demands at different phases of flight, the controller controls the battery current and rate of charging/discharging. Three case studies were used to validate and assess the performance of the hybrid system: battery fully charged (high SOC), worst case scenario and taking into account the external factors such as wind speeds and wind direction. In addition, the performance of the Artificial Neural Network Controller was compared to that of a Fuzzy Logic controller. In all cases the fuel cell act as the main power source for the PiperCub J3 aircraft. The tests were carried-out in both simulation and hardware-in-the-loop.

The greening of air transport is the driver for developing technologies to reduce the environmental impact of aviation with the aim of halving the amount of carbon dioxide (COଶ) emitted by air transport, cutting specific emissions of nitrogen oxides (NO୶) by 80% and halving perceived noise by the year 2020. Fuel Cells (FC) play an important role in the new power generation field as inherently clean, efficient and reliable source of power especially when comparing with the traditional fossil-fuel based technologies. The project investigates the feasibility of using an electric hybrid system consisting of a fuel cell and battery to power a small model aircraft (PiperCub J3). In order to meet the desired power requirements at different phases of flight efficiently, a simulation model of the complete system was first developed, consisting of a Proton Exchange Membrane hybrid fuel cell system, 6DoF aircraft model and neural network based controller. The system was then integrated in one simulation environment to run in real-time and finally was also tested in hardware-in-the-loop with real-time control. The control strategy developed is based on a neural network model identification technique; specifically Model Reference Control (MRC), since neural network is well suited to nonlinear systems. To meet the power demands at different phases of flight, the controller controls the battery current and rate of charging/discharging. Three case studies were used to validate and assess the performance of the hybrid system: battery fully charged (high SOC), worst case scenario and taking into account the external factors such as wind speeds and wind direction. In addition, the performance of the Artificial Neural Network Controller was compared to that of a Fuzzy Logic controller. In all cases the fuel cell act as the main power source for the PiperCub J3 aircraft. The tests were carried-out in both simulation and hardware-in-the-loop.

Depuis de nombreuses années, nous nous sommes efforcés de comprendre le comportement humain et nos interactions avec l'environnement sociotechnique. Grâce à l'avancée de nos connaissances dans ce domaine, nous avons contribué à la conception de technologies et de processus de travail nouveaux ou améliorés. Historiquement, une part importante du travail d'analyse des interactions sociales fut entreprise au sein des sciences sociales. Cependant, la simulation informatique a apporté un nouvel outil pour tenter de comprendre et de modéliser les comportements humains. En utilisant une approche à base d'agents, cette présentation décrit mon travail sur la construction de modèles informatiques du comportement humain pour guider la conception par la simulation. A l'aide d'exemples issus de projets des deux domaines d'application que sont la gestion des crises et de l'urgence et la gestion de l'énergie, je décris comment mon travail aborde certains problèmes centraux à la simulation sociale à base d'agents. Le premier concerne le processus par lequel nous développons ces modèles. Le second problème provient de la nature des systèmes sociotechniques. Les sociétés humaines constituent un exemple parfait de système complexe possédant des caractéristiques d'auto-organisation et d'adaptabilité, et affichant des phénomènes émergents tels que la coopération et la robustesse. Je décris comment la théorie des systèmes complexes peut être appliquée pour améliorer notre compréhension des systèmes sociotechniques, et comment nos interactions au niveau microscopique mènent à l'émergence d'une conscience mutuelle pour la résolution de problèmes. A partir de systèmes de simulation à base d'agents, je montre comment la conscience du contexte peut être modélisée. En terme de perspectives, j'expliquerai comment la hausse de la prévalence des agents artificiels dans notre société nous forcera à considérer de nouveaux types d'interactions et de comportements coopératifs.

Robotics and autonomous systems (RAS) have great potential to propel the world to future growth. Electroadhesion is a promising and potentially revolutionising material handling technology for manufacturing automation applications. There is, however, a lack of an in-depth understanding of this electrostatic adhesion phenomenon based on a confident electroadhesive pad design, manufacture, and testing platform and procedure. This Ph.D. research endeavours to obtain a more comprehensive understanding of electroadhesion based on an extensive literature review, theoretical modelling, electrostatic simulation, and experimental validation based on a repeatable pad design, manufacture, and testing platform and procedure.

O elevado consumo energético de edificações para o seu condicionamento térmico já é um dos maiores impactos ambientais da construção civil nos países desenvolvidos e, progressivamente, vem se tornando uma preocupação também nos países em desenvolvimento. Uma alternativa renovável relativamente recente, tanto para aquecimento como para resfriamento, são as bombas de calor geotérmicas (ground source heat pumps, GSHPs), que se valem da inércia térmica do solo para reduzir significativamente a demanda energética em relação ao ar. Apesar do seu grande potencial em reduzir o consumo de energia e as emissões de carbono, esta tecnologia ainda é quase inexistente na América Latina, não somente devido à falta de conhecimento ao seu respeito, mas também pela incerteza quanto à sua viabilidade frente a um custo inicial mais elevado. Embora existam muitas pesquisas sobre o funcionamento de GSHPs, grande parte do conhecimento prático quanto ao projeto, desempenho e viabilidade é baseada na experiência dos primeiros países onde a tecnologia começou a se tornar popular, e não pode ser assumida automaticamente como válida em outras situações sem uma avaliação prévia Estudos anteriores apontaram que o retorno econômico de GSHPs tende a ser maior em climas mais frios, enquanto a eficiência tende a ser maior em climas mais amenos, mas com sazonalidade. Por apresentar um bom balanço destas características, o Estado do Rio Grande do Sul apresenta potencial para um desempenho elevado de GSHPs. O objetivo deste trabalho é estimar os níveis de eficiência que esta tecnologia pode atingir em uma cidade de referência no Estado, assim como seu retorno econômico frente a ASHPs (air source heat pumps, ou seja, aparelhos de ar condicionado, ou A/C), seus limitantes e as dificuldades de projeto. Uma edificação unifamiliar residencial em Porto Alegre, projetada especialmente para estudos de sustentabilidade, foi utilizada como caso de estudo para o projeto conceitual completo de um GSHP vertical, um GSHP horizontal e um ASHP. Foram estimadas a demanda de energia e os custos de operação e de implantação para determinadas condições, que foram comparadas em um estudo de viabilidade econômica. Após, os resultados foram extrapolados por uma análise de sensitividade para condições mais genéricas de consumo. Através dos cálculos, observou-se que a tecnologia de GSHPs atingiu o dobro da eficiência dos A/C mais eficientes disponíveis para o caso de estudo. Economicamente, a tecnologia também se mostrou viável, e foram estimados limitantes econômicos a esta viabilidade. Por fim, o processo de projeto desenvolvido pode servir de auxílio a novos projetistas na área. / The high energy consumption in buildings for their thermal conditioning is already one of the greatest environmental impacts of civil construction in developed countries and, progressively, is also becoming a concern in the developing countries. A relatively recent renewable alternative, as much for heating as for cooling, are the ground source heat pumps, GSHPs, which take advantage of the thermal inertia of the ground to significantly decrease the energy demand comparing to air as a source. In spite of their huge potential for decreasing energy comsumption and carbon emissions, this technology is still almost inexistent in Latin America, due not only to lack of awareness, but mainly to uncertainties about their feasibility because of their higher investment cost. Although there are many researches about the characteristics and operation of GSHPs, great part of the practical knowledge as to the design, performance and feasibility is based in the experience from the first countries where the technology spreaded, and cannot be automatically assumed as valid without a previous evaluation. Former studies indicated that the economic profit of GSHPs tends to be higher in colder climates, while the efficiency tends to be higher in milder, but still seasonal climates. For presenting a good balance among these factors, the Rio Grande do Sul State shows potential for a high performance of GSHPs The objective of this work is to estimate the efficiency levels that can be reached by this technology in a reference city in the State, as its financial feedback compared to ASHPs (air source heat pumps, the traditional air conditioners, or A/C), its limitations and the design difficulties. A residential single-family building in Porto Alegre, specifically designed for sustainability studies, was used as case study for the complete conceptual design of a vertical GSHP, a horizontal GSHP and an ASHP. The energy demand and operation and installation costs for specific conditions were estimated and compared in a economic feasibility study. After, the results were extrapolated through a sensitivity analysis for more generic situations of consumption. Through the calculations, it was observed that the technology achieved twice the efficiency of the most efficient A/C available for the case study. Economically, the technology showed itself viable, and economic limitations were also estimated. At last, the developed design process can also serve as aid to new designers in the field.

O elevado consumo energético de edificações para o seu condicionamento térmico já é um dos maiores impactos ambientais da construção civil nos países desenvolvidos e, progressivamente, vem se tornando uma preocupação também nos países em desenvolvimento. Uma alternativa renovável relativamente recente, tanto para aquecimento como para resfriamento, são as bombas de calor geotérmicas (ground source heat pumps, GSHPs), que se valem da inércia térmica do solo para reduzir significativamente a demanda energética em relação ao ar. Apesar do seu grande potencial em reduzir o consumo de energia e as emissões de carbono, esta tecnologia ainda é quase inexistente na América Latina, não somente devido à falta de conhecimento ao seu respeito, mas também pela incerteza quanto à sua viabilidade frente a um custo inicial mais elevado. Embora existam muitas pesquisas sobre o funcionamento de GSHPs, grande parte do conhecimento prático quanto ao projeto, desempenho e viabilidade é baseada na experiência dos primeiros países onde a tecnologia começou a se tornar popular, e não pode ser assumida automaticamente como válida em outras situações sem uma avaliação prévia Estudos anteriores apontaram que o retorno econômico de GSHPs tende a ser maior em climas mais frios, enquanto a eficiência tende a ser maior em climas mais amenos, mas com sazonalidade. Por apresentar um bom balanço destas características, o Estado do Rio Grande do Sul apresenta potencial para um desempenho elevado de GSHPs. O objetivo deste trabalho é estimar os níveis de eficiência que esta tecnologia pode atingir em uma cidade de referência no Estado, assim como seu retorno econômico frente a ASHPs (air source heat pumps, ou seja, aparelhos de ar condicionado, ou A/C), seus limitantes e as dificuldades de projeto. Uma edificação unifamiliar residencial em Porto Alegre, projetada especialmente para estudos de sustentabilidade, foi utilizada como caso de estudo para o projeto conceitual completo de um GSHP vertical, um GSHP horizontal e um ASHP. Foram estimadas a demanda de energia e os custos de operação e de implantação para determinadas condições, que foram comparadas em um estudo de viabilidade econômica. Após, os resultados foram extrapolados por uma análise de sensitividade para condições mais genéricas de consumo. Através dos cálculos, observou-se que a tecnologia de GSHPs atingiu o dobro da eficiência dos A/C mais eficientes disponíveis para o caso de estudo. Economicamente, a tecnologia também se mostrou viável, e foram estimados limitantes econômicos a esta viabilidade. Por fim, o processo de projeto desenvolvido pode servir de auxílio a novos projetistas na área. / The high energy consumption in buildings for their thermal conditioning is already one of the greatest environmental impacts of civil construction in developed countries and, progressively, is also becoming a concern in the developing countries. A relatively recent renewable alternative, as much for heating as for cooling, are the ground source heat pumps, GSHPs, which take advantage of the thermal inertia of the ground to significantly decrease the energy demand comparing to air as a source. In spite of their huge potential for decreasing energy comsumption and carbon emissions, this technology is still almost inexistent in Latin America, due not only to lack of awareness, but mainly to uncertainties about their feasibility because of their higher investment cost. Although there are many researches about the characteristics and operation of GSHPs, great part of the practical knowledge as to the design, performance and feasibility is based in the experience from the first countries where the technology spreaded, and cannot be automatically assumed as valid without a previous evaluation. Former studies indicated that the economic profit of GSHPs tends to be higher in colder climates, while the efficiency tends to be higher in milder, but still seasonal climates. For presenting a good balance among these factors, the Rio Grande do Sul State shows potential for a high performance of GSHPs The objective of this work is to estimate the efficiency levels that can be reached by this technology in a reference city in the State, as its financial feedback compared to ASHPs (air source heat pumps, the traditional air conditioners, or A/C), its limitations and the design difficulties. A residential single-family building in Porto Alegre, specifically designed for sustainability studies, was used as case study for the complete conceptual design of a vertical GSHP, a horizontal GSHP and an ASHP. The energy demand and operation and installation costs for specific conditions were estimated and compared in a economic feasibility study. After, the results were extrapolated through a sensitivity analysis for more generic situations of consumption. Through the calculations, it was observed that the technology achieved twice the efficiency of the most efficient A/C available for the case study. Economically, the technology showed itself viable, and economic limitations were also estimated. At last, the developed design process can also serve as aid to new designers in the field.

The majority of drugs available on the market are intended to be administered through the oral route. To achieve the desired therapeutic effect, an orally administered drug must first reach the systemic circulation and then its site of action. The fraction of the administered drug that reaches the systemic circulation is known as oral bioavailability and it is the product of the absorption and first-pass metabolism processes occurring in both the GI tract and the liver. The factors controlling bioavailability are manifold –both drug and physiologically related - and their complex interplay is key to defining a drug’s oral bioavailability. In drug discovery and development it is therefore pivotal to anticipate and understand the bioavailability of a drug candidate; a far from simple task, considering the multifactorial nature of the process. For that reason, the overall aim of this thesis was to provide different modelling and simulation (M&S) strategies that can be used for the prediction of oral bioavailability that can be of use in drug discovery and development. The first part of this thesis was focused on the evaluation of the use of bioavailability data obtained from pre-clinical species as a predictor of the human value, in a more traditional approach. In particular, the aim was to evaluate models that can quantitatively and qualitatively provide a relationship between animal and human bioavailability, by analysing trends in a large bioavailability dataset. This section demonstrated that although pre-clinical species cannot quantitatively predict bioavailability, the data obtained from them can be used for qualitative prediction of the human value. Nevertheless, such a modelling approach does not provide a mechanistic rationale of the factors affecting the bioavailability differences. Consequently, the second part of this thesis was focused on such mechanistic predictions. Particularly, we investigated the impact that drug release patterns can have on drug absorption and intestinal first pass metabolism, taking into account the physiological differences observed across the length of the human gastrointestinal (GI) tract. These release patterns are suspected to lead to bioavailability differences due to changes in the first-pass metabolism, especially for CYP3A substrates. Therefore we investigated this phenomenon applying a physiologically-based pharmacokinetic (PBPK) M&S approach: firstly, from the discovery point of view, using PBPK models in a prospective fashion to investigating the drug-related factors that might lead to such differences and secondly, from the development point of view, to predict the mechanistic differences in absorption and metabolism of oxybutynin, a drug known for its higher bioavailability when formulated as a modified release (MR) product. The latter was done by developing and applying a novel simplified PBPK model to predict such differences. The results of this work showed that the intestinal metabolism can be significantly reduced when having MR formulations of CYP3A substrates which, in some cases, can lead to higher relative bioavailability. Additionally, this thesis provided novel methods and models that have the potential to improve bioavailability predictions when using PBPK models, in particular for drugs formulated as MR.

The interaction of a diverse set of opioid alkaloids and peptides with various opioid receptors has been examined using biochemical and pharmacological techniques. Structural information on the compounds was obtained from single crystal X-ray diffraction and nuclear magnetic resonance studies, and modelled by computational methods. The introduction of a dithiocarbazate moiety into the 7a-position of a bridged thebaine was shown to afford a degree of μ selectivity in this class of nonselective compounds. X-ray diffraction analysis of this compound and comparison with the structure of [Met5]enkephalin showed the importance of the sulphydryl moiety. The conformation of [Leu5]enkephalin, in which the amino acid methionine is replaced by leucine, at the same receptor is unlikely to be similar. A series of morphinan derivatives which had been developed as μ-antagonists were evaluated. Substitution patterns of the morphinan ring nucleus and their effect upon activity were examined. X-ray analysis of several key compounds was performed. Unexpectedly a 3-hydroxymorphinan-6-one analogue showed an ability to differentiate apparently similar opioid Kreceptors. The implications in terms of K-receptor subtypes are discussed. The opioid receptor binding characteristics of structurally diverse K-receptor ligands were examined in two different buffer systems. Electrostatic modelling of the K-ligands, based upon crystal structure coordinates, was performed. From electrostatic potential maps a requirement for ligands acting at Kreceptors is postulated. Solution conformations of the endogenous K-ligand, dynorphin A(1-8), were determined by nuclear magnetic resonance studies and compared with the wo preferring [Leu5]enkephalin. Models were proposed based upon dihedral angles determined from HCtl-NH coupling constants, amide proton-deuteron exchange and amide proton temperature coefficient data. Candidate conformations were shown to be stable under dynamic simulation conditions. Electrostatic modelling of a chosen dynorphin An-8) conformation gave results comparable with the observed electrostatic model of the K-ligands. The proposed model is discussed in terms of its suitability as a retro-model for the active site ofthe K-opioid receptor.