Title Promoter Affiliations Abstract "Sensor fusion and adaptive learning for optimization and control of Wire-Arc Additive Manufacturing – with applications to functionally graded materials" "Tegoeh Tjahjowidodo" "Sustainable Materials Processing and Recycling (SeMPeR), Manufacturing Processes and Systems (MaPS)" "A niche research trend in advanced materials is the creation of materials that are non-isotropic or graded in single or multiple directions of a component, which is referred to as functionally graded materials (FGM). Directional material properties are seen in nature in various forms and this can be utilized in engineering applications where a gradation in thermal, mechanical or tribological properties are required. FGMs can be created with wire-arc additive manufacturing (WAAM). WAAM is an additive manufacturing (AM) method that used the heat created from an arc to deposit and build material (usually metallic). Twin-WAAM (TWAM) upgrades the original WAAM process by using two wire feeders. Varying the feed rates of these wires independently enables us to produce FGMs strategically in one or more directions. The production of FGMs using TWAM is a novel yet is a complex challenge. The process needs to be optimized and, thus, requires a robust monitoring system. WAAM has a scarce literature on monitoring but monitoring strategies on other similar processes such as Direct Energy Deposition (DED) and Arc Welding can be referenced due to their similarity with WAAM [1]. The typical sensors that are used in literature are vision, acoustic, spectral, and thermal sensing. Vision and spectral sensors and video-based thermal sensors generate video feeds, while acoustic sensor signals require high sampling rates to detect frequencies at the MHz range. The variety, velocity, and volume of these sensor signals will be costly when implemented in an online in-situ monitoring system and should be managed. A potential approach is by using a low level representation of the process state which is achieved by using Deep Convolutional Neural Networks (DCNN) [2]. With a correct design and implementation, properly trained DCNNs could detect the low-level process features, predict defects and monitor the process quality. The natural continuation is then to use these monitored signals to control the process, by creating corrective actions that can reduce or eliminate the defects. Classical control algorithms such as PID or fuzzy controllers exist, but machine learning – reinforcement learning, can potentially improve the current control systems. WAAM process is very dynamic in a way that the buildup layer can affect the signals significantly. Complex parts also make the process challenging to predict. A typical controller may be able to preserve geometry and microstructure details given when they are designed to simple components, though these controller designs may not be able to adapt to more complex geometries. Reinforcement learning can be used to adapt to unexpected changes by learning from experience and improving the controller’s performance through active observation and interaction to the process[3]. This PhD project aims on the following: 1. Review the existing sensing methodologies for WAAM or similar processes, e.g. Direct Energy Deposition (DED), Fused deposition method (FDM), Arc Welding, etc. in the context of FGMs 2. A study on the potential features that are generated during the WAAM process; down-select the sensors that can measure these features; design and implement monitoring strategy 3. Data collection, analysis, and interpretation; develop ML models to predict defects and final material composition from the WAAM process 4. Demonstration of in-situ monitoring using the developed models and validation of these models with actual components 5. Development of a control system (using ML/RL methods) and comparing with benchmark control methods [1] C. Xia et al., “A review on wire arc additive manufacturing: Monitoring, control and a framework of automated system,” Journal of Manufacturing Systems, vol. 57. Elsevier B.V., pp. 31–45, Oct. 01, 2020, doi: 10.1016/j.jmsy.2020.08.008. [2] J. Günther, P. M. Pilarski, G. Helfrich, H. Shen, and K. Diepold, “Intelligent laser welding through representation, prediction, and control learning: An architecture with deep neural networks and reinforcement learning,” Mechatronics, vol. 34, pp. 1–11, Mar. 2016, doi: 10.1016/j.mechatronics.2015.09.004. [3] R. S. Sutton and A. G. Barto, Reinforcement learning: An introduction. 2018." "Increasing the convergence speed of psysically-based rendering algorithms in computer graphics using frequency-based combined adaptive sampling and adaptive filtering" "Philip Dutré" "Human-Computer Interaction (HCI)" "Physically-based rendering is one of the major research topics in the field of computer graphics. It deals with the generation of photo-realistic images depicting a virtual world. Although research in this field has a history dating back more than 25 years, generating photo-realistic images efficiently is still a problem of major significance. For each pixel in the final image a light transport equation needs to be evaluated, of which the complexity depends upon the variety of global illumination phenomena in the 3D scene. This light transport equation is typically evaluated by tracing light paths through multiple reflections between light sources and the virtual camera in the 3D scene.This research project aims at increasing the convergence speed of these physically-based rendering algorithms by proposing new combined adaptive sampling and adaptive filtering techniques, which adapt to the frequency properties of the underlying illumination signals. Based on the inherent band limits due to the scene and camera, we can robustly derive for each pixel the required number of light paths to generate during the actual rendering and the required filter size needed during post-processing.The proposed research is an important step towards a full comprehensive model describing the variety of global illumination phenomena ubiquitous in realistic scenes, as well as towards real-time physically accurate rendering that will enable both low sampling rates and low post processing times." "The use of adaptive logics for the practice and the philosophy of mathematics and the use of mathematical tools for the abstract analysis of adaptive logics" "Joke Meheus" "Department of Philosophy and moral sciences" "This project concerns an investigation into three aspects of the relation between mathematics and adaptive logics (AL's): it is investigated how (a) AL's can be used for the formal modeling of the mathematical practice, (b) AL's can solve problems of the foundations of mathematics, and © mathematical techniques can be used for the abstract study of AL's." "AI driven VR training in an adaptive user context" "Silvia Van Aken" "Immersive Lab, Onderzoekscentrum een Leven Lang Leren en Innoveren, Universiteit Antwerpen" "This TETRA project, supported by the Innovation and Enterprise Agency, will introduce AI and 3D scanning techniques in the development of VR training. Through Proofs of Concept (POCs) and user cases, it will illustrate how VR training can be made more effective to optimise motivation and knowledge transfer among employees. This project initially targets 50 SMEs that develop VR training simulations. In addition, 129 training and education centres that offer training and issue certificates belong to the target group. Finally, a third group is formed by the 13,716 Flemish SMEs and 1,454 large companies in the industrial sector. THE PROJECT HAS THE FOLLOWING OBJECTIVES: Development of 2 adaptive VR-training courses (POCs) that increase the learning efficiency with 15%, verifiable by an impact study. User manual for trainers within industry and training centres to train employees more efficiently in order to achieve the necessary behavioural change. Open source AI module (Machine Learning Algorithm) for adaptive VR simulation. Sandbox for user testing by researchers and students to measure the learning efficiency of adaptive VR training in the Immersive Lab (AP) or on the workfloor (with guidance group). User manual for 3D scanning of training objects/spaces and integration into VR workflow. Connecting stakeholders (VR and AI companies, training centres and businesses) and creating a dynamic that initiates training innovation in Flanders. Providing tools for innovative learning methods. Convince to invest more in training, measurable through survey." "Rapid and easy-to-perform FO-SPR-based assays for VWF antigen and functionality and insights into the vaccine-induced adaptive immune response against SARS-CoV-2" "Simon De Meyer" "Cardiovascular Sciences, Kulak Kortrijk Campus, Centre for Molecular and Vascular Biology, Chemistry, Kulak Kortrijk Campus" "This PhD-project encompasses two different parts that can be placed under the discipline ‘laboratory medicine’ as both concerned on the analysis of blood samples. Part 1 focusses on improving the analytical aspects of the laboratory diagnosis of von Willebrand disease (VWD), whereas part 2 involves the characterization of the vaccine-induced adaptive immune response against SARS-CoV-2. For each part, a brief description will be given. PART 1: When the blood vessel wall is damaged, circulating von Willebrand factor (VWF) will interact with the subendothelial collagen. This causes activation of the A1-domain of VWF leading to interaction between this A1-domain and the GPIbα receptors of circulating platelets. The latter ultimately leads to the formation of an initial platelet plug, which stops the bleeding. This plug is fortified with fibrin strand formation due to the activation of the coagulation cascade. VWD is the most common inherited bleeding disorder caused by mutations of the VWF gene that lead to either a quantitative deficit (i.e. type 1 and type 3 VWD) or a qualitative deficit of VWF (i.e. type 2 VWD). Diagnosis of VWD not only involves a suggestive bleeding anamnesis, but also requires a laboratory-proven abnormality within either the VWF antigen (VWF:Ag) or one of the VWF functions (i.e. VWF:RCo and VWF:CB). Unfortunately, the laboratory assays to determine these VWF parameters are complex-to-perform, not readily available and require multiple platforms. Fiber optic – surface plasmon resonance (FO-SPR) is an innovative technology to measure biomolecular interactions by the means of a simple dip in the sample. Within this PhD-project, I will try to develop new FO-SPR-based assays for accurate and easy quantification of each of the abovementioned VWF parameters. PART 2: During this PhD-project, a novel virus named SARS-CoV-2 caused the largest pandemic of the Modern Time. Although most infections are (pauci)symptomatic, there are also very severe-to-lethal infection described, even within perfectly normal health persons. In an attempt to battle the SARS-CoV-2-related burden, large-scale vaccination campaigns were initiated. As the immune response induced by an infection or after vaccination is unique for each pathogen or vaccine, it is of utmost importance to gain information regarding the vaccine-induced adaptive immune response within healthy subjects. During my PhD, we performed a prospective clinical trial (COVID-VAX-AZG) to study complete and in-depth characterization of both the vaccine-induced immune response and monitor breakthrough infections." "Run-time adaptive hierarchical neural architectures for representation learning in modular sensor networks" "Tijl De Bie, Joni Dambre" "Department of Electronics and information systems" "The human brain excels when it comes to the efficient processing of multiple correlated information streams. It continuously receives massive streams of sequential data from our hearing, vision, touch and other senses. Nevertheless, it is capable to quickly process and learn from these huge amounts of data, because it can efficiently abstract it into a more compact representation and because it optimally exploits the correlations that exist between information from its different senses. It is also very adaptive: the brain dynamically rewires itself all the time and can learn to cope with sudden and lasting changes. Unfortunately, artificial neural networks still cannot match the brain its performance when dealing with multi-sensory information. Current approaches typically focus on specific combinations of sensors and do not offer a generally applicable solution. They process the different sensor streams in separation and only combine them at the highest levels of abstraction, while the human brain appears to search for correlations already in early processing stages. Also, other recent observations in the field of neuroscience involving dynamic adaptation remain unexplored for applications in neural networks. In my research, I will therefore draw inspiration from these insights to create a generic deep architecture that can efficiently integrate information streams from multiple and noisy sensors, while being able to adapt to changes in the sensor or the environment." "Adaptive target functions for Cutting & Packing problems." "Greet Vanden Berghe" "Computer Science Technology, Ghent and Aalst Technology Campuses" "The present project aims at research into the modeling of C&P problems." "Adaptive Behaviour Training for People with Intellectual Disabilities" "Bieke Zaman" "Institute for Media Studies" "The rate of psychiatric comorbidities with affective and anxiety disorders is about 11 times higher for people with Intellectual Disabilities (ID) compared to the general population. A major reason for the poor mental health in people with ID is having limited adaptive skills putting people with ID at increased risk of social exclusion, a key determinant of poor mental health. One way to promote the inclusion and well-being of people with ID is through Adaptive Behaviour training. Adaptive behaviour is the collection of conceptual, social, and practical skills that are learned and performed by people in their everyday lives. As Virtual Reality (VR) is making an appearance in the health sector with promising results, the safe and controlled environment of a VR application can be used for training adaptive behaviour. The ADAPT-ID project will explore the extent to which digital interventions promote participation/inclusion leading to well-being and mental health for people with ID. We aim to better address mental health issues of persons with ID by developing and piloting strategies to empower them using the opportunities provided by VR." "Integrated piezoelectric metamaterials for adaptive and improved noise and vibration mitigation" "Elke Deckers" "Mecha(tro)nic System Dynamics (LMSD)" "Metamaterials (MMs) have emerged as promising noise and vibration solutions which enable reconciling the often conflicting requirements of low mass and volume and high noise and vibration attenuation. Although their potential has been widely evidenced in a variety of (academic) demonstrators, the manufacturing of MMs is far from mature while their performance is mostly limited to narrowband noise and vibration reduction. The main hypothesis in this project is that by levering on the suitability of insert injection moulding for mass-production and the versatility of additive manufacturing to create geometrically complex resonator inclusions, mass-manufacturable integrated MM solutions can be produced. Alongside passive vibro-acoustic performance broadening strategies which rely on optimally designed resonator properties and their distribution, embedding smart elements within the resonant inclusions can enable (smart) MM systems with enhanced (adaptive) broadband noise and vibration performance. Therefore this research will focus on insert injection moulding approaches and design for both passive and smart additive manufacturing based resonant inclusions, a multi-physical integrated MM modeling framework for fast performance predictions and a first-of-their-kind injection moulded (smart) MM panel with demonstrated broadband improved vibro-acoustic performance." "Standardization of the ABAS-3, an instrument for mapping adaptive behavior." "Ilse Noens" "Parenting and Special Education" "The research will result in a manual that can be used in practice, including the standardization data and the data from psychometric research."