https://www.um.edu.mt/library/oar/handle/123456789/519 The Faculty of Engineering is located at the University's main campus and offers tuition and supervision to about 477 students at both undergraduate and postgraduate levels while conducting research in all fields covered by its departments. 2026-08-09T01:29:53Z https://www.um.edu.mt/library/oar/handle/123456789/148411 Title: UV aging of PET microplastics in a custom-built weathering chamber replicating Mediterranean conditions Authors: Galea, Jack; Agius Anastasi, Anthea; Vella, Daniel A.; Briffa, Sophie M. Abstract: During their lifetime, microplastics undergo transformations and react in ways that change their behavior and properties over time. Current standard plastic weathering protocols, which were developed prior to the increased concern about plastic pollution, may not be fully suited for microplastic studies as they aim to monitor durability and understand bulk plastic behavior, with little concern about fragments or leachates produced during degradation. This work aims to age poly­(ethylene terephthalate) (PET) microplastics (MPs) in a custom-built weathering chamber replicating Mediterranean shoreline conditions (detailed in the authors’ previous work). This work studies the physicochemical changes of PET microplastic pellets (3–5 mm) following aging for a period of 90 days in dry and aquatic environments within a purposely designed weathering chamber. The chamber implemented UV-A, mechanical abrasion, and elevated temperatures to replicate the shore conditions. Micro-Raman spectroscopy was used to follow changes in the carbonyl content, crystallinity, and phenyl content of the PET MPs. Additionally, scanning electron microscopy (SEM) paired with energy-dispersive spectroscopy (EDS) was used to observe changes at the surface of the aged PET MPs. SEM showed that abrasion led to surface roughening and fibrillation. Despite changes being noted, no clear pattern emerged within the processed Raman spectroscopic results to elicit the contributions of UV radiation, salinity, and temperature. Loss of the surface layer by fragmentation and exposure of new surfaces could explain the fluctuating micro-Raman data. This work demonstrates early-stage degradation features including fluctuating chemical changes, fragmentation, abrasion, and salt crystallization under environmentally realistic Mediterranean shoreline conditions while laying methodological groundwork for future microplastic weathering research that seeks longer durations. 2025-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/148408 Title: Detection, quantification, and characterisation of microplastics in Maltese bottled water Authors: Camilleri, Josmar; Agius Anastasi, Anthea; Briffa, Sophie Marie Abstract: Plastics play a pivotal role in various industries owing to their versatility in engineering their physical, mechanical, and chemical properties while exploiting their remarkable durability, light-weight nature, and cost-effectiveness. Yet, their widespread use has led to the pollution of Earth’s water systems. Over time, plastic waste degrades into microplastics, particles smaller than 5 mm. Recent studies have highlighted the growing concerns associated with microplastics, especially in bottled beverages, including bottled water, with associated hazards still in the very early stages of being fully understood. Furthermore, the global understanding of the extent of microplastic contamination in the environment and along the food chain remains limited. This study aimed to detect, quantify, and characterise microplastics in bottled drinking water produced and sold in Malta. Samples from five brands were filtered, stained with Nile red, and quantified using fluorescence microscopy. The average microplastic concentration was found to be 35,877 ± 23,542 particles per litre, with 84% of samples exhibiting contamination, which was noted to be statistically significant. The average particle diameter was measured to be 2.3696 ± 0.0035 µm. Raman spectroscopy was used to chemically characterise 10 larger particles per brand (i.e., 50 samples), identifying the presence of cellulose, polyurethane, polymethyl methacrylate, polyethylene, and smaller quantities of other polymers. Morphological analysis classified 36 of the larger particles as fragments and 14 as fibres. Excluding laboratory-introduced contamination, the primary source of microplastic contamination in the analysed bottled water was traced to the bottle caps. Description: Supplementary material is available within this record. 2025-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/148128 Title: Development of an analytical framework for robot-inclusive homes, and of an autonomous assistive robot Abstract: Robotic assistance in domestic environments is gaining importance for supporting older adults and people with impairments. Many domestic robots depend on complex algorithms and high computational power, which raises cost and makes integration into ordinary homes difficult. At the same time, most homes are not arranged to help robots see, reach, or move well. This research addresses that mismatch by treating the person, the robot, and the home as one measurable system, and by showing that a simple robot can deliver useful assistance in a space designed to support it. The research develops a Robot-Inclusive Space framework that formalises design and evaluation across four linked measures. Human Impairment Index evaluation and task demand set an explicit upper bound on robot capability through the Robot Complexity Index. The Robot Inclusive Space Index quantifies how layout choices support observability, accessibility, and manipulability. The Space Convertibility Index estimates the effort to reach a target layout under practical limits of cost, effort, and time. The study follows a structured RIS workflow: impairments are evaluated and translated into remaining capacities, mapped to task demands, and used to identify the minimum robot features required. RIS home modifications are then defined to support both robot operation and human needs, and their feasibility is assessed in terms of cost, effort, and time. Using a teleoperated baseline, MARIS-I, the study motivates MARISII, a semi-autonomous platform sized for small homes that combines lightweight mapping, marker-aided localisation, goal-biased curvature-bounded planning, and suitable sensor placement with simple task stations consistent with the framework. Validation in physical layouts designed according to Robot-Inclusive Space principles uses consistent hardware and repeatable trials to assess reliability and efficiency. Clear sightlines, uncluttered paths, structured object placement, and marker cues keep mapping and planning lightweight and support pick and place within defined zones. Across experiments, RIS-guided design reduces planning latency, turn counts, and processor load while maintaining path quality, enabling MARIS-II to operate more reliably with lower computation. Overall, the results show that measurable improvements in observability, accessibility and manipulability reduce the need for robot-side complexity and help identify the smallest set of feasible home changes, within realistic cost, effort and time limits, for compact single-floor homes. Description: Ph.D.(Melit.) 2026-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/148036 Title: Hybrid Swin vision transformer with IMFCC for enhanced underwater acoustic noise reduction Authors: Ashok, P.; Tran, Tien Anh; Srithar, A.; Manimala, G. Abstract: Underwater wireless communication is essential for deep-sea exploration, environmental monitoring, and Autonomous Underwater Vehicle (AUV) coordination. Its performance is severely affected by ambient noise, multipath propagation, reverberation, and non-stationary interference, which degrade acoustic signal quality. Existing noise reduction techniques, such as Wiener filtering, spectral subtraction, and wavelet denoising, struggle to adapt to rapidly changing underwater conditions. Centralized Deep Learning (DL) based enhancement models are also unsuitable for distributed Underwater Wireless Sensor Networks (UWSNs) due to privacy, bandwidth, and real-time constraints. To overcome these challenges, this work proposes a Federated Learning–Driven Hybrid Swin Vision Transformer with Improved MFCCs (Fed-SVTIMFCC) for intelligent underwater acoustic noise reduction. Federated learning enables multiple underwater nodes to collaboratively train a denoising model without sharing raw acoustic data, ensuring data privacy and adaptability. The Swin Vision Transformer (SVT) captures local spectral cues and long-range dependencies to enhance reconstruction, while IMFCCs provide robust features for distinguishing noise from useful signals. Experimental results on real underwater datasets demonstrate significant improvements in SNR (26.94 dB), PSNR (33.28 dB), and SSIM (0.947), establishing Fed-SVT-IMFCC as a powerful solution for next-generation underwater communication systems. 2026-01-01T00:00:00Z