Browse
Recent Submissions
Item Embedded systems design of a low-cost flight controller for use in UAV platforms(2025-09) Fraser, D; Pretorius, A; Hepworth, J; Botha, NatashaServing as a critical interface between high-level software and low-level electronic systems, flight controllers are essential for aerial robotic applications. However, their significant cost poses a barrier to widespread adoption. While affordable, off-the-shelf flight controllers exist, they generally do not support the firmware required for research- and industrylevel implementation, such as PX4. This paper presents embedded system designs for low-cost flight controllers that are PX4 compliant. The research delves into examining the necessary components, existing open-source firmware/software, and appropriate hardware solutions. The presented designs are based on open standards and capable of operating with various platforms, applications, and components. The final design is shown to be low in cost while also demonstrating substantial potential for integration with existing applications. The flight controllers were verified with a range of tests, including the implementation of a control system that was responsible for governing the roll angle of a quadcopter. The overall flight controller performance is shown to be comparable with the performance and functionality of commercial solutions.Item Assessing Depth Anything V2 monocular depth estimation as a LiDAR alternative in robotics(2025-11) Louw, Jakobus M; Verster, Jacobus J; Dickens, John SThis paper evaluates the performance of Depth Anything V2, a deep learning-based monocular depth estimation model, as a low-cost alternative to LiDAR for robotic depth sensing. LiDAR, while widely used, is expensive, prompting the search for affordable solutions. Six datasets were recorded in indoor environments to assess the performance of the pretrained metric depth model. Qualitative analysis showed that overall relative depth is well estimated, but fine details and close-range depths in featuresparse areas are not represented well. Quantitative analysis revealed variability in performance across datasets, with mean errors ranging from 0.32 m to 0.66 m. Additionally, performance varies with distance. For objects within 2 m, 89.1% of errors are within ±0.5 m. This decreases to 77.0% for objects within 4 m and further drops to 70.8% for objects within 6 m. Depth Anything V2 demonstrates higher pixel resolution than LiDAR but with significantly reduced metric depth accuracy. While not suitable for high-precision applications like indoor navigation and obstacle avoidance, the model can still provide useful depth information in scenarios where finegrained accuracy is less critical.Item FASTSeg3D: A fast, efficient, and adaptive ground filtering algorithm for 3D point clouds in mobile sensing applications(2025-05) Oladele, DA; Markus, ED; Abu-Mahfouz, Adnan MIBackground: Accurate ground segmentation in 3D point clouds is critical for robotic perception, enabling robust navigation, object detection, and environmental mapping. However, existing methods struggle with over-segmentation, under-segmentation, and computational inefficiency, particularly in dynamic or complex environments. Methods: This study proposes FASTSeg3D, a novel two-stage algorithm for real-time ground filtering. First, Range Elevation Estimation (REE) organizes point clouds efficiently while filtering outliers. Second, adaptive Window-Based Model Fitting (WBMF) addresses over-segmentation by dynamically adjusting to local geometric features. The method was rigorously evaluated in four challenging scenarios: large objects (vehicles), pedestrians, small debris/vegetation, and rainy conditions across day/night cycles. Results: FASTSeg3D achieved state-of-the-art performance, with a mean error of <7%, error sensitivity < 10%, and IoU scores > 90% in all scenarios except extreme cases (rainy/night small-object conditions). It maintained a processing speed 10× faster than comparable methods, enabling real-time operation. The algorithm also outperformed benchmarks in F1 score (avg. 94.2%) and kappa coefficient (avg. 0.91), demonstrating superior robustness. Conclusions: FASTSeg3D addresses critical limitations in ground segmentation by balancing speed and accuracy, making it ideal for real-time robotic applications in diverse environments. Its computational efficiency and adaptability to edge cases represent a significant advancement for autonomous systems.Item DFT analysis of spin configurations and its impact on manganese oxide cathodes(2025-11) Kgasago, MP , M; Phoshoko, Katlego W; Ngoepe, P; Ledwaba, RThe electronic structure of lithium manganese oxide cathode materials is dependent on the choice of functional and spin configuration, necessitating a thorough analysis to accurately describe their structural and electronic properties. While these materials are appealing for lithium-ion batteries due to their high capacity, abundance, and low cost, accurately modelling their electronic properties remains challenging due to strong correlations in the manganese d orbitals. In this study, we employ firstprinciples calculations using the GGA+U functional to investigate the influence of spin configurations on the structural and electronic properties of these cathode materials. Our results shows that the antiferromagnetic (AFM) configuration gives a band gap consistent with experimental observations, whereas the ferromagnetic (FM) configuration produces a larger band gap. This shows the critical role of spin configuration in determining the electronic properties of these materials, indicating the importance of precise spin state modelling for accurate materials. These findings shows the importance of benchmarking beyond default ferromagnetic settings, including AFM configurations, to improve the accuracy of calculated properties in Mn-based cathode materials.Item Effects of process parameters and many‑objective optimisation of SiC/ TiC/16MnCr5 coating deposited through laser cladding using NSGA III(2025-05) Kyekyere, E; Onyono, SO; Olakanmi, EO; Prasad, RVS; Botes, A; Pityana, Sisa L; Skhosane, Besabakhe SDue to its several advantages, laser cladding has been used to enhance the surface properties of parts subjected to severe loading conditions. Despite its benefits, the performance and quality characteristics of the coatings produced depend highly on the complex relationship between the coating materials and the process variables. Hybrid reinforcements such as TiC and SiC in metal matrix composites can provide synergistic benefits for components used in extreme environments, like mining, by significantly enhancing wear resistance and mechanical properties. In this work, the effect and many-objective optimisation of the laser processing variables were carried out on TiC/SiC/16MnCr5 multi-track composite coating on A514 steel. By employing hybrid response surface modelling (RSM) and non-dominated sorting genetic algorithm III (NSGA III), the effect of scanning speed (S), powder feed rate (F), and laser power (P) on coating aspect ratio, dilution, microhardness, and wear resistance were investigated and optimised. The influence of the process variables on aspect ratio, dilution, microhardness, and wear volume loss are in this order, respectively: S > P > F; S > F > P; P > F > S; and P > S > F. The interaction of the process parameters was significant. The wear resistance and microhardness were enhanced due to the partial dissolution of carbides in the matrix. Based on the NSGA III optimisation, the optimal process parameters identified were P = 1550 W, S = 500 mm/min, and F = 7 g/min. The validation experiment revealed a close agreement with the predicted results, with errors of less than 5% for all the objectives. The optimised coating’s microstructure consisted predominantly of columnar crystals with minor regions of equiaxed dendrites. Compared to the substrate, the optimised coating’s microhardness improved by 350%, while its compressive strength was enhanced by 41%.Item Microstructure evolution and properties development of in situ processed Ni–Ti alloys by laser directed energy deposition(2025-09) Abdelwahed, M; Skhosane, Besabakhe S; Ishola, M; Casati, R; Vedani, M; Pityana, Sisa L; Taha, MAThis investigation proposes a flexible technique for in situ fabrication of Ni–Ti structures using laser directed energy deposition, in which nickel and titanium powders are separately fed and melted together during laser processing. The proposed mechanism enables controlled flows of powders facilitating a fine-tuning of the desired chemical composition, when compared to the conventional use of pre-mixed feedstocks, in an endeavor toward the laser processing of pseudoelastic Ni–Ti alloys. The results highlighted the possibility of fabricating a wide range of tailored Ni–Ti compositions and microstructures, depending on the powder flow ratios. The developed alloys were classified as either Ni-rich or Ti–rich compositions, in which the Ni-rich alloys were composed of different fractions of B2-NiTi austenite, NiTi/Ni3Ti eutectics, and Ni3Ti intermetallic with a minor presence of NiTi2/Ni2Ti4Ox. While the Ti–rich compositions were mainly dominated by NiTi austenite with a fraction of NiTi2/Ni2Ti4Ox dendrites. Under identical laser processing parameters, the findings showed that the in situ alloyed Ni47.6Ti52.4 composition exhibited a comparable microstructure and pseudoelastic behavior similar to that obtained from a laser processed pre-alloyed powder. The output of the investigation highlights the potential use of the in situ alloying mechanism as a cost-effective and flexible approach for fabricating Ni–Ti alloys with desired compositions and properties.Item Adapting SegMap: A LiDAR place recognition framework for standalone use in C++ applications(2025-09) Maweni, Thabisa; Amayo, PAutonomous mobile robots rely on accurate environmental mapping and continuous self-localisation for effective navigation, often achieved through complex algorithms that combine data from multiple sensors. Aru-SegMap is an adaptation of SegMap, a widely used 3D point cloud segment-based map representation, for modern ROS2-based and standalone C++ applications focused on localisation. SegMatch, a 3D point cloud segmentation and matching library integral to SegMap, reliably estimates a robot's position and detects loop closures. This adaptation involved modularising the original library, decoupling it from a deprecated TensorFlow C++ API and ROS1, and integrating visualisation capabilities, enabling greater flexibility and usability for continued robotics research and development. Aru_SegMap was validated using datasets of varied agricultural environments. It is functional, produces consistent segments, and provides reliable localisation.Item Formation following: Multi-sensor object detection and localization using ROS2 and Gazebo(2025-09) Van Eden, Beatrice; Maweni, Thabisa; Setati, Cecil T; Naidoo, Thegaran , Thegaran Dickens; Dickens, John SFormation control is essential in multi-robot systems, enabling efficient navigation, coordination, and task execution in dynamic environments. This work presents a vision-based approach for formation following using three robots. Each robot is identified by a unique colour and recognised using YOLOv8. The recognition output is combined with LiDAR distance measurements to facilitate waypoint navigation and formation maintenance. A set of hardcoded rules ensures proper formation control. The proposed method is tested in a structured environment, demonstrating effective multi-robot coordination.Item Investigating how spatter evolves in metal additive manufacturing processes with machine learning(2025-11) Nkomo, Brighton VIn metal additive manufacturing, laser-powder-bed fusion (LPBF) suffers from layer-to-layer instabilities; most notably molten-metal spatter and recoater streaking - that degrade surface finish and internal integrity. We investigate whether physics-informed machine-learning (PIML) can detect and predict these anomalies more efficiently than purely data-driven models. Using the Oakridge National Laboratory (ORNL) Peregrine in-situ dataset, we (i) derive physically meaningful features such as volumetric energy density, Peclet number and plume-attenuation proxies, and (ii) embed gradient penalties that enforce monotonic behaviour with respect to energy input. A lightweight PIML network attains an Root Mean Square Error (RMSE) of ≈ 3.9 × 10⁴ spatter pixels (R² = 0.94) while requiring 40 % less training data than an architecture-matched multilayer perceptron. SHapley Additive exPlanations (SHAP) analysis shows that the model’s attributions follow established heat-transfer mechanisms, confirming improved interpretability. These results demonstrate that even minimal physics supervision delivers data-efficient, trustworthy defect monitoring, at least in the case of neural networks tested in this work, paving the way for real-time, closed-loop LPBF control.Item Safety and efficacy of plant-produced trivalent virus-like particle vaccine candidate against bluetongue disease(2025-08) Mlingo, TAM; O'Kennedy, Martha M; Matsiela, M; Nkomo, N; Coetzee, P; Rametse, T; Mutowembwa, P; Heath, L; Mokoena, NBluetongue (BT) is a non-contagious, insect-transmitted disease of wild and domestic ruminants caused by bluetongue virus (BTV). Effective control of BT disease relies on vaccination against prevalent or seasonal serotypes using live attenuated or inactivated vaccines. Limitations of these vaccines offer opportunities for improvement. Transient protein production in plants has evolved as a platform that offers a unique ability to express multiple antigens and complex protein assemblies. In this study, a trivalent virus-like particle (VLP) vaccine candidate against BT was produced in Nicotiana benthamiana. Chimeric VLPs comprised of outer capsid proteins (VP2/VP5) from BTV3 and BTV4 were designed with BTV8 inner core proteins (VP3/VP7). The proteins were successfully expressed, and assembly of chimeric and homologous BTV8 VLPs was achieved. Antigens were formulated with stabilised nano alum-based adjuvant, and safety and efficacy were evaluated in merino sheep. Two groups of animals were vaccinated with a plant-produced or live attenuated vaccine, formulated with similar BTV serotypes. The third group of animals received a placebo. Both vaccines were safe and did not induce temperature reactions or BTV clinical signs in sheep. Vaccines further protected animals against challenge with serotypes 3, 4 and 8. Animals that received a placebo vaccination demonstrated typical BTV clinical signs following challenge with virulent viruses. The results demonstrated that the plant-made chimeric VLP vaccine candidate was safe and efficacious in sheep, and can be used for prophylactic immunisation against BT disease. This is a first report demonstrating the safety and efficacy of a plant-produced trivalent VLP candidate vaccine in target animals.Item The impact of fixed-Tilt PV arrays on vegetation growth through ground sunlight distribution at a solar farm in Aotearoa New Zealand(2025-10) Dhlamini, Matlotlo M; Brent, ACThe land demands of ground-mounted PV systems raise concerns about competition with agriculture, particularly in regions with limited productive farmland. Agrivoltaics, which integrates solar energy generation with agricultural use, offers a potential solution. While agrivoltaics has been extensively studied, less is known about its feasibility and impacts in complex temperate maritime climates such as Aotearoa New Zealand, in particular, the effects of PV-induced shading on ground-level light availability and vegetation. This study modelled the spatial and seasonal distribution of ground-level irradiation and Photosynthetic Photon Flux Density (PPFD) beneath fixed-tilt PV arrays at the Tauhei solar farm in the Waikato region. It quantifies and maps PPFD to evaluate light conditions and its implications for vegetation growth. The results reveal significant spatial and temporal variation over a year. The under-panel ground irradiance is lower than open-field GHI by 18% (summer), 22% (spring), 16% (autumn), and 3% (winter), and this seasonal reduction translates into PPFD gradients. This variation supports a precision agrivoltaic strategy that zones land based on irradiance levels. By aligning crop types and planting schedules with seasonal light profiles, land productivity and ecological value can be improved. These findings are highly applicable in Aotearoa New Zealand’s pasture-based systems and show that effective light management is critical for agrivoltaic success in temperate maritime climates. This is, to our knowledge, the first spatial PPFD zoning analysis for fixed-tilt agrivoltaics, linking year-round ground-light maps to crop/pasture suitability.Item The effect of lyophilisation on the physiochemical and biological properties of plant-based chitosan modified- poly lactic-co-glycolic acid microparticles(2025-06) Shatri, AMN; Lemmer, Yolandy; Mumbengegwi, DRTerminalia sericea, Lantana camara, Grewiatenax, and Corchorus tridens extracts are potent against gastrointestinal bacteria. However, metabolism by gastric acid reduces their efficacy and limits their mainstream uses. Encapsulating extracts into microparticles could improve biological properties while offering protection against gastric acid. Moreover, lyophilizing microparticle suspensions could reduce swelling and bursting, while improving microparticle stability during long-term solid storage. The study aimed to investigate the effect of lyophilization on the physicochemical and biological properties of plant-based chitosan-modified-PLGA (CMPLGA) microparticles over time. The formulated microparticles were lyophilized and analyzed for size, polydispersity index, zeta potential, and pH. The antibacterial and cytotoxicity of the lyophilized microparticles were also determined by agar disc diffusion and MTT. There was a significant increase in microparticle size after lyophilization due to aggregation. However, there was no statistically significant change in zeta potential and polydispersity index after lyophilization p < 0.05. Potent antibacterial activity of the micro-dissolutions was recorded against clinical Salmonella and Verotoxigenic Escherichia coli with the lowest recorded Minimum inhibitory concentrations of 6.25 ± 0.0 µg/ml with only minimal microparticle swelling and leakage. CMPLGA microparticles containing L. camara showed the lowest IC50 of 66.7 ± 0.005 µg/ml, indicating that microparticles are not significantly toxic to the NIH/3T3 mouse embryonic cells. The study indicates that lyophilization could be a useful method for long storage of CMPLGA microparticles as it does not preserve the surface charges and polydispersity index of the nanoparticles, while enhancing the efficacy and safety of the CMPLGA nanoparticles.Item Enrichment of extracellular vesicles using Mag-Net for the analysis of the plasma proteome(2025-07) Wu, CC; Tsantilas, KA; Park, J; Plubell, D; Sanders, JA; Naicker, Previn; Govender, I; Buthelezi, Sindisiwe G; Stoychev, SExtracellular vesicles (EVs) in plasma are composed of exosomes, microvesicles, and apoptotic bodies. We report a plasma EV enrichment strategy using magnetic beads called Mag-Net. Proteomic interrogation of this plasma EV fraction enables the detection of proteins that are beyond the dynamic range of liquid chromatography-mass spectrometry of unfractionated plasma. Mag-Net is robust, reproducible, inexpensive, and requires <100 μL plasma input. Coupled to data-independent mass spectrometry, we demonstrate the measurement of >37,000 peptides from >4,000 proteins. Using Mag-Net on a pilot cohort of patients with neurodegenerative disease and healthy controls, we find 204 proteins that differentiate (q-value < 0.05) patients with Alzheimer’s disease dementia (ADD) from those without ADD. There are also 310 proteins that differ between individuals with Parkinson’s disease and without. Using machine learning we distinguish between individuals with ADD and not ADD with an area under the receiver operating characteristic curve (AUROC) = 0.98 ± 0.06.Item Coupling of excitation energy to photochemistry in natural marine phytoplankton communities under iron stress(2025-07) Pupulewatte, H; Gorbunov, MY; Moore, CM; Selden, CD; Ryan-Keogh, Thomas J; Furbu, J; Hawley, R; Lohan, MC; Bibby, TS; Falkowski, PGOxygenic photosynthesis requires excitation energy transfer from light-harvesting complexes (LHCs) to reaction centers (RCs) to drive photochemical redox chemistry. The effective absorption cross section of RCs dynamically responds to the light environment on time scales of seconds to days, allowing rapid acclimations to changes in spectral irradiance and photoprotection under high light, thereby optimizing light absorption for photochemistry. Although energy coupling between LHC–RCs has been studied for decades in laboratory cultures, it remains poorly understood in real-world conditions, where it is potentially influenced by nutrients. In the oceans, one of the most critical micronutrients for photosynthesis is iron (Fe). To investigate the effects of Fe stress on the energetic coupling between LHC–RCs in natural phytoplankton assemblages in the Southern Atlantic Ocean, we assessed photophysiological responses using a pair of custom-built fluorometers measuring chlorophyll-a variable fluorescence and picosecond fluorescence lifetimes. Detailed analysis based on the functional absorption cross section of the oxygen-evolving complex, quantum yield of photochemistry, energetic connectivity of RCs, and the average lifetime of in vivo chlorophyll fluorescence suggested that between 10 and 25% of LHCs remain uncoupled from RCs and do not effectively contribute to photochemical charge separation. Addition of Fe to samples under trace metal-clean on-board incubations indicates relatively rapid recoupling (< 24 h) of antennae to photochemistry, followed by biophysical stabilization of recoupled complexes. Our findings highlight the crucial role of micronutrients in controlling the excitation energy transfer from LHCs to RCs in marine phytoplankton and the overall primary productivity in the real-world oceans.Item Utilizing aquifer hydraulic parameters to assess local and regional recharge potentials for enhancing water allocations in groundwater-dependent areas in De Aar, Northern Cape, South Africa(2025-09) Baloyi, L; Kanyerere, T; Muchingami, I; Pienaar, Harrison H; Igwebuike, N; Mukhawana, MBThe precise and accurate use of aquifer hydraulic parameters for assessing local and regional recharge potential for enhancing groundwater allocation planning is vital for many hydrogeological studies. The conventional approach for allocating groundwater presents a challenging scenario, as it remains uncertain whether the applied recharge estimate is local or regional recharge. The approach does not account for the extent of the aquifer recharge in terms of local and regional scale; instead, it assumes that recharge is distributed across the catchment. This study aimed to demonstrate the use of aquifer hydraulic parameters (transmissivity and storativity) to explain areas of potential recharge (local and regional) for enhancing groundwater allocation planning with a specific case study of De Aar, Northern Cape, South Africa. It argues that not integrating local and regional recharge potentials in planning for groundwater allocations can result in over- or under-allocation of groundwater resources to users. A constant discharge pumping test and recovery test matching the duration of pumping were conducted for data collection. The Flow Characteristics method was used as a diagnostic tool to understand the different aquifer flow regimes in the study area. To develop an integrated understanding of the groundwater system, a hydrogeological conceptual model was used to visualize areas with higher or lower recharge potential across local and regional scales. Results showed significant variability in transmissivity, ranging from 213 to 596 m2/d, and storativity, ranging from 0.0000297 to 0.000185. The transmissivity values suggest that groundwater moves faster; meanwhile, the storativity values suggest that the aquifer system has high water storage capacity. These results will assist water resource planners in making informed decisions on how to allocate groundwater to users. This study demonstrated that aquifer hydraulic parameters are a valuable tool for improving groundwater allocations, thereby highlighting the importance of considering areas for potential recharge, both local and regional, in planning groundwater allocation.Item The design of discrete-time systems with prescribed probability distribution and power spectrum pole(2025-04) McDonald, Andre M; Van Wyk, MA; Cilliers, Jacques EThis paper addresses the problem of designing nonlinear discrete-time dynamical systems for prospective use in low-complexity random signal generators. Drawing upon ergodic systems theory, we derive a novel method for designing nonlinear systems with zero-input responses that simultaneously satisfy two user-specified statistical metrics as specifications. These metrics are a piecewise-constant probability density function and a power spectral density expressed as a rational function with an arbitrary number of prescribed non-repeating poles. This new method configures a memoryless nonlinearity (i.e., a map) with a low-complexity hat-like structure as a recursive system with adjustable parameters, thereby yielding a simplified and explicit relationship between the system’s parameters and the statistical metrics of its response. This yields a matrix design approach that, unlike existing methods, affords the freedom to prescribe all of the poles as a set of non-repeating complex values without resorting to a numerical search over the parameter space for a candidate solution. Simulations are presented of elementary systems constructed using the novel method, and the link between the maps’ structure and the observed statistical behavior is expounded. More sophisticated systems with richer statistical metrics are constructed as examples that demonstrate the method’s versatility. We also demonstrate the independent adjustment of the center frequency and bandwidth of power spectrum modes during system design, which affords greater freedom than existing methods to select the power spectral density. We anticipate that the proposed method will find application in the construction of efficient random signal generators for Monte Carlo simulation in domains such as mobile communications, radar engineering, data encryption and optimization.Item Synthesis of iron-based metal–organic frameworks and carbon derivatives via unconventional synthetic methods and waste precursors with potential for gas storage(2025) Mosupi, Keaoleboga; Mthembu, NT; Masukume, Mike; Musyoka, NM; Langmi, HWMetal–organic frameworks (MOFs) have remarkable characteristics including high porosity as well as large internal surface areas. However, these materials have found very limited use industrially due to their high cost of production. The use of waste materials and industrial by-products to generate cheaper and environmentally friendly precursors could potentially open doors for industrial production of MOFs. Two types of Fe-based MOFs (Fe-MIL-53 and Fe-MIL-88B) were prepared using acid mine drainage (AMD) waters as a metal precursor source and waste polyethylene terephthalate-derived terephthalic acid (PET-BDC) as a linker via microwave-assisted and sonochemical-assisted synthesis procedures. Additionally, a carbonization strategy was utilized to enhance the porosity and surface area of these MOF materials. Upon carbonization, surface areas were drastically improved to above 600 m2 g−1 for both MIL-53 and MIL-88B prepared using the two unconventional methods. The obtained carbons also exhibited reasonable gas uptake capacities, with MIL-53 derived carbons having a higher hydrogen capacity of 1.32 wt% (at 77 K and 1 bar) and a carbon dioxide capacity of 2.09 mmol g−1 (at 298 K and 1 bar). The gas uptake capacities of MIL-88B derived carbons were found to be relatively low.Item Evaluation of the performance of a geogrid placed in a cement-treated subbase using accelerated pavement testing(2025-09) Smit, Michelle A; Rust, FC; Zamara, Z; Mazurowski, P; Kawalec, JRoad construction costs are high partly due to the scarcity and cost of good quality building materials. The inclusion of geosynthetics in flexible pavement structures has been shown to improve pavement performance. The study indicated the potential benefit of increasing the life of the pavement by adding a multi-shape geogrid layer to a cement-treated subbase. APT testing was conducted using the Heavy Vehicle Simulator (HVS) of the Council for Scientific and Industrial Research (CSIR) in South Africa. When the rutting data, deflection data and the stiffnesses back-calculated from FWD data are considered, it is evident that the inclusion of the geogrid into the cement-treated layer was beneficial and could increase the life of the pavement significantly. The addition of the geogrid to the structure led to an approximate 5% increase in the construction cost. This is negligible in comparison to the significant increase in pavement life. However, the results are valid for a relatively strong (10–30 million ESAL) design. Further work should include the use of the geogrid in a cement-treated base in a weaker pavement where the benefit could potentially be higher.Item A critical review on the applications of Sentinel satellite datasets for soil moisture assessment in crop production(2025-07) Mkhwenkwana, A; Matongera, TN; Blaauw, Ciara; Mutanga, OUnderstanding soil moisture dynamics in crop production is critical for optimising water resource management. The Sentinel satellite missions have significantly contributed to soil moisture monitoring by providing high-resolution, multi-sensor data. This review examines advancements in soil moisture assessment using Sentinel datasets, particularly in crop production. It highlights key challenges, evaluates their impact on monitoring accuracy, and explores potential methodological improvements. Findings indicate that Sentinel-1′s synthetic aperture radar (SAR) data, particularly VV and VH polarizations, and Sentinel-2′s multispectral indices, such as NDVI and NDMI, are widely integrated with machine learning algorithms to enhance soil moisture estimation. However, dense vegetation and complex topography reduce retrieval accuracy, necessitating sensor fusion and calibration for improved reliability. Sentinel-3 provides valuable surface temperature and land condition data for indirect soil moisture estimation, but its application remains limited due to higher uncertainty compared to SAR and multispectral approaches. Emerging trends suggest that machine and deep learning techniques, such as RF, SVR, and CNN, can enhance data fusion across Sentinel missions. Additionally, preprocessing steps such as RTC, speckle filtering, and the integration of multipolar and polarimetric data with physical backscattering models show promise in mitigating radar backscatter interference. Further development of robust retrieval models that incorporate topography, soil roughness, and texture are essential for improving soil moisture accuracy in diverse agricultural landscapes. This review underscores the need for continued methodological advancements to maximise the potential of Sentinel datasets for soil moisture monitoring in precision agriculture and water resource management.Item Green synthesis of amorphous silica nanoparticles (SiO2NPs) from sugarcane bagasse ash by sol-gel method(2025-11) September, LA; Kheswa, N; Seroka, Ntalane S; Khotseng, LSugarcane bagasse ash was utilized as a silica source in the sol-gel polymeric method used to produce amorphous silica nanoparticles (SiO2NPs). Sugarcane production process generates a huge amount of bagasse, an agricultural waste, which may constitute an environmental hazard if not properly managed. Due to sugarcane bagasse being a natural source of silica, this study produced silica nanoparticles as an alternate purpose for this waste. The present research uses L-cysteine hydrochloride monohydrate acid and citric acid to extract silica from sugarcane bagasse ash, with it being the first to utilize L-cysteine hydrochloride monohydrate in the sol-gel polymeric synthesis of silica. In order to study the extracted silica, XRD, FTIR, XRF, PIXE, SEM/EDS, TEM, and TGA were used to characterize the resulting silica xerogel in order to identify its elemental composition, surface morphology, functional groups, crystalline phases, thermal stability, and porosity. The purification techniques studied made it possible to produce silica nanoparticles with varying degrees of purity, a range of 67.25–94.58 wt% of silica xerogels were produced. Amorphous SiO2 nanoparticles were produced, with an average size of about 11–21 nm. The produced silica xerogel demonstrated porous network topologies with a high specific surface area and mesopore volume of 199.19, 469.09, 523.89 m2/g and 0.77, 0.64, 0.71 cm3/g using L-cysteine hydrochloride monohydrate, hydrochloric acid and citric acid respectively, according to the data. These materials, combined with morphological characteristics—large specific surface and pore volume—allow them to be used in a variety of technical domains, including biomedical, electronics, and optical.