AgriEngineering

This scoping review maps 224 empirical studies (205 from a structured Scopus search, 2020–2026, plus 19 from a targeted Central Asia supplement) across four digital technology domains for crop production: IoT and sensor-based systems, UAVs and remote sensing, machine learning and AI, and nanostructured agrochemicals. The review follows the PRISMA-ScR framework and pursues three research questions concerning documented effects and validation limitations (RQ1); cross-cutting barriers in human capital, data governance, and infrastructure (RQ2); and the state of empirical evidence from Central Asia and Kazakhstan relative to international findings (RQ3). Across all four domains, the strongest reported effects occur where the data-to-decision-to-action loop is closed and sustained over multiple seasons, yet most published metrics rest on single-season, single-site, or controlled-environment validation that overstates likely field portability. IoT and selected UAV and ML workflows are closest to operational readiness where maintenance, calibration, and advisory support are sustained. Nanostructured materials remain the least mature domain in agronomic terms. For Central Asia, foundational monitoring and salinity-oriented remote sensing are the most immediately transferable elements; intervention-grade ML and integrated digital systems require local calibration, extension infrastructure, and multi-season field validation that are largely still absent. The review identifies the digital skills gap, incomplete data governance, and underreported total cost of ownership as the principal institutional barriers to scaling. Policy priorities include shifting from technical pilots to multi-season agronomic proof, building intermediary service capacity, and establishing transparent data-governance frameworks before large-scale procurement. Full article

Accurate instance segmentation of oranges in complex orchard environments is crucial for obtaining clean regions of interest (ROIs). Coarse region extraction may include non-target pixels from leaves, shadows, background, and adjacent fruits, thereby increasing boundary pixel mixing in subsequent hyperspectral-assisted observation. This study proposes an improved lightweight YOLO11n-Seg method as an RGB-based visual front-end for cleaner single-fruit ROI extraction. Its contribution lies in the task-oriented integration of three complementary components: a Local Deformable Convolution Backbone (LDC-Backbone) for representing irregular and occluded fruit contours, a Boundary-Guided GSConv (BG-GSConv) module for efficiently fusing shallow boundary details with deep semantic features, and an ROI-Purity-Oriented Dice Boundary Loss for constraining mask integrity and boundary adherence. Evaluated on a complex orchard dataset, the improved model achieved a Mask mAP@0.5 of 0.962, a Mask mAP@0.5:0.95 of 0.692, a Box mAP@0.5 of 0.942, and an inference speed of 101 FPS with 3.20 M parameters. Background leakage analysis further showed that the proposed model reduced the inclusion of non-fruit pixels in extracted ROIs, supporting cleaner mask-based single-fruit region extraction. Preliminary ROI-based reflectance observation indicated that the reflectance curves obtained from the improved-model ROIs were closer to those of manually referenced pure ROIs than those obtained from the baseline extraction. These results suggest that the proposed method can serve as a real-time RGB-based front-end for cleaner single-fruit ROI extraction and later hyperspectral-assisted sampling. Complete closed-loop spectral quality modeling with paired RGB–HSI data remains a direction for future work. Full article

Agricultural pesticide wastewater represents a significant environmental and public health challenge, highlighting the need for scalable and resource-efficient treatment strategies. This review adopted a PRISMA-based methodology using the Scopus and Web of Science databases, leading to the analysis of 176 peer-reviewed studies published between 2014 and 2025. The selected literature was critically examined to assess pesticide wastewater treatment technologies, including adsorption, membrane filtration (MF), advanced oxidation processes (AOPs), biological treatments, and hybrid configurations. Particular attention was given to their treatment performance, scalability from farm to district level, resource recovery potential, economic feasibility, and life-cycle assessment (LCA) implications. Among the evaluated systems, hybrid configurations combining biological processes with AOPs or MF generally showed higher removal performance, often achieving more than 80% pesticide residue removal, while offering greater adaptability and compatibility with circular biorefinery frameworks. The review identifies key opportunities for resource recovery, including methane and hydrogen production, nutrient recycling, water reuse, and chemical reclamation, thereby supporting circular bioeconomy objectives. Overall, this review proposes an integrated, multiscale circular biorefinery perspective for sustainable pesticide wastewater management and identifies research priorities for developing resilient, safe, and resource-efficient agricultural water treatment systems. Full article

Winter feeding represents a significant cost in beef production, requiring efficient strategies that maintain productivity while minimizing environmental impact. Forty-six pregnant cows (620 ± 61 kg BW) were used to evaluate an ad libitum hay-based diet (2.02 Mcal/kg ME; HFOR; n = 23) versus a corn-based diet (2.84 Mcal/kg ME) limit-fed at 1.2% BW (HCON; n = 23) from 50 d pre-calving to 84 d post-calving. Pre- and post-calving, HCON cows consumed less (p < 0.01) dry matter, crude protein, and water than HFOR cows. While CH₄ yield per kg DMI was greater (p < 0.01) for HCON cows, total daily CH₄ emissions and CH₄ per unit of NEm intake were lower (p ≤ 0.03) compared with HFOR cows. Behavioral data showed that HCON cows had fewer (p < 0.01) meals and spent less time eating, but had greater intake per minute. Cow BW differed by treatment over time (p < 0.01), with HCON cows weighing less through early lactation, though no differences were observed from d 84 to weaning. Calf BW remained unaffected (p ≥ 0.76). In conclusion, limit-feeding a corn-based diet improves feed and water use efficiency and reduces enteric CH₄ emissions without compromising calf growth, offering a viable alternative to traditional forage-based wintering systems. Full article

Ablation and harvesting are among the most labor-intensive and physically demanding operations in oil palm cultivation, often resulting in significant drudgery and safety concerns when performed manually through climbing or pole-assisted methods. To overcome these challenges, a motorized backpack-type machine was developed and evaluated for its field performance, ergonomics, and economic feasibility. The machine met required quality standards and exhibited satisfactory performance under field conditions, achieving average ablation and harvesting capacities of 286 inflorescences per day and 4.115 t day⁻¹, with actual field capacities of 0.727 ha h⁻¹ (ablation), 0.516 ha h⁻¹ (sickle), and 0.537 ha h⁻¹ (chisel), and field efficiencies of 81.23%, 76.3%, and 79.91%, respectively. Ergonomic evaluation indicated that operation of the machine falls within a moderate workload category, thereby reducing operator fatigue compared to manual methods. Economic analysis further revealed that the cost of operation was substantially reduced to 3.02 USD t⁻¹ and 60.40 USD ha⁻¹ year⁻¹, resulting in increased harvester earnings of 174.72% and 64.83% compared to climbing and pole harvesting methods, respectively. These findings demonstrate that the motorized backpack machine is a practical, efficient, and economically viable alternative to traditional techniques and minimizes drudgery while improving productivity and profitability in oil palm plantations. Full article

The geographical origin of aniseed (Pimpinella anisum L.) represents a key quality determinant, as it directly influences the chemical composition and commercial value of its essential oil. Agronomic traits of aniseed (plant height, umbel diameter, number of umbels per plant), productivity-related traits (number of seeds, thousand-seed weight, yield per plant, plant biomass, harvest index, yield per hectare, essential oil content and yield), and physiological traits (germination energy and total germination) exhibit variations depending on geographical origin. The study proposes an integrated framework for accurate classification by combining agronomic, productivity, and physiological data with GC-MS profiles and advanced machine learning (ML) techniques. A total of 144 samples were analyzed, based on a factorial design including three locations, six fertilizer treatments, two years, and four replications. trans-Anethole was the dominant compound in all samples (89.508–101.441%). Several classification models, including artificial neural networks, random forests, MARSplines, boosted trees, interactive trees, naïve Bayes, and support vector machines, were evaluated to discriminate samples by geographical origin using agro-meteorological and GC-MS data. The results indicate that AI and ML approaches effectively captured complex non-linear relationships. Overall, the multi-model framework highlights the strong potential of machine learning for agro-food authentication, supporting improved traceability, site-specific decision-making, and quality control. Full article

This study developed and evaluated a cloud-based smart irrigation platform (DSmart Farming) integrating low-cost sensors and IoT technology for automated irrigation control in community greenhouses of Puen Jai Insee, organic group in Sa Kaeo Province. The system combined soil moisture, air temperature, and relative humidity sensors, with a LoRa32-based control unit in each greenhouse and a central web-based management application linked to a MariaDB database on a cloud server. Five vegetable crops, including cherry tomato, broccoli, cabbage, Chinese kale, and kale, were grown over two distinct seasons under four irrigation strategies in a completely randomized design with three replications: three smart irrigation treatments based on soil moisture thresholds (on/off at 40/50%, 45/55%, and 50/60%) and a farmer-managed conventional irrigation control. The smart irrigation system maintained root-zone moisture within the target range (approximately 50–60%) and moderated greenhouse microclimate, preventing daytime temperatures from exceeding 40 °C, in contrast to 40–45 °C peaks in the conventional greenhouses. Across crops, smart irrigation increased yields by 20–29% while reducing water use by 41–60% compared to conventional practice, leading to income increases of 20–56%, depending on the crop. Bacterial soft rot caused by Pectobacterium carotovorum subsp. carotovorum occurred only under conventional irrigation, whereas no soft rot or other major diseases were detected in smart-irrigated greenhouses. These results demonstrate that the DSmart Farming system can enhance water use efficiency, avoid disease incidence, and improve the productivity and profitability of organic greenhouse vegetable production in water-limited smallholder systems. Full article

Climate change poses a significant challenge to food security, as it alters crop productivity, distribution patterns, and the overall food supply. This study modelled the emergence of Amaranthus hybridus L. in bean (Phaseolus vulgaris L.) and maize (Zea mays L.) production systems in the Brazilian state of Minas Gerais, in the cities of Coimbra, Paracatu, São João del-Rei, and Uberaba, under the Coupled Model Intercomparison Project Phase 6 (CMIP6) SSP1-2.6 and SSP5-8.5 scenarios. Using Hydrothermal Time (HTT), computational modelling, and nonlinear Weibull regression, weed emergence was simulated under current and future climate scenarios for 2050 and 2070. Although biological triggers such as temperature and base water potential remain constant, higher average temperatures accelerate HTT accumulation. Thus, this results in earlier and more intense emergence flows. The highest and lowest cumulative emergence were observed in Uberaba and Paracatu, respectively. The SSP5-8.5 scenario projects high emergence windows for 2070. This reduces the time available for management interventions. The root-mean-square error (RMSE) associated with the coefficient of determination (R²) of the models validates HTT as an essential tool in computational agriculture. The integration of these models into decision-support systems is essential to mitigating productivity losses and it will increase control efficiency amid future climate uncertainties. Full article

Agricultural soils are a major source of greenhouse gas (GHG) emissions, particularly carbon dioxide (CO₂) and methane (CH₄), highlighting the need for cost-effective and field-applicable monitoring solutions. This study developed and evaluated a low-cost real-time monitoring system for soil CO₂ and CH₄ emissions by integrating surface emission chambers, low-cost gas sensors, a solar-powered energy supply, and IoT-based wireless communication. Three acrylic chambers with different heights (40, 60, and 80 cm) were fabricated to investigate the influence of chamber geometry on measurement performance. System performance was assessed through simultaneous measurements against a Biogas 5000 analyzer under simulated conditions and during field deployment in a sugarcane cultivation area in Khon Kaen Province, Thailand. Relative agreement was used to compare the developed system with the reference instrument. The results showed that relative agreement varied with chamber height for both gases. Under simulated conditions, the 80 cm chamber achieved the highest overall relative agreement for CO₂ and CH₄, underscoring the importance of sufficient headspace volume in chamber-based measurements. Field experiments confirmed the system’s capability for continuous CO₂ monitoring in an agricultural environment. However, CH₄ emissions were not detected during the study period, likely due to drought-induced, well-aerated soil conditions. The developed system demonstrated stable autonomous operation, low energy consumption, and ease of installation, making it suitable for long-term field applications. Overall, the proposed platform provides a practical and scalable approach for real-time soil GHG monitoring and offers strong potential for integration into precision agriculture and climate-smart farming systems to support GHG mitigation strategies. Full article

Integrating semi-transparent photovoltaics (STPVs) into greenhouses offers a dual-use solution for land efficiency, although matching electricity generation with crop spectral needs remains a challenge. To address this, this study assesses the optical and microclimatic impact of Cadmium Telluride (CdTe, 50% transparency) and amorphous Silicon (a-Si, 20%) technologies compared to a conventional control in a semi-arid Mediterranean climate. Spectral analysis revealed that CdTe aligned with chlorophyll absorption peaks, preserving a transparency window that yielded a 66% relative gain in biologically useful radiation over the blue-blocking a-Si. Furthermore, while both technologies significantly reduced Photosynthetically Active Radiation (PAR), this shading served as a protective filter against supra-optimal irradiance, stabilizing the internal microclimate. In the control prototype, extreme vapour pressure deficits (VPDs approaching 9.0 kPa) drove maximum reference evapotranspiration (ET₀) above 4.6 mm/day. In contrast, the STPV systems effectively capped ET₀ at approximately 3.09 mm/day (CdTe) and 1.64 mm/day (a-Si) through their radiative attenuation, despite internal VPDs still reaching 6.5–7.0 kPa during peak summer. This decoupling resulted in drastic average ET₀ reductions of 31.4% and 61.3%, respectively, while mitigating soil overheating by up to 17.8%. These findings demonstrate that specific STPV technologies transcend mere shading to function as passive climate resilience tools, naturally enforcing water conservation and physically disarming atmospheric aridity in high-radiation environments. Full article

The drying kinetics of three varieties of cassava were evaluated in a tray dryer, using a completely randomized design with a three-factor factorial arrangement: temperature (50, 60, and 70 °C), air velocity (1, 2, and 3 m/s), and variety (“Blanca Mona”, “Ica Negrita”, “Venezolana”), with three replicates per treatment. The results obtained were used to construct drying curves, which showed that this process occurred in the decreasing period. The drying curves were adjusted to mathematical models, and the Page model was the best fit to the experimental data with R²_adj values closer to 1 and RSS values less than 0.0086. The effective diffusivities (D_eff) in cassava flours were represented by the Arrhenius equation with values ranging from 5.24 × 10⁻¹⁰ to 1.58 × 10⁻⁹ m²/s. The activation energy (E_a) recorded values between 20.34 and 28.32 kJ/mol. The flours from the three cassava varieties were obtained under the best drying conditions (70 °C and 3 m/s). The physicochemical characterization of fresh roots and flours from three cassava varieties revealed significant genotype-dependent differences in their proximal composition. Blanca Mona exhibited the highest ash content and the lowest total carbohydrates among fresh roots, while Ica Negrita stood out for its superior crude fiber content in flour. Venezolana flour stood out for its higher protein content (3.86 ± 0.04 g/100 g) and significant fiber content (1.39 ± 0.39 g/100 g), making it the flour with the best nutritional profile and greatest potential for food applications. Therefore, tray drying is recommended as one of the suitable methods for cassava flour production. Full article

China boasts abundant straw resources but grapples with notable challenges in straw processing: returning straw to fields can lead to soil compaction and aggravated pests/diseases, while baled straw for off-field storage and transportation tends to scatter. Additionally, domestic netting technology for square bales remains underdeveloped, and imported equipment is ill-suited for small-scale farmers. To tackle these issues, this study developed an integrated straw baling and netting machine by modifying the 9YFSG-2.2 square straw baler. It integrates a conveying mechanism, an offset crank–connecting rod compression mechanism (300 mm crank, 885 mm connecting rod), a two-stage gear-driven net-wrapping mechanism (with hollowed-out large gears for weight reduction), and a sensor-controlled net-cutting device, forming a complete workflow of “straw pick-up–shredding–conveying–compaction–net wrapping–net cutting”. Via coupled simulation using RecurDyn 2019, EDEM 2020, and ANSYS Workbench 2018, straw particles were modeled as 28-mm-long segments (composed of three 7 mm spheres). Simulations showed straw compaction in 0.48 s, with the compression chamber and plate having equivalent stresses of 0.2767 MPa and 173.44 MPa and maximum deformations of 0.0012 mm and 0.66 mm—both well below structural steel’s yield strength. Field tests in Xinxiang, Henan (straw moisture 30.03%), yielded results exceeding standards: 99.4% bale formation rate, 96% regular bale rate, 93% drop resistance rate, 170 kg/m³ bale density, and 12 s per bale efficiency. Controlling netting time further boosted efficiency and reduced consumption, successfully realizing integrated straw baling and netting. Full article

Precise mapping of rice cultivars is of great significance for crop management and food security evaluation. Nevertheless, differentiating between Indica and Japonica rice remains a formidable task, mainly due to subtle discrepancies in spectral characteristics and scattered planting distributions. This study evaluated the synergistic effect of spatial resolution and red edge information in rice variety classification using PlanetScope (PS) and Sentinel-2 (S2) images from the Tillering and Jointing stage, Heading and Flowering stage in Huai’an, Jiangsu Province. Multiple feature schemes were constructed, including spectral bands, vegetation indices, and texture features, with and without red-edge variables. A total of eight feature schemes have been constructed, including spectral bands, vegetation index, texture features, and red edge features. The feature scheme division is based on the participation of different sensors, growth periods, and red edges. We fine-tune three classification models, Random Forest (RF), Light Gradient Boosting Machine (LightGBM), and TabNet, to enhance classification performance. Additionally, we employ Shapley Additive Explanations (SHAP) to quantitatively measure the contribution of each feature to the prediction of distinct rice varieties. Results demonstrate that classification accuracy of different sensors reach the highest at the Heading and Flowering stage. The overall accuracy of PS scheme is 98.14%, the F1 scores of Japonica and Indica rice are 97.67% and 98.41%, the overall accuracy of S2 scheme is 97.87%, and the F1 scores of Japonica and Indica rice are 98.62% and 98.68, respectively. Incorporating red-edge features leads to a notable improvement in F1-scores for both Indica and Japonica rice under all experimental configurations. Although PS only has one red edge band set, its classification performance is similar to S2, and the boundaries between different rice variety recognition results and between non rice and rice plots are more refined compared to S2. Feature attribution analysis reveals that red-edge indices exert a dominant influence on the decision-making process of the models, especially during the Heading–Flowering period. These findings suggest that high-accuracy discrimination of rice varieties relies heavily on the synergistic optimization of phenological timing, red-edge spectral information, and spatial resolution, rather than merely increasing spectral dimensionality. The optimization direction for high-precision rice variety mapping in the future should prioritize the collaborative mechanism of phenological period, red edge data, and spatial resolution, rather than being limited to simple stacking in the spectral dimension. Full article

Yield monitoring in forage production is typically limited to chopping or baling operations, where spatial resolution is often reduced by windrow merging. This study evaluated the feasibility of estimating mass flow rate (MFR) and generating spatial yield maps at the mowing stage using sensors integrated into a windrower. Conditioning roll speed, swath shield impact force, and the displacement of spring-loaded vanes (fingers) in the crop flow were evaluated during alfalfa harvest and calibrated against measured MFR. Model performance was assessed using cross-validation, and spatial fidelity was evaluated using experimental variograms and kriged yield maps. The average MFR was 19 kg·s⁻¹ with a range of 4 to 55 kg·s⁻¹. Conditioning roll speed provided the most robust and transferable predictor of MFR (R² = 0.89, RMSE = 3.4 kg·s⁻¹), consistently outperforming impact force (R² = 0.70, RMSE = 1.9 kg·s⁻¹) and finger displacement (R² = 0.82, RMSE = 4.3 kg·s⁻¹), which were more sensitive to machine dynamics and sensor placement. Validation of the roll-speed model using an independent dataset resulted in an R² = 0.87 and RMSE of 2.62 kg·s⁻¹. Yield maps derived from roll-speed-based models exhibited clear spatial structure with correlation lengths of approximately 25–40 m, whereas the finger displacement model exhibited higher nugget effects. Yield mapping with the forage harvester showed reduced spatial fidelity compared to mowing stage estimates, as windrow merging prior to chopping caused spatial averaging that diminished recoverable fine-scale yield variability. These results demonstrate that yield monitoring at the mowing stage enabled yield estimates to complement downstream harvest data and improve characterization of within-field yield variability. Full article

The analysis of soil carbon helps various sectors, including agriculture, in the context of monitoring soil health. In precision agriculture, decisions are made on the basis of site-specific information and thus have the potential to increase crop productivity more than is possible with traditional high-input agriculture. Site-specific information-based nutrition management, pest and disease management, and water management are the main areas of interest in the era of precision agriculture. Soil organic carbon (SOC) is one of the main components of the carbon cycle and impacts soil physical and chemical properties. Soil color is considered an indicator of soil carbon. In relation to soil physical properties, soil color has been used to determine SOC level and classification throughout history in a qualitative manner, and recently, researchers have shown interest in relating soil color data to quantify soil chemical properties. From spectroscopy-based color analysis to image-based color analysis, research has shown strong relationships between SOC and color properties. Therefore, with the improvement of technology to create smaller and portable sensors, the potential exists to automate the processes of soil chemical analysis to use them in precision agriculture. Two of the major limitations of these methodologies in research are the number of known soil samples required to calibrate a model (the majority of the models require more than 100 samples) and the use of expensive spectrometers with complex processes. Thus, the potential of individual farmers to deploy these methods is limited. This research was conducted to develop a methodology with complete guidelines and a set of tools to allow farmers to analyze SOC themselves. Furthermore, by encouraging farmers to analyze their farmland soils for SOC and update the data, the research enables them to potentially use this information to manage their agronomic practices, including the addition of organic fertilizer to reduce soil carbon pool inefficiencies and decisions regarding the mode of tillage and water management. During this research, three sensors and different combinations of sensors were used to capture soil surface color, temperature, and reflectance and were considered for model development. The highest-model-fit equation was obtained from the thermal image and red, green, and blue (RGB) image combinations (R² = 0.65 and MSE = 0.0335). The variables used for X from the color models were hue values and redness (a), and those from the thermal image minimum and maximum temperature data were used. Finally, using a regression equation along with the image data and SOC data from the chemical analysis, a farmer-feedback-based SOC prediction model was developed. Full article

Population growth, climate change, and increasing pressure on water and nitrogen resources pose major challenges for sustainable maize production. Maize yield is highly sensitive to inter-annual weather variability, yet many prediction approaches still rely on simple linear relationships and rarely integrate SPAD (Soil Plant Analysis Development)-based crop diagnostics with machine learning in multi-year nitrogen × irrigation experiments. In a three-year field experiment (2018–2020) in Hungary, we evaluated how basal and top-dressing fertilization and supplemental irrigation under contrasting water supply conditions affected the chlorophyll status and grain yield of a maize hybrid. Relative chlorophyll content was monitored using SPAD measurements at key phenological stages (V6, V12, and R1), and a multilayer perceptron (MLP) model was developed to improve yield prediction and to identify informative combinations of input variables. Five alternative scenarios (SC1–SC5) were tested by combining SPAD values with the fertilization rate, irrigation status, and crop year in different configurations, and model performance was assessed using root mean square deviation (RMSD), mean absolute error (MAE), normalized root mean square error (NRMSE), correlation (r, r²), Nash–Sutcliffe efficiency (NSE), Kling–Gupta efficiency (KGE), Kendall’s tau, and the index of agreement (d). Overall, SC4 (SPAD + fertilization + crop year + irrigation) achieved the best agreement with observed yields across most indices (e.g., r ≈ 0.93, NSE ≈ 0.86, KGE ≈ 0.90), whereas SC2 (SPAD + fertilization) produced the lowest prediction error on the independent test subset, indicating the most robust generalization. Basal fertilization with 60 and 120 kg N ha⁻¹ significantly increased yield in 2019 and 2020, while irrigation generally enhanced yield except for the 30 kg N ha⁻¹ top dressing applied at the V6–V12 stages. These results demonstrate that coupling SPAD measurements with MLP modeling and multi-criteria performance evaluation can support more efficient, site-specific nitrogen and irrigation decisions and help stabilize maize yields under variable climatic conditions. Full article

The application of boron in soybeans in Oxisols of the Brazilian Cerrado is frequently integrated into complex tank fertilizer mixtures with multiple components via foliar application. This study investigated the interactive effects of varying spray application rates (40, 70, 100, and 130 L ha⁻¹) and adjuvant types (organosilicone surfactant; methylated seed oil; and a water control) on boron deposition and the resulting physiological status. The organosilicone surfactant provided superior technical stability and deposition efficiency, allowing for a reduction in application rates to volumes between 40 and 70 L ha⁻¹ maintaining a stable foliar B status across the evaluated range. In contrast, the performance of the methylated oil was strictly dependent on physical deposition, being effective only at intermediate rates, while the use of water alone represented a high risk of technical failure at reduced volumes. Furthermore, the NDRE index proved to be more responsive and robust than NDVI for monitoring delivery efficiency in high-density canopies, as it avoided signal saturation. Finally, Multivariate Analysis helped to observe that soybean yield in the Cerrado is primarily governed by the mitigation of water and thermal stress (TVDI), with optimized boron application acting as a key facilitator of reproductive success and yield stability under these environmental constraints. Full article

Tomato leaf diseases represent a persistent threat to global food security, causing annual crop losses of 20% to 40%. Although deep learning models achieve accuracies exceeding 95% in centralized settings, their deployment across distributed farms is constrained by data privacy concerns, communication bottlenecks, and heterogeneous data quality. This paper proposes Personalized, Clustered, and Communication-Efficient Federated Learning (PCE-FL), a framework that integrates three synergistic components: (1) server-side client clustering to group farms with similar data distributions for personalized model training; (2) federated knowledge distillation to reduce communication overhead by over 91%; and (3) reputation-based aggregation to ensure robustness against unreliable contributions. Extensive experiments on realistic non-IID simulations of the PlantVillage tomato dataset Dirichlet($\alpha \in \{1.0,0.5,0.1\}$) demonstrate that PCE-FL achieves 89.1% accuracy under extreme heterogeneity ($\alpha =0.1$), surpassing FedAvg by 10.9 and IFCA by 4.8 percentage points, while maintaining a 91% reduction in communication cost. All improvements are statistically significant ($p<0.001$). These results advance the practical deployment of privacy-preserving collaborative AI in resource-constrained agricultural environments. Full article

The study addresses the challenge of accurately simulating and visualising the kinematics of agricultural machinery during field operations. The research is motivated by the current lack of comprehensive guidelines for selecting optimal movement and turning modes under varying forward speeds, working widths, and field geometries. A spreadsheet-based environment was utilised to perform simultaneous kinematic simulation and trajectory visualisation. Turning manoeuvres were modelled using smooth composite curves, consisting of straight segments, clothoids, and circular arcs, with trajectories represented in a Cartesian coordinate system through geometric transformations including translation, rotation, and mirror symmetry. Continuity between curve elements was ensured by dimensional chains linking abscissas, ordinates, and direction angles at their start and end points. The influence of key operational factors—forward speed, angular turning velocity, working direction, and field boundaries—was evaluated for a range of turn types, including semicircle, pear-shaped, figure-eight, side exit, U-turn, and P-turn manoeuvres. Field experiments conducted on selected patterns confirmed that the proposed approach can reproduce actual trajectories with sufficient practical accuracy. These results demonstrate that spreadsheet-based kinematic modelling is a robust and accessible tool for optimising tractor–implement movement, enhancing operational planning, and providing a reliable framework for further research into machinery performance under complex field conditions. Full article