The logic behind distributed machine learning is deceptively simple. In the traditional model, all data from sensors, drones, and farm management systems is shipped to a central server where a single model is trained. That architecture carries well-known weaknesses: a single point of failure, bandwidth latency, and acute privacy risks, since consolidating records about crop production, animal health, and economic yields creates a tempting target for massive breaches. Distributed learning inverts this model. Each participant trains a model locally on their own devices, and only the learned parameters, not the raw data, travel across the network. A central server, or in some architectures no server at all, then combines these updates into a shared global model. The result is enhanced privacy, security, efficiency, and scalability, qualities that matter enormously in a sector where farmers increasingly worry about data asymmetry and how agricultural technology providers store and exploit their information.

To capture this rapidly evolving landscape, the researchers followed the systematic literature review guidelines of Kitchenham and colleagues, ensuring the process was replicable, unbiased, thorough, and transparent. They searched six major databases, including IEEE Xplore, ACM Digital Library, Scopus, ScienceDirect, SpringerLink, and Wiley Online Library, covering publications from 2015 to 2025. The initial automated search returned 1,864 articles. After screening against strict exclusion criteria, only journal articles were considered, and titles, abstracts, and full texts were evaluated, 43 studies survived. A quality assessment on an eight-point scale, scored independently by all three authors and reconciled through consensus meetings, trimmed the set to 41 primary studies. Notably, no study published before 2020 met the criteria, a striking signal that distributed learning in agriculture is a genuinely recent phenomenon that has accelerated sharply in the past five years.

The review’s findings reveal a field with clear centers of gravity. Crop management dominates the application landscape, appearing far more frequently than any other domain, with livestock monitoring, soil and water management, and general smart farming frameworks trailing behind. Within these domains, the researchers identified seventeen distinct tasks, ranging from disease prediction and classification, crop classification, and crop recommendation to weed segmentation, yield prediction, resource optimization, evapotranspiration estimation, bird detection, milk quality prediction, and even prawn growth inspection. Disease prediction and classification emerged as the most common objective, reflecting the enormous value of catching plant pathogens early. One study, a federated explainable AI framework, tackled the widest range of tasks simultaneously, spanning disease prediction, crop health monitoring, yield prediction, resource optimization, and anomaly detection.

Among the distributed learning paradigms, federated learning stands overwhelmingly supreme, appearing across every one of the seventeen identified tasks. It is followed by decentralized peer-to-peer systems, hierarchical distributed learning, and blockchain-based approaches, with two studies proposing hybrids that combine federated learning with blockchain for applications such as pest detection. Hierarchical models insert intermediate edge servers between local devices and the cloud to reduce communication overhead and latency. Decentralized peer-to-peer architectures remove the central coordinator entirely, with nodes exchanging updates only with their direct neighbors, making performance highly dependent on network topology. Blockchain-based systems replace the central server with an immutable distributed ledger, using smart contracts to automatically verify, validate, and aggregate model updates. Each approach embodies a trade-off between control and resilience: centralized systems are simpler and faster to manage but collapse if the main server fails, while decentralized designs keep running even when individual devices drop offline, at the cost of greater network traffic and slower synchronization.

The architectural details expose an interesting tension. Although federated learning distributes the data, most implementations still rely on centralized client-server topologies, which appeared in 27 of the reviewed studies, compared with 12 using hierarchical tree architectures and only 2 using peer-to-peer or mesh designs. Governance follows architecture: centralized authority and server-dictated models dominate, in which a single controlling entity manages data flows and decision-making. Yet the review uncovered notable exceptions, including a client-server system that adopted collaborative, consensus-based governance by combining its architecture with blockchain, requiring agreement among local nodes before updating the global model. Aggregation strategies cluster around Federated Averaging, used in 27 studies, followed by custom weighted averaging and Federated Proximal Aggregation, with more exotic variants such as SCAFFOLD, FedBN, and QFedAvg appearing once or twice each.

On the data side, images reign as the most common input, which aligns with the dominance of deep learning algorithms, particularly convolutional neural networks such as YOLO, ResNet, VGG, MobileNet, and Inception, alongside transformers like Vision Transformer and Swin Transformer. Classical machine learning methods, including support vector machines, random forests, decision trees, and k-nearest neighbors, handle tabular, sensor, and time-series data. Implementation leans on a familiar toolkit: TensorFlow leads with 12 studies, followed by Keras, Flower, and PyTorch, with TensorFlow Federated, PySyft, MATLAB, scikit-learn, iFogSim, Horovod, and Open MPI filling out the ecosystem. Hardware ranges from Raspberry Pi, Jetson, and Arduino edge devices to NVIDIA GPUs and cloud platforms, underscoring that image-heavy distributed learning still demands serious computing resources.

Perhaps the most valuable contribution of the review is its honest accounting of what still goes wrong. Three challenges recur across the literature. Privacy leakage persists because shared model updates, gradients and weights, carry mathematical footprints of private data, enabling sophisticated attacks such as gradient reconstruction, model inversion, and membership inference. Communication bottlenecks arise because continuous model updates from thousands of devices can strain network transmission far beyond local computing speeds, stretching training from hours to weeks, especially under unstable bandwidth and intermittent rural connectivity. Data heterogeneity, the notorious non-IID problem, means each farm’s data is statistically unique and skewed, so a global model can underperform local models and quietly favor majority clients, creating unfair bias against smaller participants.

The solutions documented in the literature are correspondingly diverse. Against privacy threats, researchers deploy differential privacy, which injects noise into data and updates, along with homomorphic encryption, secure multi-party computation, gradient encryption, and blockchain frameworks augmented with decentralized identifiers and verifiable credentials. To tame communication costs, the field turns to gradient compression, quantization, binary compression, knowledge distillation, adaptive and asynchronous transmission, lightweight architectures, and model pruning. For non-IID data, promising remedies include advanced aggregation algorithms such as FedProx and SCAFFOLD, adaptive weighting, data refinement techniques like resampling, zero-padding, and augmentation, and ensemble or prototype-based learning to boost robustness. The authors conclude by proposing a multi-dimensional taxonomy, categorizing systems by domain, architecture, learning method, infrastructure, and challenges, and call for reference architectures and for extending distributed learning beyond crops into livestock monitoring and precision aquaculture. As farming becomes ever more data-driven, this review makes clear that the future of agricultural AI may depend less on how much data is collected, and more on how intelligently, and safely, it is shared.

Subject of Research: Systematic literature review of distributed machine learning methods, applications, architectures, and challenges in precision agriculture

Article Title: Distributed machine learning in precision agriculture: methods, applications, and challenges

Article References: Krisnawijaya, N. N. K., Tekinerdogan, B., & Catal, C. (2026). Distributed machine learning in precision agriculture: methods, applications, and challenges. Smart Agricultural Technology, 15, Article 102527. https://doi.org/10.1016/j.atech.2026.102527

Image Credits: AI Generated

DOI: 10.1016/j.atech.2026.102527

Keywords: distributed machine learning, federated learning, precision agriculture, smart farming, data privacy, crop management, non-IID data, edge computing, blockchain, systematic literature review, differential privacy, agricultural IoT

The scale of the evidence base is striking. Following the PRISMA 2020 reporting framework, the team searched ScienceDirect, PubMed, IEEE Xplore, Google Scholar, and MDPI for English-language peer-reviewed work published between January 2021 and December 2025, retrieving 13,926 records. After deduplication, title and abstract screening, and full-text eligibility assessment of 198 articles, 95 studies survived the cut. Quality ran high: 51.58 percent of included papers appeared in first-quartile journals and another 36.84 percent in second-quartile venues, with Journal Citation Reports impact factors ranging from 2.2 to 14.0. The topical map shows where the intelligence is concentrating. Environmental monitoring accounted for 23.15 percent of publications, greenhouse control and protected cultivation for 21.05 percent, disease and pest detection for 21.05 percent, hydroponics for 16.84 percent, and yield prediction with production forecasting for 13.68 percent. Methodologically, classical machine learning anchored 25.26 percent of the studies, IoT and AIoT sensor systems 22.11 percent, and deep learning and computer vision 13.68 percent, a sign that pragmatic, sensor-driven systems currently outweigh frontier models in the field.

At the foundation sits machine learning, in which computers learn patterns from data rather than following explicitly programmed rules. The review organizes these algorithms into supervised, unsupervised, and reinforcement learning families and shows how workhorse models such as decision trees, support vector machines, random forests, and artificial neural networks now digest soil nutrient profiles, fertilizer inputs, and crop characteristics to make site-specific recommendations. In one line of work, random forests, support vector machines, and neural networks predicted optimal water requirements for maize, soybean, and tomato across diverse climates and soils, outperforming conventional irrigation scheduling. Transparency is emerging as a design priority. When researchers applied SHAP and LIME, techniques that reveal which variables drive a model’s predictions, to 15 greenhouse-grown cabbage plants tracked over an 85-day period, they could see that leaf count and plant height strengthened forecasts of nitrogen, phosphorus, and potassium levels while days after planting and average leaf area weakened them. Such explainable AI, the authors argue, is becoming essential for earning the trust of farmers expected to act on algorithmic advice.

Above that layer operates deep learning, multi-layer neural networks that automatically discover hierarchical features in raw data without hand-engineered rules. Convolutional neural networks, the workhorses of agricultural computer vision, learn to recognize edges, textures, and lesions in images streaming from cameras, drones, satellites, and field sensors, enabling real-time monitoring of plant health, growth analysis, and yield estimation. The review documents multimodal systems that fuse hyperspectral and X-ray imaging to predict seed viability non-destructively, alongside architectures such as VGG16, YOLO, Mask R-CNN, and transformer-based models that classify weeds, detect plant stress, and recognize growth stages. The techniques even reach into farm physics: one team built a three-dimensional computational fluid dynamics model to optimize airflow velocity, temperature distribution, and relative humidity across the cultivation trays of an indoor vertical farm, reporting improved airflow uniformity, better thermal control, and higher energy efficiency. Hybrid designs push further still. Convolutional layers coupled with bidirectional long short-term memory networks capture spatial patterns and temporal dynamics together, and a grey-wolf-optimized BiLSTM model forecast greenhouse temperatures with a coefficient of determination of 0.97.

Binding these computational layers together is the Internet of Things, networks of sensors, microcontrollers, and communication modules that measure soil moisture, temperature, humidity, pH, electrical conductivity, light intensity, and nutrient levels, then stream the data over Wi-Fi, LoRa, Zigbee, or cellular links to cloud and edge computing platforms. When AI sits atop those streams, researchers call it AIoT: sensors feed predictive models, models command actuators such as pumps, valves, LED arrays, and climate systems, and the farm adjusts itself in real time. The review catalogs the hardware ecosystem, from Arduino, ESP32, and Raspberry Pi controllers to LoRaWAN gateways and mobile dashboards. One LoRaWAN-based subsurface drip irrigation system, triggered when soil moisture fell to 12 percent, enhanced tomato seedling growth while cutting water use by 10 percent and lifting yield to 1,243 grams per plant. Edge computing is shrinking latency and energy costs further, and digital twins, virtual replicas of physical farms, now let growers simulate crop growth, microclimate behavior, and management scenarios before committing a single resource in the real world.

Nowhere is the transformation more visible than in controlled-environment agriculture, the umbrella term for greenhouses, vertical farms, and plant factories where light, temperature, humidity, and nutrition are engineered rather than endured. The review describes greenhouses that have evolved into quasi-autonomous organisms: one IoT-enabled system used artificial neural networks and regression models combined with image analysis to predict lettuce growth, harvest timing, and crop quality from environmental variables, while another deployed long short-term memory networks to forecast actuator behavior from live sensor data and adapt the climate automatically. Vertical farming, which stacks crops in layers to squeeze productivity from land-scarce cities, has embraced the digital twin approach, with one framework using a genetic algorithm to continuously retune RGB LED lighting against measured plant performance; the adaptive strategy consistently beat static configurations. Hydroponics, the soilless cultivation of roots bathed in nutrient solution, offers up to 90 percent water savings through closed-loop recycling, and machine learning and deep learning models now predict lettuce growth, flag abnormal plant conditions, and, with explainable AI, optimize even Thai basil production.

Water may be AI’s most dramatic success story. As climate change tightens the screws on global freshwater supplies, the review found smart irrigation systems consistently outperforming manual scheduling. An Arduino-based autonomous irrigation rig pairing soil-moisture and air-humidity sensors with an adaptive machine-learning controller achieved a water-use efficiency of 87.97 percent alongside higher plant survival, more uniform growth, and better visual quality than conventional watering. An AI-integrated framework combining IoT sensor networks with predictive analytics reached roughly 80 percent decision accuracy and simulated water-use efficiency gains of up to 25 percent, while a K-nearest-neighbors model predicting irrigation needs hit 98.3 percent accuracy in field trials. In a cloud-deployed greenhouse system for cherry tomatoes, models spanning multiple linear regression, support vector machines, random forests, extreme gradient boosting, and recurrent architectures predicted optimal irrigation with coefficients of determination around 0.82 to 0.83, translating into 17.8 percent heavier individual fruits and a 20.7 percent improvement in water-use efficiency under real growing conditions.

Computer vision is closing in on the pathogens, pests, and post-harvest logistics that drain harvests. An enhanced Faster R-CNN with multiscale feature fusion detected strawberry diseases at 92.18 percent mean average precision, and a pruned YOLOv5s model classified melon leaf diseases at 96.7 percent mAP with real-time inference. The most striking figure belongs to a hydroponic lettuce system called CNN-WOPNet, which married a ParNet-attention-enhanced convolutional network to a Walrus Optimization Algorithm and achieved 99.54 percent classification accuracy, 99.60 percent precision, and a 99.61 percent F1-score in identifying leaf diseases under extreme environmental conditions. Robotics is following the same trajectory. A tomato-monitoring robot using RGB-D cameras, LiDAR, and Faster R-CNN located fruit at 88.6 percent accuracy even with obscured samples, a YOLOv5-powered machine graded greenhouse mushrooms by size with 96 percent accuracy, and a vision-guided sowing robot cut vegetable planting operation time by 51 percent. Meanwhile, hybrid CNN-LSTM models fusing spatial and temporal features are sharpening soybean yield forecasts built from climate indicators and hydrological records, improving logistics planning and reducing waste.

But the review refuses to celebrate uncritically, and its quality appraisal exposes a soft underbelly beneath the headline numbers. Most included studies validated their models only through train-test splits or k-fold cross-validation on datasets from a single location, season, or production system; independent external validation, the gold standard for proving that a model generalizes, was rare. That means the published accuracies may overestimate real-world applicability. The authors also flag a metrics obsession: studies routinely report accuracy, precision, recall, F1-score, intersection-over-union, mean average precision, and RMSE, while computational efficiency, inference time, robustness under field conditions, and long-term operational reliability go largely unmeasured. Only a small fraction of the systems were ever tested in genuine agricultural environments, with many remaining at the laboratory or prototype stage. The team rated the overall risk of bias as low to moderate, stemming from dataset selection, controlled validation environments, limited external testing, and a publication bias that favors high-performing models, a reminder that failure cases, which matter most to practicing farmers, rarely make it into print.

The path forward, the authors argue, runs through hybridization and hardening. Multimodal systems combining machine learning, deep learning, and IoT still represent only 5.26 percent of the literature, yet they consistently post the strongest results, and AI-enabled edge devices are already delivering real-time pest and disease detection with high accuracy, low latency, and reduced energy consumption. Scaling those wins will demand standardized benchmarks, field-scale trials across locations and seasons, energy-efficient models that run on inexpensive hardware, explainability to win farmer trust, and robust data security for farms that have become, in effect, distributed computing networks. The review also ties the technology to crop genetics: AI-driven high-throughput phenotyping using RGB, hyperspectral, thermal, fluorescence, and 3D imaging is accelerating the identification of yield, stress, and disease-resistance traits that genomic tools can then target for breeding. What emerges is a portrait of agriculture in mid-transformation, no longer purely a craft of soil and weather and not yet a fully autonomous industry, but unmistakably becoming an information science. The success of that transition will help decide who eats in 2050.