Published Paper Details:

Precision agriculture is moving from isolated sensing and threshold-based automation toward integrated platforms that jointly interpret soil condition, irrigation-water quality, weather variability, crop response, and disease pressure. This paper presents a structured integrative review and an implementable framework for a Plant and Ground Monitoring System (PGMS) that combines calibrated field sensing, edge intelligence, cloud analytics, remote sensing, machine-learning inference, and a multilingual web dashboard. The review synthesizes evidence on soil nutrient monitoring, crop and fertilizer recommendation, evapotranspiration-informed irrigation, image-based plant disease diagnosis, and edge-cloud deployment. The resulting architecture uses an ESP32-S3 field node to acquire soil moisture, soil pH/electrical conductivity/nutrients, irrigation-water pH/electrical conductivity, atmospheric measurements, and leaf imagery; local rules provide safe operation during connectivity loss, while cloud services perform time-series analysis and decision fusion. Reported literature demonstrates approximately 97% crop-classification accuracy in one hybrid MLP-LSTM study, soil-parameter deviations below 13.26% against laboratory measurements, and water savings reported up to 70% in reviewed smart-irrigation deployments. These outcomes are not treated as results of the proposed prototype; instead, they define evidence-based performance targets and a staged validation protocol. The review identifies interoperability, field robustness, calibration drift, uncertainty communication, and secure deployment as the principal barriers to practical adoption. PGMS addresses these gaps through multimodal fusion, confidence-aware recommendations, offline control, and traceable human override.