Agriculture, an essential driver of economic expansion, is faced by the issue of sustaining an increasing global population in the context of climatic uncertainty and limited resources. As a result, “Smart Farming”, which uses cutting-edge artificial intelligence (AI) to support autonomous decision-making, has become more popular. This article explores how the Internet of Things (IoT), AI, machine learning (ML), remote sensing, and variable-rate technology (VRT) work together to transform agriculture. Using sophisticated algorithms to predict soil conditions, improving agricultural yield projections, diagnosing water stress from sensor data, and identifying plant diseases and weeds through image recognition, crop mapping, and AI-guided crop selection are some of the main applications investigated. Furthermore, the precision with which VRT applies water, pesticides, and fertilizers optimizes resource utilization, enhancing sustainability and efficiency. To effectively meet the world’s food demands, this study forecasts a sustainable agricultural future that combines AI-driven approaches with conventional methods.

Soil salinization stands as a prominent global environmental challenge, necessitating enhanced assessment methodologies. This study is dedicated to refining soil salinity assessment in the Lake Urmia region of Iran, utilizing multi-year data spanning from 2015 to 2018. To achieve this objective, soil salinity was measured at 915 sampling points during the 2015–2018 timeframe. Simultaneously, remote sensing data were derived from surface reflectance data over the same study period. Four distinct scenarios were considered such as a newly developed spectral index (Scenario I), the newly developed index combined with other salt-based spectral indices from the literature (Scenario II), indirect spectral indices based on vegetation and soil characteristics (Scenario III), and the amalgamation of both direct and indirect spectral indices (Scenario IV). Linear Regression (LR), Support Vector Machine (SVM), and Random Forest (RF) were employed to assess soil salinity. The measured data divided to 75% of the data as the calibration dataset, while the remaining 25% constituted the validation dataset. The findings revealed a correlation between soil salinity and spectral indices from the literature, with a range of -0.53 to 0.51, while the newly developed spectral index exhibited a stronger correlation (r = 0.59). Furthermore, RF yielded superior results when using the newly developed spectral index (Scenario I). Overall, SVM emerged as the most effective model (ME = -9.678, R² = 0.751, and RPIQ = 1.78) when integrating direct and indirect spectral indices (Scenario IV). This study demonstrates the efficacy of combining machine learning techniques with a blend of newly developed and existing spectral indices from the literature for the monitoring of soil salinity, particularly in arid and semi-arid regions.

Variable-rate technology (VRT) may reduce input costs, increase crop productivity and quality, and help to protect the environment. The present study was conducted to evaluate the performance of a variable-rate fertilizer applicator for rice (Oryza sativa L.). Three replications were conducted, each of which was divided into four plots. Field performance of the system was assessed at different nitrogen levels (N1 to N4, i.e., 75, 125, 175, 225 kg ha⁻¹), growth stages (tillering, panicle initiation, heading), and heights (40, 60, 80, 100 cm) of the sensor from the crop canopy. Fertilizer rate was at minimum 12.59 kg ha⁻¹ at 10 rpm of drive-shaft rotational speed and at maximum 50.41 kg ha⁻¹ at 40 rpm. The system response time was within the range of 3.53 to 4.93 s, with overall error ranging between 0.83% to 4.92%. Across different growth stages, when fertilizer rate was increased from N1 to N4, NDVI increased from 0.49 to 0.69. Hence, drive-shaft rotational speed is decreased from 25 to 7 rpm to shift the application rate from 30.83 to 9.15 kg ha⁻¹. There was a 45% reduction in total fertilizer rate applied by the system, with respect to the recommended rate.

A study was conducted with the goal of developing an algorithm for use in sensors to monitor available soil N. For this purpose, three different soils were selected. The soils were studied for electrical conductivity (EC) at four different moisture levels and four levels of N. The selection of moisture levels was based on optimum moisture levels between tillage moisture and field capacity. The results revealed a significant relationship between electrical conductivity and moisture level of the soil as well as between electrical conductivity and soil N content. Based on these relations, a polynomial model was developed between the EC of each selected soil sample and moisture content as well as N levels. The regression model for moisture content-based EC determination had coefficients of determination of 0.985, 0.988, and 0.981 for clay loam, sandy loam, and sandy loam soils, respectively. Similarly, the regression model for N content-based EC determination had coefficients of determination of 0.9832, 0.9, and 0.99 for clay loam, sandy loam, and sandy loam soils, respectively. An algorithm developed using a polynomial relationship between the EC of each selected soil sample at all moisture and N levels can be used to develop a sensor for site-specific N application.

Environmental factors such as drought significantly influence vegetation growth, coverage, and ecosystem functions. Hence, monitoring the spatiotemporal vegetation responses to drought in a high temporal and adequate spatial resolution is essential, mainly at the local scale. This study was conducted to investigate the aspatial and spatial relationships between vegetation growth status and drought in the southeastern South Dakota, USA. For this purpose, Landsat 8 OLI images from the months of April through September for the years 2016–2021, with cloud cover of less than 10%, were acquired. After that, radiometric calibration and atmospheric correction were performed on all of the images. Some spectral indices were calculated using the Band Math toolbox in ENVI 5.3 (Environment for Visualizing Images v. 5.3). In the present study, the extracted spectral indices from Landsat 8 OLI images were the Normalized Difference Vegetation Index (NDVI) and the Normalized Multiband Drought Index (NMDI). The results showed that the NDVI values for the month of July in different years were at maximum value at mostly pixels. Based on the statistical criteria, the best regression models for explaining the relationship between NDVI and NMDI_Soil were polynomial order 2 for 2016 to 2019 and linear for 2021. The developed regression models accounted for 96.7, 95.7, 96.2, 88.4, and 32.2% of vegetation changes for 2016, 2017, 2018, 2019, and 2021, respectively. However, there was no defined trend between NDVI and NMDI_Soil observed in 2020. In addition, pixel-by-pixel analyses showed that drought significantly impacted vegetation coverage, and 69.6% of the pixels were negatively correlated with the NDVI. It was concluded that the Landsat satellite images have potential information for studying the relationships between vegetation growth status and drought, which is the primary step in site-specific management.

Considerable variation in soil often occurs within and across production fields, which can significantly impact farming input management strategies. Optimizing resource utilization while enhancing crop productivity is critical for achieving Sustainable Development Goals (SDGs). This paper proposes a low-cost retrofittable Variable Rate Applicator Controller (VRAC) designed to leverage soil variability and facilitate the adoption of Variable Rate Technologies. The controller operates using a Raspberry Pi platform, RTK—Global Navigation Satellite System (GNSS), a stepper motor, and an anti-slip wheel encoder. The VRAC allows precise, on-the-fly control of the Variable Rate application of farming inputs utilizing an accurate GNSS to pinpoint geographic coordinates in real time. A wheel encoder measures accurate distance travel, providing a real-time calculation of speed with a slip-resistant wheel design for precise RPM readings. The Raspberry Pi platform processes the data, enabling dynamic adjustments of variability based on predefined maps, while the motor driver controls the motor’s RPM. It is designed to be plug-and-play, user-friendly, and accessible for a broader range of farming practices, including seeding rates, dry fertilizer, and liquid fertilizer application. Data logging is performed from various field sensors. The controller exhibits an average of 0.864 s for rate changes from 267 to 45, 45 to 241, 241 to 128, 128 to 218, and 218 to 160 kg/ha at speeds of 8, 11, 16, 19, 24, and 32 km/h. It has an average coefficient of variation of 4.59, an accuracy of 97.17%, a root means square error (RMSE) of 4.57, an R square of 0.994, and an average standard deviation of 1.76 kg for seeding discharge. The cost-effectiveness and retrofitability of this technology offer an increase in precision agriculture adoption to a broader range of farmers and promote sustainable farming practices.