UTILIZING GIS, GPS, REMOTE SENSING, AND AI IN THE STUDY OF SOIL CHARACTERISTICS

Background. Modern agriculture faces numerous challenges associated with climate change, economic factors, and increasing demands for production efficiency. The implementation of advanced technologies, particularly Geographic Information Systems (GIS), Remote Sensing (RS), Global Navigation Satellite Systems (GNSS/GPS), and Artificial Intelligence (AI), allows for the optimization of agrotechnical processes and improved productivity in precision farming. Methods. This study examines the application methods of GIS, GPS, RS, and AI in precision agriculture. It employs the analysis of satellite and aerial imagery, spatial modelling techniques, geostatistics, and machine learning for yield prediction and optimization of management decisions. Additionally, the use of sensor systems for field data collection and their integration into digital agricultural platforms is analysed. Results. The study implemented a comprehensive model for assessing soil characteristics by combining GIS, GPS, remote sensing, and artificial intelligence methods. The results confirmed the effectiveness of using digital maps and satellite images for spatial interpolation of soil parameters (such as potassium, moisture, and humus content), yield mapping, and real-time crop monitoring. GPS navigation ensured high accuracy in machinery positioning and soil sampling, while machine learning algorithms (particularly LAI-based models and Random Forest) demonstrated yield prediction accuracy above 80 %. A crop rotation model built using Python libraries enabled the development of an optimal five-year rotation plan, considering soil types, climatic conditions, and potential yield. Variability maps and zoning results served as the basis for scenario-based field management at the enterprise level. Conclusions. The integration of GIS, GPS, RS, and AI into agricultural practices significantly enhances the accuracy of soil analysis and the efficiency of agroprocess management. The developed model enables the automation of decision-making processes based on large volumes of spatial and field data, contributing to cost reduction, increased productivity, and preservation of soil fertility. The implementation experience in the Kyiv region has demonstrated its practical applicability and potential for scaling within the framework of modern precision agriculture.