Soil moisture is a highly significant variable in hydrological issues, playing a crucial role in hydrological modeling, quantitative meteorological forecasting, and the analysis of issues such as drought, forest fires, climate change, and water resource management. In recent years, remote sensing has gained significant attention for soil moisture estimation due to its speed, regular coverage, extensive reach, and cost-effectiveness. However, one of the primary concerns for remote sensing specialists in soil moisture estimation is the provision of ground-truth data. ‏In this research, the design and development of a ground-based soil moisture sensor were undertaken, which is capable of measuring soil moisture, collecting spatial data, and displaying and storing this information to provide ground-truth soil moisture data. Using this sensor, along with remote sensing imagery and artificial intelligence (AI) methods, it is possible to estimate soil moisture across a vast area. This process involves the simultaneous use of the sensor for sampling a small area during the satellite pass over the targeted region, which is then used to train the selected AI model. The trained model subsequently estimates soil moisture over the desired area using remote sensing imagery. ‏Factors such as soil texture, soil electrolytes, and temperature impact the sensor’s measurements. Therefore, evaluating these influencing factors and ensuring appropriate environmental conditions during laboratory testing and sensor calibration are critically important. The laboratory process will reveal the specific conditions and extent to which these factors affect the sensor’s output. In the sensor calibration process, a third-degree polynomial regression model was developed to determine the soil’s gravimetric moisture content, achieving an accuracy of 0.95% for gravimetric soil moisture content and a coefficient of determination of 97%.