Review on Applications of artificial intelligence in Indian agriculture: future and challenges

Abstract

DOI: https://doi.org/10.5281/zenodo.17863581

Soil moisture measurement technologies, including capacitance and time-domain reflectometry (TDR) sensors, function by assessing the dielectric properties of soil. These sensors emit an electric field, the behavior of which—its strength and propagation—is influenced by the soil's dielectric constant. Artificial Intelligence (AI) has the capacity to process extensive meteorological datasets derived from satellites, ground-based instruments, and historical archives. In addition to forecasting, AI is utilized for agricultural field monitoring, where drones equipped with digital imaging tools identify weed growth, and decision-making systems control site-specific herbicide application using GPS guidance. Such technologies enable farmers to conduct rapid, accurate, and economically efficient field assessments, enhancing productivity by maximizing the use of existing resources.

Furthermore, technological integration allows for real-time monitoring of crop conditions—covering aspects such as soil moisture, irrigation efficiency, pest control, and nutrient management. However, in rural or low-resource environments, there may be reluctance among farmers to adopt AI-driven solutions if the rationale behind automated decisions is opaque or difficult to interpret. To facilitate adoption, it is crucial to foster trust through transparent explanations of AI outputs.

Successful implementation of AI in agriculture also depends on substantial training initiatives for farmers, extension agents, and other key stakeholders. This includes developing intuitive user interfaces, delivering practical training sessions, and creating robust support systems to improve digital competency and promote widespread technology adoption.