Integrating machine learning and drone technology for precision agriculture: A smart solution for automated irrigation and crop management

• The study integrated IoT-enabled soil sensors, localized weather forecasting, and autonomous drones across 150 smallholder farms in Tamale, Bolgatanga, and Wa using a mixed-methods research design. • A Random Forest machine learning model accurately forecasted irrigation requirements, enabling drones to deliver precision-targeted water applications. • Policy recommendations include implementing government subsidies, enhancing farmer training programs, and introducing regulatory reforms to support widespread adoption. • The model aligns with SDGs 2 (Zero Hunger), 6 (Clean Water and Sanitation), and 13 (Climate Action), offering a scalable and replicable approach for dryland farming across sub-Saharan Africa. Agriculture is vital to Ghana’s economy, contributing approximately 20 % to GDP and employing 45 % of the workforce. However, the agricultural sector’s reliance on rain-fed farming, particularly in northern Ghana, exposes it to climate variability, erratic rainfall, and prolonged droughts which lead to chronic food insecurity and economic losses. With only 2 % of farmland irrigated, traditional methods exacerbate water scarcity and low productivity. This study proposes an innovative machine learning (ML) and drone-based precision irrigation system to optimize water use, enhance crop yields, and build climate resilience in northern Ghana. The study deployed internet of things (IoT) soil sensors, weather forecasts, and autonomous drones across 150 smallholder farms in Tamale, Bolgatanga, and Wa through a mixed-methods approach. A random forest ML model predicted irrigation needs, while drones delivered targeted water applications. Results showed a 50.6 % increase in crop yields and a 30–40 % reduction in water usage compared to traditional methods. However, stakeholder interviews and factor analysis identified barriers such as high costs, limited digital literacy, and policy gaps. The study recommends government subsidies, farmer training, and regulatory reforms to facilitate adoption. This scalable model contributes to Sustainable Development Goals (SDGs 2, 6, and 13) and offers a replicable framework for other arid regions in sub-Saharan Africa.