Use of Artificial Intelligence in Controlled Environment Agriculture: A Review

Controlled environment agriculture (CEA) refers to the practice of growing crops within regulated, enclosed systems where factors such as light, temperature, humidity, and nutrients can be precisely controlled. These systems generate vast amounts of operational and biological data as sensors continuously monitor plant and environmental conditions. Artificial intelligence (AI), a field focused on creating systems that learn from data and make informed decisions, can then be used to process this huge data set, including environmental adjustments to forecasting crop growth and yield, ultimately improving productivity while reducing waste. Technologies such as machine learning, image processing, robotics, deep learning, Internet of Things (IoT), and neural networks are embedded in AI to address agricultural challenges. AI monitors agricultural parameters such as crop genetics, health and disease, livestock, soil moisture, soil quality and moisture, climate, autonomous machinery, resource efficiency, weed control, genetics, harvest timing and market forecasting to enable real-time monitoring of factors like weather, temperature, water usage, and soil conditions, thus enabling farmers to minimize losses and boost yields. Individual farmers are likely to operate with more than 75 million interconnected devices, with an average farm producing 4.1 million data points on average per day by 2050. This review critically examines the successes and challenges of AI in redefining CEA productivity and sustainability. As AI technologies continue to evolve, their integration into CEA systems will become increasingly vital in building resilient, data-driven, and adaptive agricultural systems for the future. Ultimately, AI-enhanced CEA convergence represents a critical frontier for research and innovation in agri-food systems.