AI-Driven Robotics in Agriculture

Goals

This VIP project aims to advance vertical precision agriculture by developing a hybrid AI-physiological crop modeling framework that integrates machine learning, non-invasive plant monitoring, and mechanistic models to optimize nutrient uptake, biomass growth, and crop health. Hyperspectral imaging will enable real-time plant monitoring, providing insights into biomass accumulation and growth patterns, while ion-selective sensors track nutrient uptake without harming the plant. This approach enhances efficiency, sustainability, and resilience in vertical hydroponic farming.

Issues Involved or Addressed

Agriculture faces increasing challenges from population growth, climate change, and resource scarcity, challenging traditional farming and limiting food security. Vertical hydroponic farming offers a transformative solution by maximizing resource efficiency, optimizing land use, and enabling year-round crop production. However, advanced tools are needed to identify complex, non-linear relationships in agricultural systems for precise nutrient management and plant growth predictions. This project aims to develop a hybrid AI-physiological crop modeling framework that integrates machine learning, physiological models, and robotics to optimize growth conditions, nutrient cycling, and yield prediction. Traditional crop assessments rely on destructive sampling, which disrupts the growing process and reduces experimental capacity and efficiency. To overcome this, we use non-invasive hyperspectral imaging (HSI) to monitor plant morphology and nutrition in real time while utilizing spectral data to detect nutrient deficiencies, water stress, and diseases early. We will leverage Vision Transformer (ViT) architectures to analyze raw hyperspectral images and detect physiologically meaningful spatial-spectral patterns. The hybrid AI-physiological crop modeling framework will integrate deep learning, biochemical models, and robotics to track plant pigmentation, yield, and nutrient uptake dynamics, improving yield, optimizing resource use, and reinforcing urban food security.

Partners/Sponsors

Local indoor farming communities, USDA, and NSF-funded initiatives on AI-driven precision agriculture.