Prof Georgios Leontidis (University of Aberdeen), Dr Aiden Durrant (University of Aberdeen), Dr Oorbessy (Reshmi) Gaju (University of Lincoln), Rory McLeod (Intelligent Growth Solutions Limited)

NutriPredict will leverage plant physiology and AI to optimise nutrient management in Controlled Environment Agriculture (CEA), a fast-growing sector that is vital for sustainable food production. As global demand for resource- efficient farming grows, this project addresses the need for precise, data-driven nutrient delivery to improve yield, quality, and sustainability.

The student will develop a causal, multimodal machine learning system that integrates sensor data, environmental variables, and physiological insights to predict nutrient needs and detect early deficiencies. The project will use data from multiple crops, combined with imaging, sensor streams, and AI modelling, to simulate nutrient uptake, monitor growth, and generate actionable recommendations.

The student will be based at the University of Aberdeen, embedded in a machine learning group that includes several other students working on agri-food challenges. In addition, they will benefit from co-supervision in plant modelling from colleagues at the University of Lincoln’s Lincoln Institute of Agri-Food Technology. On a day-to-day basis, the student will be responsible for setting up experiments to collect new data, organising and pre-processing historical and new datasets, expanding their knowledge of multimodal deep learning systems and their integration with crop models, and developing and evaluating deep learning pipelines that can operate in sparse-data scenarios. They will have access to comprehensive training spanning plant science, machine learning theory, and causal inference, alongside hands-on skills in sensing technologies, data engineering, and stakeholder co-design.

The student will be embedded in a unique environment that includes two universities and an innovative industry partner. Extensive training will be provided according to the student’s background, with anticipated skills development covering advanced machine learning and a deep understanding of crop modelling.

A strong background in computer science and programming in Python/PyTorch, as well as an interest in agri-food systems, is required.