Cluster D: Integration

Project D1: Modelling interactions of climate, growth, and quality using machine learning

This project aims to model and predict interactions between shoot and root environment, genotype, growth, and quality using machine learning. By combining high-definition visual imagery with data from sensors measuring climate, light, irrigation, and plant properties, we can not only accurately predict plant growth and quality in order to optimise growth recipes, but also assess the impact of any planned or unplanned changes to these conditions, detect anomalous growth patterns, and improve our biological understanding of plant growth.

Project D2: Optimisation of operation and design of a vertical farm

The scientific challenge is to create an algorithm that adjusts the climate settings in such a way that costs, energy use, yield, and quality are well balanced according to the grower’s preferences. This requires an integrated model that relates the effects of climate settings on yield, quality, energy, and costs. Furthermore, it requires an algorithm that is flexible depending on how individual growing companies weigh the importance of energy, yield, quality, and costs. This project aims to optimise the operation and design of a vertical farm in terms of energy, yield, quality, and costs.