The Business Unit for Greenhouse Horticulture and Flower Bulbs at Wageningen University & Research (WUR) is constantly reviewing its crop models and has said that it hopes to produce its first cucumber cultivation algorithm in time for 2024.
Various models describe reality using calculations based on rules and parameters, for example, how much does a crop evaporate at a certain combination of greenhouse temperature, humidity, and light intensity? Not only should the physics be described correctly in the model, but crop parameters also need to be adapted to their environment. This is difficult because every crop is different, and a crop adapts to the conditions in which it is growing: a greenhouse crop in a constantly high humid environment will behave differently than a crop grown in harsh conditions.
The ultimate goal of the project is to develop a computer which is just as good, or better, at growing as a real grower. In order to achieve this, artificial intelligence must know which decisions a grower makes and be able to obtain the right information to make such decisions itself. For example, leaves are an important indicator for the growth of the crop for growers, but how can a camera with image recognition correctly count the number of leaves in a dense crop?
WUR has been working on the AGROS (Evolution to sustainable AGRicultural Operation Systems) project since 2020, and the work is funded by the Dutch Ministry of Agriculture, Nature and Food Quality.