In modern greenhouses, the level of automation for any crop production is already high. Climate control and irrigation in modern greenhouses are in general automated, however, decisions on setpoints are in still made by the grower. Spacing of plants in pots as they become larger can be done by robots or some well-designed mechanical system, however, also here decisions are made by humans until now.
During greenhouse production decisions on what climate and irrigation to apply, when to space and at what plant distance, and whether the tomato plants are ready for harvest, are made by human operators. Based on their knowledge and experience they make decisions to influence growing process and to allow for optimal resource use. These decisions are based on the observed growing process. It means that the current crop growth must be compared with an expected future growth to determine if the growing process is still on track. To do so, observations are generally made by visual inspection by an experienced human grower. This is a time-consuming process, requires loads of experience and is subjective. To assist or replace the human by automated decision process, machine learning could be used to optimize crop production and resource use, computer vision could be used to quantify the crop growth.
Teams are challenged to develop software that performs a conversion of images to numerical crop Parameters with computer vision analysis (part A) and to develop software that find autonomously the maximum net profit (part B).
Earlier editions of the Autonomous Greenhouse Challenge have been carried out at Wageningen University & Research, the Netherlands, between 2018-2022, with teams growing cucumbers, cherry tomatoes, and lettuce crops autonomously. During these editions we have shown that computer algorithms can increase greenhouse crop production, save energy, and yield higher net profits. We have demonstrated that artificial intelligence could be superior to human intelligence, and could potentially control indoor farming in the future.
For more information visit: www.autonomousgreenhouses.com
