Open Access

Dominik Walczuch, Tim Nitzsche, Tim Seidel, Julius Schöning

• The increased demand for agricultural land and the more frequent occurrence of extreme weather conditions lead to increased greenhouses usage in agriculture. Greenhouses offer an efficient alternative to traditional agriculture as all environmental parameters are controlled, leading to a higher yield per land use. In older greenhouses, the different environmental factors are set manually, resulting in unfavorable climate conditions due to over- and undershoots of the factors. Optimizing environmental conditions, closed-loop control systems (CLCS) are used to reduce the susceptibility to over- and undershoots as well as to disturbance variables. The greenhouse actuators respond to various input values bringing changes to the environmental parameters. For controlling the actuators, artificial intelligence (AI) offers the potential to control the environmental parameters more accurately than a simple CLCS. With the help of predicting the influence of actuators regarding the greenhouse climate, AI-based climate systems might outperform CLCS systems and human experts. This paper provides an overview of the different architectures in which AI is used for controlling complex systems and discusses its potential for greenhouses. The results will show that AI offers the possibility for yield increase and resources saving.