A cyber-physical intelligent agent for irrigation scheduling in horticultural crops

Irrigation is one of the most important activities in crop management. The aim of irrigation scheduling is to supply the correct amount of water to the crops and define appropriate irrigation times. The rational behavior of intelligent agents added to internet of things technologies could be a strategy for improving the performance of automatic irrigation systems. This paper presents the design and implementation of a cyber-physical crop irrigation system, based on the intelligent agent concept. The system allows the acquisition of field information through sensors, the application of water according to a decision system and the activation of solenoid valves. Communication capabilities and the performance measurement, the environment, the actuation and sensing systems (PEAS) are explained. Monitoring in the farm or remotely from other locations is achieved with a central station and its internet connectivity. The performance evaluation of the system was made using crop modelling software and an experimental crop in a field located in the large-scale irrigation and drainage district of Chicamocha and Firavitoba (Usochicamocha) in Boyaca - Colombia. The developed system allowed maintaining soil moisture close to field capacity with several irrigation strategies, and avoiding losses and excesses in the use of water resources.