A multi-agent system framework for autonomous crop irrigation

This article presents the design and development of an innovative multi-agent system (MAS) composed of various subsystems of agents grouped by similarity in their tasks. The proposed MAS will manage the information captured from wireless sensor networks for knowledge discovery and decision making in rural environments on actuator networks. The proposed MAS has been built on the Cloud Computing paradigm to provide greater flexibility and scalability in the management of both small and large projects. The use of these “intelligent” technologies in rural areas can improve the efficiency and effectiveness of agricultural production systems, analyzing the data collected by sensors, extracting patterns to predict events affecting the crop and anticipate such events giving localized responses with automated actuators. The proposed architecture has been tested in an agricultural environment in order to optimize irrigation in a corn crop. Thanks to the wireless sensor network (WSN), information has been obtained about the terrain of the crop and its climatic conditions, extracting information about the needs of cultivated corn and taking efficient irrigation decisions based on those needs, reducing water consumption by 17.16% compared to traditional automotive irrigation.