A survey on intelligent agents and multi-agents for irrigation scheduling

Irrigation is very important for ensuring food security and reducing crop production vulnerability caused by the lack of rain. Sustainable irrigation is the rational practice of all the activities related to water application on the crops. In irrigation, the rationality of intelligent agents can be used to reach soil water content near the field capacity to increase yields and reduce waste of water. Rationality in artificial intelligence is the capability of the intelligent agents to decide their actions. This paper discusses how incorporating intelligent agents on irrigation systems allows significant advances in respect of current irrigation approaches. This paper review not only focuses on intelligent reactive systems as usual, but rather discloses developments in systems that incorporate other behaviors such as proactivity, planning, learning, social abilities, organization, coordination and negotiation. From the literature review, it is found that the use of soil, plant and environmental sensors, as well as reasoning, learning and communication capabilities, provides innovative technological support to improve sustainability in irrigated agriculture. The review also shows that intelligent agents can adequately consider the timing and the amount of water to apply according to the spatio-temporal variations of the soil-plant-atmosphere system. It is concluded that significant improvements in water savings and crop yield can be achieved incorporating artificial intelligence into precision irrigation. Further research is needed on irrigation scheduling based on multi-agent systems at different scales of agricultural production systems.