Methods for building ontologies to ensure coordinated behavior of autonomous intellectual systems in agriculture

Abstract

Ontologies are the main tool for representing domain knowledge. The need to use ontologies to ensure consistent behavior of autonomous intelligent systems in agriculture is determined by the need to create semantic interactiveness of agricultural systems and decision support systems, in which automatic knowledge inference occurs based on semantic processing of data from a sensor network. The purpose of the study is to develop methods for constructing ontologies to ensure consistent behavior of autonomous intelligent systems in agriculture. The paper presents the architecture of an intelligent control system for a heterogeneous human-machine team, consisting of an autonomous robot for active plant protection and an unmanned aerial vehicle (UAV) for monitoring the condition of the garden from the air. A multiagent architecture for constructing ontologies has been developed to ensure consistent behavior of a human-machine team. In particular, a multiagent architecture for constructing an ontology when a robot detects a new object is presented, as well as a scheme for exchanging knowledge between team members. The results obtained can be used in the development of decision support systems, in which automatic knowledge inference occurs based on semantic processing of data from a sensor network.