On an Approach to Knowledge Management and the Development of the Knowledge-Вased Multi-Agent System

The chapter discusses the architecture of Knowledge-Вased Multi-Agent System (KBMAS) and describes software agent models. purpose functional organization system agents used for planning management computing resources KBMAS are considered. An approach to applied agent’s development that integrates knowledge-based reasoning mechanisms with neural network models is proposed. structure problem-oriented Solver, including groups reactive cognitive solve complex ill-formalized problems, interaction diagram states transitions node given. control scheme shown includes methods determining availability microservices by agents, reliability assurances coordinated operation system’s nodes. method reinforcement learning, rules (productions), queries knowledge base described. Methods distribution in nodes, as well construction an optimal logical Distributed Knowledge Base, which has minimal information connectivity ensures effective on multicomputers,