Multi-agent systems for decentralized control and adaptive interaction between humans and machines for industrial environments

Abstract

The digitization of industrial processes in production systems and logistics reveals new concepts of control and new techniques of information technology to solve complex problems of systems to be decentralized. The Internet of Things and the so called Industry 4.0 accompanied by their definition of architecture on the one hand and multi-agent systems with their decentralized system structure to handle complex tasks and adapting to changing conditions on the other hand are not matched as of yet. Furthermore, the human worker is not yet an integrated part of these kinds of systems and may be overstrained by the vast amount of information resulting from more and more digitally available information. This paper proposes a strategy to integrate methods and architecture of IoT with cyber-physical systems and multi-agent systems. It elaborates on the processing of data and environmental conditions to provide the human worker with context-based information reduction which helps to minimize the information overload. Two use cases in the field of intralogistics which make use of the new concept are presented.