An Agent-based Intelligent Fms Control System

This paper describes the development of an agent-based framework for the intelligent control of flexible manufacturing systems. An FMS is a complex manufacturing system, in which the control system is responsible for the planning, control and monitoring functions. Traditionally, manufacturing systems are commonly established with hierarchical control architecture. This approach is not flexible enough. Any change in the hardware may lead to re-writing a major part of the control software. Besides, it is not robust enough to tackle any machine breakdown incidence. It is the motivation of this paper to present a multi-agent control system with good expandability and be able to cope with dynamic changes in the manufacturing environment. In this regard, the system control software architecture is established based on intelligent agents. In this paper, a systematic approach has been adopted in the development of the agent-based control framework. The approach is based on the object-oriented modelling technique. With the help of the system function model and dynamic model, a framework for the agent-based control architecture is proposed. According to the proposed control architecture, the structure of individual agents and the protocols are also established. The colour Petri-net model is then established to represent the agent-based control system, with which dynamic simulation can be carried out to study the dynamic behaviour of the FMS. To evaluate the performance of the developed system, advanced data mining and searching algorithm has been developed and implemented.