An intelligent fuzzy-Petri reasoning supervisory control for FMS manufacturing plants

Abstract

The organization level of complex automated robotic and manufacturing systems, whose dynamics is discrete-event dominated, is revisited and some innovative results towards enhancing machine intelligence derived. These comprise a fuzzy-Petri-net reasoning emulator implementing the fuzzy-Petri rule-based decision-making with an appropriate knowledge base support. This fuzzy-Petri reformulation follows the original concept of entropy-based, intelligent machines of G.N. Saridis. Its generic feature in the sense of an improved fuzzy-Petri reasoning based organizing coordination controller is crucial. A software tool emulating the fuzzy-Petri-net reasoning is discussed. Its version employing knowledge-base with 10 rules takes only a few milliseconds on a standard PC to be simulated. Thus it is believed a feasible yet efficient real-time reasoning for supervisory control functions has been developed. Some computer simulations results for FMS applications are given that illustrate these novel developments.