Applications of Petri Nets to Human-in-the-Loop Control for Discrete Automation Systems

For discrete automation systems, certain human operations may violate desired safety requirements and result in catastrophic failure. For such humanin-the-loop systems, this paper proposes a systematic approach to developing supervisory agents which guarantee that manual operations meet required safety specifications. In the present approach, Petri nets (PN) are applied to construct a system model and synthesize a desired supervisor. Applications to 1) a rapid thermal process (RTP) in semiconductor manufacturing controlled over the Internet (Lee and Hsu, 2003) and 2) a two-robot remote surveillance system (Lee et al., 2005) are provided to demonstrate the practicability of the developed supervisory control approach. The technique developed in this paper is significant in industrial applications. 1.1 General Review Basically, an automated process is inhe rently a discrete event system (DES). The Petri net (PN) has been developed as a powerful tool for modelling, analysis, simulation, and control of DES. PN was named after Carl A. Petri (1962), who created a net-like mathematical tool for describing relations between the conditions and the events. PN was further developed to meet the need in specifying process synchronization, asynchronous events, concurrent operations, and conflicts or resource sharing for a variety of industrial automated systems at the discrete-event level. Starting in the late of 1970’s, researchers investigated possible industrial applications of PN in discrete-event systems as in the survey/tutorial papers of Murata (1989), Zurawski and Zhou (1994), David and Alla (1994), and Zhou and Jeng (1998).