This research addresses problems associated with the design and analysis of discrete-event (DE) supervisory controllers of continuous- time (CT) systems (hybrid systems) often encountered in manufacturing situations and distributed computing. The approach involves development of a modeling methodology and representational forms of hybrid systems which permit straight-forward simulation and analysis of such hybrid systems. Appropriate mathematical structures for the analysis and design of these hybrid systems are developed. Petri-Nets (PNs) serve as the starting point for the modeling of the DE controller component and classical state model (geometric) control theory techniques are developed for the CT component of the hybrid systems. Modifications to existing PN and CT modeling methodologies address several key problems associated with hybrid systems including characterization of information flow within the hybrid model and between the CT and DE systems. The new modeling technique will permit meaningful re-interpretation of classical PN properties and CT system properties.
