Structured design of flexibly automated manufacturing cells through semantic models and petri nets in a virtual reality environment

Abstract

The work presented in this paper outlines a design framework for flexibly automated manufacturing cells in a Virtual Reality environment. Design encompassing structure/configuration and control/operation aspects of manufacturing cells is supported through different levels of analysis. The cornerstone is semantic models that are constructed for stereotypical manufacturing cell objects and their relations but are extensible for new ones. VRML 3D models of these objects are created and introduced into the virtual world. Core methods are developed for object functional control. A major class of methods at object level refers to trivial and non-trivial kinematics for stereotypical objects such as robots, machine tools and conveyors. The main method at cell level pertains to controlling interactions among different manufacturing tasks and is based on definition of control-interpreted Petri Nets at two levels of detail: generic nets, which are less detailed and directly derive from semantic models, and specific nets, which add further detail that cannot be included in a practical way in the semantic nets. Furthermore, special attention is directed to the user within the virtual environment, by providing mostly visual tools that enhance user perception as well as information and manipulation tools that facilitate interaction with the virtual manufacturing cell being designed. The virtual environment was developed from scratch using relatively high-level 3D graphics development tools and open libraries and is thus amenable to extensions.