Automating virtualization of machinery for enabling efficient virtual engineering methods

Abstract

Virtual engineering as a new working method in product development should make it much easier to validate the development progress and facilitate team communication. Work steps are brought forward and start with the virtual components instead of real ones. To validate mechanical and electrical CAD as well as programming, automated virtualization systems should create the virtual twin of the machine at the push of a button. For this purpose, generic intelligence is added to enable complex interactive virtual models that can be used for training, monitoring and many other applications. Advanced applications are for example training and support applications, especially in combination with augmented reality and remote collaboration. We propose a system that combines virtual reality, virtual engineering and artificial intelligence methods for the product development process. Geometry analysis algorithms are used to process mechanical CAD data and thus, for example, to automatically parameterize kinematic simulations. In combination with electrical CAD data and the simulations of electric circuits as well as the original machine program allow simulating the behavior of the machine and the user interaction with it. This article will describe the virtualization method in detail and present various use-cases in special machine construction. It will also propose a novel method to use causal discovery in complex machine simulations.