Ontology-Driven Metamodel Validation in Cyber-Physical Systems

This paper describes the development and use of an ontology to validate model integration and metamodel description in the cyber-physical systems engineering domain. A primary goal of the Defense Advanced Research Program Agency (DARPA)’s Adaptive Vehicle Make (AVM) program is to reduce product cost by reducing the design space early in the engineering life cycle, to enable a high-value focus on reasonable alternatives. The OpenMETA tool suite uses semantic constructs for model integration across engineering disciplines, and facilitates trade analysis across those disciplines (such as electrical, mechanical, thermal, fluid, and cyber). Independently, through expert interviews, the authors developed a lightweight ontology to represent important trades in the early system engineering design process, and have been using that ontology as both a communication and validation mechanism for the semantics embedded in the OpenMETA tools. The results are informing the tool development, providing an independent validation mechanism, and creating a layer of the ontology that is extremely useful to both people and machines that was not initially obvious. The ontology-based approach described in this paper facilitates the identification and encoding of the important relationships between design decisions and subsequent reasonable decision alternatives, and serves as an independent validation of the semantics and relationships encoded in the OpenMETA tool suite, both at very low cost.