dc.creator | Lattmann, Zsolt | |
dc.date.accessioned | 2020-08-22T00:38:30Z | |
dc.date.available | 2018-04-29 | |
dc.date.issued | 2016-04-29 | |
dc.identifier.uri | https://etd.library.vanderbilt.edu/etd-04292016-141027 | |
dc.identifier.uri | http://hdl.handle.net/1803/12240 | |
dc.description.abstract | Product design has become more complex in recent decades. Several design methods and processes have been developed to reduce the design complexity to a manageable level. Model-based systems engineering leverages these methods (e.g., v-model, virtual integration) to manage design complexity and to reduce development time and costs for software and hardware development. Most of these design methods and processes were primarily developed for either software or hardware development. In Cyber-Physical Systems, computational elements (hardware and software) are tightly integrated with physical processes and physical components and often interact with the physical system.
There are hundreds of distinct tools used to design and analyze different aspects of a complex system design, which shows that there is no single tool that can deal with all aspects of a complex design problem. An adequate model of Cyber-Physical Systems must: (1) capture domain interactions (e.g., electrical power and mechanical systems), (2) incorporate multiple aspects and domain models for each component, (3) support a wide variety of analysis techniques, (4) able to reuse existing models from libraries, and (5) extract sufficient information from model libraries to support architecture exploration for product families.
We present the key concepts and functions of an Analysis-Driven Rapid Design Process for Cyber-Physical Systems to address all aforementioned challenges. We define and contribute to three platforms to improve efficiency and quality of the design process: a Model Integration Platform, Tool Integration Platform, and Execution Integration Platform. The Model Integration Platform (a) uses heterogeneous component models, (b) keeps the multi-domain models consistent, (c) tracks model dependencies, and (d) facilitates importing from existing model libraries. The Tool Integration Platform accommodates a variety of analysis tools with the flexibility to add new tools in the future. The Execution Integration Platform provides an analysis tool independent framework for analysis execution and organization of analysis results. Finally, we propose a design flow to achieve an analysis-driven rapid iterative design process; we demonstrate and evaluate this design flow utilizing two case studies: (a) an oscillator circuit design and (b) a ground vehicle driveline design. | |
dc.format.mimetype | application/pdf | |
dc.subject | model-based engineering | |
dc.subject | cyber-physical systems | |
dc.subject | rapid design process | |
dc.title | An Analysis-Driven Rapid Design Process for Cyber-Physical Systems | |
dc.type | dissertation | |
dc.contributor.committeeMember | Janos Sztipanovits | |
dc.contributor.committeeMember | Sandeep Neema | |
dc.contributor.committeeMember | Xenofon Koutsoukos | |
dc.contributor.committeeMember | Gautam Biswas | |
dc.contributor.committeeMember | Theodore Bapty | |
dc.type.material | text | |
thesis.degree.name | PHD | |
thesis.degree.level | dissertation | |
thesis.degree.discipline | Electrical Engineering | |
thesis.degree.grantor | Vanderbilt University | |
local.embargo.terms | 2018-04-29 | |
local.embargo.lift | 2018-04-29 | |
dc.contributor.committeeChair | Gabor Karsai | |