Model Based Design and Evaluation of Measurement Systems

Abstract

As modern engineering challenges grow more complex, the demand for novel methods of sensing the world around us and more accurate models to explain that world grow in tandem. This dissertation seeks to advance the state of the art in radiation tolerant robotics, underwater imaging, and artificial hearts through both model-based design and sensor evaluation, with a particular emphasis on physics-based models in the design and evaluation of measurement systems. First, this work proposes a method to evaluate an encoder’s performance and failure modes in a high-radiation environment. This work is useful for designing robots that can withstand ionizing radiation with only minor compromises to cost and dexterity. Second, this research develops an acoustic model and a physical prototype of a novel short-range sonar system. A device based on this research would help a diver maintain vision in zero-visibility would dramatically improve the diver’s effectiveness, allowing for faster repairs and more complete inspections. Finally, this work develops a translation of a lumped parameter model of the human circulatory system into the physical domain. This “mock circulatory loop” supports the development of next-generation artificial hearts by providing a realistic dynamic load for prototypes. This work fills the gap between in-silico simulations and blood compatibility testing.