Thesis: Ph. D., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2014. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 161-172). / Engineering systems of the modern day are increasingly complex, often involving numerous components, countless mathematical models, and large, globally-distributed design teams. These features all contribute uncertainty to the system design process that, if not properly managed, can escalate into risks that seriously jeopardize the design program. In fact, recent history is replete with examples of major design setbacks due to failure to recognize and reduce risks associated with performance, cost, and schedule as they emerge during the design process. The objective of this thesis is to develop methods that help quantify, understand, and mitigate the effects of uncertainty in the design of engineering systems. The design process is viewed as a stochastic estimation problem in which the level of uncertainty in the design parameters and quantities of interest is characterized probabilistically, and updated through successive iterations as new information becomes available. Proposed quantitative measures of complexity and risk can be used in the design context to rigorously estimate uncertainty, and have direct implications for system robustness and reliability. New local sensitivity analysis techniques facilitate the approximation of complexity and risk in the quantities of interest resulting from modifications in the mean or variance of the design parameters. A novel complexity-based sensitivity analysis method enables the apportionment of output uncertainty into contributions not only due to the variance of input factors and their interactions, but also due to properties of the underlying probability distributions such as intrinsic extent and non-Gaussianity. Furthermore, uncertainty and sensitivity information are combined to identify specfic strategies for uncertainty mitigation and visualize tradeoffs between available options. These approaches are integrated with design budgets to guide decisions regarding the allocation of resources toward improving system robustness and reliability. The methods developed in this work are applicable to a wide variety of engineering systems. In this thesis, they are demonstrated on a real-world aviation case study to assess the net cost-benet of a set of aircraft noise stringency options. This study reveals that uncertainties in the scientific inputs of the noise monetization model are overshadowed by those in the scenario inputs, and identifies policy implementation cost as the largest driver of uncertainty in the system. / by Qinxian He. / Ph. D.

Thesis: S.M., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2016. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 85-87). / In numerous scientific and engineering fields, sensitivity analysis tools are essential for design optimization as well as uncertainty quantification. For instance, adjoint algorithms are common place in aerospace engineering when it comes to optimize the shape of an airfoil, the configuration of a rocket or to quantify the impact of a manufacturing imperfection on the performance of a product. The quantities of interest are long-time averaged outputs such as the average drag on a plane wing. However, these conventional methods fail to compute the right sensitivity when the physical model exhibits chaos. This is the case of many turbulent fluid flows and atmospheric modelisations. A recently developed method, Least Squares Shadowing or simply LSS, tackles this problem and proposes an alternative approach to compute the desired sensitivities. The results are very promising and this thesis is intended to lay the mathematical foundations of this new algorithm. A latter part is dedicated to some improvements of LSS which make it faster and more reliable. / by Mario Chater. / S.M.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2014. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 115-122). / Extravehicular Activity (EVA) is a critical component of human spaceflight. Working in gas-pressurized space suits, however, causes fatigue, unnecessary energy expenditure, and injury. The problem of injury is particularly acute and is exacerbated with the additional hours astronauts spend training inside the suit, especially underwater in NASA's Neutral Buoyancy Laboratory (NBL). Although space suit performance and improved system designs have been investigated, relatively little is known about how the astronaut moves and interacts with the space suit, what factors lead to injury, and how to prevent injury. At the outset of this research effort there were no technologies suitable to evaluate human movement and contact within the space suit during dynamic movements. The objective of this thesis is to help understand human-space suit interaction and design hardware to assess and ultimately mitigate injury. This is accomplished through two specific aims. The first specific aim is to use data mining techniques to uncover trends in space suit configuration, training environment, and anthropometry, which may lead to injury. Two groups of subjects were analyzed: those whose reported shoulder injury incidence is specifically attributable to the NBL or working in the space suit, and those whose shoulder problems began in active duty, meaning working in the suit could have been a contributing factor. The first statistical model correctly identifies 39% of injured subjects, while the second model correctly identifies 68% of injured subjects. For both models, percent of training incidence in the space suit planar hard upper torso (HUT) was the most important predictor variable. Frequency of training and recovery between training were also identified as significant metrics. These variables can be monitored and modified operationally to reduce the impacts on the astronaut's health. Several anthropometric dimensions were also found to have explanatory power for injury. Expanded chest depth was included in both models, while bi-deltoid breadth was relevant for identifying injured NBL subjects and shoulder circumference was relevant for identifying injured Active subjects. These dimensions may be targeted as particularly important to accommodate in future designs of the HUT or any advanced concept space suits. Finally, for the NBL subjects, previous record of injury was found to be an important factor. Further descriptive analysis implies that analyzing the HUT style and size together may be critical for future detailed studies on fit and accommodation. These results quantitatively elucidate the underlying mechanisms of shoulder injuries for astronauts working inside the space suit. The second specific aim is to develop a wearable pressure sensing capability to quantitatively measure areas on the body's surface that the space suit impacts during normal EVA movement. A low-pressure sensing system was designed and constructed for the upper body during dynamic movements inside the space suit environment. Sensors were designed to measure between 5-60 kPa with approximately 1 kPa resolution. The sensors are constructed from hyper-elastic silicone imbedded with a microfluidic channel. The channel is filled with liquid conductive metal, galinstan, such that an applied pressure corresponds to a change in resistance of the liquid metal. The system of 12 pressure sensors accommodates anthropometry from a 50th percentile female to a 95th percentile male upper body dimensions with near shirt-sleeve mobility. The wiring was intentionally designed to achieve the best trade between flexibility, resistance, and stretch ability, but ultimately was the greatest limitation in system durability. The electronics architecture utilizes onboard data storage with more than 4 hours of use. The entire system was designed with extreme environments in mind, where considerations of shock, battery hazards, and material properties in mixed gas, pressurized atmosphere were minimized to ensure user safety. The pressure sensing system was used in a human subject experiment to characterize human-suit interaction. Three experienced subjects were asked to perform a series of 3 isolated joint movements and 2 functional tasks, all focused on upper body movement. Movements were repeated 12 times each and pressure responses were evaluated both by quantifying peak pressure and full profile responses. Comparing subjective feedback to the quantitative pressure data allows a sense of the variability of movement and minor changes in loading on the body while performing suited motions. Users generally felt they were consistent for all movements. However, using a nonparametric H-test, 53% of movements were found to be biomechanically inconsistent (p < 0.05). This experiment provided the first "window" inside the suit to evaluate contact pressures and sequential indexing of the person inside the suit for realistic EVA movement. It cannot be extrapolated how changes in contact pressure would affect a subject's propensity for injury as injuries accumulate over long time scales. However, changes in pressure may be due to alterations in biomechanical strategies or fatigue, both of which could be precursors for injury and discomfort. This work focuses on the upper body, but the methods may be extended to the full body as future work. It provides solutions that could be applied beyond the field of aerospace to assess human-garment interactions and recommending armor protection for defense applications to alleviate fall impacts for medical applications. The contributions to the field include the development of a protection system that assesses and prevents injury inside gas-pressurized space suits. / by Allison Anderson. / Ph. D.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2015. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 151-160). / Reinforcement learning (RL) has great potential in robotic systems as a tool for developing policies and controllers in novel situations. However, the cost of realworld samples remains prohibitive as most RL algorithms require a large number of samples before learning near-optimal or even useful policies. Simulators are one way to decrease the number of required real-world samples, but imperfect models make deciding when and how to trust samples from a simulator difficult. Two frameworks are presented for efficient RL through the use of simulators. The first framework considers scenarios where multiple simulators of a target task are available, each with varying levels of fidelity. It is designed to limit the number of samples used in each successively higher-fidelity/cost simulator by allowing a learning agent to choose to run trajectories at the lowest level simulator that will still provide it with useful information. Theoretical proofs of this framework's sample complexity are given and empirical results are demonstrated on a robotic car with multiple simulators. The second framework focuses on problems represented with continuous states and actions, as are common in many robotics domains. Using probabilistic model-based policy search algorithms and principles of optimal control, this second framework uses data from simulators as prior information for the real-world learning. The framework is tested on a propeller-driven inverted pendulum and on a drifting robotic car. These novel frameworks enable RL algorithms to find near-optimal policies in physical robot domains with fewer expensive real-world samples than previous transfer approaches or learning without simulators. / by Mark Johnson Cutler. / Ph. D.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1997. / Includes bibliographical references (p. 67-68). / by Julian Sell. / M.S.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1991. / Includes bibliographical references (leaves 107-111). / by Laurent Paul Bonnaure. / M.S.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 2004. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Includes bibliographical references (p. 79-80). / This research effort presents the design and implementation of a supervisory controller for a 3DOF helicopter. This safety critical system is used in undergraduate laboratories in the Department of Aeronautics and Astronautics at MIT. There already exists a framework for designing a supervisory safety controller for motions about one axis. It is based on an analytical description of the safety region in state space. However, this framework cannot be easily extended to more complicated systems such as a 3DOF helicopter. In this thesis we present a different approach which uses a real-time simulation of linearized plant dynamics with a feedback law to ensure the system's safety. We describe the development of the system model, the design and implementation of the supervisory safety controller, integration of the safety controller as part of a remote laboratory and its evaluation based on its performance during laboratory exercises. / by Mariya A. Ishutkina. / S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2014. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 111-113). / Modern rocket engine turbopumps utilize cavitating inducers to meet mass and volume requirements. Rotating cavitation and higher order cavitation instabilities have frequently been observed during inducer testing and operation and can cause severe asymmetric loading on the inducer blades and shaft, potentially leading to failure of the inducer. To date no broadly applicable design method exists to characterize and suppress the onset of cavitation instabilities. This thesis presents the development of a body force model for cavitating inducers with the goal of enabling interrogation of the onset of rotating cavitation and higher order cavitation instabilities and characterization of the governing uid dynamic mechanisms. Building on body force models of gas turbine compressors for compressor stability, the model introduces an additional force component, the binormal force, to capture the strong radial flows observed in inducer ow fields. The body forces were defined and the methodology was successfully validated for two test inducers, a helical inducer and a more advanced design resembling the Space Shuttle Main Engine Low Pressure Oxidizer Pump. The head rise characteristic of each test inducer was captured with less than 4% error across the operating range and the extent of the upstream backflow region was predicted to within 18% at every operating condition. Several challenges with the blade passage model were encountered during the course of the research and the diagnostics performed to investigate them are detailed. An extension of the body force model to two-phase flows was formulated and preliminary calculations with the extended model are presented. The preliminary two-phase results are encouraging and pave the way for future assessment of rotating cavitation instabilities. / by William Alarik Sorensen. / S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Aeronautics and Astronautics, 2017. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 149-158). / The future of human space exploration will involve extra-vehicular activities (EVA) on foreign planetary surfaces (i.e. Mars), an activity that will have significantly different characteristics than exploration scenarios on Earth. These activities become challenging due to restricted vision and limitations placed on sensory feedback from altered gravity and the space suit. The use of a bulky, pressurized EVA suit perceptually disconnects human explorers from the hostile environment, increasing navigation workload and risk of collision associated with traversing through unfamiliar terrain. Due to the hazardous nature of this work, there is a critical need to design interfaces for optimizing task performance and minimizing risks; in particular, an information presentation device that can aid in obstacle avoidance during surface exploration and way-finding. Multi-modal displays are being considered as cues to multiple sensory modalities enhance cognitive processing through taking advantage of multiple sensory resources, and are believed to communicate risk more efficiently than unimodal cues. This thesis presents a wearable multi-modal interface system to examine human performance when visual, vibratory, and visual-vibratory cues are provided to aid in ground obstacle avoidance. The wearable system applies vibrotactile cues to the feet and visual cues through augmented reality glasses to convey obstacle location and proximity. An analysis of obstacle avoidance performance with the multi-modal device was performed with human subjects in a motion capture space. Metrics included completion time, subjective workload, head-down time, collisions, as well as gait parameters. The primary measures of performance were collision frequency and head-down time, as these both must be minimized in an operational environment. Results indicate that information displays enhance task performance, with the visual-only display promoting the least head-down time over tactile-only or visual-tactile displays. Head-down time was the highest for trials without a display. Results provide implications for presenting information during physically active tasks such as suited obstacle avoidance. / by Alison Eve Gibson. / S.M.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Aeronautics and Astronautics, 1985. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND AERO / Bibliography: leaves 93-94. / by Christopher Mark Greiner. / M.S.