Matching Items (33)
Description
Earth's topographic surface forms an interface across which the geodynamic and geomorphic engines interact. This interaction is best observed along crustal margins where topography is created by active faulting and sculpted by geomorphic processes. Crustal deformation manifests as earthquakes at centennial to millennial timescales. Given that nearly half of Earth's human population lives along active fault zones, a quantitative understanding of the mechanics of earthquakes and faulting is necessary to build accurate earthquake forecasts. My research relies on the quantitative documentation of the geomorphic expression of large earthquakes and the physical processes that control their spatiotemporal distributions. The first part of my research uses high-resolution topographic lidar data to quantitatively document the geomorphic expression of historic and prehistoric large earthquakes. Lidar data allow for enhanced visualization and reconstruction of structures and stratigraphy exposed by paleoseismic trenches. Lidar surveys of fault scarps formed by the 1992 Landers earthquake document the centimeter-scale erosional landforms developed by repeated winter storm-driven erosion. The second part of my research employs a quasi-static numerical earthquake simulator to explore the effects of fault roughness, friction, and structural complexities on earthquake-generated deformation. My experiments show that fault roughness plays a critical role in determining fault-to-fault rupture jumping probabilities. These results corroborate the accepted 3-5 km rupture jumping distance for smooth faults. However, my simulations show that the rupture jumping threshold distance is highly variable for rough faults due to heterogeneous elastic strain energies. Furthermore, fault roughness controls spatiotemporal variations in slip rates such that rough faults exhibit lower slip rates relative to their smooth counterparts. The central implication of these results lies in guiding the interpretation of paleoseismically derived slip rates that are used to form earthquake forecasts. The final part of my research evaluates a set of Earth science-themed lesson plans that I designed for elementary-level learning-disabled students. My findings show that a combination of concept delivery techniques is most effective for learning-disabled students and should incorporate interactive slide presentations, tactile manipulatives, teacher-assisted concept sketches, and student-led teaching to help learning-disabled students grasp Earth science concepts.
ContributorsHaddad, David Elias (Author) / Arrowsmith, Ramon (Thesis advisor) / Reynolds, Stephen (Committee member) / Semken, Steven (Committee member) / Shirzaei, Manoochehr (Committee member) / Whipple, Kelin (Committee member) / Zielke, Olaf (Committee member) / Arizona State University (Publisher)
Created2014
Description
Aquifers host the largest accessible freshwater resource in the world. However, groundwater reserves are declining in many places. Often coincident with drought, high extraction rates and inadequate replenishment result in groundwater overdraft and permanent land subsidence. Land subsidence is the cause of aquifer storage capacity reduction, altered topographic gradients which can exacerbate floods, and differential displacement that can lead to earth fissures and infrastructure damage. Improving understanding of the sources and mechanisms driving aquifer deformation is important for resource management planning and hazard mitigation.
Poroelastic theory describes the coupling of differential stress, strain, and pore pressure, which are modulated by material properties. To model these relationships, displacement time series are estimated via satellite interferometry and hydraulic head levels from observation wells provide an in-situ dataset. In combination, the deconstruction and isolation of selected time-frequency components allow for estimating aquifer parameters, including the elastic and inelastic storage coefficients, compaction time constants, and vertical hydraulic conductivity. Together these parameters describe the storage response of an aquifer system to changes in hydraulic head and surface elevation. Understanding aquifer parameters is useful for the ongoing management of groundwater resources.
Case studies in Phoenix and Tucson, Arizona, focus on land subsidence from groundwater withdrawal as well as distinct responses to artificial recharge efforts. In Christchurch, New Zealand, possible changes to aquifer properties due to earthquakes are investigated. In Houston, Texas, flood severity during Hurricane Harvey is linked to subsidence, which modifies base flood elevations and topographic gradients.
Poroelastic theory describes the coupling of differential stress, strain, and pore pressure, which are modulated by material properties. To model these relationships, displacement time series are estimated via satellite interferometry and hydraulic head levels from observation wells provide an in-situ dataset. In combination, the deconstruction and isolation of selected time-frequency components allow for estimating aquifer parameters, including the elastic and inelastic storage coefficients, compaction time constants, and vertical hydraulic conductivity. Together these parameters describe the storage response of an aquifer system to changes in hydraulic head and surface elevation. Understanding aquifer parameters is useful for the ongoing management of groundwater resources.
Case studies in Phoenix and Tucson, Arizona, focus on land subsidence from groundwater withdrawal as well as distinct responses to artificial recharge efforts. In Christchurch, New Zealand, possible changes to aquifer properties due to earthquakes are investigated. In Houston, Texas, flood severity during Hurricane Harvey is linked to subsidence, which modifies base flood elevations and topographic gradients.
ContributorsMiller, Megan Marie (Author) / Shirzaei, Manoochehr (Thesis advisor) / Reynolds, Stephen (Committee member) / Tyburczy, James (Committee member) / Semken, Steven (Committee member) / Werth, Susanna (Committee member) / Arizona State University (Publisher)
Created2018
Description
This dissertation is focused on developing an algorithm to provide current state estimation and future state predictions for biomechanical human walking features. The goal is to develop a system which is capable of evaluating the current action a subject is taking while walking and then use this to predict the future states of biomechanical features.
This work focuses on the exploration and analysis of Interaction Primitives (Amor er al, 2014) and their relevance to biomechanical prediction for human walking. Built on the framework of Probabilistic Movement Primitives, Interaction Primitives utilize an EKF SLAM algorithm to localize and map a distribution over the weights of a set of basis functions. The prediction properties of Bayesian Interaction Primitives were utilized to predict real-time foot forces from a 9 degrees of freedom IMUs mounted to a subjects tibias. This method shows that real-time human biomechanical features can be predicted and have a promising link to real-time controls applications.
This work focuses on the exploration and analysis of Interaction Primitives (Amor er al, 2014) and their relevance to biomechanical prediction for human walking. Built on the framework of Probabilistic Movement Primitives, Interaction Primitives utilize an EKF SLAM algorithm to localize and map a distribution over the weights of a set of basis functions. The prediction properties of Bayesian Interaction Primitives were utilized to predict real-time foot forces from a 9 degrees of freedom IMUs mounted to a subjects tibias. This method shows that real-time human biomechanical features can be predicted and have a promising link to real-time controls applications.
ContributorsClark, Geoffrey Mitchell (Author) / Ben Amor, Heni (Thesis advisor) / Si, Jennie (Committee member) / Berisha, Visar (Committee member) / Arizona State University (Publisher)
Created2018
Description
Reinforcement learning (RL) is a powerful methodology for teaching autonomous agents complex behaviors and skills. A critical component in most RL algorithms is the reward function -- a mathematical function that provides numerical estimates for desirable and undesirable states. Typically, the reward function must be hand-designed by a human expert and, as a result, the scope of a robot's autonomy and ability to safely explore and learn in new and unforeseen environments is constrained by the specifics of the designed reward function. In this thesis, I design and implement a stateful collision anticipation model with powerful predictive capability based upon my research of sequential data modeling and modern recurrent neural networks. I also develop deep reinforcement learning methods whose rewards are generated by self-supervised training and intrinsic signals. The main objective is to work towards the development of resilient robots that can learn to anticipate and avoid damaging interactions by combining visual and proprioceptive cues from internal sensors. The introduced solutions are inspired by pain pathways in humans and animals, because such pathways are known to guide decision-making processes and promote self-preservation. A new "robot dodge ball' benchmark is introduced in order to test the validity of the developed algorithms in dynamic environments.
ContributorsRichardson, Trevor W (Author) / Ben Amor, Heni (Thesis advisor) / Yang, Yezhou (Committee member) / Srivastava, Siddharth (Committee member) / Arizona State University (Publisher)
Created2018
Description
The dynamic Earth involves feedbacks between the solid crust and both natural and anthropogenic fluid flows. Fluid-rock interactions drive many Earth phenomena, including volcanic unrest, seismic activities, and hydrological responses. Mitigating the hazards associated with these activities requires fundamental understanding of the underlying physical processes. Therefore, geophysical monitoring in combination with modeling provides valuable tools, suitable for hazard mitigation and risk management efforts. Magmatic activities and induced seismicity linked to fluid injection are two natural and anthropogenic processes discussed in this dissertation.
Successful forecasting of the timing, style, and intensity of a volcanic eruption is made possible by improved understanding of the volcano life cycle as well as building quantitative models incorporating the processes that govern rock melting, melt ascending, magma storage, eruption initiation, and interaction between magma and surrounding host rocks at different spatial extent and time scale. One key part of such models is the shallow magma chamber, which is generally directly linked to volcano’s eruptive behaviors. However, its actual shape, size, and temporal evolution are often not entirely known. To address this issue, I use space-based geodetic data with high spatiotemporal resolution to measure surface deformation at Kilauea volcano. The obtained maps of InSAR (Interferometric Synthetic Aperture Radar) deformation time series are exploited with two novel modeling schemes to investigate Kilauea’s shallow magmatic system. Both models can explain the same observation, leading to a new compartment model of magma chamber. Such models significantly advance the understanding of the physical processes associated with Kilauea’s summit plumbing system with potential applications for volcanoes around the world.
The unprecedented increase in the number of earthquakes in the Central and Eastern United States since 2008 is attributed to massive deep subsurface injection of saltwater. The elevated chance of moderate-large damaging earthquakes stemming from increased seismicity rate causes broad societal concerns among industry, regulators, and the public. Thus, quantifying the time-dependent seismic hazard associated with the fluid injection is of great importance. To this end, I investigate the large-scale seismic, hydrogeologic, and injection data in northern Texas for period of 2007-2015 and in northern-central Oklahoma for period of 1995-2017. An effective induced earthquake forecasting model is developed, considering a complex relationship between injection operations and consequent seismicity. I find that the timing and magnitude of regional induced earthquakes are fully controlled by the process of fluid diffusion in a poroelastic medium and thus can be successfully forecasted. The obtained time-dependent seismic hazard model is spatiotemporally heterogeneous and decreasing injection rates does not immediately reduce the probability of an earthquake. The presented framework can be used for operational induced earthquake forecasting. Information about the associated fundamental processes, inducing conditions, and probabilistic seismic hazards has broad benefits to the society.
Successful forecasting of the timing, style, and intensity of a volcanic eruption is made possible by improved understanding of the volcano life cycle as well as building quantitative models incorporating the processes that govern rock melting, melt ascending, magma storage, eruption initiation, and interaction between magma and surrounding host rocks at different spatial extent and time scale. One key part of such models is the shallow magma chamber, which is generally directly linked to volcano’s eruptive behaviors. However, its actual shape, size, and temporal evolution are often not entirely known. To address this issue, I use space-based geodetic data with high spatiotemporal resolution to measure surface deformation at Kilauea volcano. The obtained maps of InSAR (Interferometric Synthetic Aperture Radar) deformation time series are exploited with two novel modeling schemes to investigate Kilauea’s shallow magmatic system. Both models can explain the same observation, leading to a new compartment model of magma chamber. Such models significantly advance the understanding of the physical processes associated with Kilauea’s summit plumbing system with potential applications for volcanoes around the world.
The unprecedented increase in the number of earthquakes in the Central and Eastern United States since 2008 is attributed to massive deep subsurface injection of saltwater. The elevated chance of moderate-large damaging earthquakes stemming from increased seismicity rate causes broad societal concerns among industry, regulators, and the public. Thus, quantifying the time-dependent seismic hazard associated with the fluid injection is of great importance. To this end, I investigate the large-scale seismic, hydrogeologic, and injection data in northern Texas for period of 2007-2015 and in northern-central Oklahoma for period of 1995-2017. An effective induced earthquake forecasting model is developed, considering a complex relationship between injection operations and consequent seismicity. I find that the timing and magnitude of regional induced earthquakes are fully controlled by the process of fluid diffusion in a poroelastic medium and thus can be successfully forecasted. The obtained time-dependent seismic hazard model is spatiotemporally heterogeneous and decreasing injection rates does not immediately reduce the probability of an earthquake. The presented framework can be used for operational induced earthquake forecasting. Information about the associated fundamental processes, inducing conditions, and probabilistic seismic hazards has broad benefits to the society.
ContributorsZhai, Guang (Author) / Shirzaei, Manoochehr (Thesis advisor) / Garnero, Edward (Committee member) / Clarke, Amanda (Committee member) / Tyburczy, James (Committee member) / Li, Mingming (Committee member) / Arizona State University (Publisher)
Created2018
Description
The movement between tectonic plates is accommodated through brittle (elastic) displacement on the plate boundary faults and ductile permanent deformation on the fault borderland. The elastic displacement along the fault can occur in the form of either large seismic events or aseismic slip, known as fault creep. Fault creep mainly occurs at the deep ductile portion of the crust, where the temperature is high. Nonetheless, aseismic creep can also occur on the shallow brittle portion of the fault segments that are characterized by frictionally weak material, elevated pore fluid pressure, or geometrical complexity. Creeping segments are assumed to safely release the accumulated strain(Kodaira et al., 2004; Rice, 1992)(Kodaira et al., 2004; Rice, 1992)(Kodaira et al., 2004; Rice, 1992)(Kodaira et al., 2004; Rice, 1992)(Kodaira et al., 2004; Rice, 1992) on the fault and also impede propagation of the seismic rupture. The rate of aseismic slip on creeping faults, however, might not be steady in time and instead consist of successive periods of acceleration and deceleration, known as slow slip events (SSEs). SSEs, which aseismically release the strain energy over a period of days to months, rather than the seconds to minutes characteristic of a typical earthquake, have been interpreted as earthquake precursors and as possible triggering factor for major earthquakes. Therefore, understanding the partitioning of seismic and aseismic fault slip and evolution of creep is fundamental to constraining the fault earthquake potential and improving operational seismic hazard models. Thanks to advances in tectonic geodesy, it is now possible to detect the fault movement in high spatiotemporal resolution and develop kinematic models of the creep evolution on the fault to determine the budget of seismic and aseismic slip.
In this dissertation, I measure the decades-long time evolution of fault-related crustal deformation along the San Andrea Fault in California and the northeast Japan subduction zone using space-borne geodetic techniques, such as Global Positioning System (GPS) and Interferometric Synthetic Aperture Radar (InSAR). The surface observation of deformation combined with seismic data set allow constraining the time series of creep distribution on the fault surface at seismogenic depth. The obtained time-dependent kinematic models reveal that creep in both study areas evolves through a series of SSEs, each lasting for several months. Using physics-based models informed by laboratory experiments, I show that the transient elevation of pore fluid pressure is the driving mechanism of SSEs. I further investigate the link between SSEs and evolution of seismicity on neighboring locked segments, which has implications for seismic hazard models and also provides insights into the pattern of microstructure on the fault surface. I conclude that while creeping segments act as seismic rupture barriers, SSEs on these zones might promote seismicity on adjacent seismogenic segments, thus change the short-term earthquake forecast.
In this dissertation, I measure the decades-long time evolution of fault-related crustal deformation along the San Andrea Fault in California and the northeast Japan subduction zone using space-borne geodetic techniques, such as Global Positioning System (GPS) and Interferometric Synthetic Aperture Radar (InSAR). The surface observation of deformation combined with seismic data set allow constraining the time series of creep distribution on the fault surface at seismogenic depth. The obtained time-dependent kinematic models reveal that creep in both study areas evolves through a series of SSEs, each lasting for several months. Using physics-based models informed by laboratory experiments, I show that the transient elevation of pore fluid pressure is the driving mechanism of SSEs. I further investigate the link between SSEs and evolution of seismicity on neighboring locked segments, which has implications for seismic hazard models and also provides insights into the pattern of microstructure on the fault surface. I conclude that while creeping segments act as seismic rupture barriers, SSEs on these zones might promote seismicity on adjacent seismogenic segments, thus change the short-term earthquake forecast.
ContributorsKhoshmanesh, Mostafa (Author) / Shirzaei, Manoochehr (Thesis advisor) / Arrowsmith, Ramon (Committee member) / Garnero, Edward (Committee member) / Tyburczy, James (Committee member) / Whipple, Kelin (Committee member) / Arizona State University (Publisher)
Created2018
Description
This thesis presents a family of adaptive curvature methods for gradient-based stochastic optimization. In particular, a general algorithmic framework is introduced along with a practical implementation that yields an efficient, adaptive curvature gradient descent algorithm. To this end, a theoretical and practical link between curvature matrix estimation and shrinkage methods for covariance matrices is established. The use of shrinkage improves estimation accuracy of the curvature matrix when data samples are scarce. This thesis also introduce several insights that result in data- and computation-efficient update equations. Empirical results suggest that the proposed method compares favorably with existing second-order techniques based on the Fisher or Gauss-Newton and with adaptive stochastic gradient descent methods on both supervised and reinforcement learning tasks.
ContributorsBarron, Trevor (Author) / Ben Amor, Heni (Thesis advisor) / He, Jingrui (Committee member) / Levihn, Martin (Committee member) / Arizona State University (Publisher)
Created2019
Description
Automated driving systems are in an intensive research and development stage, and the companies developing these systems are targeting to deploy them on public roads in a very near future. Guaranteeing safe operation of these systems is crucial as they are planned to carry passengers and share the road with other vehicles and pedestrians. Yet, there is no agreed-upon approach on how and in what detail those systems should be tested. Different organizations have different testing approaches, and one common approach is to combine simulation-based testing with real-world driving.
One of the expectations from fully-automated vehicles is never to cause an accident. However, an automated vehicle may not be able to avoid all collisions, e.g., the collisions caused by other road occupants. Hence, it is important for the system designers to understand the boundary case scenarios where an autonomous vehicle can no longer avoid a collision. Besides safety, there are other expectations from automated vehicles such as comfortable driving and minimal fuel consumption. All safety and functional expectations from an automated driving system should be captured with a set of system requirements. It is challenging to create requirements that are unambiguous and usable for the design, testing, and evaluation of automated driving systems. Another challenge is to define useful metrics for assessing the testing quality because in general, it is impossible to test every possible scenario.
The goal of this dissertation is to formalize the theory for testing automated vehicles. Various methods for automatic test generation for automated-driving systems in simulation environments are presented and compared. The contributions presented in this dissertation include (i) new metrics that can be used to discover the boundary cases between safe and unsafe driving conditions, (ii) a new approach that combines combinatorial testing and optimization-guided test generation methods, (iii) approaches that utilize global optimization methods and random exploration to generate critical vehicle and pedestrian trajectories for testing purposes, (iv) a publicly-available simulation-based automated vehicle testing framework that enables application of the existing testing approaches in the literature, including the new approaches presented in this dissertation.
One of the expectations from fully-automated vehicles is never to cause an accident. However, an automated vehicle may not be able to avoid all collisions, e.g., the collisions caused by other road occupants. Hence, it is important for the system designers to understand the boundary case scenarios where an autonomous vehicle can no longer avoid a collision. Besides safety, there are other expectations from automated vehicles such as comfortable driving and minimal fuel consumption. All safety and functional expectations from an automated driving system should be captured with a set of system requirements. It is challenging to create requirements that are unambiguous and usable for the design, testing, and evaluation of automated driving systems. Another challenge is to define useful metrics for assessing the testing quality because in general, it is impossible to test every possible scenario.
The goal of this dissertation is to formalize the theory for testing automated vehicles. Various methods for automatic test generation for automated-driving systems in simulation environments are presented and compared. The contributions presented in this dissertation include (i) new metrics that can be used to discover the boundary cases between safe and unsafe driving conditions, (ii) a new approach that combines combinatorial testing and optimization-guided test generation methods, (iii) approaches that utilize global optimization methods and random exploration to generate critical vehicle and pedestrian trajectories for testing purposes, (iv) a publicly-available simulation-based automated vehicle testing framework that enables application of the existing testing approaches in the literature, including the new approaches presented in this dissertation.
ContributorsTuncali, Cumhur Erkan (Author) / Fainekos, Georgios (Thesis advisor) / Ben Amor, Heni (Committee member) / Kapinski, James (Committee member) / Shrivastava, Aviral (Committee member) / Arizona State University (Publisher)
Created2019
Description
In the field of machine learning, reinforcement learning stands out for its ability to explore approaches to complex, high dimensional problems that outperform even expert humans. For robotic locomotion tasks reinforcement learning provides an approach to solving them without the need for unique controllers. In this thesis, two reinforcement learning algorithms, Deep Deterministic Policy Gradient and Group Factor Policy Search are compared based upon their performance in the bipedal walking environment provided by OpenAI gym. These algorithms are evaluated on their performance in the environment and their sample efficiency.
ContributorsMcDonald, Dax (Author) / Ben Amor, Heni (Thesis director) / Yang, Yezhou (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2018-12
Description
Lyric classification and generation are trending in topics in the machine learning community. Long Short-Term Networks (LSTMs) are effective tools for classifying and generating text. We explored their effectiveness in the generation and classification of lyrical data and proposed methods of evaluating their accuracy. We found that LSTM networks with dropout layers were effective at lyric classification. We also found that Word embedding LSTM networks were extremely effective at lyric generation.
ContributorsTallapragada, Amit (Author) / Ben Amor, Heni (Thesis director) / Caviedes, Jorge (Committee member) / Computer Science and Engineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2019-05