Matching Items (31)
Description
Lunar Reconnaissance Orbiter (LRO) and MErcury Surface, Space ENvironment, GEochemistry, and Ranging (MESSENGER) spacecraft missions provide new data for investigating the youngest impact craters on Mercury and the Moon, along with lunar volcanic end-members: ancient silicic and young basaltic volcanism. The LRO Wide Angle Camera (WAC) and Narrow Angle Camera (NAC) in-flight absolute radiometric calibration used ground-based Robotic Lunar Observatory and Hubble Space Telescope data as standards. In-flight radiometric calibration is a small aspect of the entire calibration process but an important improvement upon the pre-flight measurements. Calibrated reflectance data are essential for comparing images from LRO to missions like MESSENGER, thus enabling science through engineering. Relative regolith optical maturation rates on Mercury and the Moon are estimated by comparing young impact crater densities and impact ejecta reflectance, thus empirically testing previous models of faster rates for Mercury relative to the Moon. Regolith maturation due to micrometeorite impacts and solar wind sputtering modies UV-VIS-NIR surface spectra, therefore understanding maturation rates is critical for interpreting remote sensing data from airless bodies. Results determined the regolith optical maturation rate on Mercury is 2 to 4 times faster than on the Moon. The Gruithuisen Domes, three lunar silicic volcanoes, represent relatively rare lunar lithologies possibly similar to rock fragments found in the Apollo sample collection. Lunar nonmare silicic volcanism has implications for lunar magmatic evolution. I estimated a rhyolitic composition using morphologic comparisons of the Gruithuisen Domes, measured from NAC 2-meter-per-pixel digital topographic models (DTMs), with terrestrial silicic dome morphologies and laboratory models of viscoplastic dome growth. Small, morphologically sharp irregular mare patches (IMPs) provide evidence for recent lunar volcanism widely distributed across the nearside lunar maria, which has implications for long-lived nearside magmatism. I identified 75 IMPs (100-5000 meters in dimension) in NAC images and DTMs, and determined stratigraphic relationships between units common to all IMPs. Crater counts give model ages from 18-58 Ma, and morphologic comparisons with young lunar features provided an additional age constraint of <100 Ma. The IMPs formed as low-volume basaltic eruptions significantly later than previous evidence of lunar mare basalt volcanism's end (1-1.2 Ga).
ContributorsBraden, Sarah E (Author) / Robinson, Mark S (Thesis advisor) / Bell, James F. (Committee member) / Christensen, Philip R. (Committee member) / Clarke, Amanda B (Committee member) / Lawrence, Samuel J (Committee member) / Arizona State University (Publisher)
Created2013
Description
Planetary surface studies across a range of spatial scales are key to interpreting modern and ancient operative processes and to meeting strategic mission objectives for robotic planetary science exploration. At the meter-scale and below, planetary regolith conducts heat at a rate that depends on the physical properties of the regolith particles, such as particle size, sorting, composition, and shape. Radiometric temperature measurements thus provide the means to determine regolith properties and rock abundance from afar. However, heat conduction through a matrix of irregular particles is a complicated physical system that is strongly influenced by temperature and atmospheric gas pressure. A series of new regolith thermal conductivity experiments were conducted under realistic planetary surface pressure and temperature conditions. A new model is put forth to describe the radiative, solid, and gaseous conduction terms of regolith on Earth, Mars, and airless bodies. These results will be used to infer particle size distribution from temperature measurements of the primitive asteroid Bennu to aid in OSIRIS-REx sampling site selection. Moving up in scale, fluvial processes are extremely influential in shaping Earth's surface and likely played an influential role on ancient Mars. Amphitheater-headed canyons are found on both planets, but conditions necessary for their development have been debated for many years. A spatial analysis of canyon form distribution with respect to local stratigraphy at the Escalante River and on Tarantula Mesa, Utah, indicates that canyon distribution is most closely related to variations in local rock strata, rather than groundwater spring intensity or climate variations. This implies that amphitheater-headed canyons are not simple markers of groundwater seepage erosion or megaflooding. Finally, at the largest scale, volcanism has significantly altered the surface characteristics of Earth and Mars. A field campaign was conducted in Hawaii to investigate the December 1974 Kilauea lava flow, where it was found that lava coils formed in an analogous manner to those found in Athabasca Valles, Mars. The location and size of the coils may be used as indicators of local effusion rate, viscosity, and crustal thickness.
ContributorsRyan, Andrew J (Author) / Christensen, Philip R. (Thesis advisor) / Bell, James F. (Committee member) / Whipple, Kelin X (Committee member) / Ruff, Steven W (Committee member) / Asphaug, Erik I (Committee member) / Arizona State University (Publisher)
Created2018
Description
The seasonal deposition of CO2 on the polar caps is one of the most dynamic processes on Mars and is a dominant driver of the global climate. Remote sensing temperature and albedo data were used to estimate the subliming mass of CO2 ice on south polar gullies near Sisyphi Cavi. Results showed that column mass abundances range from 400 - 1000 kg.m2 in an area less than 60 km2 in late winter. Complete sublimation of the seasonal caps may occur later than estimated by large-scale studies and is geographically dependent. Seasonal ice depth estimates suggested variations of up to 1.5 m in depth or 75% in porosity at any one time. Interannual variations in these data appeared to correlate with dust activity in the southern hemisphere. Correlation coefficients were used to investigate the relationship between frost-free surface properties and the evolution of the seasonal ice in this region. Ice on high thermal inertia units was found to disappear before any other ice, likely caused by inhibited deposition during fall. Seasonal ice springtime albedo appeared to be predominantly controlled by orientation, with north-facing slopes undergoing brightening initially in spring, then subliming before south-facing slopes. Overall, the state of seasonal ice is far more complex than globally and regionally averaged studies can identify.
The discovery of cryovolcanic features on Charon and the presence of ammonia hydrates on the surfaces of other medium-sized Kuiper Belt Objects suggests that cryovolcanism may be important to their evolution. A two-dimensional, center-point finite difference, thermal hydraulic model was developed to explore the behavior of cryovolcanic conduits on midsized KBOs. Conduits on a Charon-surrogate were shown to maintain flow through over 200 km of crust and mantle down to radii of R = 0.20 m. Radii higher than this became turbulent due to high viscous dissipation and low thermal conductivity. This model was adapted to explore the emplacement of Kubrik Mons. Steady state flow was achieved with a conduit of radius R = 0.02 m for a source chamber at 2.3 km depth. Effusion rates computed from this estimated a 122 - 163 Myr upper limit formation timescale.
The discovery of cryovolcanic features on Charon and the presence of ammonia hydrates on the surfaces of other medium-sized Kuiper Belt Objects suggests that cryovolcanism may be important to their evolution. A two-dimensional, center-point finite difference, thermal hydraulic model was developed to explore the behavior of cryovolcanic conduits on midsized KBOs. Conduits on a Charon-surrogate were shown to maintain flow through over 200 km of crust and mantle down to radii of R = 0.20 m. Radii higher than this became turbulent due to high viscous dissipation and low thermal conductivity. This model was adapted to explore the emplacement of Kubrik Mons. Steady state flow was achieved with a conduit of radius R = 0.02 m for a source chamber at 2.3 km depth. Effusion rates computed from this estimated a 122 - 163 Myr upper limit formation timescale.
ContributorsMount, Christopher (Author) / Christensen, Philip R. (Thesis advisor) / Desch, Steven J (Committee member) / Bell, James F. (Committee member) / Clarke, Amanda B (Committee member) / Whipple, Kelin X (Committee member) / Arizona State University (Publisher)
Created2019
Description
Interpreting the petrogenesis of materials exposed on the surface of planets and asteroids is fundamental to understanding the origins and evolution of the inner Solar System. Temperature, pressure, fO2, and bulk composition directly influence the petrogenetic history of planetary surfaces and constraining these variables with remote sensing techniques is challenging. The integration of remote sensing data with analytical investigations of natural samples, lab-based spectroscopy, and thermodynamic modelling improves our ability to interpret the petrogenesis of planetary materials.
A suite of naturally heated carbonaceous chondrite material was studied with lab-based spectroscopic techniques, including visible near-infrared and Fourier transform infrared reflectance spectroscopy. Distinct mineralogic, and thus spectroscopic, trends are observed with increasing degree of thermal metamorphism. Characterization of these spectral trends yields a set of mappable parameters that will be applied to remotely sensed data from the OSIRIS-REx science payload. Information about the thermal history of the surface of the asteroid Bennu will aid in the selection of a sampling site, ensuring OSIRIS-REx collects a pristine regolith sample that has not experienced devolatilization of primitive organics or dehydration of phyllosilicates.
The evolution of mafic magma results in distinct major element chemical trends. Mineral assemblages present in evolved volcanic rocks are indicators of these processes. Using laboratory spectroscopic analyses of a suite of evolved volcanic rocks from the Snake River Plain, Idaho, I show that these evolutionary trends are reflected in the spectral signatures of ferromagenesian and feldspar minerals.
The Athena science package on the Mars Exploration Rover Spirit allows for the in situ investigation of bulk chemistry, texture, and mineralogy on the surface of Mars. Using the bulk composition of the Irvine and Backstay volcanic rocks, thermodynamic modeling was performed to further constrain the formation conditions of Martian volcanics. Irvine and Backstay compositions exhibit dramatic variations in modal mineralogy with changing fO2. Using these results, I show that the observed Mini-TES spectra of Irvine and Backstay can be adequately reproduced, and additional constraints can be placed on their primary fO2.
A suite of naturally heated carbonaceous chondrite material was studied with lab-based spectroscopic techniques, including visible near-infrared and Fourier transform infrared reflectance spectroscopy. Distinct mineralogic, and thus spectroscopic, trends are observed with increasing degree of thermal metamorphism. Characterization of these spectral trends yields a set of mappable parameters that will be applied to remotely sensed data from the OSIRIS-REx science payload. Information about the thermal history of the surface of the asteroid Bennu will aid in the selection of a sampling site, ensuring OSIRIS-REx collects a pristine regolith sample that has not experienced devolatilization of primitive organics or dehydration of phyllosilicates.
The evolution of mafic magma results in distinct major element chemical trends. Mineral assemblages present in evolved volcanic rocks are indicators of these processes. Using laboratory spectroscopic analyses of a suite of evolved volcanic rocks from the Snake River Plain, Idaho, I show that these evolutionary trends are reflected in the spectral signatures of ferromagenesian and feldspar minerals.
The Athena science package on the Mars Exploration Rover Spirit allows for the in situ investigation of bulk chemistry, texture, and mineralogy on the surface of Mars. Using the bulk composition of the Irvine and Backstay volcanic rocks, thermodynamic modeling was performed to further constrain the formation conditions of Martian volcanics. Irvine and Backstay compositions exhibit dramatic variations in modal mineralogy with changing fO2. Using these results, I show that the observed Mini-TES spectra of Irvine and Backstay can be adequately reproduced, and additional constraints can be placed on their primary fO2.
ContributorsHaberle, Christopher William (Author) / Christensen, Philip R. (Thesis advisor) / Garvie, Laurence A. J. (Committee member) / Bell, James F. (Committee member) / Ruff, Steven W. (Committee member) / Hervig, Richard L. (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
Previous workers hypothesized that lunar Localized Pyroclastic Deposits (LPDs) represent products of vulcanian-style eruptions, since some have low proportions of juvenile material. The objective of the first study is to determine how juvenile composition, calculated using deposit and vent volumes, varies among LPDs. I used Lunar Reconnaissance Orbiter Camera Narrow Angle Camera (LROC NAC) digital terrain models (DTMs) to generate models of pre-eruption surfaces for 23 LPDs and subtracted them from the NAC DTMs to calculate deposit and vent volumes. Results show that LPDs have a wide range of juvenile compositions and thinning profiles, and that there is a positive relationship between juvenile material proportion and deposit size. These findings indicate there is greater diversity among LPDs than previously understood, and that a simple vulcanian eruption model may only apply to the smallest deposits.
There is consensus that martian outflow channels were formed by catastrophic flooding events, yet many of these channels exhibit lava flow features issuing from the same source as the eroded channels, leading some authors to suggest that lava may have served as their sole agent of erosion. This debate is addressed in two studies that use Context Camera images for photogeologic analysis, geomorphic mapping, and cratering statistics: (1) A study of Mangala Valles showing that it underwent at least two episodes of fluvial activity and at least three episodes of volcanic activity during the Late Amazonian, consistent with alternating episodes of flooding and volcanism. (2) A study of Maja Valles finds that it is thinly draped in lava flows sourced from Lunae Planum to the west, rendering it analogous to the lava-coated Elysium outflow systems. However, the source of eroded channels in Maja Valles is not the source of the its lava flows, which instead issue from south Lunae Planum. The failure of these lava flows to generate any major channels along their path suggests that the channels of Maja Valles are not lava-eroded.
Finally, I describe a method of locating sharp edges in out-of-focus images for application to automated trajectory control systems that use images from fixed-focus cameras to determine proximity to a target.
There is consensus that martian outflow channels were formed by catastrophic flooding events, yet many of these channels exhibit lava flow features issuing from the same source as the eroded channels, leading some authors to suggest that lava may have served as their sole agent of erosion. This debate is addressed in two studies that use Context Camera images for photogeologic analysis, geomorphic mapping, and cratering statistics: (1) A study of Mangala Valles showing that it underwent at least two episodes of fluvial activity and at least three episodes of volcanic activity during the Late Amazonian, consistent with alternating episodes of flooding and volcanism. (2) A study of Maja Valles finds that it is thinly draped in lava flows sourced from Lunae Planum to the west, rendering it analogous to the lava-coated Elysium outflow systems. However, the source of eroded channels in Maja Valles is not the source of the its lava flows, which instead issue from south Lunae Planum. The failure of these lava flows to generate any major channels along their path suggests that the channels of Maja Valles are not lava-eroded.
Finally, I describe a method of locating sharp edges in out-of-focus images for application to automated trajectory control systems that use images from fixed-focus cameras to determine proximity to a target.
ContributorsKeske, Amber (Author) / Christensen, Philip R. (Thesis advisor) / Robinson, Mark S (Committee member) / Clarke, Amanda B (Committee member) / Whipple, Kelin X (Committee member) / Bell, James F. (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