Search NASA⌕ Search

SEARCH · Search NASA

Results for “data augmentation”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
NASA NTRS1993

DMS augmented monitoring and diganosis application (DMS AMDA) prototype

The Data Management System Augmented Monitoring and Diagnosis Application (DMS AMDA) is currently under development at NASA Ames Research Center (ARC). It will provide automated monitoring and diagnosis capabilities for the Space Station Freedom (SSF) Data Management System (DMS) in the Control Center Complex (CCC) at NASA Johnson Space Center. Several advanced automation applications are under development for use in the CCC for other SSF subsystems. The DMS AMDA, however, is the first application to utilize digraph failure analysis techniques and the Extended Realtime FEAT (ERF) application as the core of its diagnostic system design, since the other projects were begun before the digraph tools were available. Model-based diagnosis and expert systems techniques will provide additional capabilities and augment ERF where appropriate. Utilization of system knowledge captured in the design phase of a system in digraphs should result in both a cost savings and a technical advantage during implementation of the diagnostic software. This paper addresses both the programmatic and technical considerations of this approach, and describes the software design and initial prototyping effort.

Patterson-Hine, F. A.↗
NASA NTRS1971

A method for determining the response of space shuttle to atmospheric turbulence. Volume 2: Computer program description and usage instructions

A description is given of a computer program which calculates the response of an elastic aerospace vehicle to random and discrete turbulence. The program uses vibration modes as generalized coordinates. Inputs to the program include modal information, generalized aerodynamic forces, flight data, stability augmentation system description, and description of the disturbing function. Program outputs include tabulations of the ratio of root-mean-square response to root-mean-square turbulence and response upward zero crossings per second, and SC4020 plots of transfer functions, response power spectral densities, and Nyquist stability traces.

Huntington, R.↗
NASA NTRS1981

Inventory of wetlands and agricultural land cover in the upper Sevier River Basin, Utah

The use of color infrared aerial photography in the mapping of agricultural land use and wetlands in the Sevier River Basin of south central utah is described. The efficiency and cost effectiveness of utilizing LANDSAT multispectral scanner digital data to augment photographic interpretations are discussed. Transparent overlays for 27 quadrangles showing delineations of wetlands and agricultural land cover were produced. A table summarizing the acreage represented by each class on each quadrangle overlay is provided.

Jaynes, R. A.↗
NASA NTRS1989

The meteor radar as a tool for upper atmosphere research

Meteor radar provide measurements of the upper mesosphere and lower thermosphere neutral wind field by using the reflection of electromagnetic waves from meteor trails. These radars are relatively inexpensive and provide an excellent means of monitoring the mean winds and tides in the 80 to 100 km region. Recently new techniques were developed to detect meteor echoes from other ground based radar systems operating in the HF/VHF frequency range. The meteor echo information augments the data that are routinely collected by these radars. These new techniques are discussed.

Avery, S. K.↗
NASA NTRS1991

Automatic analysis of stereoscopic satellite image pairs for determination of cloud-top height and structure

Results are presented on an automatic stereo analysis of cloud-top heights from nearly simultaneous satellite image pairs from the GOES and NOAA satellites, using a massively parallel processor computer. Comparisons of computer-derived height fields and manually analyzed fields show that the automatic analysis technique shows promise for performing routine stereo analysis in a real-time environment, providing a useful forecasting tool by augmenting observational data sets of severe thunderstorms and hurricanes. Simulations using synthetic stereo data show that it is possible to automatically resolve small-scale features such as 4000-m-diam clouds to about 1500 m in the vertical.

Hasler, A. F.↗
NASA NTRS1992

Accurate far-infrared rotational frequencies of carbon monoxide

This study presents high-resolution measurements of the pure rotational absorption spectrum of CO in its ground state for the range J arcsec - 5-37. A least-squares fit to this data set, augmented by previous microwave measurements of the J arcsec = 0-4 rotational transitions in the literature, determined accurate values for the molecular constants. A table of calculated CO rotational frequencies is provided for the range J arcsec = 0-45.

Varberg, Thomas D.↗
NASA NTRS1996

Deforestation and Biogenic Trace Emissions from Brazilian Cerrado

The overall goal of NASA's SCAR (Smoke, Cloud and Radiation) Program is to obtain physical and chemical properties of the smoke produced by biomass burning and the effects of the smoke on the earth's radiation balance and climate. It is a joint project with the Brazilian government and their organizations, including INPE (Instituto Nacional Pesquisas Espaciais) who actively participate in all activities. Appropriate estimates of the biomass buming in the tropics is therefore essential to determine its effect on the atmosphere and on climate. The SCAR series of experiments is designed with that purpose. The present study of evaluating the burnt-out areas is to augment the data collected to date to help evaluate the effect of biomass burning.

Sinha, Ravi↗
NASA NTRS2002

Modeling and Simulation of a Helicopter Slung Load Stabilization Device

This paper addresses the problem of simulation and stabilization of the yaw motions of a cargo container slung load. The study configuration is a UH-60 helicopter carrying a 6ft x 6 ft x 8 ft CONEX container. This load is limited to 60 KIAS in operations and flight testing indicates that it starts spinning in hover and that spin rate increases with airspeed. The simulation reproduced the load yaw motions seen in the flight data after augmenting the load model with terms representing unsteady load yaw moment effects acting to reinforce load oscillations, and augmenting the hook model to include yaw resistance at the hook. The use of a vertical fin to stabilize the load is considered. Results indicate that the CONEX airspeed can be extended to 110 kts using a 3x5 ft fin.

Cicolani, Luigi S.↗
NASA NTRS2006

Working Group Reports and Presentations: Virtual Worlds and Virtual Exploration

Scientists and engineers are continually developing innovative methods to capitalize on recent developments in computational power. Virtual worlds and virtual exploration present a new toolset for project design, implementation, and resolution. Replication of the physical world in the virtual domain provides stimulating displays to augment current data analysis techniques and to encourage public participation. In addition, the virtual domain provides stakeholders with a low cost, low risk design and test environment. The following document defines a virtual world and virtual exploration, categorizes the chief motivations for virtual exploration, elaborates upon specific objectives, identifies roadblocks and enablers for realizing the benefits, and highlights the more immediate areas of implementation (i.e. the action items). While the document attempts a comprehensive evaluation of virtual worlds and virtual exploration, the innovative nature of the opportunities presented precludes completeness. The authors strongly encourage readers to derive additional means of utilizing the virtual exploration toolset.

LAmoreaux, Claudia↗
NASA NTRS2011

Prognostics of Power MOSFET

This paper demonstrates how to apply prognostics to power MOSFETs (metal oxide field effect transistor). The methodology uses thermal cycling to age devices and Gaussian process regression to perform prognostics. The approach is validated with experiments on 100V power MOSFETs. The failure mechanism for the stress conditions is determined to be die-attachment degradation. Change in ON-state resistance is used as a precursor of failure due to its dependence on junction temperature. The experimental data is augmented with a finite element analysis simulation that is based on a two-transistor model. The simulation assists in the interpretation of the degradation phenomena and SOA (safe operation area) change.

Celaya, Jose Ramon↗
NASA NTRS2015

An Active Learning Framework for Hyperspectral Image Classification Using Hierarchical Segmentation

Augmenting spectral data with spatial information for image classification has recently gained significant attention, as classification accuracy can often be improved by extracting spatial information from neighboring pixels. In this paper, we propose a new framework in which active learning (AL) and hierarchical segmentation (HSeg) are combined for spectral-spatial classification of hyperspectral images. The spatial information is extracted from a best segmentation obtained by pruning the HSeg tree using a new supervised strategy. The best segmentation is updated at each iteration of the AL process, thus taking advantage of informative labeled samples provided by the user. The proposed strategy incorporates spatial information in two ways: 1) concatenating the extracted spatial features and the original spectral features into a stacked vector and 2) extending the training set using a self-learning-based semi-supervised learning (SSL) approach. Finally, the two strategies are combined within an AL framework. The proposed framework is validated with two benchmark hyperspectral datasets. Higher classification accuracies are obtained by the proposed framework with respect to five other state-of-the-art spectral-spatial classification approaches. Moreover, the effectiveness of the proposed pruning strategy is also demonstrated relative to the approaches based on a fixed segmentation.

classification↗
NASA NTRSDate not supplied

Coupling Online Conductivity with Offline Ion Chromatography Measurements using the Particle-into-Liquid Sampler

A particle-into-liquid sampler has been combined with a flow through conductivity cell to provide a continuous, non-destructive, online measurement in support of offline ion chromatography analysis. The conductivity measurement provides a rapid assessment of the total ion concentration augmenting the slower batch data from the offline analysis and is developed primarily to assist airborne measurements, where fast time response is essential. A model of the conductivity was developed for measured ions and excellent closure is derived for laboratory-generated aerosols. The PILS-conductivity measurement was extensively tested during the NASA Cloud, Aerosol and Monsoon Processes: Philippines Experiment (CAMP2Ex) across nineteen research flights and the conductivity data was found to augment the temporal capability of the PILS instrument for assessing ionic species, allowing sub-minute variability to be resolved. Through the sampling of a diverse range of ambient aerosol, including biomass burning, fresh and aged urban pollution, the conductivity measurement offered additional useful information to untangle the composition of complex aerosol mixtures, specifically for assessing acidic aerosol conditions.

Ewan Colin Crosbie↗
NASA NTRSDate not supplied

Impact Ice Adhesion at NASA Glenn: Current Experimental Methods and Supporting Measurements

When examining the literature on the adhesion strength of impact ice, there have been a wide range of methodologies tried to measure the required stresses to induce interfacial delamination. Utilizing the Icing Research Tunnel at the NASA Glenn Research Center to generate the impact ice required for this work, several different mechanical tests have been and are being developed to determine the stresses along the interface between ice and coupon. This set of tests includes the technical mature modified lap joint test which has been used to conduct ice adhesion studies through a wide sweep of icing conditions. To conduct in situ ice adhesion measurements inside of the Icing Research Tunnel, several new experiments are currently being developed to make ice adhesion measurements during and immediately after ice accretion. In addition to these experimental methods, several supporting measurement techniques have been developed to allow for a better understanding on the influence of icing cloud conditions on the mechanical behavior of impact ice. Digital image correlation has been successfully implemented to augment the data generated by the modified lap joint test with full field surface displacement and strain measurements which allow for insight into the deformation processes present during a test. Both optical microscopy of impact ice samples along with ice replication techniques have been used to study the grain structure of the impact ice. This has led to a deeper understanding of the results from the modified lap joint method and how the structure of impact ice changes as it is accreted during an icing spray. The freezing process of impact ice generated by supercooled liquid water is not a volume conserving process, which leads to the presence of residual strains along the interface between ice and substrate. These strains have been observed using both a simplified flat geometry and a representative airfoil. The data gathered by these experimental adhesion methods and supporting measurements allows for a comprehensive understanding on the behavior of impact ice which will be critical to the development of future ice shedding models.

Ice Adhesion↗
NASA NTRSDate not supplied

Impact Ice Adhesion at NASA Glenn: Current Experimental Methods and Supporting Measurements

When examining the literature on the adhesion strength of impact ice, there have been a wide range of methodologies tried to measure the required stresses to induce interfacial delamination. Utilizing the Icing Research Tunnel at the NASA Glenn Research Center to generate the impact ice required for this work, several different mechanical tests have been and are being developed to determine the stresses along the interface between ice and coupon. This set of tests includes the technical mature modified lap joint test which has been used to conduct ice adhesion studies through a wide sweep of icing conditions. To conduct in situ ice adhesion measurements inside of the Icing Research Tunnel, several new experiments are currently being developed to make ice adhesion measurements during and immediately after ice accretion. In addition to these experimental methods, several supporting measurement techniques have been developed to allow for a better understanding on the influence of icing cloud conditions on the mechanical behavior of impact ice. Digital image correlation has been successfully implemented to augment the data generated by the modified lap joint test with full field surface displacement and strain measurements which allow for insight into the deformation processes present during a test. Both optical microscopy of impact ice samples along with ice replication techniques have been used to study the grain structure of the impact ice. This has led to a deeper understanding of the results from the modified lap join method and how the structure of impact ice changes as it is accreted during an icing spray. The freezing process of impact ice generated by supercooled liquid water is not a volume conserving process, which leads to the presence of residual strains along the interface between ice and substrate. These strains have been observed using both a simplified flat geometry and a representative airfoil. The data gathered by these experimental adhesion methods and supporting measurements allows for a compressive understanding on the behavior of impact ice which will be critical to the development of future ice shedding models.

Ice Adhesion↗
NASA NTRSDate not supplied

Laser Beam Welding for in-Space Joining Demonstrated Under Vacuum on the Ground and By Parabolic Flight Experiments

High energy density electron beam welding enabled the first and, to date, only American weld performed in space during the M551 experiment on Skylab in 1973. Though welding is critical to 90% of durable goods manufacturing in America, there is not yet regular and reliable application of welding processes to the In-space Servicing, Assembly, and Manufacturing (ISAM) sector. Fundamental studies are needed to develop basic capabilities and to enhance fundamental process knowledge, which will support follow-on efforts to mature in-space welding for use in commercial, defense, and other aerospace applications. The current work seeks to build on past flights and improve understanding and quantification of materials joining in space conditions using laser beam welding (LBW). LBW offers several advantages over the electron beam welding of Skylab, amongst others: reduced electromagnetic interference, less exposure of operators to ionizing radiation, and flexible delivery through optical fibers supporting unique workpieces and joints. Since there is no orbital laboratory to mature laser beam welding for space, the current effort addresses maturation of laser beam welding through parabolic flights augmented with data collection to enable numerical modeling efforts that capture the physical effects of the space environment. This team has retrofitted an LBW experimental apparatus that can simulate the vacuum and, during a parabolic flight, the reduced gravity & microgravity conditions of in-space welding. A team of Capstone students modified the setup, originally developed by NASA Langley Research Center for electron beam free-form fabrication, to replace its electron gun with a 1 kW, 1070 nm Yb fiber laser. The apparatus was further instrumented with temperature sensing and high-speed welding cameras to monitor and record changes in the thermal state of the workpiece, the melt pool, the laser penetration level, and the development and orientation of spatter & plumes. The system will operate autonomously, demonstrating its utility to uncrewed missions. During upcoming parabolic flight campaigns expected summer 2024, this LBW equipment will weld common aerospace alloys of aluminum, stainless steel, and titanium under conditions representative of the space environment. This data will guide future computational modeling efforts of laser welding in space and help to qualify in-space welding as a viable ISAM technique.

laser beam welding↗
NASA NTRSDate not supplied

From Simulation to Reality With Random Noise

The challenging environment of autonomous vehicle (AV) navigation necessitates certain functions be performed by deep neural networks. Optimizing these models involves collecting vast quantities of domain-specific training data and ensuring that the dataset is representative of expected conditions. High-fidelity simulation plays a vital role in making this process feasible, allowing a wide range of scenarios to be explored at low cost. However, learning from simulation introduces subtle biases into models, which can degrade real-world performance in unpredictable ways. This effect can be mitigated with learning schemes specialized to bridge distributional shifts (transfer learning). Given the complex nature of these methods, the underlying models, and their environments, meaningfully evaluating performance is notstraight forward. Many unrelated factors can effect an improvement in generalization accuracy, but a full ablation analysis is often difficult. To tease out signal from noise, it is necessary to understand how transfer learning performance is affected by noise itself. The goals of this paper are (i) to establish a domain randomization baseline for a simple classification transfer learning task and (ii) to validate the RRAV testbed as a platform for further research in sim-to-real learning. We generate imagery from a simulation of NASA Ames Research Center and train a small convolutional neural network (ConvNet) to classify position relative to a centerline. Further models are trained with different types of noise progressively added to the data. The models are deployed aboard the on-site test vehicle to test real-world performance. In our experiments, we find that such naive domain randomization raises sim-to-real accuracy from 64% to 79%, while training directly on real data yields an 89% accuracy ceiling. These results suggest that the isolated mechanism of domain randomization can significantly improve generalization.

simulation↗
NASA NTRS1971

Aerodynamic characteristics of a large scale model with a swept wing and augmented jet flap

Data of tests of a large-scale swept augmentor wing model in the 40- by 80-foot wind tunnel are presented. The data includes longitudinal characteristics with and without a horizontal tail as well as results of preliminary investigation of lateral-directional characteristics. The augmentor flap deflection was varied from 0 deg to 70.6 deg at isentropic jet thrust coefficients of 0 to 1.47. The tests were made at a Reynolds number from 2.43 to 4.1 times one million.

Falarski, M. D.↗
NASA NTRS1972

The induced magnetic field of the moon - Conductivity profiles and inferred temperature.

Electromagnetic induction in the moon driven by fluctuations of the interplanetary magnetic field is used to determine the lunar bulk electrical conductivity. The earlier data are now augmented by an order of magnitude. The present data clearly show the north-south and east-west transfer function difference as well as the high-frequency rollover suggested earlier. The difference is shown to be compatible over the midfrequency range (0.001 to 0.01 Hz) with a noise source associated with the compression of the local remanent field by solar wind dynamic pressure fluctuations. The rollover of the transfer functions is shown to result from higher order magnetic multipole radiation; electric multipoles appear supressed, although a vestigial TM interaction may still be present. Models for two, three, and four layer; current layer, double current layer, and core plus current layer moons are generated by inversion of the data, using a theory that incorporates higher-order multipoles.

Sonett, C. P.↗