Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “temporal learning”

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 937 records · Page 52

Assessment of the Advanced Very High Resolution Radiometer (AVHRR) for Snowfall Retrieval in High Latitudes Using CloudSat and Machine Learning

Precipitation retrieval is a challenging topic, especially in high latitudes (HL), and current precipitation products face ample challenges over these regions. This study investigates the potential of the Advanced Very High Resolution Radiometer (AVHRR) for snowfall retrieval in HL using CloudSat radar information and machine learning (ML). With all the known limitations, AVHRR observations should be considered for HL snowfall retrieval because 1) AVHRR data have been continuously collected for about four decades on multiple platforms with global coverage, and similar observations will likely continue in the future; 2) current passive microwave satellite precipitation products have several issues over snow and ice surfaces; and 3) good coincident observations between AVHRR and CloudSat are available for training ML algorithms. Using ML, snowfall rate was retrieved from AVHRR’s brightness temperature and cloud probability, as well as auxiliary information provided by numerical reanalysis. The results indicate that the ML-based retrieval algorithm is capable of detection and estimation of snowfall with comparable or better statistical scores than those obtained from the Atmospheric Infrared Sounder (AIRS) and two passive microwave sensors contributing to the Global Precipitation Measurement (GPM) mission constellation. The outcomes also suggest that AVHRR-based snowfall retrievals are spatially and temporally reasonable and can be considered as a quantitatively useful input to the merged precipitation products that require frequent sampling or long-term records.

Mohammad Reza↗

Two-Layer Dynamic Recycling Model (2L-DRM): Learning from Moisture Tracking Models of Different Complexity

Atmospheric moisture tracking models are used to identify and quantify sources and sinks of water in the atmospheric branch of the hydrologic cycle. These models are primarily used to investigate the origin of moisture resulting in precipitation for particular regions around the globe. Moisture tracking models vary widely in their level of complexity, depending on the number of physical processes represented. Complex models are comprehensive in their physical representation, but computationally much more expensive than simple models, which only focus on specific physical processes and use simplifying assumptions. We present the mathematical derivation of the new two-layer dynamical recycling model (2L-DRM), a simple analytical moisture tracking model that relaxes the vertically-integrated formulation of the original one-layer DRM (Dominguez et al., 2006). By comparing the simple DRM to a very complex moisture tracking model that uses water vapor tracers embedded within the Weather Research and Forecast regional climate model (WRF-WVT) for the North American Monsoon region, we pinpoint the absence of vertical wind shear as the main deficiency in the simple DRM. When comparing both simple models (DRM and 2L-DRM) to the WRF-WVT (which we treat as ‘truth’), the 2L-DRM better captures the spatial extent, the net amount and the temporal variability of precipitation that originates from oceanic and local terrestrial sources. The 2L-DRM is well suited to study the large-scale climatological sources of moisture, and for these applications, performs on-par with the much more complex and computationally demanding WRF-WVT model.

Dominguez, Francina↗

Accounting for Training Data Error in Machine Learning Applied to Earth Observations

Remote sensing, or Earth Observation (EO), is increasingly used to understand Earth system dynamics and create continuous and categorical maps of biophysical properties and land cover, especially based on recent advances in machine learning (ML). ML models typically require large, spatially explicit training datasets to make accurate predictions. Training data (TD) are typically generated by digitizing polygons on high spatial-resolution imagery, by collecting in situ data, or by using pre-existing datasets. TD are often assumed to accurately represent the truth, but in practice almost always have error, stemming from (1) sample design, and (2) sample collection errors. The latter is particularly relevant for image-interpreted TD, an increasingly commonly used method due to its practicality and the increasing training sample size requirements of modern ML algorithms. TD errors can cause substantial errors in the maps created using ML algorithms, which may impact map use and interpretation. Despite these potential errors and their real-world consequences for map-based decisions, TD error is often not accounted for or reported in EO research. Here we review the current practices for collecting and handling TD. We identify the sources of TD error, and illustrate their impacts using several case studies representing different EO applications (infrastructure mapping, global surface flux estimates, and agricultural monitoring), and provide guidelines for minimizing and accounting for TD errors. To harmonize terminology, we distinguish TD from three other classes of data that should be used to create and assess ML models: training reference data, used to assess the quality of TD during data generation; validation data, used to iteratively improve models; and map reference data, used only for final accuracy assessment. We focus primarily on TD, but our advice is generally applicable to all four classes, and we ground our review in established best practices for map accuracy assessment literature. EO researchers should start by determining the tolerable levels of map error and appropriate error metrics. Next, TD error should be minimized during sample design by choosing a representative spatio-temporal collection strategy, by using spatially and temporally relevant imagery and ancillary data sources during TD creation, and by selecting a set of legend definitions supported by the data. Furthermore, TD error can be minimized during the collection of individual samples by using consensus-based collection strategies, by directly comparing interpreted training observations against expert-generated training reference data to derive TD error metrics, and by providing image interpreters with thorough application-specific training. We strongly advise that TD error is incorporated in model outputs, either directly in bias and variance estimates or, at a minimum, by documenting the sources and implications of error. TD should be fully documented and made available via an open TD repository, allowing others to replicate and assess its use. To guide researchers in this process, we propose three tiers of TD error accounting standards. Finally, we advise researchers to clearly communicate the magnitude and impacts of TD error on map outputs, with specific consideration given to the likely map audience.

58 GEOSCIENCES↗

Exploring the Neurocircuitry Underpinning Predictability of Threat in Soldiers with PTSD Compared to Deployment Exposed Controls

Background: Prior work examining emotional dysregulation observed in posttraumatic stress disorder (PTSD) has primarily been limited to fear-learning processes specific to anticipation, habituation, and extinction of threat. In contrast, the response to threat itself has not been systematically evaluated. Objective: To explore potential disruption in fear conditioning neurocircuitry in service members with PTSD, specifically in response to predictable versus unpredictable threats. Method: In the current study, active-duty U.S. Army soldiers with (PTSD group; n = 38) and without PTSD (deployment-exposed controls; DEC; n = 40), participated in a fear-conditioning study in which threat predictability was manipulated by presenting an aversive unconditioned stimulus (UCS) that was either preceded by a conditioned stimulus (i.e., predictable) or UCS alone (i.e., unpredictable). Threat expectation, skin conductance response (SCR), and functional magnetic resonance imaging (fMRI) signal to predictable and unpredictable threats (i.e., UCS) were assessed. Results: Both groups showed greater threat expectancy and diminished threat-elicited SCRs to predictable compared to unpredictable threat. Significant group differences were observed within the amygdala, hippocampus, insula, and superior and middle temporal gyri. Contrary to our predictions, the PTSD group showed a diminished threat-related response within each of these brain regions during predictable compared to unpredictable threat, whereas the DEC group showed increased activation. Conclusion: Although, the PTSD group showed greater threat-related diminution, hypersensitivity to unpredictable threat cannot be ruled out. Furthermore, pre-trauma, trait-like factors may have contributed to group differences in activation of the neurocircuitry underpinning fear conditioning.

60 APPLIED LIFE SCIENCES↗

Short‐Term Hourly Weather Forecasting Using PredRNN With Image Preprocessing

Global weather forecast models are vital tools with numerous applications, including public safety, agriculture, and transportation. Recent advancements in artificial intelligence (AI) and deep learning (DL) have shown the potential to enhance weather forecasting accuracy and speed. In this study, we developed a short-term hourly weather forecast framework with a wavelet transform function for data preprocessing and a spatiotemporal DL model, PredRNN, for predicting five surface atmospheric variables, including wind speed and direction, mean sea level pressure (MSLP), temperature, and precipitation. The framework demonstrated promising results. It produces global forecasts at 0.25° (∼25 km) with a 1-day lead time RMSE of 1.8 m/s for wind components, 180 Pa for MSLP, and 1.8 K for temperature. Although our model does not surpass state-of-the-art AI weather forecast models across all metrics, it outperforms these models in precipitation forecasting and wind prediction at short lead times and achieves comparable accuracy for MSLP. Its native hourly forecasting capability, together with training on widely accessible GPU hardware, contributes meaningfully to the advancement of accessible DL weather forecasting methods. Our work highlights the importance of integrating temporal components and data transformation techniques to improve the predictability and accuracy of weather forecasts.

Tran, Hoang [Pacific Northwest National Laboratory↗

The U.S. Forest Service's analysis of cumulative effects to wildlife: A study of legal standards, current practice, and ongoing challenges on a National Forest

Cumulative effects analysis (CEA) allows natural resource managers to understand the status of resources in historical context, learn from past management actions, and adapt future activities accordingly. U.S. federal agencies are required to complete CEA as part of environmental impact assessment under the National Environmental Policy Act (NEPA). Past research on CEA as part of NEPA has identified significant deficiencies in CEA practice, suggested methodologies for handling difficult aspects of CEA, and analyzed the rise in litigation over CEA in U.S. courts. This article provides a review of the literature and legal standards related to CEA as it is done under NEPA and then examines current practice on a U.S. National Forest, utilizing qualitative methods in order to provide a detailed understanding of current approaches to CEA. Research objectives were to understand current practice, investigate ongoing challenges, and identify impediments to improvement. Methods included a systematic review of a set of NEPA documents and semi-structured interviews with practitioners, scientists, and members of the public. Findings indicate that the primary challenges associated with CEA include: issues of both geographic and temporal scale of analysis, confusion over the purpose of the requirement, the lack of monitoring data, and problems coordinating and disseminating data. Improved monitoring strategies and programmatic analyses could support improved CEA practice.

54 ENVIRONMENTAL SCIENCES↗

A Multi-Satellite Assimilation and Modeling Platform to Construct a Global and Complete view of the Hydrologic Cycle

What captivates me about the research field of satellite remote sensing of the earth system science is the opportunity to help human beings by addressing challenging questions related to the current and future availability of natural resources for domestic, agricultural, and industrial needs. For example, a continuously varying climate poses a threat to the water resource, not only in developing countries, but also in economically well-developed regions, creating the urgent need to characterize and predict its spatial and temporal availability. Specifically, my curiosity is driven by the challenge of merging cutting-edge space technology and earth observations (i.e., remote sensing) with state-of-the-art models for the purpose of improving our scientific knowledge about the variability and the change of the hydrologic cycle.The only practical way to observe the land surface processes (e.g., hydrologic cycle) on continental to global scales is via satellite remote sensing. Though remote sensing can make spatially comprehensive measurements of various components of the land surface system, it cannot provide direct information on the entire system, and the measurements represent only a snapshot in time. Physical based models may be used to continuously predict the temporal and spatial earth processes, but these predictions are often poor, due to model initialization, parameter and meteorological forcing errors, and inadequate model physics and/or resolution. Thus, an attractive prospect is to combine the strengths of land surface models and observations (and minimize the weaknesses of both) to provide a superior land surface state estimate. This is the goal of land surface data assimilation. Data assimilation provides a better estimate of the environmental states than either models or observations could individually do. The broad concept of data assimilation can be applied to various disciplines such as hydrology, ecology, environmental hazards, agriculture and economy.My research is targeted to integrating multiple satellite information having multi-sensor and multi-resolution assimilation within today's state-of-the-art hydrologic models. Multi-sensor and multi-resolution assimilation techniques represent a necessary milestone in future earth science applications because they offer the capability of comprehensively integrating disparate types of earth observations via deep-learning techniques. A multi-satellite assimilation and modeling platform will ultimately provide a robust and complete dataset to more fully understand earth's dynamics. In fact, the multi-satellite assimilation platforms build a comprehensive description of the earth processes that is geared toward accurately representing the complexity of natural and anthropogenic interactions in land surface processes. Most importantly, these platforms are suitable as decision support tools for applications across different aspects of earth system science.My recent work has focused on recent multi-sensor data assimilation techniques targeted at improving snow, soil moisture, groundwater, and terrestrial water storage hydrological states.

Girotto, Manuela↗

Improved GEO Derived LW Broadband Fluxes for the Edition5 CERES SYN1Deg Data Product

The NASA Clouds and the Earth's Radiant Energy System (CERES) project provides over 24 years of accurately observed top-of-the-atmosphere (TOA) and surface flux data record for climate monitoring and diagnostic studies. The major level-3 data products include SSF1deg, SYN1deg, FluxByCldTyp, and EBAF data. The CERES SYN1deg product provides 1° gridded observed TOA fluxes and computed surface fluxes based on hourly GEO clouds and derived broadband TOA fluxes. The hourly GEO derived fluxes fill in the temporal gaps between the Terra and Aqua CERES observations. In order to maintain a climate quality dataset the GEO cloud and flux retrievals are consistent over the record. The CERES project is preparing for Edition 5 product providing an opportunity to improve the GEO flux retrieval algorithms. To achieve this, the GEO satellite channel radiances are calibrated against the Aqua-MODIS calibration reference. For SW the GEO channel radiances are then converted to broadband radiances using empirical and theoretical models. The SW broadband radiances are then converted to fluxes using the same CERES angular directional models. For LW the GEO channel radiances are directly converted to LW fluxes utilizing empirical models. The resulting GEO derived broadband fluxes are then normalized to the CERES fluxes regionally by regressing the instantaneous coincident flux pairs. The normalized GEO fluxes are then tied to the CERES instrument calibration, which is very stable over time. This study will attempt to improve the GEO derived LW fluxes based on machine learning technique for the future Edition5 SYN1deg data product.

Moguo Sun↗

Simulating water dynamics related to pedogenesis across space and time: Implications for four-dimensional digital soil mapping

Digital soil mapping (DSM) relies on machine-learning and geostatistics to represent soil property observations across space. DSM techniques are powerful but often empirical, being limited to the quality and density of point samples. Water dynamics are closely related to soil variability, and the physics that govern water movement are well known. Hydrological properties can hence be simulated by physical models through space and time, unveiling key characteristics about soils. We propose the use of hydrologic models to map soils across the surface (2D), depth (1D), and time (1D)–which provides a 4D approach to digital soil mapping (4DSM). The Distributed Hydrology Soil Vegetation Model (DHSVM) was applied to a watershed currently under pasture. Moisture sensors and wells were installed at different depths in the watershed on summit, sideslope and toeslope positions to validate the model. DHSVM simulations of soil moisture distribution and depth to saturation were performed during the hydrological year (October 2008-September 2009). Clusters of similar pixels based on soil moisture values were determined using Dynamic Time Warping (DTW) to align temporal data and K-means. Clustering was performed both seasonally and for the entire year. Temporal patterns simulated by DHSVM matched measurements given by moisture sensors and wells. Seasonal clusters differed from the annual cluster. Distinct clusters were observed for each season and with depth, showing that spatiotemporal soil variability is lost when statically assessing soils. Spatiotemporal clusters corroborated field observations of fragipan occurrence not explicitly spatially mapped by Soil Survey Geographic Database (SSURGO). If a connection can be made between water and soils, static and dynamic soil variability can be predicted using physically based hydrologic models. Hydrologic models can benefit soil mapping by enabling reliable 4D simulation of water dynamics, which are fundamental to soil variability and soil classification and directly relate to biological, physical and chemical soil processes not captured by typical soil sampling protocols.

54 ENVIRONMENTAL SCIENCES↗

What can we learn from N 2 O isotope data? – Analytics, processes and modelling

The isotopic composition of nitrous oxide (N 2 O) provides useful information for evaluating N 2 O sources and budgets. Due to the co–occurrence of multiple N 2 O transformation pathways, it is, however, challenging to use isotopic information to quantify the contribution of distinct processes across variable spatiotemporal scales. Here, we present an overview of recent progress in N 2 O isotopic studies and provide suggestions for future research, mainly focusing on: analytical techniques; production and consumption processes; and interpretation and modelling approaches. Comparing isotope–ratio mass spectrometry (IRMS) with laser absorption spectroscopy (LAS), we conclude that IRMS is a precise technique for laboratory analysis of N 2 O isotopes, while LAS is more suitable for in situ/inline studies and offers advantages for site–specific analyses. When reviewing the link between the N 2 O isotopic composition and underlying mechanisms/processes, we find that, at the molecular scale, the specific enzymes and mechanisms involved determine isotopic fractionation effects. In contrast, at plot–to–global scales, mixing of N 2 O derived from different processes and their isotopic variability must be considered. We also find that dual isotope plots are effective for semi–quantitative attribution of co–occurring N 2 O production and reduction processes. More recently, process–based N 2 O isotopic models have been developed for natural abundance and 15 N–tracing studies, and have been shown to be effective, particularly for data with adequate temporal resolution. Here, despite the significant progress made over the last decade, there is still great need and potential for future work, including development of analytical techniques, reference materials and inter–laboratory comparisons, further exploration of N 2 O formation and destruction mechanisms, more observations across scales, and design and validation of interpretation and modelling approaches. Synthesizing all these efforts, we are confident that the N 2 O isotope community will continue to advance our understanding of N 2 O transformation processes in all spheres of the Earth, and in turn to gain improved constraints on regional and global budgets.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

An Overview of MODIS Calibration and Characterization

The MODerate Resolution Imaging Spectroradiometer (MODIS) is one of the key instruments for the NASA s Earth Observing System (EOS). It is currently operating on both EOS Terra and Aqua satellites. The MODIS is a major advance over its heritage sensors in terms of its spectral, spatial, and temporal resolutions with frequent global observations and a broad range of science applications. There are 20 reflective solar bands (RSB) with center wavelengths ranging from 0.41 to 2.l microns and 16 thermal emissive bands (TEB) from 3.7 to 14.4 microns. The absolute radiometric accuracy requirements (lsigma) at the typical spectral radiance levels are plus or minus 2% for the RSB reflectance factors and plus or minus 5% for the RSB radiance products. With few exceptions, the TEB requirements are plus or minus 1%. To verify that the instruments met their specified design requirements both Terra and Aqua MODIS underwent extensive pre-launch calibration and characterization at various levels, including system-level thermal vacuum testing. On-orbit calibration and characterization are performed by the on-board calibrators: a solar diffuser (SD) and a solar diffuser stability monitor (SDSM), a V-groove flat panel blackbody (BB), and a spectro-radiometric calibration assembly (SRCA). In this paper, we present an overview of MODIS calibration and characterization activities, methodologies, and lessons learned from pre-launch testing and on-orbit operations. Key issues to be discussed include our on-orbit efforts of monitoring detectors noise characterization, tracking solar diffuser and optics degradation, and updating sensor s response versus scan-angle. The MODIS experience has provided invaluable lessons that are being used in designing and testing the Visible Infrared Imaging Radiometer Suite (VIIRS), a direct follow-on to the MODIS that will be flown on the National Polar-Orbit Operational Environmental Satellite System (NPOESS) missions.

Xiong, X.↗

Astronaut Photography of the Earth: A Long-Term Dataset for Earth Systems Research, Applications, and Education

The NASA Earth observations dataset obtained by humans in orbit using handheld film and digital cameras is freely accessible to the global community through the online searchable database at https://eol.jsc.nasa.gov, and offers a useful compliment to traditional ground-commanded sensor data. The dataset includes imagery from the NASA Mercury (1961) through present-day International Space Station (ISS) programs, and currently totals over 2.6 million individual frames. Geographic coverage of the dataset includes land and oceans areas between approximately 52 degrees North and South latitudes, but is spatially and temporally discontinuous. The photographic dataset includes some significant impediments for immediate research, applied, and educational use: commercial RGB films and camera systems with overlapping bandpasses; use of different focal length lenses, unconstrained look angles, and variable spacecraft altitudes; and no native geolocation information. Such factors led to this dataset being underutilized by the community but recent advances in automated and semi-automated image geolocation, image feature classification, and web-based services are adding new value to the astronaut-acquired imagery. A coupled ground software and on-orbit hardware system for the ISS is in development for planned deployment in mid-2017; this system will capture camera pose information for each astronaut photograph to allow automated, full georegistration of the data. The ground system component of the system is currently in use to fully georeference imagery collected in response to International Disaster Charter activations, and the auto-registration procedures are being applied to the extensive historical database of imagery to add value for research and educational purposes. In parallel, machine learning techniques are being applied to automate feature identification and classification throughout the dataset, in order to build descriptive metadata that will improve search capabilities. It is expected that these value additions will increase interest and use of the dataset by the global community.

Stefanov, William L.↗

Towards adaptive, streaming analysis of x-ray tomography data

Temporal and spatial resolution of chemical imaging methodologies such as x-ray tomography are rapidly increasing, leading to more complex experimental procedures and fast growing data volumes. Automated analysis pipelines and big data analytics are becoming essential to effectively evaluate the results of such experiments. Offering those data techniques in an adaptive, streaming environment can further substantially improve the scientific discovery process, by enabling experimental control and steering based on the evaluation of emerging phenomena as they are observed by the experiment. Pacific Northwest National Laboratory (PNNL)’ Chemical Imaging Initiative (CII - http://imaging.pnnl.gov/ ) has worked since 2011 towards developing a framework that allows users to rapidly compose and customize high throughput experimental analysis pipelines for multiple instrument types. The framework, named ‘Rapid Experimental Analysis’ (REXAN) Framework [1], is based on the idea of reusable component libraries and utilizes the PNNL developed collaborative data management and analysis environment ‘Velo’, to provide a user friendly analysis and data management environment for experimental facilities. This article will, discuss the capabilities established for X-Ray tomography, discuss lessons learned, and provide an overview of our more recent work in the Analysis in Motion Initiative (AIM - http://aim.pnnl.gov/ ) at PNNL to provide REXAN capabilities in a streaming environment.

analysis framework↗

Time-warping invariant quantum recurrent neural networks via quantum-classical adaptive gating

Adaptive gating plays a key role in temporal data processing via classical recurrent neural networks (RNNs), as it facilitates retention of past information necessary to predict the future, providing a mechanism that preserves invariance to time warping transformations. This paper builds on quantum RNNs (QRNNs), a dynamic model with quantum memory, to introduce a novel class of temporal data processing quantum models that preserve invariance to time-warping transformations of the (classical) input-output sequences. The model, referred to as time warping-invariant QRNN (TWI-QRNN), augments a QRNN with a quantum–classical adaptive gating mechanism that chooses whether to apply a parameterized unitary transformation at each time step as a function of the past samples of the input sequence via a classical recurrent model. The TWI-QRNN model class is derived from first principles, and its capacity to successfully implement time-warping transformations is experimentally demonstrated on examples with classical or quantum dynamics.

97 MATHEMATICS AND COMPUTING↗

Apex Reference Manual 3.0 Beta

Apex is a toolkit for constructing software that behaves intelligently and responsively in demanding task environments. Reflecting its origin at NASA where Apex continues to be developed, current applications include: a) Providing autonomous mission management and tactical control capabilities for unmanned aerial vehicles including an autonomous surveillance helicopter and a simulation prototype of an unmanned fixed-wing aircraft to be used for wildfire mapping; b) Simulating human air traffic controllers, pilots and astronauts to help predict how people might respond to changes in equipment or procedures; and c) Predicting the precise duration and sequence of routine human behaviors based on a human-computer interaction engineering technique called CPM-GOMS. Among Apex s components are a set of implemented reasoning services, such as those for reactive planning and temporal pattern recognition; a software architecture that embeds and integrates these services and allows additional reasoning elements to be added as extensions; a formal language for specifying agent knowledge; a simulation environment to facilitate prototyping and analysis; and Sherpa, a set of tools for visualizing autonomy logic and runtime behavior. In combination, these are meant to provide a flexible and usable framework for creating, testing, and deploying intelligent agent software. Overall, our goal in developing Apex is to lower economic barriers to developing intelligent software agents. New ideas about how to extend or modify the system are evaluated in terms of their impact in reducing the time, expertise, and inventiveness required to build and maintain applications. For example, potential enhancements to the AI reasoning capabilities in the system are reviewed not only for usefulness and distinctiveness, but also for their impact on the readability and general usability of Apex s behavior representation language (PDL) and on the transparency of resulting behavior. A second central part of our approach is to iteratively refine Apex based on lessons learned from as diverse a set of applications as possible. Many applications have been developed by users outside the core development team including engineers, researchers, and students. Usability is thus a central concern for every aspect of Apex visible to a user, including PDL, Sherpa, the Apex installation process, APIs, and user documentation. Apex users vary in their areas of expertise and in their familiarity with autonomy technology. Focusing on usability, a development philosophy summarized by the project motto "Usable Autonomy," has been important part of enabling diverse users to employ Apex successfully and to provide feedback needed to guide iterative, user-centered refinement.

Freed, Michael A.↗

Using a Support Vector Machine and a Land Surface Model to Estimate Large-Scale Passive Microwave Temperatures over Snow-Covered Land in North America

A support vector machine (SVM), a machine learning technique developed from statistical learning theory, is employed for the purpose of estimating passive microwave (PMW) brightness temperatures over snow-covered land in North America as observed by the Advanced Microwave Scanning Radiometer (AMSR-E) satellite sensor. The capability of the trained SVM is compared relative to the artificial neural network (ANN) estimates originally presented in [14]. The results suggest the SVM outperforms the ANN at 10.65 GHz, 18.7 GHz, and 36.5 GHz for both vertically and horizontally-polarized PMW radiation. When compared against daily AMSR-E measurements not used during the training procedure and subsequently averaged across the North American domain over the 9-year study period, the root mean squared error in the SVM output is 8 K or less while the anomaly correlation coefficient is 0.7 or greater. When compared relative to the results from the ANN at any of the six frequency and polarization combinations tested, the root mean squared error was reduced by more than 18 percent while the anomaly correlation coefficient was increased by more than 52 percent. Further, the temporal and spatial variability in the modeled brightness temperatures via the SVM more closely agrees with that found in the original AMSR-E measurements. These findings suggest the SVM is a superior alternative to the ANN for eventual use as a measurement operator within a data assimilation framework.

Modeling↗

A framework for activity detection in wide-area motion imagery

Wide-area persistent imaging systems are becoming increasingly cost effective and now large areas of the earth can be imaged at relatively high frame rates (1-2 fps). The efficient exploitation of the large geo-spatial-temporal datasets produced by these systems poses significant technical challenges for image and video analysis and data mining. In recent years there has been significant progress made on stabilization, moving object detection and tracking and automated systems now generate hundreds to thousands of vehicle tracks from raw data, with little human intervention. However, the tracking performance at this scale, is unreliable and average track length is much smaller than the average vehicle route. This is a limiting factor for applications which depend heavily on track identity, i.e. tracking vehicles from their points of origin to their final destination. In this paper we propose and investigate a framework for wide-area motion imagery (W AMI) exploitation that minimizes the dependence on track identity. In its current form this framework takes noisy, incomplete moving object detection tracks as input, and produces a small set of activities (e.g. multi-vehicle meetings) as output. The framework can be used to focus and direct human users and additional computation, and suggests a path towards high-level content extraction by learning from the human-in-the-loop.

99 GENERAL AND MISCELLANEOUS↗

Predicting Unreinforced Fabric Mechanical Behavior with Recurrent Neural Networks

Unreinforced woven fabrics are widely employed in various high-performance applications, including parachute deployment systems, airbags, and ballistic armor. The analysis of such materials is inherently complex due to the multiscale structure of these materials, and the dependence of macroscale behavior on changes that occur at lower scales. Previously, NASA’s Multiscale Analysis Tool (NASMAT) showed its capability in predicting unreinforced fabric behavior at the macroscale by capturing finite rotations that occur at the mesoscale. Though effective, the tool can face high computational cost for large, complex problems, motivating the need for the development of a surrogate model that can capture the same behavior. A recurrent neural network (RNN) was developed and trained on virtual NASMAT data to mimic the physics-based solutions while improving the computational runtime. The architecture of the RNN to best simulate the fabric behavior was carefully crafted based on heuristic knowledge of predicting physics-based temporal data, manual hyperparameter case studies, and Hyperband optimization.. The resultant model was able to predict a variety of stress-strain curves for fabrics with different mesoscale geometries, and was further validated by comparing to experimental data for the K706 style Kevlar plain-weave fabric, demonstrating the ability of the model to effectively capture the geometric changes in the fabric without explicitly calculating them, as is done in NASMAT. Furthermore, the tool showed its ability to improve on the runtime by a factor of 10 for fabric solutions compared to the multiscale tool, which would further enable the simulation of complex loading scenarios on unreinforced fabrics.

Fabric↗