Search NASA⌕ Search

SEARCH · Search NASA

Results for “Distributed training”

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 451 records · Page 25
All sourcesNASA NTRSDOE OSTI
NASA NTRS2006

Probabilistic Classification Using Elemental Abundance Distributions and Lossless Image Compression in Apollo 17 Lunar Dust Samples from Mare Serenitatis

We have previously outlined a strategy for the detection of fossils [Storrie-Lombardi and Hoover, 2004] and extant microbial life [Storrie-Lombaudi and Hoover, 20051 during robotic missions to Mars using co-registered structural and chemical signatures. Data inputs included image lossless compression indices to estimate relative textural complexity and elemental abundance distributions. Two exploratory classification algorithms (principal component analysis and hierarchical cluster analysis) provide an initial tentative classification of all targets. Nonlinear stochastic neural networks are then trained to produce a Bayesian estimate of algorithm classification accuracy. The strategy previously has been successful in distinguishing regions of biotic and abiotic alteration of basalt glass from unaltered samples. [Storrie-Lombardi and Fisk, 2004; Storrie-Lombardi and Fisk, 2004] Such investigations of abiotic versus biotic alteration of terrestrial mineralogy on Earth are compromised by .the difficulty finding mineralogy completely unaffected by the ubiquitous presence of microbial life on the planet. The renewed interest in lunar exploration offers an opportunity to investigate geological materials that may exhibit signs of aqueous alteration, but are highly unlikely to contain contaminating biological weathering signatures. We here present an extension of our earlier data set to include lunar dust samples obtained during the Apollo 17 mission. Apollo 17 landed in the Taurus-Littrow Valley in Mare Serenitatis. Most of the rock samples from this region of the lunar highlands are basalts comprised primarily of plagioclase and pyroxene and selected examples of orange and black volcanic glass. SEM images and elemental abundances (C6, N7, O8, Na11, Mg12, Al13, Si14, P15, S16, Cll7, K19, Ca20, Fe26) for a series of targets in the lunar dust samples are compared to the extant cyanobacteria, fossil trilobites, Orgueil meteorite, and terrestrial basalt targets previously discussed. The data set provides a first step in producing a quantitative probabilistic methodology for geobiological analysis of returned lunar samples or in situ exploration.

Storrie-Lombardi, Michael C.↗
NASA NTRS2011

Markov Chain Monte Carlo Bayesian Learning for Neural Networks

Conventional training methods for neural networks involve starting al a random location in the solution space of the network weights, navigating an error hyper surface to reach a minimum, and sometime stochastic based techniques (e.g., genetic algorithms) to avoid entrapment in a local minimum. It is further typically necessary to preprocess the data (e.g., normalization) to keep the training algorithm on course. Conversely, Bayesian based learning is an epistemological approach concerned with formally updating the plausibility of competing candidate hypotheses thereby obtaining a posterior distribution for the network weights conditioned on the available data and a prior distribution. In this paper, we developed a powerful methodology for estimating the full residual uncertainty in network weights and therefore network predictions by using a modified Jeffery's prior combined with a Metropolis Markov Chain Monte Carlo method.

Goodrich, Michael S.↗
NASA NTRS2016

An A-Train Climatology of Extratropical Cyclone Clouds

Extratropical cyclones (ETCs) are the main purveyors of precipitation in the mid-latitudes, especially in winter, and have a significant radiative impact through the clouds they generate. However, general circulation models (GCMs) have trouble representing precipitation and clouds in ETCs, and this might partly explain why current GCMs disagree on to the evolution of these systems in a warming climate. Collectively, the A-train observations of MODIS, CloudSat, CALIPSO, AIRS and AMSR-E have given us a unique perspective on ETCs: over the past 10 years these observations have allowed us to construct a climatology of clouds and precipitation associated with these storms. This has proved very useful for model evaluation as well in studies aimed at improving understanding of moist processes in these dynamically active conditions. Using the A-train observational suite and an objective cyclone and front identification algorithm we have constructed cyclone centric datasets that consist of an observation-based characterization of clouds and precipitation in ETCs and their sensitivity to large scale environments. In this presentation, we will summarize the advances in our knowledge of the climatological properties of cloud and precipitation in ETCs acquired with this unique dataset. In particular, we will present what we have learned about southern ocean ETCs, for which the A-train observations have filled a gap in this data sparse region. In addition, CloudSat and CALIPSO have for the first time provided information on the vertical distribution of clouds in ETCs and across warm and cold fronts. We will also discuss how these observations have helped identify key areas for improvement in moist processes in recent GCMs. Recently, we have begun to explore the interaction between aerosol and cloud cover in ETCs using MODIS, CloudSat and CALIPSO. We will show how aerosols are climatologically distributed within northern hemisphere ETCs, and how this relates to cloud cover.

clouds↗
NASA NTRS2021

Machine Learning Based Path Planning for Improved Rover Navigation

Enhanced AutoNav (ENav), the baseline surface navigation software for NASA’s Perseverance rover, sorts a list of candidate paths for the rover to traverse, then uses the Approximate Clearance Evaluation (ACE) algorithm to evaluate whether the most highly ranked paths are safe. ACE is crucial for maintaining the safety of the rover, but is computationally expensive. If the most promising candidates in the list of paths are all found to be infeasible, ENav must continue to search the list and run time-consuming ACE evaluations until a feasible path is found. In this paper, we present two heuristics that, given a terrain heightmap around the rover, produce cost estimates that more effectively rank the candidate paths before ACE evaluation. The first heuristic uses Sobel operators and convolution to incorporate the cost of traversing high-gradient terrain. The second heuristic uses a machine learning (ML) model to predict areas that will be deemed untraversable by ACE. We used physics simulations to collect training data for the ML model and to run Monte Carlo trials to quantify navigation performance across a variety of terrains with various slopes and rock distributions. Compared to ENav's baseline performance, integrating the heuristics can lead to a significant reduction in ACE evaluations and average computation time per planning cycle, increase path efficiency, and maintain or improve the rate of successful traverses. This strategy of targeting specific bottlenecks with ML while maintaining the original ACE safety checks provides an example of how ML can be infused into planetary science missions and other safety-critical software.

Yue, Yisong↗
NASA NTRSDate not supplied

Developing and Testing a Physics Guided Machine Learning NeuralNetwork to Predict Tonal Noise Emitted by a Propeller

Artificial neural networks offer a highly nonlinear and adaptive model for predicting complex interactions between input-output parameters. However, these networks require large datasets which often exceed practical considerations in modeling experimental results. To alleviate the dataset size requirement, a method known as physics guided machine learning has been applied to construct several neural networks for predicting propeller tonal noise in the time domain over a broad range of flight conditions. Three space-filling designs, namely, Latin-Hypercube, Sphere-Packing, and Grid-Space, were used to distribute points throughout the input parameter space encompassing nondimensional flight conditions and observer geometry. Each neural network’s performance was validated by conditions outside of the training set and compared to the Propeller Analysis System tool from the NASA Aircraft Noise Prediction Program. Compared to the Grid-Space input design, the Latin-Hypercube and the Sphere-Packing designs provided a better representation of the domain for training. Regarding the network archetype, a fully connected perceptron was found to outperform the partially connected perceptron in their ability to predict tonal noise for small datasets. The black-box nature of these neural networks was also explored to understand how the networks constructed the waveform and understand why some network designs produce better models.

Propeller noise↗
DOE OSTI2024

Digital Twin + AI: Control Room of the Future

A digital twin enhances power grid control room operations by providing real-time monitoring, predictive insights, simulation capabilities, remote control, training opportunities, data integration, and decision support. This technology empowers control room operators to effectively manage the grid, optimize performance, and ensure reliable and efficient energy distribution.

control room of the future↗
DOE OSTI2026

Generalized parton distributions from symbolic regression

AI/ML informed Symbolic Regression is the next stage of scientific modeling. We utilize a highly customizable symbolic regression package “PySR” to model the x and t dependence of the flavor isovector combination Hu-d(x, t, ) at = 0. These PySR models were trained on GPD results provided by both Lattice QCD and phenomenological sources GGL, GK, and VGG. We demonstrate, for the first time, the consistency and systematic convergence of Symbolic Regression by quantifying the disparate models through their Taylor expansion coefficients. In addition to PySR penalizing models with higher complexity and mean-squared error, we implement schemes that test specific physics hypotheses, including force-factorized x and t dependence and Regge behavior in PySR GPDs. We show that PySR can identify factorizing GPD sources based on their response to the Force-Factorized model. Here, knowing the precise behavior of the GPDs, and their uncertainties in a wide range in x and t, crucially impacts our ability to concretely and quantitatively predict hadronic spatial distributions and their derived quantities.

Dotson, Andrew [New Mexico State University, Las C↗
NASA NTRSDate not supplied

Predicting the Seawater Chemistry of an Ocean World Using Machine Learning on Isotopic Measurements of Volatile CO2

Introduction: Given the long time intervals required for data transmission to and from ocean worlds targets, low bandwidth for data transmission, time required for data processing and analysis, and potentially extreme radiation environments (e.g., Europa), it is clear that ocean worlds missions will need more autonomous flight instruments and software in order to achieve established science goals. Protracted time intervals for data analysis (e.g., Europa Lander) strongly motivates the development of rapid, consistent and streamlined methods for interpreting data from flight mass spectrometers to e.g., determine how mass spectra from a plume or surface liquid/ice relates to the surface/subsurface. Since mass spectrometry also has the potential to correctly identify biosignatures[1], it is imperative that such methods for interpreting data are consistent and accurate. We used 848 isotope ratio mass spectra from laboratory analyses of CO2 that interacted with ocean worlds-relevant seawaters as a ‘training’ dataset for ‘unsupervised’ machine learning. In unsupervised learning, characteristics of the data are not labeled or linked, and any similarities found only result from the neural network. CO2 isotopologues analyzed for this dataset mimic the remote measurements of CO2 by a flight mass spectrometer, and are detailed in Theiling [2]. From this dataset, we used measured features of the spectra, such as retention time, intensity, and (isotopologue) mass ratios as inputs for our autoencoder neural network. Our neural network was trained to find similarities in these and other spectral features for seawaters of a particular composition and amount of initial CO2. Successful training then created an output of these similarities for various seawaters, which included MgSO4, Na2SO4, NaCl, MgCl2, KCl, and NaHCO3, and combinations of these salts. We then applied dimensionality reduction techniques such as Principal Component Analysis (PCA), T-Distributed Stochastic Neighbor Embedding (TSNE), and Uniform Manifold Approximation and Projection (UMAP) to demonstrate latent data features as a two-dimensional projection in a unitless, high-dimensional space. In this projection, a data point represents the combined effect of spectral features such as intensity, retention time, and isotope ratio. Our initial UMAP demonstrates data clustering (organization of the data by the neural network) based on the amount of CO2 that had initially interacted with each seawater. Further training using more ‘supervised’ learning techniques demonstrate strong clustering of preliminary data based on initial CO2 concentration, seawater chemical composition, and ionic strength (salinity). Our preliminary work therefore suggests that machine learning has the potential to identify compositional variants of an ocean world seawater based on mass spectra from volatile CO2 measurements. Acknowledgments: This work was funded through a Strategic Task Group at NASA Goddard Space Flight Center. The training dataset was collected through funding from the Oklahoma Space Grant Consortium. References: [1] Pappalardo, R. et al. (2013) Astrobiology, 13, 740–773. [2] Theiling (2020) Icarus, 114216.

Europa↗
NASA NTRS2001

Possible Evidence for an Event Horizon in Cyg XR-1

The X-ray emitting component in the Cyg XR-1/HDE226868 system is a leading candidate for identification as a stellar-mass sized black hole. The positive identification of a black hole as predicted by general relativity requires the detection of an event horizon surrounding the point singularity. One signature of such an event horizon would be the existence of dying pulse trains emitted by material spiraling into the event horizon from the last stable orbit around the black hole. We observed the Cyg XR-1 system at three different epochs in a 1400 - 3000 A bandpass with 0.1 ms time resolution using the Hubble Space Telescope's High Speed Photometer. Repeated excursions of the detected flux by more than three standard deviations above the mean are present in the UV flux with FWHM 1 - 10 ms. If any of these excursions are pulses of radiation produced in the system (and not just stochastic variability associated with the Poisson distribution of detected photon arrival times), then this short a timescale requires that the pulses originate in the accretion disk around Cyg XR-1. Two series of pulses with characteristics similar to those expected from dying pulse trains were detected in three hours of observation.

Dolan, Joseph F.↗
NASA NTRS2002

Second Generation RLV Space Vehicle Concept

NASA has a long history of conducting development programs and projects in a consistent fashion. Systems Engineering within those programs and projects has also followed a given method outlined by such documents as the NASA Systems Engineering Handbook. The relatively new NASA Space Launch Initiative (SLI) is taking a new approach to developing a space vehicle, with innovative management methods as well as new Systems Engineering processes. With the program less than a year into its life cycle, the efficacy of these new processes has yet to be proven or disproven. At $776M for phase 1, SLI represents a major portion of the NASA focus; however, the new processes being incorporated are not reflected in the training provided by NASA to its engineers. The NASA Academy of Program and Project Leadership (APPL) offers core classes in program and project management and systems engineering to NASA employees with the purpose of creating a "knowledge community where ideas, skills, and experiences are exchanged to increase each other's capacity for strong leadership". The SLI program is, in one sense, a combination of a conceptual design program and a technology program. The program as a whole doesn't map into the generic systems engineering project cycle as currently, and for some time, taught. For example, the NASA APPL Systems Engineering training course teaches that the "first step in developing an architecture is to define the external boundaries of the system", which will require definition of the interfaces with other systems and the next step will be to "define all the components that make up the next lower level of the system hierarchy" where fundamental requirements are allocated to each component. Whereas, the SLI technology risk reduction approach develops architecture subsystem technologies prior to developing architectures. The higher level architecture requirements are not allowed to fully develop and undergo decomposition and allocation down to the subsystems before the subsystems must develop allocated requirements based on the highest level of requirements. In the vernacular of the project cycles prior to the mid 1990's, the architecture definition portion of the program appears to be at a generic Phase A stage, while the subsystems are operating at Phase B. Even the management structure of the SLI program is innovative in its approach to Systems Engineering and is not reflected in the APPL training modules. The SLI program has established a Systems Engineering office as an office separate from the architecture development or the subsystem technology development, while that office does have representatives within these other offices. The distributed resources of the Systems Engineering Office are co-located with the respective Project Offices. This template is intended to provide systems engineering as an integrated function at the Program Level. the program management of SLI and the MAT agree that "program/project managers and the systems engineering team must work closely together towards the single objective of delivering quality products that meet the customer needs". This paper will explore the differences between the methods being taught by NASA, which represent decades of ideas, and those currently in practice in SLI. Time will tell if the innovation employed by SLI will prove to be the model of the future. For now, it is suggested that the training of the present exercise the flexibility of recognizing the new processes employed by a major new NASA program.

Bailey, Michelle↗
NASA NTRS1995

SEL's Software Process-Improvement Program

The goals and operations of the Software Engineering Laboratory (SEL) is reviewed. For nearly 20 years the SEL has worked to understand, assess, and improve software and the development process within the production environment of the Flight Dynamics Division (FDD) of NASA's Goddard Space Flight Center. The SEL was established in 1976 with the goals of reducing: (1) the defect rate of delivered software, (2) the cost of software to support flight projects, and (3) the average time to produce mission-support software. After studying over 125 projects of FDD, the results have guided the standards, management practices, technologies, and the training within the division. The results of the studies have been a 75 percent reduction in defects, a 50 percent reduction in cost, and a 25 percent reduction in development time. Over time the goals of SEL have been clarified. The goals are now stated as: (1) Understand baseline processes and product characteristics, (2) Assess improvements that have been incorporated into the development projects, (3) Package and infuse improvements into the standard SEL process. The SEL improvement goal is to demonstrate continual improvement of the software process by carrying out analysis, measurement and feedback to projects with in the FDD environment. The SEL supports the understanding of the process by study of several processes including, the effort distribution, and error detection rates. The SEL assesses and refines the processes. Once the assessment and refinement of a process is completed, the SEL packages the process by capturing the process in standards, tools and training.

Basili, Victor↗
NASA NTRS1976

Long-latency evoked potentials to irrelevant, deviant stimuli

Occasional shifts of loudness in a repetitive train of clicks elicited a late-positive wave (P3a) in nonattending subjects which peaked at a mean latency of 258 msec and had a frontocentral scalp distribution; P3a was typically preceded by an 'N2' component at 196 msec. The P3a wave was distinguishable from the longer-latency (378 msec) parietocentrally distributed 'P3b' wave that was evoked by the same stimulus in an actively attending subject, thus confirming the findings of Squires et al. (1975). Infrequently presented single sounds did not produce large or consistent N2-P3a components; the critical condition for the generation of an N2-P3a wave seemed to be that the infrequent sounds represent a deviation (intensity increment or decrement) from a repetitive background. Furthermore, increasing the repetition rate of the background clicks drastically reduced N1-P2 amplitude but had little effect on the amplitude of N2-P3a. This suggests that N2-P3a is not simply a delayed N1-P2 'vertex potential', but rather reflects the operation of a 'mismatch' detector, which registers deviations from an ongoing auditory background.

Snyder, E.↗
NASA NTRS1998

Measures and Interpretations of Vigilance Performance: Evidence Against the Detection Criterion

Operators' performance in a vigilance task is often assumed to depend on their choice of a detection criterion. When the signal rate is low this criterion is set high, causing the hit and false alarm rates to be low. With increasing time on task the criterion presumably tends to increase even further, thereby further decreasing the hit and false alarm rates. Virtually all of the empirical evidence for this simple interpretation is based on estimates of the bias measure Beta from signal detection theory. In this article, I describe a new approach to studying decision making that does not require the technical assumptions of signal detection theory. The results of this new analysis suggest that the detection criterion is never biased toward either response, even when the signal rate is low and the time on task is long. Two modifications of the signal detection theory framework are considered to account for this seemingly paradoxical result. The first assumes that the signal rate affects the relative sizes of the variances of the information distributions; the second assumes that the signal rate affects the logic of the operator's stopping rule. Actual or potential applications of this research include the improved training and performance assessment of operators in areas such as product quality control, air traffic control, and medical and clinical diagnosis.

Balakrishnan, J. D.↗
NASA NTRS2004

Applications of Intelligent Tutoring Systems to Human-Robotic Exploration of Mars

Space missions with small crews extending over several years with time-delay preventing normal conversations with people on earth will raise many challenges for training. Of special interest are possible three-year missions to Mars, requiring refresher instruction and learning new skills based on unexpected problems with machines and environmental conditions. For example, the crew will be required to monitor and repair more complex life support systems for air and water recycling than we even know how to build today. Highly educated astronauts, often with several doctorate degrees, require a very different mode of interaction than we have developed for school children or even typical college students. Explanation methods may need to differ-using analogies and techniques from different domains-depending on whether the astronaut is an astrophysicist, a pilot, or a geologist.Virtual reality (e.g., for Hubble repair missions) and "integrated" simulations (involving role-playing and emphasizing failure scenarios) are the most common advanced forms of instruction used in space flight today. The emphasis is on collaborative, embodied interaction with the same workstations and tools used in practice (e.g., a cockpit simulator). Otherwise, computerized instructional technology used by NASA is not model-based or tutorial in nature. This discussion will review some of the key instructional methods used at NASA over the past two decades and consider why ITS methods have not been exploited. Some of the problems and opportunities for training for Mars missions are examined, including how using robots in exploration activities will help but raise new training problems. These ideas will be illustrated with examples from the BrahmsVE system in which a browser- based virtual reality display with avatars allows interacting with a distributed multiagent system, in which agents can be people, robots, or software programs. Using BrahmsVE may provide a way for astronauts to interact with proxies of people who serve as instructional coaches on Mars.

Clancey, William J.↗
NASA NTRS2012

Shuttle Case Study Collection Website Development

As a continuation from summer 2012, the Shuttle Case Study Collection has been developed using lessons learned documented by NASA engineers, analysts, and contractors. Decades of information related to processing and launching the Space Shuttle is gathered into a single database to provide educators with an alternative means to teach real-world engineering processes. The goal is to provide additional engineering materials that enhance critical thinking, decision making, and problem solving skills. During this second phase of the project, the Shuttle Case Study Collection website was developed. Extensive HTML coding to link downloadable documents, videos, and images was required, as was training to learn NASA's Content Management System (CMS) for website design. As the final stage of the collection development, the website is designed to allow for distribution of information to the public as well as for case study report submissions from other educators online.

Ransom, Khadijah S.↗
NASA NTRSDate not supplied

Atmosphere Modeling and Performance Sensitivity for the Mars Sample Return Earth Entry System

The Capture, Containment, and Return System (CCRS) mission is a key element of the joint NASA-European Space Agency (ESA) planned Mars Sample Return (MSR) Campaign. The CCRS assembled Earth Entry System (EES) will enter the mission’s final segment in its Approach, Entry, Descent, and Landing (AEDL) Phase. The EES AEDL aims to deliver a highly reliable, safe, and accurate return while maintaining strict containment assurance targets established by the campaign. As currently designed, the EES would be the first fully passive sample return capsule with no parachute or onboard control system, prompting a significant effort in Earth atmosphere characterization and modeling. Earth’s atmosphere, specifically winds, have a strong influence on EES flight mechanics and landing footprint during its free fall landing. This paper describes Earth atmosphere modeling, atmosphere characterization, and performance sensitivities incorporated into the teams’s efforts to ensure AEDL success. By utilizing high resolution balloon radiosondes, analyzing wind structural and distributional compositions, and investigating flight mechanics sensitivities, the AEDL atmosphere team has been able to better understand and simulate local wind conditions at the Utah Test and Training Range (UTTR). Utilizing tools such as horizontal turbulent kinetic energy, vertical wind shear, or integrated wind, the team have been able to reveal valuable information about wind profiles in a deeper context than previously conducted, ultimately improving understanding of AEDL flight mechanics sensitivity and EES design.

Kaustubh Ray↗
DOE OSTI2025

Creating the Distributed Energy Resources Education Center (DEREC)

The built environment in the United States consumes 40% of the energy generated and emits roughly the same percentage of total carbon footprint. Distributed energy resources (DER), small or modular energy generation and storage technologies, present the nation with an opportunity to substantially improve those metrics while securing the nation’s energy independence. As opportunities increase for implementing such technologies, they also continue to evolve and often outpace the nation’s traditional building practices. In an effort to effectively and proactively incorporate distributed energy resources into the nation’s energy supply, Southface Energy Institute convened with national and regional partners to create the Distributed Energy Resources Education Center (DEREC). Using national model codes and their regionally amended versions as a collective starting point, the DEREC team collaborated with industry experts and identified impediments to effective implementation of DERs, developing discipline-specific curriculum to eliminate those impediments. The center, developed in collaboration with Interstate Renewable Energy Committee (IREC) and National Buildings Institute (NBI), leverages existing DER education content as well as new and dynamic training materials and online courses that collectively engage the many roles necessary for DER implementations, including designers, code officials, builders and skilled trades, and building owners who specify, inspect, build, operate, and maintain buildings with DERs.

14 SOLAR ENERGY↗
NASA NTRS2014

OMMYDCLD: a New A-train Cloud Product that Co-locates OMI and MODIS Cloud and Radiance Parameters onto the OMI Footprint

Clouds cover approximately 60% of the earth's surface. When obscuring the satellite's field of view (FOV), clouds complicate the retrieval of ozone, trace gases and aerosols from data collected by earth observing satellites. Cloud properties associated with optical thickness, cloud pressure, water phase, drop size distribution (DSD), cloud fraction, vertical and areal extent can also change significantly over short spatio-temporal scales. The radiative transfer models used to retrieve column estimates of atmospheric constituents typically do not account for all these properties and their variations. The OMI science team is preparing to release a new data product, OMMYDCLD, which combines the cloud information from sensors on board two earth observing satellites in the NASA A-Train: Aura/OMI and Aqua/MODIS. OMMYDCLD co-locates high resolution cloud and radiance information from MODIS onto the much larger OMI pixel and combines it with parameters derived from the two other OMI cloud products: OMCLDRR and OMCLDO2. The product includes histograms for MODIS scientific data sets (SDS) provided at 1 km resolution. The statistics of key data fields - such as effective particle radius, cloud optical thickness and cloud water path - are further separated into liquid and ice categories using the optical and IR phase information. OMMYDCLD offers users of OMI data cloud information that will be useful for carrying out OMI calibration work, multi-year studies of cloud vertical structure and in the identification and classification of multi-layer clouds.

OMMYDCLD↗