Engineering Papers⌕ Search

SEARCH · Engineering Papers

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 1,189 records · Page 66
All sourcesNASA NTRSDOE OSTI
NASA NTRS1989

Performance evaluation of a six-axis generalized force-reflecting teleoperator

Work in real-time distributed computation and control has culminated in a prototype force-reflecting telemanipulation system having a dissimilar master (cable-driven, force-reflecting hand controller) and a slave (PUMA 560 robot with custom controller), an extremely high sampling rate (1000 Hz), and a low loop computation delay (5 msec). In a series of experiments with this system and five trained test operators covering over 100 hours of teleoperation, performance was measured in a series of generic and application-driven tasks with and without force feedback, and with control shared between teleoperation and local sensor referenced control. Measurements defining task performance included 100-Hz recording of six-axis force/torque information from the slave manipulator wrist, task completion time, and visual observation of predefined task errors. The task consisted of high precision peg-in-hole insertion, electrical connectors, velcro attach-de-attach, and a twist-lock multi-pin connector. Each task was repeated three times under several operating conditions: normal bilateral telemanipulation, forward position control without force feedback, and shared control. In shared control, orientation was locally servo controlled to comply with applied torques, while translation was under operator control. All performance measures improved as capability was added along a spectrum of capabilities ranging from pure position control through force-reflecting teleoperation and shared control. Performance was optimal for the bare-handed operator.

Hannaford, B.↗
NASA NTRS1991

Performance evaluation of a six-axis generalized force-reflecting teleoperator

Work in real-time distributed computation and control has culminated in a prototype force-reflecting telemanipulation system having a dissimilar master (cable-driven force-reflecting hand controller) and a slave (PUMA 560 robot with custom controller), an extremely high sampling rate (1000 Hz), and a low loop computation delay (5 msec). In a series of experiments with this system and five trained test operators covering more than 100 hours of teleoperation, performance was measured in a series of generic and application-driven tasks with and without force feedback, and with control shared between teleoperation and local sensor referenced control. All performance measures improved as capability was added along a spectrum of capabilities ranging from pure position control through force-reflecting teleoperation and shared control. Performance was optimal for the bare-handed operator.

Hannaford, Blake↗
NASA NTRS1992

Analysis of fault-tolerant neurocontrol architectures

The fault-tolerance of analog parallel distributed implementations of a multivariable aircraft neurocontroller is analyzed by simulating weight and neuron failures in a simplified scheme of analog processing based on the functional architecture of the ETANN chip (Electrically Trainable Artificial Neural Network). The neural information processing is found to be only partially distributed throughout the set of weights of the neurocontroller synthesized with the backpropagation algorithm. Although the degree of distribution of the neural processing, and consequently the fault-tolerance of the neurocontroller, could be enhanced using Locally Distributed Weight and Neuron Approaches, a satisfactory level of fault-tolerance could only be obtained by retraining the degrated VLSI neurocontroller. The possibility of maintaining neurocontrol performance and stability in the presence of single weight of neuron failures was demonstrated through an automated retraining procedure of the neurocontroller based on a pre-programmed choice and sequence of the training parameters.

Troudet, T.↗
NASA NTRS2003

New Programs Utilizing Light Scattering and Flow Imaging Techniques for Macromolecular Crystal Growth and Fluid Dynamics Studies

Dr. Phil Segre, a physicist by training, is a recent addition to the Biotech group, SD46, having joined NASA in August of 2000. Over the past two years he has been developing a laboratory for the study of macromolecular and protein crystal growth. The main apparatus for this work is a Dynamic Light Scattering apparatus, DLS, which is capable of making highly precise measurements of size distributions of both protein solutions and protein crystals. With Drs. Chernov and Thomas (USRA), he has begun a collaboration studying the affects of protein impurities on protein crystal growth and subsequent crystal quality. One of the hypotheses behind the differences between Earth and space grown protein crystals is that the absorption of harmful impurities is reduced in space due to the absence of convective flows. Using DLS measurements we are examining crystal growth with varying amounts of impurities and testing whether there is a strong physical basis behind this hypothesis. With Dr. Joe Ng of UAH he has been collaborating on a project to examine the folding/unfolding dynamics of large RNA complexes. A detailed understanding of this process is necessary for the handling of RNA in biotech applications, and the DLS instrument gives details and results beyond that of other instruments. With Prof. Jim McClymer of the University of Maine (summer faculty visitor to NASA in 2001, 2002), we have been studying the crystallization process in model colloidal suspensions whose behavior in some cases can mimic that of much smaller protein solutions. An understanding of the self-assembly of colloids is the first step in the process of engineering novel materials for photonic and light switching applications. Finally, he has begun an investigation into the physics of particle sedimentation. In addition to the DLS instrument he also has an instrument (called PIV) that can measure flow fields of fluids. The applications are to the dynamics of protein crystal motions both on earth and in low-gravity.

Source record↗
DOE OSTI2022

Deep-learning-based workflow for boundary and small target segmentation in digital rock images using UNet++ and IK-EBM

We report three-dimensional (3D) X-ray micro-computed tomography (μCT) has been widely used in petroleum engineering because it can provide detailed pore structural information for a reservoir rock, which can be imported into a pore-scale numerical model to simulate the transport and distribution of multiple fluids in the pore space. The partial volume blurring (PVB) problem is a major challenge in segmenting raw μCT images of rock samples, which impacts boundaries and small targets near the resolution limit. We developed a deep-learning (DL)-based workflow for accurate and fast partial volume segmentation. The DL model's performance depends primarily on the training data quality and model architecture. This study employed the entropy-based-masking indicator kriging (IK-EBM) to segment 3D Berea sandstone images as training datasets. The comparison between IK-EBM and manual segmentation using a 3D synthetic sphere pack, which had a known ground truth, showed that IK-EBM had higher accuracy on partial volume segmentation. We then trained and tested the UNet++ model, a state-of-the-art supervised encoder-decoder model, for binary (i.e., void and solid) and four-class segmentation. We compared the UNet++ with the commonly used U-Net and wide U-Net models and showed that the UNet++ had the best performance in terms of pixel-wise and physics-based evaluation metrics. Specifically, boundary-scaled accuracy demonstrated that the UNet++ architecture outperformed the regular U-Net architecture in the segmentation of pixels near boundaries and small targets, which were subjected to the PVB effect. Feature map visualization illustrated that the UNet++ bridged the semantic gaps between the feature maps extracted at different depths of the network, thereby enabling faster convergence and more accurate extraction of fine-scale features. The developed workflow significantly enhances the performance of supervised encoder-decoder models in partial volume segmentation, which has extensive applications in fundamental studies of subsurface energy, water, and environmental systems.

02 PETROLEUM↗
NASA NTRS1973

Radar signal return from near-shore surface and shallow subsurface features, Darien Province, Panama

The AN/APQ-97 radar imagery over eastern Panama is analyzed. The imagery was directed toward extraction of geologic and engineering data and the establishment of operational parameters. Subsequent investigations emphasized landform identification and vegetation distribution. The parameters affecting the observed return signal strength from such features are considered. Near-shore ocean phenomena were analyzed. Tidal zone features such as mud flats and reefs were identified in the near range, but were not detectable in the far range. Surface roughness dictated the nature of reflected energy (specular or diffuse). In surf zones, changes in wave train orientation relative to look direction, the slope of the surface, and the physical character of the wave must be considered. It is concluded that the establishment of the areal extent of the tidal flats, distributary channels, and reefs is practical only in the near to intermediate range under minimal low tide conditions.

Hanson, B. C.↗
NASA NTRS1996

Studies of Vestibular Neurons in Normal, Hyper- and Hypogravity

During the past year, pre-, in- and postflight studies were conducted in association with the Axon project for Bion 10 (Cosmos 2229). Recordings were made during pre- and postflight studies, from 118 horizontal semicircular canal afferents and 27 vestibular nucleus neurons in 7 rhesus monkeys; 137 pulse rotation protocols alone were executed (548 acceleration and deceleration responses were curve fit). Usable data was obtained from 127 horizontal afferents concerning their spontaneous discharge. Curve fits and analysis was made of sinusoidal and sum of sinusoidal responses from 42 and 35 horizontal afferents, respectively. Also recordings were made from neurons inflight from the two flight animals. The mean spontaneous rate varied from 128 spikes/sec. during preflight to 92 spikes/sec during postflight (day 5) - a change of 28%. In direct contrast to the results of Cosmos 2044, the best fitted neural adaptation operator (k) and the gain of the pulse response were decreased during post flight when compared to preflight. Surprisingly, the best fitted gain and k values for the sum of sines were slightly elevated during post flight tests. The gain and phase of single sine responses were compared for pre- and post flight tests and compared to a larger population of afferents. In contrast to Cosmos 2044 results where on the first day of post flight testing the gains of the best fitted sine response were skewed toward the higher values of the Miles and Braitman distribution, the gain of the best fitted sine responses during the first day of post flight testing (day 2) during Cosmos 2229 were exactly on the mode of the Miles and Braitman distribution. Thus, at least for the periodic stimuli, (pulses and sine waves) we found no change in gain and neural adaptation during post flight testing following Cosmos 2229. This conclusion is different from the one derived following the Cosmos 2044 flight. Cosmos flight 2229 differed from Cosmos flight 2044 in several significant ways: For example, during preflight, (1) The animals preflight training was different (less well trained on the gaze task) and (2) the animals were exposed to more experimental manipulations (surgical and rotational). Inflight, (1) the animals were required to make a pointing gesture (motor response) in association with eye movements to obtain reward, (2) the inflight diet was different (more balanced), (3) the feeder for one of the animals clogged following 9 days of flight resulting in evident dehydration and probably less head motion exposure in that monkey and (4) there was limited video taping of the monkeys in space. During postflight, (1) we were unable to test the flight animals until 26 hours postflight as compared to 14.5 hours during Cosmos 2044, (2) the animals received significantly more exposure to motion stimuli during postflight testing than during Cosmos 2044. These differences in the vestibular environment will require analysis of several parameters other than just neural and eye movement responses. For example, computer programs will have to be written and used to recover and quantify the number of head movements made by each animal during flight. This activity is critical to the production of neural adaptation and increased gain.

Correia, Manning J.↗
NASA NTRS2005

Size Distribution of Genesis Solar Wind Array Collector Fragments Recovered

Genesis launched in 2001 with 271 whole and 30 half hexagonally-shaped collectors mounted on 5 arrays, comprised of 9 materials described in [1]. The array collectors were damaged during re-entry impact in Utah in 2004 [2], breaking into many smaller pieces and dust. A compilation of the number and approximate size of the fragments recovered was compiled from notes made during the field packaging performed in the Class 10,000 cleanroom at Utah Test and Training Range [3].

Allton, J. H.↗
DOE OSTI2022

Adversarial autoencoder ensemble for fast and probabilistic reconstructions of few-shot photon correlation functions for solid-state quantum emitters

Second-order photon correlation measurements [g (2) (τ) functions] are widely used to classify single-photon emission purity in quantum emitters or to measure the multiexciton quantum yield of emitters that can simultaneously host multiple excitations – such as quantum dots – by evaluating the value of g (2) (τ = 0). Accumulating enough photons to accurately calculate this value is time consuming and could be accelerated by fitting of few-shot photon correlations. Here, we develop an uncertainty-aware, deep adversarial autoencoder ensemble (AAE) that reconstructs noise-free g (2) (τ) functions from noise-dominated, few-shot inputs. The model is trained with simulated g (2) (τ) functions that are facilely generated by Poisson sampling time bins. The AAE reconstructions are performed orders-of-magnitude faster, with reconstruction errors and estimates of g (2) (τ = 0) that are lower in variance and similar in accuracy compared to Maximum likelihood estimation and Levenberg-Marquardt least-squares fitting approaches, for simulated and experimentally measured few-shot g (2) (τ) functions (~100 two-photon events) of InP/ZnS/ZnSe and CdS/CdSe/CdS quantum dots. The deep-ensemble model comprises eight individual autoencoders, allowing for probabilistic reconstructions of noise-free g (2) (τ) functions, and we show that the predicted variance scales inversely with number of shots, with comparable uncertainties to computationally intensive Markov chain Monte Carlo sampling. Furthermore, this work demonstrates the advantage of machine learning models to perform uncertainty-aware, fast, and accurate reconstructions of simple Poisson-distributed photon correlation functions, allowing for on-the-fly reconstructions and accelerated materials characterization of solid-state quantum emitters.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2021

An adaptive approach to machine learning for compact particle accelerators

Abstract Machine learning (ML) tools are able to learn relationships between the inputs and outputs of large complex systems directly from data. However, for time-varying systems, the predictive capabilities of ML tools degrade if the systems are no longer accurately represented by the data with which the ML models were trained. For complex systems, re-training is only possible if the changes are slow relative to the rate at which large numbers of new input-output training data can be non-invasively recorded. In this work, we present an approach to deep learning for time-varying systems that does not require re-training, but uses instead an adaptive feedback in the architecture of deep convolutional neural networks (CNN). The feedback is based only on available system output measurements and is applied in the encoded low-dimensional dense layers of the encoder-decoder CNNs. First, we develop an inverse model of a complex accelerator system to map output beam measurements to input beam distributions, while both the accelerator components and the unknown input beam distribution vary rapidly with time. We then demonstrate our method on experimental measurements of the input and output beam distributions of the HiRES ultra-fast electron diffraction (UED) beam line at Lawrence Berkeley National Laboratory, and showcase its ability for automatic tracking of the time varying photocathode quantum efficiency map. Our method can be successfully used to aid both physics and ML-based surrogate online models to provide non-invasive beam diagnostics.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Particle Filter Based Inference Testing

The primary intent of PAR-FIT (Particle Filter based Inference Testing) is to provide hard inductive evidence that a machine learning model is capable and proven for an individual test input. By examining training data used to form the underlying model functional correlation, an estimate of the reliability that a model will make the correct prediction can be made. The Sequential Probability Ratio Test is used to derive a qualitative evaluation for reliability based on hypothesis testing. The PAR-FIT framework achieves this by implementing a particle filter and the sequential probability ratio test algorithms on the machine learning model training data to determine relevancy of new individual test samples to the training dataset. The kernel function evaluates the local proximity and density of training data used to derive a prediction outcome. Particles are used to probabilistically determine which training data to evaluate for proximity. For test samples that are within a close proximity to and surrounded by multiple training data points, the evaluated reliability of the prediction is high. For test samples that are anomalies not represented by the training dataset, in low density data clusters, or are far from existing data points, the evaluated reliability is low as insufficient training evidence exists to suggest the model is capable of making the correct prediction. Sequential Probability Ratio Test is further used to determine when a hypothesis on whether a signal can be rejected or accepted for use. The ratio test collects sequence information from the particle filter to test whether the signal is anomalous or normal via hypothesis testing of the underlying distributions.

Chen, Edward [Idaho National Laboratory (INL), Ida↗
NASA NTRS2017

Creating Aerosol Types Form CHemistry (CATCH): a New Algorithm to Extend the Link Between Remote Sensing and Models

Current remote sensing methods can identify aerosol types within an atmospheric column, presenting an opportunity to incrementally bridge the gap between remote sensing and models. Here a new algorithm was designed for Creating Aerosol Types from CHemistry (CATCH). CATCH-derived aerosol types—dusty mix, maritime, urban, smoke, and fresh smoke—are based on first-generation airborne High Spectral Resolution Lidar (HSRL-1) retrievals during the Ship-Aircraft Bio-Optical Research (SABOR) campaign, July/August 2014. CATCH is designed to derive aerosol types from model output of chemical composition. CATCH-derived aerosol types are determined by multivariate clustering of model-calculated variables that have been trained using retrievals of aerosol types from HSRL-1. CATCH-derived aerosol types (with the exception of smoke) compare well with HSRL-1 retrievals during SABOR with an average difference in aerosol optical depth (AOD) <0.03. Data analysis shows that episodic free tropospheric transport of smoke is underpredicted by the Goddard Earth Observing System- with Chemistry (GEOS-Chem) model. Spatial distributions of CATCH-derived aerosol types for the North American model domain during July/August 2014 show that aerosol type-specific AOD values occurred over representative locations: urban over areas with large population, maritime over oceans, smoke, and fresh smoke over typical biomass burning regions. This study demonstrates that model-generated information on aerosol chemical composition can be translated into aerosol types analogous to those retrieved from remote sensing methods. In the future, spaceborne HSRL-1 and CATCH can be used to gain insight into chemical composition of aerosol types, reducing uncertainties in estimates of aerosol radiative forcing.

K W Dawson↗
NASA NTRSDate not supplied

Ice-Over-Water Cloud Properties in an Artificial Neural Network Approach

Clouds are a crucial component of the atmospheric energy system, particularly the radiative balance within, above, and below the troposphere. The vertical distribution of cloud mass and phase determines layer heating rates, the loss of radiation to space, and the amount of radiative heating at the surface. Thus, it is important to know how clouds are distributed both vertically and horizontally at all times of day. Satellite remote sensing is the only approach available to monitor clouds day and night around the globe. In this paper several artificial neural network (ANN) algorithms, employing several Aqua MODIS infrared channels, profiles of relative humidity and temperature from GMAO numerical weather analyses, and the retrieved total cloud visible optical depth, are trained to detect multilayer ice-over-water cloud systems and to retrieve some of their properties as identified by a year of 2008 Aqua MODIS data matched with CloudSat and CALIPSO (CC) cloud profiles. The CC lidar and radar profiles provide the vertical structure that serves as output truth for the multilayer algorithm. The neural networks were trained using one year (2008) of cloud top height data from the CC dataset, with correlation around 0.94 (0.95) and MAE as low as 0.82 (0.81) km for nonpolar regions during the day (night). Applying the trained ANN to independent year 2009 MODIS data resulted in a combined ML and single layer hit rate of 86.4% (85.1%) for nonpolar regions during the day (night). Since the ANN is trained using near-nadir MODIS pixels, infrared radiance corrections were developed as a function of view zenith angle from MODIS and applied to off-nadir pixels when processing MODIS swath data. The multilayer amount derived with the ANN is relatively invariant with increasing view zenith angle compared to the multilayer amount without the corrections.

MODIS↗
NASA NTRS2024

Synergistic Retrievals of Ice in High Clouds From Elastic Backscatter Lidar, Ku-band Radar and Submillimeter Wave Radiometer Observations

In this study, we investigate the synergy of elastic backscatter lidar, Ku-band radar, and sub-millimeter-wave radiometer measurements in the retrieval of ice from satellite observations. The synergy is analyzed through the generation of a large dataset of IceWater Content (IWC) profiles and simulated lidar, radar and radiometer observations. The characteristics of the instruments e.g. frequencies, sensitivities, etc. are set based on the expected characteristics of instruments of the Atmosphere Observing System (AOS) mission. A hold-out validation methodology is used to assess the accuracy of the IWC profiles retrieved from various combinations of observations from the three instruments. Specifically, the IWC and associated observations are randomly divided into two datasets, one for training and the other for evaluation. The training dataset is used to train the retrieval algorithm, while the evaluation dataset is used to assess the retrieval performance. The dataset of IWC profiles is derived from CloudSat reflectivity and CALIOP lidar observations. The retrieval of the ice water content IWC profiles from the computed observations is achieved in two steps. In the first step, a class, out of 18 potential classes characterized by different vertical distribution of IWC, is estimated from the observations. The 18 classes are predetermined based on the k-Means clustering algorithm. In the second step, the IWC profile is estimated using an Ensemble Kalman Smoother (EKS) algorithm that uses the estimated class as a priori information. The results of the study show that the synergy of lidar, radar, and radiometer observations is significant in the retrieval of the IWC profiles. Nevertheless, it should be mentioned that this synergy was found under idealized conditions, and additional work might be required to materialize it in practice. The inclusion of the lidar backscatter observations in the retrieval process has a larger impact on the retrieval performance than the inclusion of the radar observations. As ice clouds have a significant impact on atmospheric radiative processes, this work is relevant to ongoing efforts to reduce uncertainties in climate analyses and projections.

Mircea Grecu↗
NASA NTRS1981

Geological applications and training in remote sensing

Some of the experiences, methods, and opinions developed during 15 years of teaching an introductory course in remote sensing at several universities in the Southern California area are related. Although the course is offered in Geology departments, every class includes significant numbers of students from other disciplines including geography, computer science, biology, and environmental science. The instructor or teaching assistant provides a few hours of tutorial lectures (outside of regular class time) on basic geology for these nongeologists. This approach is successful because the grade distribution for nongeologists is similar to that for geologists. The schedule for a typical one-semester course is given.

Sabins, F. F., Jr.↗
NASA NTRS2011

Collaborative Clustering for Sensor Networks

Traditionally, nodes in a sensor network simply collect data and then pass it on to a centralized node that archives, distributes, and possibly analyzes the data. However, analysis at the individual nodes could enable faster detection of anomalies or other interesting events, as well as faster responses such as sending out alerts or increasing the data collection rate. There is an additional opportunity for increased performance if individual nodes can communicate directly with their neighbors. Previously, a method was developed by which machine learning classification algorithms could collaborate to achieve high performance autonomously (without requiring human intervention). This method worked for supervised learning algorithms, in which labeled data is used to train models. The learners collaborated by exchanging labels describing the data. The new advance enables clustering algorithms, which do not use labeled data, to also collaborate. This is achieved by defining a new language for collaboration that uses pair-wise constraints to encode useful information for other learners. These constraints specify that two items must, or cannot, be placed into the same cluster. Previous work has shown that clustering with these constraints (in isolation) already improves performance. In the problem formulation, each learner resides at a different node in the sensor network and makes observations (collects data) independently of the other learners. Each learner clusters its data and then selects a pair of items about which it is uncertain and uses them to query its neighbors. The resulting feedback (a must and cannot constraint from each neighbor) is combined by the learner into a consensus constraint, and it then reclusters its data while incorporating the new constraint. A strategy was also proposed for cleaning the resulting constraint sets, which may contain conflicting constraints; this improves performance significantly. This approach has been applied to collaborative clustering of seismic and infrasonic data collected by the Mount Erebus Volcano Observatory in Antarctica. Previous approaches to distributed clustering cannot readily be applied in a sensor network setting, because they assume that each node has the same view of the data set. A view is the set of features used to represent each object. When a single data set is partitioned across several computational nodes, distributed clustering works; all objects have the same view. But when the data is collected from different locations, using different sensors, a more flexible approach is needed. This approach instead operates in situations where the data collected at each node has a different view (e.g., seismic vs. infrasonic sensors), but they observe the same events. This enables them to exchange information about the likely cluster membership relations between objects, even if they do not use the same features to represent the objects.

Wagstaff. Loro :/↗
DOE OSTI2023

UAS Edge Computing of Energy Infrastructure Damage Assessment

Energy infrastructure assessments are needed within 72 hours of natural disasters, and previous data collection methods have proven too slow. We demonstrate a scalable end-to-end solution using a prototype unmanned aerial system that performs on-the-edge detection, classification (i.e., damaged or undamaged), and geo-location of utility poles. The prototype is suitable for disaster response because it requires no local communication infrastructure and is capable of autonomous missions. Collections before, during, and after Hurricane Ida in 2021 were used to test the system. The system delivered an F1 score of 0.65 operating with a 2.7 s/frame processing speed with the YOLOv5 large model and an F1 score of 0.55 with a 0.48 s/frame with the YOLOv5 small model. Geo-location uncertainty in the bottom half of the frame was ~8 m, mostly driven by error in camera pointing measurement. With additional training data to improve performance and detect additional types of features, a fleet of similar drones could autonomously collect actionable post-disaster data.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS1993

Learning class descriptions from a data base of spectral reflectance of soil samples

Consideration is given to a program developed to learn class descriptions from positive and negative training examples of spectral reflectance data of bare soils. It is a combination of 'learning by example' and the generate-and-test paradigm and is designed to provide a robust learning environment that can handle error-prone data. The program was tested by having it learn class descriptions of various categories of organic carbon content, iron oxide content, and particle size distribution in soils. These class descriptions were then used to classify an array of targets. The program found the sequence of relationships between bands that contained the most important information to distinguish the classes. Physical explanations for the class descriptions obtained are presented.

Kimes, D. S.↗