SEARCH · Engineering Papers

Results for “Training Analysis”

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.

SourceResearch connectionFrom yearThrough year

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

At least 505 records · Page 28

All sources NASA NTRS DOE OSTI

NASA NTRS1977

Development of the engineering design integration (EDIN) system: A computer aided design development

The EDIN (Engineering Design Integration) System which provides a collection of hardware and software, enabling the engineer to perform man-in-the-loop interactive evaluation of aerospace vehicle concepts, was considered. Study efforts were concentrated in the following areas: (1) integration of hardware with the Univac Exec 8 System; (2) development of interactive software for the EDIN System; (3) upgrading of the EDIN technology module library to an interactive status; (4) verification of the soundness of the developing EDIN System; (5) support of NASA in design analysis studies using the EDIN System; (6) provide training and documentation in the use of the EDIN System; and (7) provide an implementation plan for the next phase of development and recommendations for meeting long range objectives.

Glatt, C. R.↗

NASA NTRS1980

ICAP: An Interactive Cluster Analysis Procedure for analyzing remotely sensed data

An Interactive Cluster Analysis Procedure (ICAP) was developed to derive classifier training statistics from remotely sensed data. The algorithm interfaces the rapid numerical processing capacity of a computer with the human ability to integrate qualitative information. Control of the clustering process alternates between the algorithm, which creates new centroids and forms clusters and the analyst, who evaluate and elect to modify the cluster structure. Clusters can be deleted or lumped pairwise, or new centroids can be added. A summary of the cluster statistics can be requested to facilitate cluster manipulation. The ICAP was implemented in APL (A Programming Language), an interactive computer language. The flexibility of the algorithm was evaluated using data from different LANDSAT scenes to simulate two situations: one in which the analyst is assumed to have no prior knowledge about the data and wishes to have the clusters formed more or less automatically; and the other in which the analyst is assumed to have some knowledge about the data structure and wishes to use that information to closely supervise the clustering process. For comparison, an existing clustering method was also applied to the two data sets.

Wharton, S. W.↗

NASA NTRS1990

Neural net diagnostics for VLSI test

This paper discusses the application of neural network pattern analysis algorithms to the IC fault diagnosis problem. A fault diagnostic is a decision rule combining what is known about an ideal circuit test response with information about how it is distorted by fabrication variations and measurement noise. The rule is used to detect fault existence in fabricated circuits using real test equipment. Traditional statistical techniques may be used to achieve this goal, but they can employ unrealistic a priori assumptions about measurement data. Our approach to this problem employs an adaptive pattern analysis technique based on feedforward neural networks. During training, a feedforward network automatically captures unknown sample distributions. This is important because distributions arising from the nonlinear effects of process variation can be more complex than is typically assumed. A feedforward network is also able to extract measurement features which contribute significantly to making a correct decision. Traditional feature extraction techniques employ matrix manipulations which can be particularly costly for large measurement vectors. In this paper we discuss a software system which we are developing that uses this approach. We also provide a simple example illustrating the use of the technique for fault detection in an operational amplifier.

Lin, T.↗

NASA NTRS1995

GPS survey of the western Tien Shan

There were two major developments in 1994 in our collaborative GPS experiment in the Tien Shan of the Former Soviet Union (FSU). Both were motivated by our expectation that we will ultimately obtain better science at lower cost if we involve our colleagues in the FSU more deeply in (1) the collection and (2) the analysis of data. As an experimental test of the concept of having our local collaborators carry out the field work semi-autonomously, we sent 6 MIT receivers to the Tien Shan for a period of 3 months. To enable our collaborators to have the capability for data analysis, we provided computers for two data analysis centers and organized a two-week training session. This report emphasizes the rationale for deeper involvement of FSU scientists, describes the training sessions, discusses the data collection, and presents the results. We also discuss future plans. More detailed discussion of background, general scientific objectives, discussions with collaborators, and results for the campaigns in 1992 and 1993 have been given in previous reports.

Hager, Bradford H.↗

NASA NTRS1998

Neural Network and Regression Approximations in High Speed Civil Transport Aircraft Design Optimization

Nonlinear mathematical-programming-based design optimization can be an elegant method. However, the calculations required to generate the merit function, constraints, and their gradients, which are frequently required, can make the process computational intensive. The computational burden can be greatly reduced by using approximating analyzers derived from an original analyzer utilizing neural networks and linear regression methods. The experience gained from using both of these approximation methods in the design optimization of a high speed civil transport aircraft is the subject of this paper. The Langley Research Center's Flight Optimization System was selected for the aircraft analysis. This software was exercised to generate a set of training data with which a neural network and a regression method were trained, thereby producing the two approximating analyzers. The derived analyzers were coupled to the Lewis Research Center's CometBoards test bed to provide the optimization capability. With the combined software, both approximation methods were examined for use in aircraft design optimization, and both performed satisfactorily. The CPU time for solution of the problem, which had been measured in hours, was reduced to minutes with the neural network approximation and to seconds with the regression method. Instability encountered in the aircraft analysis software at certain design points was also eliminated. On the other hand, there were costs and difficulties associated with training the approximating analyzers. The CPU time required to generate the input-output pairs and to train the approximating analyzers was seven times that required for solution of the problem.

Patniak, Surya N.↗

NASA NTRS2011

Software Certification - Coding, Code, and Coders

We describe a certification approach for software development that has been adopted at our organization. JPL develops robotic spacecraft for the exploration of the solar system. The flight software that controls these spacecraft is considered to be mission critical. We argue that the goal of a software certification process cannot be the development of "perfect" software, i.e., software that can be formally proven to be correct under all imaginable and unimaginable circumstances. More realistically, the goal is to guarantee a software development process that is conducted by knowledgeable engineers, who follow generally accepted procedures to control known risks, while meeting agreed upon standards of workmanship. We target three specific issues that must be addressed in such a certification procedure: the coding process, the code that is developed, and the skills of the coders. The coding process is driven by standards (e.g., a coding standard) and tools. The code is mechanically checked against the standard with the help of state-of-the-art static source code analyzers. The coders, finally, are certified in on-site training courses that include formal exams.

training↗

NASA NTRS2016

Promoting A-Priori Interoperability of HLA-Based Simulations in the Space Domain: The SISO Space Reference FOM Initiative

Distributed and Real-Time Simulation plays a key-role in the Space domain being exploited for missions and systems analysis and engineering as well as for crew training and operational support. One of the most popular standards is the 1516-2010 IEEE Standard for Modeling and Simulation (M&S) High Level Architecture (HLA). HLA supports the implementation of distributed simulations (called Federations) in which a set of simulation entities (called Federates) can interact using a Run-Time Infrastructure (RTI). In a given Federation, a Federate can publish and/or subscribes objects and interactions on the RTI only in accordance with their structures as defined in a FOM (Federation Object Model). Currently, the Space domain is characterized by a set of incompatible FOMs that, although meet the specific needs of different organizations and projects, increases the long-term cost for interoperability. In this context, the availability of a reference FOM for the Space domain will enable the development of interoperable HLA-based simulators for related joint projects and collaborations among worldwide organizations involved in the Space domain (e.g. NASA, ESA, Roscosmos, and JAXA). The paper presents a first set of results achieved by a SISO standardization effort that aims at providing a Space Reference FOM for international collaboration on Space systems simulations.

Moller, Bjorn↗

NASA NTRS2015

Flight Test Series 3: Flight Test Report

This document is a flight test report from the Operational perspective for Flight Test Series 3, a subpart of the Unmanned Aircraft System (UAS) Integration in the National Airspace System (NAS) project. Flight Test Series 3 testing began on June 15, 2015, and concluded on August 12, 2015. Participants included NASA Ames Research Center, NASA Armstrong Flight Research Center, NASA Glenn Research Center, NASA Langley Research center, General Atomics Aeronautical Systems, Inc., and Honeywell. Key stakeholders analyzed their System Under Test (SUT) in two distinct configurations. Configuration 1, known as Pairwise Encounters, was subdivided into two parts: 1a, involving a low-speed UAS ownship and intruder(s), and 1b, involving a high-speed surrogate ownship and intruder. Configuration 2, known as Full Mission, involved a surrogate ownship, live intruder(s), and integrated virtual traffic. Table 1 is a summary of flights for each configuration, with data collection flights highlighted in green. Section 2 and 3 of this report give an in-depth description of the flight test period, aircraft involved, flight crew, and mission team. Overall, Flight Test 3 gathered excellent data for each SUT. We attribute this successful outcome in large part from the experience that was acquired from the ACAS Xu SS flight test flown in December 2014. Configuration 1 was a tremendous success, thanks to the training, member participation, integration/testing, and in-depth analysis of the flight points. Although Configuration 2 flights were cancelled after 3 data collection flights due to various problems, the lessons learned from this will help the UAS in the NAS project move forward successfully in future flight phases.

Unmanned↗

NASA NTRS2017

Behavioral Adaptations of Female Mice on the International Space Station

Adult female mice were sent to the International Space Station (ISS) as part of an early life science mission utilizing NASA's Rodent Habitat. Its primary purpose was to provide further insight into the influence of a microgravity environment on various aspects of mammalian physiology and well-being as part of an ongoing program of research aimed ultimately at understanding and ameliorating the deleterious influences of space on the human body. The present study took advantage of video collected from fixed, in-flight cameras within the habitat itself, to assess behavioral adaptations observed among in-flight mice aboard the ISS and differences in behavior with respect to a control group on the ground. Data collection consisted of several behavioral measures recorded by a trained observer with the assistance of interactive behavior analysis software. Specific behavioral measures included frequencies of conspecific interactionsociability, time spent feeding and conducting hygienic behavior, and relative durations of thigmotactic behavior, which is commonly used as an index of anxiety. Data were used to test tentative hypotheses that such behaviors differ significantly across mice under microgravity versus 1g conditions, and the assumption that the novel experience of microgravity itself may represent an initially anxiogenic stimulus which an animal will eventually acclimate to, perhaps through habituation.

Behavior in space↗

NASA NTRS2017

Behavioral Adaptations of Female Mice on the International Space Station

rodent behavior↗

NASA NTRSDate not supplied

Development of an Inertial Sensor-based Methodology for Spacesuited Geology Task Assessments during Simulated Lunar Extravehicular Activities

Lunar surface exploration during Artemis missions will require the specific skill set of geology sampling. Apollo astronauts had extensive training and used specialized tools to collect lunar rocks, core samples, pebbles, sand, and dust. The inflexibility of the pressurized Apollo spacesuits forced sampling to be taken at a standstill posture. However, new exploration spacesuits are expected to incorporate advanced materials and joint bearings, allowing for greater mobility and a wider range of functional postures. Thus, science and exploration during Artemis missions will likely involve a variety of standing, squatting, and kneeling postures. In preparation for future lunar exploration missions, NASA provides geologic training to astronauts and other mission personnel. This professional training with a spacesuit in simulated lunar environments will enhance performance and reduce risk of injury to astronauts on the lunar surface. However, anecdotally, untrained or newly trained people wearing prototype planetary spacesuits have been observed to performing motions differently than a trained geologist would when conducting the same geology sampling tasks. Therefore, a tool for evaluating geology postures at extravehicular activity (EVA) training facilities becomes required. In this paper, we introduce a novel inertial measurement unit (IMU)-based method of geology task assessments in spacesuited conditions during simulated lunar EVAs. As a case study, two subjects (one geologist and one non-geologist) participated and donned the Mark III prototype planetary spacesuit during offloading with the spreader bar gimbal in NASA’s Active Response Gravity Offload System (ARGOS). For automated geology task assessments, the spacesuit was instrumented with three wireless IMUs (APDM Opal, OR, USA): one on the chest and one each on the left and right ankle bearings. Then subjects performed geology tasks using various tools (rake, trench, hammer chisel, scoop, and drive tube) for 45 minutes each. The chest IMU measured the torso tilt angle in the sagittal plane. We used an ensemble learning method with the ankle IMUs to discriminate between standing and kneeling activities. IMU data were processed using custom MATLAB (Mathworks, MA, USA) software. In our case study, the developed method was able to discriminate differences in standing and kneeling activity levels between subjects who were all highly experienced with spacesuited testing. Our preliminary data showed one subject maintained the constant and lower range of the upper body tilt angle while both standing and kneeling, while the other subject showed more variation of the upper body tilt angle and preferred bending the upper body rather than changing from standing to kneeling posture and vice versa. While geology experience may be a factor, these results need further investigation as suit sizing and ARGOS offloading configurations have been proven to have a significant influence on suited ARGOS tasks. Also, more subjects will be needed to complete these tasks for validation. IMU-based geology task assessments can provide useful information for geology training programs. Additionally, our IMU-based posture analysis can provide new insights into how to evaluate spacesuited geology task characteristics of astronauts during simulated lunar EVAs.

Kyoung Jae Kim↗

NASA NTRS2011

Fast Image Texture Classification Using Decision Trees

Texture analysis would permit improved autonomous, onboard science data interpretation for adaptive navigation, sampling, and downlink decisions. These analyses would assist with terrain analysis and instrument placement in both macroscopic and microscopic image data products. Unfortunately, most state-of-the-art texture analysis demands computationally expensive convolutions of filters involving many floating-point operations. This makes them infeasible for radiation- hardened computers and spaceflight hardware. A new method approximates traditional texture classification of each image pixel with a fast decision-tree classifier. The classifier uses image features derived from simple filtering operations involving integer arithmetic. The texture analysis method is therefore amenable to implementation on FPGA (field-programmable gate array) hardware. Image features based on the "integral image" transform produce descriptive and efficient texture descriptors. Training the decision tree on a set of training data yields a classification scheme that produces reasonable approximations of optimal "texton" analysis at a fraction of the computational cost. A decision-tree learning algorithm employing the traditional k-means criterion of inter-cluster variance is used to learn tree structure from training data. The result is an efficient and accurate summary of surface morphology in images. This work is an evolutionary advance that unites several previous algorithms (k-means clustering, integral images, decision trees) and applies them to a new problem domain (morphology analysis for autonomous science during remote exploration). Advantages include order-of-magnitude improvements in runtime, feasibility for FPGA hardware, and significant improvements in texture classification accuracy.

Thompson, David R.↗

DOE OSTI2022

Analysis of human performance differences between students and operators when using the Rancor Microworld simulator

Here, from within the umbrella of the Simplified Human Error Experimental Program (SHEEP) framework, this paper analyzes human performance differences between professional and student operators when using a simplified simulator (i.e., Rancor Microworld). This paper represents a crucial step in understanding the fidelity of the simplified simulators and student operators within the SHEEP study. This paper explores a randomized factorial experimental design that features two independent variables: participant type and event class. Six human performance measurements are considered in the experiment. The experiment is conducted using 20 professional reactor operators employed at actual nuclear power plants (NPPs), along with 20 trained students. The experimental data are analyzed via statistical analysis methods. Finally, this paper examines the differences in human performance between actual operators and students when using Rancor Microworld.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2023

Interpretable boosted-decision-tree analysis for the Majorana Demonstrator

The Majorana Demonstrator is a leading experiment searching for neutrinoless double-beta decay with high purity germanium detectors (HPGe). Machine learning provides a new way to maximize the amount of information provided by these detectors, but the data-driven nature makes it less interpretable compared to traditional analysis. An interpretability study reveals the machine's decision-making logic, allowing us to learn from the machine to feedback to the traditional analysis. In this work, we have presented the first machine learning analysis of the data from the Majorana Demonstrator; this is also the first interpretable machine learning analysis of any germanium detector experiment. Two gradient boosted decision tree models are trained to learn from the data, and a game-theory-based model interpretability study is conducted to understand the origin of the classification power. By learning from data, this analysis recognizes the correlations among reconstruction parameters to further enhance the background rejection performance. By learning from the machine, this analysis reveals the importance of new background categories to reciprocally benefit the standard Majorana analysis. This model is highly compatible with next-generation germanium detector experiments like LEGEND since it can be simultaneously trained on a large number of detectors.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2020

On the Road to Achieving Safety Excellence at the Idaho Cleanup Project: The Initial Challenges, Solutions, and Positive Results of Striving to be the Best and Safest in the DOE Complex - 20518

When Fluor Idaho, LLC (Fluor Idaho) assumed the Idaho Cleanup Project Core contract (ICP Core) in mid-2016, it inherited the successful multi-faceted, employee-owned safety programs of two prior contractors supporting cleanup operations at the Idaho National Laboratory Site. Fluor Idaho incorporated the best of both safety programs along with Fluor Corporate's 106-year legacy of safety initiatives and successes. Fluor Idaho integrated the separate programs and cultures under a One Fluor approach - one project, one mission, One Fluor. Job changes, management changes, and procedures, processes, and system changes presented multiple challenges. Those challenges combined with the severe Winter of 2016/2017 resulted in numerous slips, trips, and falls first aids and reportable injuries. The road to safety excellence started off bumpy and stayed that way for some time. Since then, Fluor Idaho has successfully implemented several key initiatives including employee engagement; management engagement and time in the field; communications improvements; safety culture sustainment; injury/illness prevention; organizational learning; and DOE Voluntary Protection Program (VPP) readiness and preparation. Management became more engaged by making more workplace visits and increasing their supervision of 'life critical work' or complex activities. Additional training became mandatory for first-line supervisors and front-line workers. Most recently, Fluor Idaho increased its injury and illness prevention initiatives extensively. In the last year, things have turned around in a dramatic way. This shift is directly attributable to the significant number of initiatives launched in late 2017 and 2018. These initiatives addressed safety performance in the five key areas of the DOE VPP and promoted sustainability of the company's safety culture. The five key areas are: management leadership, employee involvement, work-site analysis, hazard prevention and control, and safety and health training. Employees and managers alike recommitted to making safety the highest value. The result of this laser-like focus on safety and accident and injury prevention is that injuries and first aids are a fraction of what they were. During FY 2019, employees worked more than 4.2 million hours without a serious injury or lost work-day and 1 million hours without a recordable injury. Fluor Idaho is proud of its safety trend. The company invested significant resources in improving its safety record, and that investment is paying off. Other contractors in the DOE Complex can learn and benefit from this discussion of the challenges Fluor Idaho faced, as well as its innovative safety initiatives that are producing results that are being recognized by our U.S. Department of Energy (DOE) customer. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2025

Application-specific machine-learned interatomic potentials: exploring the trade-off between DFT convergence, MLIP expressivity, and computational cost

Machine-learned interatomic potentials (MLIPs) are revolutionizing computational materials science and chemistry by offering an efficient alternative to ab initio molecular dynamics (MD) simulations. However, fitting high-quality MLIPs remains a challenging, time-consuming, and computationally intensive task where numerous trade-offs have to be considered, e.g., How much and what kind of atomic configurations should be included in the training set? Which level of ab initio convergence should be used to generate the training set? Which loss function should be used for fitting the MLIP? Which machine learning architecture should be used to train the MLIP? The answers to these questions significantly impact both the computational cost of MLIP training and the accuracy and computational cost of subsequent MLIP MD simulations. In this study, we use a configurationally diverse beryllium dataset and quadratic spectral neighbor analysis potential. We demonstrate that joint optimization of energy versus force weights, training set selection strategies, and convergence settings of the ab initio reference simulations, as well as model complexity can lead to a significant reduction in the overall computational cost associated with training and evaluating MLIPs. This opens the door to computationally efficient generation of high-quality MLIPs for a range of applications which demand different accuracy versus training and evaluation cost trade-offs.

36 MATERIALS SCIENCE↗

NASA NTRS1972

Automatic classification of soils and vegetation with ERTS-1 data

Preliminary results of a test of a computerized analysis method using ERTS 1 data are presented. The method consisted of a four-spectral-band supervised, maximum likelihood, Gaussian classifier with training statistics derived through a combination of clustering and manual methods. The multivariate analysis method leads to the assignment of each resolution element of the data to one of a preselected set of discrete classes. The data frame was an area over the Texas-Oklahoma border including Lake Texoma. The study suggests that multispectral scanner data coupled with machine processing shows promise for earth surface cover surveys. Futhermore, the processing time is short and consequently the costs are low; a full frame can be analyzed completely within 48 hours.

Landgrebe, D. A.↗

DOE OSTI2022

Identifying Critical Infrastructure in Imagery Data Using Explainable Convolutional Neural Networks

To date, no method utilizing satellite imagery exists for detailing the locations and functions of critical infrastructure across the United States, making response to natural disasters and other events challenging due to complex infrastructural interdependencies. This paper presents a repeatable, transferable, and explainable method for critical infrastructure analysis and implementation of a robust model for critical infrastructure detection in satellite imagery. This model consists of a DenseNet-161 convolutional neural network, pretrained with the ImageNet database. The model was provided additional training with a custom dataset, containing nine infrastructure classes. The resultant analysis achieved an overall accuracy of 90%, with the highest accuracy for airports (97%), hydroelectric dams (96%), solar farms (94%), substations (91%), potable water tanks (93%), and hospitals (93%). Critical infrastructure types with relatively low accuracy are likely influenced by data commonality between similar infrastructure components for petroleum terminals (86%), water treatment plants (78%), and natural gas generation (78%). Local interpretable model-agnostic explanations (LIME) was integrated into the overall modeling pipeline to establish trust for users in critical infrastructure applications. The results demonstrate the effectiveness of a convolutional neural network approach for critical infrastructure identification, with higher than 90% accuracy in identifying six of the critical infrastructure facility types.

97 MATHEMATICS AND COMPUTING↗