Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Operator learning”

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

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

At least 703 records · Page 39
All sourcesNASA NTRSDOE OSTI
NASA NTRS2020

Probabilistic Modeling of Heavy Machinery Using Machine Learning

NASA Glenn Research Center’s facility operations seeks to leverage its extensive instrumentation and historical data with machine learning to increase system efficiencies. The ultimate goal of this effort is to probabilistically model the behavior of Glenn’s central air service compressors for optimal decision making and planning. This project is a first step in that direction. We propose a multimodel approach that uses high-dimensional models to ask simple questions about complex dependent structures, and low-dimensional models to ask complex questions about simple dependent structures. While the low-dimensional models make strong assumptions, they can be visualized and they can be insightful. We show good fits for univariate models of compressor sensors, and preliminary work on high-dimensional multivariate models.

Machine learning↗
DOE OSTI2024

Development of algorithms for augmenting and replacing conventional process control using reinforcement learning

Here, this work seeks to allow for the online operation and training of model-free reinforcement learning (RL) agents but limit the risk to system equipment and personnel. The parallel implementation of RL alongside more conventional process control (CPC) allows for the RL algorithm to learn from CPC. The past performance of both methods are assessed on a continuous basis allowing for a transition from CPC to RL and, if needed, transitioning back to CPC from RL. This allows for the RL algorithm to slowly and safely assume control of the process without significant degradation in control performance. It is shown that the RL can derive a near optimal policy even when coupled with a suboptimal CPC. It is also demonstrated that the coupled RL-CPC algorithm learns at a faster rate than traditional RL methods of exploration while the algorithm’s performance does not deteriorate below CPC, even when exposed to an unknown operating condition.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2019

Automated control of microfluidic devices based on machine learning

A system is provided to automatically monitor and control the operation of a microfluidic device using machine learning technology. The system receives images of a channel of a microfluidic device collected by a camera during operation of the microfluidic device. Upon receiving an image, the system applies a classifier to the image to classify the operation of the microfluidic device as normal, in which no adjustment to the operation is needed, or as abnormal, in which an adjustment to the operation is needed. When an image is classified as normal, the system may make no adjustment to the microfluidic device. If, however, an image is classified as abnormal, the system may output an indication that the operation is abnormal, output an indication of a needed adjustment, or control the microfluidic device to make the needed adjustment.

42 ENGINEERING↗
DOE OSTI2021

Automated control of microfluidic devices based on machine learning

A system is provided to automatically monitor and control the operation of a microfluidic device using machine learning technology. The system receives images of a channel of a microfluidic device collected by a camera during operation of the microfluidic device. Upon receiving an image, the system applies a classifier to the image to classify the operation of the microfluidic device as normal, in which no adjustment to the operation is needed, or as abnormal, in which an adjustment to the operation is needed. When an image is classified as normal, the system may make no adjustment to the microfluidic device. If, however, an image is classified as abnormal, the system may output an indication that the operation is abnormal, output an indication of a needed adjustment, or control the microfluidic device to make the needed adjustment.

Giera, Brian↗
NASA NTRSDate not supplied

Towards Autonomous Lunar Resource Excavation via Reinforcement Learning

To continue on a sustainable and flexible path, NASA needs to address the challenge of collecting and moving large amounts of regolith at the destination. NASA’s Regolith Advanced Surface Systems Operations Robot (RASSOR) is principally designed to mine and deliver regolith for In-Situ Resource Utilization (ISRU) processing. RASSOR’s design enables it to efficiently collect and deposit regolith, return collected material for processing, and myriad related ISRU activities. To reliably perform these operations on the lunar surface, RASSOR software and sensory systems need to be robust and maximize the information extracted from a reduced sensor payload. Herein, we present preliminary findings from the Intelligent Capabilities Enhanced RASSOR project. We created reduced-order simulation environments to develop autonomous trenching controllers via reinforcement learning and prototype state estimation architectures. The goal of reinforcement learning is for an agent to learn a policy (task strategy) through interactions with an environment. When the agent performs an action, a change occurs in environment state and a numerical reward is received which informs the agent whether the action performed was good or not. Since reinforcement learning algorithms learn through trial-and-error, a simulation is a desirable first environment for development and learning. We developed two simulations, the first is a 2D excavation simulation developed to facilitate parameter selection, and a 3D simulation developed using a game physics engine, to simulate simplified soil interactions and increase the fidelity of the dynamic models of the robotic agents. The development of this 3D simulation has enabled the training of additional sensing capabilities and research both at the granular mechanics and operations levels. We experimented with various virtual sensor payloads to identify a combination that enabled efficient excavation operation and learning. Our reward function is based on how much material is excavated per step. A penalty is also received for leaving the dig site and to smooth the acceleration of the drum arms. We implemented pseudo time-of-flight sensors to report distance from each drum to ground and the height above ground which was found to be more efficient than existing solutions. Our findings suggest that reinforcement learning for autonomous operations has learned viable trenching strategies within 3000 training episodes in our simplified 2D environment and helped identify desirable sensing capabilities, arrangements, and considerations such as the positioning of time-of-flight sensors. Future work includes expanding our simulation to more complex environments and scenarios, and transfer learning from simulation to RASSOR 2.0 hardware for deployment in the Regolith Test Bin at NASA's Kennedy Space Center.

rassor↗
DOE OSTI2024

SRF Cavity Instability Detection with Machine Learning at CEBAF

During the operation of the Continuous Electron Beam Accelerator Facility (CEBAF), one or more unstable superconducting radio-frequency (SRF) cavities often cause beam loss trips while the unstable cavities themselves do not necessarily trip off. The present RF controls for the legacy cavities report at only 1 Hz, which is too slow to detect fast transient instabilities during these trip events. These challenges make the identification of an unstable cavity out of the hundreds installed at CEBAF a difficult and time-consuming task. To tackle these issues, a fast data acquisition system (DAQ) for the legacy SRF cavities has been developed, which records the sample at 5 kHz. A Principal Component Analysis (PCA) approach is being developed to identify anomalous SRF cavity behavior. We will discuss the present status of the DAQ system and PCA model, along with initial performance metrics. Overall, our method offers a practical solution for identifying unstable SRF cavities, contributing to increased beam availability and machine reliability.

Carpenter, A.↗
DOE OSTI2024

SRF Cavity Instability Detection with Machine Learning at CEBAF

During the operation of the Continuous Electron Beam Accelerator Facility (CEBAF), one or more unstable superconducting radio-frequency (SRF) cavities often cause beam loss trips while the unstable cavities themselves do not necessarily trip off. The present RF controls for the legacy cavities report at only 1 Hz, which is too slow to detect fast transient instabilities during these trip events. These challenges make the identification of an unstable cavity out of the hundreds installed at CEBAF a difficult and time-consuming task. To tackle these issues, a fast data acquisition system (DAQ) for the legacy SRF cavities has been developed, which records the sample at 5 kHz. A Principal Component Analysis (PCA) approach is being developed to identify anomalous SRF cavity behavior. We will discuss the present status of the DAQ system and PCA model, along with initial performance metrics. Overall, our method offers a practical solution for identifying unstable SRF cavities, contributing to increased beam availability and machine reliability.

Carpenter, A.↗
NASA NTRS1997

Goldstone-Apple Valley Radio Telescope System Theory of Operation

The purpose of this learning module is to enable learners to describe how the Goldstone-Apple Valley Radio Telescope (GAVRT) system functions in support of Apple Valley Science and Technology Center's (AVSTC) client schools' radio astronomy activities.

Radio Telescope↗
NASA NTRS2011

Imbalanced Learning for Functional State Assessment

This paper presents results of several imbalanced learning techniques applied to operator functional state assessment where the data is highly imbalanced, i.e., some function states (majority classes) have much more training samples than other states (minority classes). Conventional machine learning techniques usually tend to classify all data samples into majority classes and perform poorly for minority classes. In this study, we implemented five imbalanced learning techniques, including random undersampling, random over-sampling, synthetic minority over-sampling technique (SMOTE), borderline-SMOTE and adaptive synthetic sampling (ADASYN) to solve this problem. Experimental results on a benchmark driving lest dataset show thai accuracies for minority classes could be improved dramatically with a cost of slight performance degradations for majority classes,

Li, Feng↗
NASA NTRS2020

Terra MODIS: 20 years of on-orbit calibration and performance

Since its launch in December 1999, Terra Moderate Resolution Imaging Spectroradiometer (MODIS) has successfully operated for more than 20 years, with its observations generating a broad range of science data products that have greatly enabled the remote sensing community and users worldwide in their studies of many key geophysical parameters of the Earth’s systems. MODIS collects data in 36 spectral bands, covering wavelengths from 0.41 to 14.4 μm, which are calibrated by a set of onboard calibrators (OBCs). Also contributing to the sensor’s mission-long on-orbit calibration and characterization are near-monthly scheduled lunar observations and multiple time series of the sensor’s responses over select ground targets at a variety of scan angles. To a large extent, the quality of MODIS data products relies strongly on the dedicated efforts to operate and calibrate the instrument, to derive and update calibration parameters, and to develop and implement new calibration strategies and algorithms in response to on-orbit changes of the sensor’s characteristics and its OBC functions. We provide an overview of the Terra MODIS on-orbit operation and calibration activities over the last 20 years, including changes made to extend and preserve the instrument and OBC functions and their operation strategies. It also illustrates the sensor’s on-orbit performance with results derived from its OBC, lunar observations, and select ground targets and discusses major changes in sensor characteristics and corrections applied to the L1B algorithms as well as calibration lookup table updates. To date, the Terra MODIS instrument and its OBCs continue to operate and function normally. Except for those identified prelaunch, most spectral bands and detectors continue to meet their specified calibration requirements. Also discussed in our paper are lessons learned from Terra MODIS operation and calibration, as well as future efforts to further extend and maintain the quality of its long-term data records.

Terra↗
NASA NTRS1996

Surface and Atmospheric Parameter Retrieval From AVIRIS Data: The Importance of Non-Linear Effects

AVIRIS data represent a new and important approach for the retrieval of atmospheric and surface parameters from optical remote sensing data. Not only as a test for future space systems, but also as an operational airborne remote sensing system, the development of algorithms to retrieve information from AVIRIS data is an important step to these new approaches and capabilities. Many things have been learned since AVIRIS became operational, and the successive technical improvements in the hardware and the more sophisticated calibration techniques employed have increased the quality of the data to the point of almost meeting optimum user requirements. However, the potential capabilities of imaging spectrometry over the standard multispectral techniques have still not been fully demonstrated. Reasons for this are the technical difficulties in handling the data, the critical aspect of calibration for advanced retrieval methods, and the lack of proper models with which to invert the measured AVIRIS radiances in all the spectral channels. To achieve the potential of imaging spectrometry, these issues must be addressed. In this paper, an algorithm to retrieve information about both atmospheric and surface parameters from AVIRIS data, by using model inversion techniques, is described. Emphasis is put on the derivation of the model itself as well as proper inversion techniques, robust to noise in the data and an inadequate ability of the model to describe natural variability in the data. The problem of non-linear effects is addressed, as it has been demonstrated to be a major source of error in the numerical values retrieved by more simple, linear-based approaches. Non-linear effects are especially critical for the retrieval of surface parameters where both scattering and absorption effects are coupled, as well as in the cases of significant multiple-scattering contributions. However, sophisticated modeling approaches can handle such non-linear effects, which are especially important over vegetated surfaces. All the data used in this study were acquired during the 1991 Multisensor Airborne Campaign (MAC-Europe), as part of the European Field Experiment on a Desertification-threatened Area (EFEDA), carried out in Spain in June-July 1991.

Green Robert O.↗
NASA NTRS2014

Behavioral Health and Performance at NASA JSC: Recent Successes and Future Plan for BHP Research and Operations

The Behavioral Health and Performance discipline at NASA Johnson Space Center is organized into two distinct Divisions (Biomedical Research and Environmental Science Division and Space and Clinical Operations Division) but is integrated and interrelated in its day‐to‐day work. Ongoing operations supporting NASA's spaceflight goals benefit from the research portfolios that address risks to mission success. Similarly, these research portfolios are informed by operations to ensure investigations stay relevant given the dynamic environment of spaceflight. There are many success stories that can be presented where initial work begun as a BHP Research project, and funded through the Human Research Program, was fully implemented in operations or addressed an operational need. Examples include improving effectiveness of the debriefings used within Mission Control by the Mission Operations Directorate and countermeasures for fatigue management. There is also ongoing collaboration with research and operations for developing selection methods for future generation astronauts, and to enhance and inform the current family support function. The objective of this panel is to provide examples of recent success stories, describe areas where close collaboration is benefitting ongoing research and operations, and summarize how this will come together as NASA plans for the one year ISS mission - a unique opportunity for both BHP operations and research to learn more about preparing and supporting crewmembers for extended missions in space. The proposed panel will be comprised of six presentations, each describing a unique aspect of research or operations and the benefits to current and future spaceflight.

Leveton, L. B.↗
NASA NTRS1995

Advanced Transportation System Studies Technical Area 2 (TA-2) Heavy Lift Launch Vehicle Development Contract: Technical Results - Volume 2

The sections in this report include: Single Stage to Orbit (SSTO) Design Ground-rules; Operations Issues and Lessons Learned; Vertical-Takeoff/Landing Versus Vertical-Takeoff/Horizontal-Landing; SSTO Design Results; SSTO Simulation Results; SSTO Assessment Results; SSTO Sizing Tool User's Guide; SSto Turnaround Assessment Report; Ground Operations Assessment First Year Executive Summary; Health Management System Definition Study; Major TA-2 Presentations; First Lunar Outpost Heavy Lift Launch Vehicle Design and Assessment; and the section, Russian Propulsion Technology Assessment Reports.

Source record↗
DOE OSTI2021

Hard Rock Drilling Optimization Software

The main objective of the developed software is to reduce the cost per foot during drilling, in other words, optimize the drilling operational parameters in achieving optimum ROP while avoiding critical operational parameters due to either low ROP, drillstring vibration, accelerated cutter wear, or low MSE. The developed software can also be used for post-well analysis to provide insight and lessons learned for future drilling operations. Several functions are available in the software to help the user perform drilling analysis, optimization, and simulation.

15 GEOTHERMAL ENERGY↗
NASA NTRS1995

Space Mechanisms Lessons Learned Study. Volume 1: Summary

Hundreds of satellites have been launched to date. Some have operated extremely well and others have not. In order to learn from past operating experiences, a study was conducted to determine the conditions under which space mechanisms (mechanically moving components) have previously worked or failed. The study consisted of: (1) an extensive literature review that included both government contractor reports and technical journals; (2) communication and visits (when necessary) to the various NASA and DOD centers and their designated contractors (this included contact with project managers of current and prior NASA satellite programs as well as their industry counterparts); (3) requests for unpublished information to NASA and industry; and (4) a mail survey designed to acquire specific mechanism experience. The information obtained has been organized into two volumes. Volume 1 provides a summary of the lessons learned, the results of a needs analysis, responses to the mail survey, a listing of experts, a description of some available facilities and a compilation of references. Volume 2 contains a compilation of the literature review synopsis.

Shapiro, Wilbur↗
NASA NTRS2006

Human Factors Lessons Learned on the International Space Station

Experience on International Space Station (ISS) provides many important lessons for future space flight. NASA human factors engineers have been systematically collecting lessons learned from crew debriefs, as well as working with ground support teams to continuously improve crew operations. This paper describes the methods for collecting data from debriefs, lessons learned through that process, and an example of a technology development task funded through the Space Human Factors Engineering (SHFE) program element in response to an identified operational need. Each ISS increment crew spends many hours after the flight answering questions from the various subsystem leads. The Flight Crew Integration subsystem lead asks questions specific to human factors and habitability issues. In addition, crew comments on many other subsystems provide insight into interface designs, operability and maintainability. The debrief comments are unique to each crew, and must be categorized to provide operational lessons learned. Personal identifiers are removed and comments aggregated to separate consistent issues from personal preferences. Examples will be given, and the procedure for incorporating the lessons into requirements and guidelines for the next human space vehicle will be described. In flight, very few astronauts are medical doctors. Written medical procedures during flight need to be easy to follow and quick to understand. The problem was analyzed as part of a SHFE task. Organization was analyzed and reorganizations were created and tested. Results will be reported. The ISS is a very important analog for planning future long-term missions. Collection of data from debriefs, studying the lessons learned and focusing on requirements for future missions are examples of the accomplishments through the SHFE program.

Woolford, Barbara↗