Results for “successive training” — Page 2

NASA NTRSDate not supplied

MULTI-OBJECTIVE REINFORCEMENT LEARNING FOR LOW-THRUST TRANSFER DESIGN BETWEEN LIBRATION POINT ORBITS

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to construct low-thrust transfers between periodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a traditional optimization formulation.

algorithm↗

NASA NTRSDate not supplied

MULTI-OBJECTIVE REINFORCEMENT LEARNING FOR LOW-THRUST TRANSFER DESIGN BETWEEN LIBRATION POINT ORBITS

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to construct low-thrust transfers between periodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a gradient descent optimization scheme to validate the performance of MRPPO with the moving reference modification.

Christopher J. Sullivan↗

NASA NTRSDate not supplied

MULTI-OBJECTIVE REINFORCEMENT LEARNING FOR LOW-THRUST TRANSFER DESIGN BETWEEN LIBRATION POINT ORBITS

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective reinforcement learning algorithm used to construct low-thrust transfers between periodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a gradient descent optimization scheme to validate the performance of MRPPO with the moving reference modification

Christopher John Sullivan↗

NASA NTRS2021

Multi-objective Reinforcement Learning for Low-thrust Transfer Design Between Libration Point Orbits

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective rein- forcement learning algorithm used to construct low-thrust transfers between pe- riodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a gradient descent optimization scheme to validate the performance of MRPPO with the moving reference modification.

Anderson, Rodney L.↗

NASA NTRS2021

Multi-objective Reinforcement Learning for Low-thrust Transfer Design Between Libration Point Orbits

Multi-Reward Proximal Policy Optimization (MRPPO) is a multi-objective rein- forcement learning algorithm used to construct low-thrust transfers between pe- riodic orbits in multi-body systems. Previous implementations of MRPPO have relied on a predefined reference transfer to successfully train each policy. In this paper, an algorithmic modification labeled the ‘moving reference’, is introduced to autonomously construct these reference trajectories during training. With this modification, MRPPO is used to recover various low-thrust transfers between two periodic orbits in the Earth-Moon circular restricted three-body problem to solve a multi-objective optimization problem. These results are then compared with the solutions recovered via a gradient descent optimization scheme to validate the performance of MRPPO with the moving reference modification.

Anderson, Rodney L↗

NASA NTRSDate not supplied

The Contribution of Pilots to Resilience in Normal Operations: A Survey Approach

Much of our knowledge about human performance in flight safety has come from the analysis of undesired events, whether accidents, incidents, or crew behaviors identified via flight exceedance monitoring or observational techniques. In recent years, there has been an acknowledgement that operational personnel are not merely sources of “human error”, but also make a unique human contribution to safe outcomes. In a few celebrated cases, this takes the form of “heroic saves”, but on many more occasions, operational personnel contribute to safety through everyday, often-unnoticed actions that turn potentially hazardous situations into non-events. An emerging approach to safety, frequently referred to as “Safety II,” proposes that the positive human contribution is an important and largely untapped source of safety information. Some airlines have successfully trained observers to identify and record the positive behaviors exhibited by the crew over the course of a flight. In other cases, flight crew are interviewed about good practices or positive behaviors. However, each of these methods are relatively limited in scale and resource intensive. A survey could provide a relatively low-cost approach to systematically gather this information on a larger scale. The primary purpose of the research was to develop and assess a surveys methodology for assessing crews' activities in normal flights and the operational perturbations encountered during normal operations. We hope that such a survey could be both a research tool as well as a safety management aid for the aviation industry. We collected responses concerning revenue flights from two groups of airline pilots (N = 25 & N= 65). The results indicated that relatively few flights proceeded exactly as in the original flight plan. Pilots routinely anticipated and adapted to changing circumstances. We will review the challenges encountered in developing the survey and summarize preliminary findings from two administrations of the survey to airline pilots.

human contribution safety↗

NASA NTRSDate not supplied

The Contribution of Pilots to Resilience in Normal Operations. Part II: A closer look at briefings: Anticipation and Monitoring Also Known as Planning and Coordination

Much of our knowledge about human performance in flight safety has come from the analysis of undesired events, whether accidents, incidents, or crew behaviors identified via flight exceedance monitoring or observational techniques. In recent years, there has been an acknowledgement that operational personnel are not merely sources of “human error”, but also make a unique human contribution to safe outcomes. In a few celebrated cases, this takes the form of “heroic saves”, but on many more occasions, operational personnel contribute to safety through everyday, often-unnoticed actions that turn potentially hazardous situations into non-events. An emerging approach to safety, frequently referred to as “Safety II,” proposes that the positive human contribution is an important and largely untapped source of safety information. Some airlines have successfully trained observers to identify and record the positive behaviors exhibited by the crew over the course of a flight. In other cases, flight crew are interviewed about good practices or positive behaviors. However, each of these methods are relatively limited in scale and resource intensive. A survey could provide a relatively low-cost approach to systematically gather this information on a larger scale. The primary purpose of the research was to develop and assess a surveys methodology for assessing crews' activities in normal flights and the operational perturbations encountered during normal operations. We hope that such a survey could be both a research tool as well as a safety management aid for the aviation industry. We collected responses concerning revenue flights from two groups of airline pilots (N = 25 & N= 65). The results indicated that relatively few flights proceeded exactly as in the original flight plan. Pilots routinely anticipated and adapted to changing circumstances. We will review the challenges encountered in developing the survey and summarize preliminary findings from two administrations of the survey to airline pilots.

human contribution safety↗

NASA NTRS1994

Autogenic-feedback training as a treatment for airsickness in high-performance military aircraft: Two case studies

The purpose of this paper is to present a detailed description of the physiological and performance responses of two military pilots undergoing a treatment for motion sickness. The treatment used, Autogenic-Feedback Training (AFT), is an operant conditioning procedure where subjects are taught to control several of their autonomic responses and thereby suppress their motion sickness symptoms. Two male, active duty military pilots (U.S. Navy and U. S. Marine Corps), ages 30 and 35, were each given twelve 30-minute training sessions. The primary criterion for success of training was the subject's ability to tolerate rotating chair motion sickness tests for progressively longer periods of time and at higher rotational velocities. A standardized diagnostic scale was used during motion sickness to assess changes in the subject's perceived malaise. Physiological data were obtained from one pilot during tactical maneuvers in an F-18 aircraft after completion of his training. A significant increase in tolerance to laboratory-induced motion sickness tests and a reduction in autonomic nervous system (ANS) response variability was observed for both subjects after training. Both pilots were successful in applying AFT for controlling their airsickness during subsequent qualification tests on F-18 and T-38 aircraft and were returned to active duty flight status.

Cowings, Patricia S.↗

NASA NTRS2020

CloudSat - From the A-Train to the C-Train

The CloudSat satellite was launched from Vandenberg Air Force Base, in California, on April 28, 2006. It was launched with the CALIPSO satellite, a joint mission between the National Aeronautics and Space Administration (NASA) and the Centre National D'Etudes Spatiales (CNES). CloudSat and CALIPSO were designed to fly in formation with each other, to combine their complimentary scientific measurements. The two missions were maneuvered into the Earth Science Afternoon Constellation (A-Train) shortly after launch and have flown in formation together since then. Originally designed to fly for 22 months, the CloudSat satellite has far exceeded expectations. Inflight anomalies on CloudSat necessitated changes to both the daily concept of operations and formation flying operations with CALIPSO. In 2011, a severe battery anomaly restricted CloudSat science data collection and maneuverability to the sunlit portions of the orbit. In 2017, CloudSat suffered a reaction wheel failure, triggering the criteria established to exit the A-Train. CloudSat successfully exited the A-Train in early 2018, suspending formation flying and joint science observations with CALIPSO. The joint CloudSat/CALIPSO science team continued collaboration with discussions focused on resuming joint observations. As both satellites were aging and CALIPSO was nearing the exhaustion of satellite propellant, the decision was made to lower CloudSat to the CALIPSO graveyard orbit, to be joined by CALIPSO as soon as feasible. With NASA’s approval, CloudSat was maneuvered to the lower orbit over the course of several months in mid-2018. Once CloudSat was in place, and with approval from CNES, CALIPSO began the delicate dance of lowering their orbit and re-aligning the satellite ground tracks. CloudSat and CALIPSO resumed formation flying in the newly dubbed C-Train (CloudSat and CALIPSO) orbit in the Fall of 2018. The ingenuity of the CloudSat and CALIPSO teams made the concept of the C-Train a reality. This paper will discuss the decision processes and the operational challenges the CloudSat flight team faced while designing and implementing the C-Train.

Witkowski, M.M.↗

NASA NTRS2009

21st Century Extravehicular Activities: Synergizing Past and Present Training Methods for Future Spacewalking Success

Neil Armstrong's understated words, "That's one small step for man, one giant leap for mankind." were spoken from Tranquility Base forty years ago. Even today, those words resonate in the ears of millions, including many who had yet to be born when man first landed on the surface of the moon. By their very nature, and in the the spirit of exploration, extravehicular activities (EVAs) have generated much excitement throughout the history of manned spaceflight. From Ed White's first space walk in June of 1965, to the first steps on the moon in 1969, to the expected completion of the International Space Station (ISS), the ability to exist, live and work in the vacuum of space has stood as a beacon of what is possible. It was NASA's first spacewalk that taught engineers on the ground the valuable lesson that successful spacewalking requires a unique set of learned skills. That lesson sparked extensive efforts to develop and define the training requirements necessary to ensure success. As focus shifted from orbital activities to lunar surface activities, the required skill-set and subsequently the training methods, changed. The requirements duly changed again when NASA left the moon for the last time in 1972 and have continued to evolve through the Skylab, Space Shuttle; and ISS eras. Yet because the visits to the moon were so long ago, NASA's expertise in the realm of extra-terrestrial EVAs has diminished. As manned spaceflight again shifts its focus beyond low earth orbit, EVA success will depend on the ability to synergize the knowledge gained over 40+ years of spacewalking to create a training method that allows a single crewmember to perform equally well, whether performing an EVA on the surface of the Moon, while in the vacuum of space, or heading for a rendezvous with Mars. This paper reviews NASA's past and present EVA training methods and extrapolates techniques from both to construct the basis for future EVA astronaut training.

Moore, Sandra K.↗

NASA NTRS1979

Salyut-6--Soyuz-29: Our commentary, the crew and the station

A newspaper article written by an assistant flight director which stresses the importance of cosmonaut training is presented. The preliminary stage of training is discussed, and the training of a crewmaker on a real spacecraft is briefly covered. The success in training foreign cosmonauts by Soviets is also discussed. The author feels that training is even more important today than it was seven or eight years ago because of the increased complexity of the experiments and the length of time a cosmonaut spends in space now.

Kravets, V.↗

DOE OSTI2025

Stacked networks improve physics-informed training: Applications to neural networks and deep operator networks

Physics-informed neural networks and operator networks have shown promise for effectively solving equations modeling physical systems. However, these networks can happen to be difficult or impossible to train accurately. Here, we present a novel multifidelity framework for stacking physics-informed neural networks and operator networks that facilitates training. We successively build a chain of networks, where the output at one step can act as a low-fidelity input for training a longer chain, gradually increasing the expressivity of the learnt model. The equations imposed at each step of the iterative process can be the same or different (akin to simulated annealing). The iterative (stacking) nature of the proposed method allows us to learn progressively features of a solution which could have been hard to learn directly. Through benchmark problems including a nonlinear pendulum, the wave equation, and the viscous Burgers equation, we show how stacking can be used to improve the accuracy and reduce the required size of physics-informed neural networks and operator networks.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1998

Neural Network-Based Sensor Validation for Turboshaft Engines

Sensor failure detection, isolation, and accommodation using a neural network approach is described. An auto-associative neural network is configured to perform dimensionality reduction on the sensor measurement vector and provide estimated sensor values. The sensor validation scheme is applied in a simulation of the T700 turboshaft engine in closed loop operation. Performance is evaluated based on the ability to detect faults correctly and maintain stable and responsive engine operation. The set of sensor outputs used for engine control forms the network input vector. Analytical redundancy is verified by training networks of successively smaller bottleneck layer sizes. Training data generation and strategy are discussed. The engine maintained stable behavior in the presence of sensor hard failures. With proper selection of fault determination thresholds, stability was maintained in the presence of sensor soft failures.

Moller, James C.↗

NASA NTRS2007

Exploring and Visualizing A-Train Instrument Data

The succession of US and international satellites that follow each other in close succession, known as the A-Train, affords an opportunity to atmospheric researchers that no single platform could provide: Increasing the number of observations at any given geographic location.. . a more complete "virtual science platform". However, vertically and horizontally, co-registering and regridding datasets from independently developed missions, Aqua, Calipso, Cloudsat, Parasol, and Aura, so that they can be inter-compared can be daunting to some, and may be repeated by many. Scientists will individually spend much of their time and resources acquiring A-Train datasets of interest residing at various locations, developing algorithms to match up and graph datasets along the A-Train track, and search through large amounts of data for areas and/or phenomena of interest. The aggregate amount of effort that can be expended on repeating pre-science tasks could climb into the tens of millions of dollars. The goal of the A-Train Data Depot (ATDD) is to enable free movement of remotely located A-Train data so that they are combined to create a consolidated vertical view of the Earth's Atmosphere along the A-Train tracks. The innovative approach of analyzing and visualizing atmospheric profiles along the platforms track (i.e., time) is accomplished by through the ATDDs Giovanni data analysis and visualization tool. Giovanni brings together data from Aqua (MODIS, AIRS, AMSR-E), Cloudsat (cloud profiling radar) and Calipso (CALIOP, IIR), as well as the Aura (OMI, MLS, HIRDLS, TES) to create a consolidated vertical view of the Earth's Atmosphere along the A-Train tracks. This easy to learn and use exploration tool will allow users to create vertical profiles of any desired A-Train dataset, for any given time of choice. This presentation shows the power of Giovanni by describing and illustrating how this tool facilitates and aids A-Train science and research. A web based display system Giovanni provides users with the capability of creating co-located profile images of temperature and humidity data from the MODIS, MLS and AIRS instruments for a user specified time and spatial area. In addition, Cloud and Aerosol profiles may also be displayed for the Cloudsat and Caliop instruments. The ability to modify horizontal and vertical axis range, data range and dynamic color range is also provided. Two dimensional strip plots of MODIS, AIRS, OM1 and POLDER parameters, co-located along the Cloudsat reference track, can also be plotted along with the Cloudsat cloud profiling data. Center swath pixels for the same parameters can also be shown as line plots overlaying the Cloudsat or Calipso profile images. Images and subsetted data produced in each analysis run may be downloaded. Users truly can explore and discover data specific to their needs prior to ever transferring data to their analysis tools.

Kempler, S.↗

NASA NTRS2006

A-Train Data Depot: Integrating and Visualizing Atmospheric Measurements Along the A-Train Tracks

The succession of US and international satellites that follow each other, seconds to minutes apart, across the local afternoon equator crossing is called the ATrain. The A-Train consists of the following satellites, in order of equator crossing: OCO, EOS Aqua, CloudSat, CALIPSO, PARASOL, and EOS Aura. Flying in such formation increases the number of observations, validates observations, and enables coordination between science observations, resulting in a more complete virtual science platform (Kelly, 2000) The goal of this project is to create the first ever A-Train virtual data portal/center, the A-Train Data Depot, to process, archive, access, visualize, analyze and correlate distributed atmosphere measurements from various A-Train instruments along A-Train tracks. The A-Train Data Depot (ATDD) will enable the free movement of remotely located A-Train data so that they are combined to create a consolidated vertical view of the Earth s Atmosphere along the A-Train tracks. Once the infrastructure of the ATDD is in place, it will be easily evolved to serve data from all A-Train data measurements: one stop shopping. The innovative approach of analyzing and visualizing atmospheric profiles along the platforms track (i.e., time) will be accommodated by reusing the GSFC Atmospheric Composition Data and Information Services Center (ACDISC) visualization and analysis tool, GIOVANNI, existing data reduction tools, on-line archwing for fast data access, and Cooperative Institute for Research in the Atmosphere (CRA) data co-registration tools. Initial measurements utilized include CALIPSO lidar backscatter, CloudSat radar reflectivity, clear air relative humidity, water vapor and temperature from AIRS, and cloud properties and aerosols from both MODIS. This will be followed by associated measurements from MLS, OMI, HIRDLS, and TES. Given the independent nature of instrument/platform development, the ATDD project has been met with many interesting challenges that, once resolved, will provide a much greater understanding of the relative flight dynamics and data co-registration of the suite of A-Train instruments, thus greatly increasing the accuracy of A-Train data analysis. Some of these challenges will be discussed. The project s resulting visualizations and analysis illustrate the importance of managing data so that measurements from various missions can be combined to enhance the understanding of the atmosphere. A-Train data management coordination, as performed here, is extremely significant in facilitating the A-Train science of clouds, precipitation, aerosol and chemistry.

Kempler, Steven↗

NASA NTRS2006

A-Train Data Depot: Integrating, Visualizing, and Extracting Cloudsat, CALIPSO, MODIS, and AIRS Atmospheric Measurements Along the A-Train Tracks

The succession of US and international Earth observing satellites that follow each other, seconds to minutes apart, across the local afternoon equator crossing is called the ATrain. The A-Train consists of the following satellites, in order of equator crossing: OCO, EOS Aqua, CloudSat, CALIPSO, PARASOL, and EOS Aura. Flying in such formation increases the number of observations, validates observations, and enables coordination between science observations, resulting in a more complete virtual science platform (Kelly, 2000). The goal of this project is to create the first ever A-Train virtual data portal/center, the A-Train Data Depot (ATDD), to process, archive, access, visualize, analyze and correlate distributed atmosphere measurements from various A-Train instruments along A-Train tracks. The ATDD will enable the free movement of remotely located A-Train data so that they are combined to create a consolidated vertical view of the Earth's Atmosphere along the A-Train tracks. Once the infrastructure of the ATDD is in place, it will be easily evolved to serve data from all A-Train data measurements: one stop shopping. The innovative approach of analyzing and visualizing atmospheric profiles along the platforms track (i.e., time) will be accommodated by reusing the GSFC Atmospheric Composition Data and Information Services Center (ACDISC) visualization and analysis tool, GIOVANNI, existing data reduction tools, on-line archiving for fast data access, access to remote data without unnecessary data transfers, and data retrieval by users finding data desirable for further study. Initial measurements utilized include CALIPSO lidar backscatter, CloudSat radar reflectivity, clear air relative humidity, water vapor and temperature from AIRS, and cloud properties and aerosols from both MODIS. This will be foilowed by associated measurements from TVILS, =MI, HIRDLS, sad TES. Given the independent nature of instrumentlplatform development, the ATDD project has been met with many interesting challenges that, once resolved, will provide a much greater understanding of the relative flight dynamics and data co-registration of the suite of A-Train instruments, thus greatly increasing the accuracy of A-Train data analysis. Some of these challenges will be illustrated and discussed. The project's early visualizations and analysis efforts illustrate the importance of managing data so that measurements from various missions can be combined to enhance the understanding of the atmosphere. A-Train data management coordination, as performed here, is extremely significant in facilitating the A-Train science of clouds, precipitation, aerosol and chemistry.

Kempler, Steven↗

NASA NTRS2001

Phase 1 research program overview

The Phase 1 research program was unprecedented in its scope and ambitious in its objectives. The National Aeronautics and Space Administration committed to conducting a multidisciplinary long-duration research program on a platform whose capabilities were not well known, not to mention belonging to another country. For the United States, it provided the first opportunity to conduct research in a long-duration space flight environment since the Skylab program in the 1970's. Multiple technical as well as cultural challenges were successfully overcome through the dedicated efforts of a relatively small cadre of individuals. The program developed processes to successfully plan, train for and execute research in a long-duration environment, with significant differences identified from short-duration space flight science operations. Between August 1994 and June 1998, thousands of kilograms of research hardware was prepared and launched to Mir, and thousands of kilograms of hardware and data products were returned to Earth. More than 150 Principal Investigators from eight countries were involved in the program in seven major research disciplines: Advanced Technology; Earth Sciences; Fundamental Biology; Human Life Sciences; International Space Station Risk Mitigation; Microgravity; and Space Sciences. Approximately 75 long-duration investigations were completed on Mir, with additional investigations performed on the Shuttle flights that docked with Mir. The flight phase included the participation of seven US astronauts and 20 Russian cosmonauts. The successful completion of the Phase 1 research program not only resulted in high quality science return but also in numerous lessons learned to make the ISS experience more productive. The cooperation developed during the program was instrumental in its success. c2001 AIAA. Published by Elsevier Science Ltd.

short duration↗

DOE OSTI2025

Remote quantification of Cm(III) and HNO 3 by fluorescence spectroscopy and chemometrics

A unique approach to remotely quantify Cm(III) (0–100 µg mL −1 ) in HNO 3 (1–12 M) using steady-state laser fluorescence spectroscopy and multivariate regression models was developed. Photoluminescence is amenable to remote measurements using fiber-optic cables and is sensitive to numerous lanthanide and actinide species. In-line measurements can provide feedback to support complex processing in harsh environments (e.g., hot cells) to help guide and optimize radiochemical separations. In this work, Cm(III) spectra were acquired remotely in a glove box as a function of HNO 3 concentration to better understand spectral characteristics and evaluate the utility of multivariate regression models in this system. Furthermore, the Cm(III) fluorescence peak shape, width, position, and intensity changed significantly as a function of HNO 3 concentration, likely because of the displacement of emission quenching inner-sphere water molecules and complexation with nitrate ions. Despite significant covariance and nonlinearity in the data, a D-optimal design strategy successfully minimized training set sample size and was used to build effective partial least squares regression models for Cm(III) and HNO 3 concentrations without a priori knowledge of solution conditions. Chemometrics for modeling complex fluorescence spectra are promising and may find widespread applicability for online analysis in numerous chemical systems found in the nuclear field.

Actinide↗