Search NASA⌕ Search

SEARCH · Search NASA

Results for “TRANSFER OF 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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS2009

Benefits to the Simulation Training Community of a New ANSI Standard for the Exchange of Aero Simulation Models

The American Institute of Aeronautics Astronautics (AIAA) Modeling and Simulation Technical Committee is in final preparation of a new standard for the exchange of flight dynamics models. The standard will become an ANSI standard and is under consideration for submission to ISO for acceptance by the international community. The standard has some a spects that should provide benefits to the simulation training community. Use of the new standard by the training simulation community will reduce development, maintenance and technical refresh investment on each device. Furthermore, it will significantly lower the cost of performing model updates to improve fidelity or expand the envelope of the training device. Higher flight fidelity should result in better transfer of training, a direct benefit to the pilots under instruction. Costs of adopting the standard are minimal and should be paid back within the cost of the first use for that training device. The standard achie ves these advantages by making it easier to update the aerodynamic model. It provides a standard format for the model in a custom eXtensible Markup Language (XML) grammar, the Dynamic Aerospace Vehicle Exchange Markup Language (DAVE-ML). It employs an existing XML grammar, MathML, to describe the aerodynamic model in an input data file, eliminating the requirement for actual software compilation. The major components of the aero model become simply an input data file, and updates are simply new XML input files. It includes naming and axis system conventions to further simplify the exchange of information.

Hildreth, Bruce L.↗
DOE OSTI2026

Transfer learning for analysis of collective and non-collective Thomson scattering spectra

Thomson scattering (TS) diagnostics provide reliable, minimally perturbative measurements of fundamental plasma parameters, such as electron density (⁠n e ) and electron temperature (⁠T e ⁠). Deep neural networks can provide accurate estimates of ⁠n e and T e when conventional fitting algorithms may fail, such as when TS spectra are dominated by noise, or when fast analysis is required for real-time operation. Although deep neural networks typically require large training sets, transfer learning can improve model performance on a target task with limited data by leveraging pre-trained models from related source tasks, where select hidden layers are further trained using target data. We present five architecturally diverse deep neural networks, pre-trained on synthetic TS data and adapted for experimentally measured TS data, to evaluate the efficacy of transfer learning in estimating n e and T e in both the collective and non-collective scattering regimes. We evaluate errors in n e and T e estimates as a function of training set size for models trained with and without transfer learning, and we observe decreases in model error from transfer learning when the training set contains ≲ 200 experimentally measured spectra.

Artificial neural networks↗
DOE OSTI2024

An innovative heterogeneous transfer learning framework to enhance the scalability of deep reinforcement learning controllers in buildings with integrated energy systems

Deep Reinforcement Learning (DRL)-based control shows enhanced performance in the management of integrated energy systems when compared with Rule-Based Controllers (RBCs), but it still lacks scalability and generalisation due to the necessity of using tailored models for the training process. Transfer Learning (TL) is a potential solution to address this limitation. However, existing TL applications in building control have been mostly tested among buildings with similar features, not addressing the need to scale up advanced control in real-world scenarios with diverse energy systems. This paper assesses the performance of an online heterogeneous TL strategy, comparing it with RBC and offline and online DRL controllers in a simulation setup using EnergyPlus and Python. The study tests the transfer in both transductive and inductive settings of a DRL policy designed to manage a chiller coupled with a Thermal Energy Storage (TES). The control policy is pre-trained on a source building and transferred to various target buildings characterised by an integrated energy system including photovoltaic and battery energy storage systems, different building envelope features, occupancy schedule and boundary conditions (e.g., weather and price signal). The TL approach incorporates model slicing, imitation learning and fine-tuning to handle diverse state spaces and reward functions between source and target buildings. Results show that the proposed methodology leads to a reduction of 10% in electricity cost and between 10% and 40% in the mean value of the daily average temperature violation rate compared to RBC and online DRL controllers. Moreover, online TL maximises self-sufficiency and self-consumption by 9% and 11% with respect to RBC. Conversely, online TL achieves worse performance compared to offline DRL in either transductive or inductive settings. However, offline Deep Reinforcement Learning (DRL) agents should be trained at least for 15 episodes to reach the same level of performance as the online TL. Therefore, the proposed online TL methodology is effective, completely model-free and it can be directly implemented in real buildings with satisfying performance.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS1984

Use of simplifier scenarios for CRM training

Cockpit resource management (CRM) at Metro Airlines is discussed. The process by which the program of CRM training was initiated is mentioned. Management aspects of various flying scenarios are considered. The transfer of training from the classroom to the field is assessed.

Weatherly, D.↗
NASA NTRS1991

Human factors considerations for training astronauts to function effectively in multiple environments

This paper reviews some of the basic issues involved in training individuals to function appropriately under the several conditions that comprise the aerospace environment. The topic of transfer of training is examined in some detail, and the use of high-fidelity simulators in various training programs is discussed. Both current and classical techniques used to train astronauts are noted, and some relatively new and innovative training techniques and methods are described. Particularly, the paper discusses an important aspect of functioning appropriately in a given environment that is based on how well the operator calibrates his motor activity for that specific environment. The role of motor-sensory feedback for the acquisition of motor skills is discussed in the context of training.

Cohen, Malcolm M.↗
NASA NTRS2016

Estimating the Rate of Occurrence of Renal Stones in Astronauts

Changes in urine chemistry, during and post flight, potentially increases the risk of renal stones in astronauts. Although much is known about the effects of space flight on urine chemistry, no inflight incidence of renal stones in US astronauts exists and the question "How much does this risk change with space flight?" remains difficult to accurately quantify. In this discussion, we tackle this question utilizing a combination of deterministic and probabilistic modeling that implements the physics behind free stone growth and agglomeration, speciation of urine chemistry and published observations of population renal stone incidences to estimate changes in the rate of renal stone presentation. The modeling process utilizes a Population Balance Equation based model developed in the companion IWS abstract by Kassemi et al. (2016) to evaluate the maximum growth and agglomeration potential from a specified set of urine chemistry values. Changes in renal stone occurrence rates are obtained from this model in a probabilistic simulation that interrogates the range of possible urine chemistries using Monte Carlo techniques. Subsequently, each randomly sampled urine chemistry undergoes speciation analysis using the well-established Joint Expert Speciation System (JESS) code to calculate critical values, such as ionic strength and relative supersaturation. The Kassemi model utilizes this information to predict the mean and maximum stone size. We close the assessment loop by using a transfer function that estimates the rate of stone formation from combining the relative supersaturation and both the mean and maximum free stone growth sizes. The transfer function is established by a simulation analysis which combines population stone formation rates and Poisson regression. Training this transfer function requires using the output of the aforementioned assessment steps with inputs from known non-stone-former and known stone-former urine chemistries. Established in a Monte Carlo system, the entire renal stone analysis model produces a probability distribution of the stone formation rate and an expected uncertainty in the estimate. The utility of this analysis will be demonstrated by showing the change in renal stone occurrence predicted by this method using urine chemistry distributions published in Whitson et al. 2009. A comparison to the model predictions to previous assessments of renal stone risk will be used to illustrate initial validation of the model.

gravitational physiology↗
NASA NTRSDate not supplied

Speech-Enhanced and Context Dependent Alerts: Future Implications for Spacecraft Design Abstract

In the future, NASA missions will involve many different space vehicles, habitats, and surface assets working together to provide safe and productive living and working environments for crew. Because these systems will be provided by multiple commercial companies working with NASA, it will be very different from missions of the past, bringing new challenges. One of the challenges is related to whether NASA should move beyond simple tone annunciation alerting systems, to more advanced systems that include speech. The other is related to determining the level of consistency required of safety-critical alert systems across spacecraft. Two studies were completed to address these important issues. The first study investigated the advantages and disadvantages of a tone+speech alert relative to the traditional tone-only alert. Results indicate that speech-enhanced alerts initially take longer to silence (the default action to which NASA personnel are trained), due to the need to listen to the entire message, but ultimately provided for faster understanding of the alert situation. Speech-enhanced alerts were also preferred by a large majority of crew-like study participants. An unexpected finding from this first study was that participants took longer to respond to tone-only alerts when they were mixed with speech- enhanced alerts. Participants waited to hear the speech message even for alerts they were trained to know did not contain speech components. This performance error is believed to be due to negative transfer of training. A second study focused on task and alert performance using a common set of tones across two contexts (“vehicles”) versus performance with a different set of tones for each context (“vehicle”). Participants were able to manage two different alert sets successfully; results indicate that discriminability of the two alert sets played a major role in their success. Implications for the design of spacecraft alerts are discussed and future areas of research are identified.

alerts↗
NASA NTRS2023

Speech-Enhanced and Context Dependent Alerts: Future Implications for Spacecraft Design

In the future, NASA missions will involve many different space vehicles, habitats, and surface assets working together to provide safe, productive environments for crew. Because these systems will be provided by multiple commercial companies working with NASA, it will be very different from missions of the past, bringing new challenges. One of the challenges is related to whether NASA should move beyond simple tone annunciation alerting systems, to more advanced systems that include speech. The other is related to determining the level of consistency required of safety-critical alert systems across spacecraft. Two studies were completed to address these important issues. The first study investigated the advantages and disadvantages of a tone+speech alert relative to the traditional tone-only alert. Results indicate that speech-enhanced alerts initially take longer to silence (the default action to which NASA personnel are trained), due to the need to listen to the entire message, but ultimately provided for faster understanding of the alert situation. Speech-enhanced alerts were also preferred by a large majority of crew-like study participants. An unexpected finding was that participants took longer to respond to tone-only alerts that were heard in the same session as speech-enhanced alerts. Participants waited to hear a speech message even for alerts they were trained to know did not contain speech components. This performance error is believed to be due to negative transfer of training. A second study focused on task and alert performance using a common set of tones across two contexts (e.g., vehicles, habitats, suits) versus performance with a different set of tones for each context. Participants were able to manage two different alert sets successfully; results indicate that discriminability of the two alert sets played a major role in their success. Implications for the design of spacecraft alerts are discussed, and future areas of research are identified.

alerts↗
DOE OSTI2022

Training data selection for accuracy and transferability of interatomic potentials

Abstract Advances in machine learning (ML) have enabled the development of interatomic potentials that promise the accuracy of first principles methods and the low-cost, parallel efficiency of empirical potentials. However, ML-based potentials struggle to achieve transferability, i.e., provide consistent accuracy across configurations that differ from those used during training. In order to realize the promise of ML-based potentials, systematic and scalable approaches to generate diverse training sets need to be developed. This work creates a diverse training set for tungsten in an automated manner using an entropy optimization approach. Subsequently, multiple polynomial and neural network potentials are trained on the entropy-optimized dataset. A corresponding set of potentials are trained on an expert-curated dataset for tungsten for comparison. The models trained to the entropy-optimized data exhibited superior transferability compared to the expert-curated models. Furthermore, the models trained to the expert-curated set exhibited a significant decrease in performance when evaluated on out-of-sample configurations.

36 MATERIALS SCIENCE↗
NASA NTRS2010

Training Enhances Both Locomotor and Cognitive Adaptability to a Novel Sensory Environment

During adaptation to novel gravitational environments, sensorimotor disturbances have the potential to disrupt the ability of astronauts to perform required mission tasks. The goal of this project is to develop a sensorimotor adaptability (SA) training program to facilitate rapid adaptation. We have developed a unique training system comprised of a treadmill placed on a motion-base facing a virtual visual scene that provides an unstable walking surface combined with incongruent visual flow designed to enhance sensorimotor adaptability. The goal of our present study was to determine if SA training improved both the locomotor and cognitive responses to a novel sensory environment and to quantify the extent to which training would be retained. Methods: Twenty subjects (10 training, 10 control) completed three, 30-minute training sessions during which they walked on the treadmill while receiving discordant support surface and visual input. Control subjects walked on the treadmill but did not receive any support surface or visual alterations. To determine the efficacy of training all subjects performed the Transfer Test upon completion of training. For this test, subjects were exposed to novel visual flow and support surface movement, not previously experienced during training. The Transfer Test was performed 20 minutes, 1 week, 1, 3 and 6 months after the final training session. Stride frequency, auditory reaction time, and heart rate data were collected as measures of postural stability, cognitive effort and anxiety, respectively. Results: Using mixed effects regression methods we determined that subjects who received SA training showed less alterations in stride frequency, auditory reaction time and heart rate compared to controls. Conclusion: Subjects who received SA training improved performance across a number of modalities including enhanced locomotor function, increased multi-tasking capability and reduced anxiety during adaptation to novel discordant sensory information. Trained subjects maintained their level of performance over six months.

Bloomberg, J. J.↗
NASA NTRS2017

Airspace Technology Demonstration II (ATD-2): American Airlines CLT Ramp Training RTC/RMTC: Baseline Integrated Arrival/ Departure/Surface (IADS) - ATD-2 Tech Transfer 1 - Phase 1

This training material was created to train CLT (Charlotte Douglas International Airport) ramp controllers on the ATD-2 (Airspace Technology Demonstration-2) system. It includes an overview of ATD-2 and the RTC/RMTC (Ramp Traffic Console/Ramp Manager Traffic Console). It discusses many interface details, data exchange and integration, metering modes, and scheduler advisories. It concludes with interactive exercises on all topics of the training course.

RTC/RMTC↗
NASA NTRS2012

Physiological Based Simulator Fidelity Design Guidance

The evolution of the role of flight simulation has reinforced assumptions in aviation that the degree of realism in a simulation system directly correlates to the training benefit, i.e., more fidelity is always better. The construct of fidelity has several dimensions, including physical fidelity, functional fidelity, and cognitive fidelity. Interaction of different fidelity dimensions has an impact on trainee immersion, presence, and transfer of training. This paper discusses research results of a recent study that investigated if physiological-based methods could be used to determine the required level of simulator fidelity. Pilots performed a relatively complex flight task consisting of mission task elements of various levels of difficulty in a fixed base flight simulator and a real fighter jet trainer aircraft. Flight runs were performed using one forward visual channel of 40 deg. field of view for the lowest level of fidelity, 120 deg. field of view for the middle level of fidelity, and unrestricted field of view and full dynamic acceleration in the real airplane. Neuro-cognitive and physiological measures were collected under these conditions using the Cognitive Avionics Tool Set (CATS) and nonlinear closed form models for workload prediction were generated based on these data for the various mission task elements. One finding of the work described herein is that simple heart rate is a relatively good predictor of cognitive workload, even for short tasks with dynamic changes in cognitive loading. Additionally, we found that models that used a wide range of physiological and neuro-cognitive measures can further boost the accuracy of the workload prediction.

Schnell, Thomas↗
NASA NTRS1987

Attention theory and training research

This study used elements of attention theory as a methodological basis to decompose a complex training task in order to improve training efficiency. The complex task was a microcomputer flight simulation where subjects were required to control the stability of their own helicopter while acquiring and engaging enemy helicopers in a threat enviroment. Subjects were divided into whole-task, part-task, and part/open loop adaptive task groups in a transfer of training paradigm. The effect of reducing mental workload at the early stages of learning was examined with respect to the degree that subordinate elements of the complex task could be automated through practice of consistent, learnable stimulus-response relationships. Results revealed trends suggesting the benefit of isolating consistently mapped sub-tasks for part-task training and the presence of a time-sharing skill over and above the skill required for the separate subtasks.

Connelly, James G., Jr.↗