SEARCH · Search NASA

Results for “Learning algorithms”

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 199 records · Page 11

All sources NASA NTRS DOE OSTI

NASA NTRS1980

Adaptive/learning control of large space structures - System identification techniques

Techniques developed for the control of aircraft under changing operating conditions are used to develop a learning control system structure for a multi-configuration, flexible space vehicle. A configuration identification subsystem that is to be used with a learning algorithm and a memory and control process subsystem is developed. Adaptive gain adjustments can be achieved by this learning approach without prestoring of large blocks of parameter data and without dither signal inputs which will be suppressed during operations for which they are not compatible. The Space Shuttle Solar Electric Propulsion (SEP) experiment is used as a sample problem for the testing of adaptive/learning control system algorithms.

Thau, F. E.↗

NASA NTRS1992

Application of fuzzy logic-neural network based reinforcement learning to proximity and docking operations: Translational controller results

The reinforcement learning techniques developed at Ames Research Center are being applied to proximity and docking operations using the Shuttle and Solar Maximum Mission (SMM) satellite simulation. In utilizing these fuzzy learning techniques, we also use the Approximate Reasoning based Intelligent Control (ARIC) architecture, and so we use two terms interchangeable to imply the same. This activity is carried out in the Software Technology Laboratory utilizing the Orbital Operations Simulator (OOS). This report is the deliverable D3 in our project activity and provides the test results of the fuzzy learning translational controller. This report is organized in six sections. Based on our experience and analysis with the attitude controller, we have modified the basic configuration of the reinforcement learning algorithm in ARIC as described in section 2. The shuttle translational controller and its implementation in fuzzy learning architecture is described in section 3. Two test cases that we have performed are described in section 4. Our results and conclusions are discussed in section 5, and section 6 provides future plans and summary for the project.

Jani, Yashvant↗

NASA NTRS2018

Atmospheric Chemistry Modeling Using a Regression Forest Model

Atmospheric chemistry is central to many environmental issues such as air pollution, climate change, and stratospheric ozone loss. Chemistry Transport Models (CTM) are a central tool for understanding these issues, whether for research or for forecasting. These models split the atmosphere in a large number of grid-boxes and consider the emission of compounds into these boxes and their subsequent transport, deposition, and chemical processing. The chemistry is represented through a series of simultaneous ordinary differential equations, one for each compound. Given the difference in life-times between the chemical compounds (milli-seconds for O1D to years for CH4) these equations are numerically stiff and solving them consists of a significant fraction of the computational burden of a CTM. We have investigated a machine learning approach to solving the differential equations instead of solving them numerically. From an annual simulation of the GEOS-Chem model we have produced a training dataset consisting of the concentration of compounds before and after the differential equations are solved, together with some key physical parameters for every grid-box and time-step. From this dataset we have trained a machine learning algorithm (regression forest) to be able to predict the concentration of the compounds after the integration step based on the concentrations and physical state at the beginning of the time step. We have then included this algorithm back into the GEOS-Chem model, bypassing the need to integrate the chemistry. This machine learning approach shows many of the characteristics of the full simulation and has the potential to be substantially faster. There are a wide range of application for such an approach - generating boundary conditions, for use in air quality forecasts, chemical data assimilation systems, centennial scale climate simulations etc. We discuss our approches' speed and accuracy, and highlight some potential future directions for improving this approach.

Keller, Christoph A.↗

NASA NTRS2019

Atmospheric Chemistry Modeling Using Machine Learning

Atmospheric chemistry models are a central tool to study the impact of chemical constituents on the environment, vegetation and human health. These models split the atmosphere in a large number of grid-boxes and consider the emission of compounds into these boxes and their subsequent transport, deposition, and chemical processing. The chemistry is represented through a series of simultaneous ordinary differential equations, one for each compound. Given the difference in life-times between the chemical compounds (milli-seconds for O1D to years for CH4) these equations are numerically stiff and solving them consists of a significant fraction of the computational burden of a chemistry model. We have investigated a machine learning approach to emulate the chemistry instead of solving the differential equations numerically. From a one-month simulation of the GEOS-Chem model we have produced a training dataset consisting of the concentration of compounds before and after the differential equations are solved, together with some key physical parameters for every grid-box and time-step. From this dataset we have trained a machine learning algorithm (regression forest) to be able to predict the concentration of the compounds after the integration step based on the concentrations and physical state at the beginning of the time step. We have then included this algorithm back into the GEOS-Chem model, bypassing the need to integrate the chemistry. This machine learning approach shows many of the characteristics of the full simulation and has the potential to be substantially faster. There are a wide range of application for such an approach - generating boundary conditions, for use in air quality forecasts, chemical data assimilation systems, etc. We discuss speed and accuracy of our approach, and highlight some potential future directions for improving it.

Keller, Christoph A.↗

NASA NTRS2019

Machine Learning Application to Atmospheric Chemistry Modeling

Atmospheric chemistry models are a central tool to study the impact of chemical constituents on the environment, vegetation and human health. These models split the atmosphere in a large number of grid-boxes and consider the emission of compounds into these boxes and their subsequent transport, deposition, and chemical processing. The chemistry is represented through a series of simultaneous ordinary differential equations, one for each compound. Given the difference in life-times between the chemical compounds (milli-seconds for O (sup 1) D (Deuterium) to years for CH4) these equations are numerically stiff and solving them consists of a significant fraction of the computational burden of a chemistry model. We have investigated a machine learning approach to emulate the chemistry instead of solving the differential equations numerically. From a one-month simulation of the GEOS-Chem model we have produced a training dataset consisting of the concentration of compounds before and after the differential equations are solved, together with some key physical parameters for every grid-box and time-step. From this dataset we have trained a machine learning algorithm (regression forest) to be able to predict the concentration of the compounds after the integration step based on the concentrations and physical state at the beginning of the time step. We have then included this algorithm back into the GEOS-Chem model, bypassing the need to integrate the chemistry. This machine learning approach shows many of the characteristics of the full simulation and has the potential to be substantially faster. There are a wide range of application for such an approach - generating boundary conditions, for use in air quality forecasts, chemical data assimilation systems, etc. We discuss speed and accuracy of our approach, and highlight some potential future directions for improving it.

Keller, Christoph A.↗

DOE OSTI2025

High-Throughput Discovery Illuminates Design Principles and Limits for Long-Lived Charged Species in Organic Electrolytes

The chemical stability of charged molecules in all-organic redox flow batteries (RFBs) is required for the prolonged operation of these devices. Molecular engineering and electrolyte optimization are used to mitigate parasitic reactions and extend the lifetimes of the charge carriers. However, how much can structural variation extend the lifetime? To probe this query, we designed a high-throughput kinetic study of the radical cation of N-methylphenothiazinium, guided by statistical sampling and learning algorithms. Using Argonne’s autonomous discovery facility, we conducted over 6,000 kinetic experiments with robotic sample preparation, parallel kinetic measurements, and machine learning inputs, testing 188 solvent molecules selected from a space of over 540 candidates from 11 chemical classes. Algorithmic selections guided us to stable solvent candidates, which were further tested in high concentration with and without supporting electrolyte. Our findings reveal the inherent difficulty of exceeding the current state of the art through solvent variation. The desired stability is statistically rare and poorly predictable. Among the many tested, only three solvents significantly outperformed our baseline, acetonitrile─and none by more than a factor of 3─suggesting a general challenge in achieving the necessary techno-economic targets. Furthermore, we suggest that self-discharge through solvent homolysis is the cause of the observed limitations. Several structural motifs contribute to >1,000 h half-life stability including molecular simplicity, symmetry, oxidation complement, and strategic fluorination. Importantly, this workflow establishes effective assays for diagnosing and predicting oxidative stress for highly stable liquid electrolytes in all batteries.

Batteries↗

DOE OSTI2024

Impurity gas detection for SNF canisters using probabilistic deep learning and acoustic sensing *

Abstract Monitoring impurity gases in spent nuclear fuel (SNF) canisters is a novel structural health monitoring approach for SNF in dry storage. The SNF canisters are sealed containers that do not facilitate visual access to the inside. Acoustic sensing can be deployed by taking advantage of the pathways unobstructed by internal hardware. Although the ultrasonic time-of-flight measurement can provide valuable information, it is limited in its ability to discern the concentration of only one impurity gas. As such, deep learning algorithms, particularly convolutional neural networks (CNNs), offer a promising solution. In this study, CNN-based probabilistic deep learning models were implemented to detect and quantify multiple impurity gases in helium. An experimental platform was established to simulate canister conditions, and ultrasonic test data were collected. The presence of argon and air in helium at concentrations ranging from 0% to 1.2% at increments of 0.05% was considered. The multi-layer perceptron, decision tree, and logistic regression classifiers achieved high accuracies when distinguishing pure helium from helium with impurities. CNN with dropout layers and CNN using maximum likelihood estimation showed a similar performance, indicating their ability to capture uncertainties. The ensemble CNN model exhibited improved predictions and the ability to balance individual gas concentration by integrating 1D- and 2D-CNN models. These findings contribute probabilistic deep learning solutions for impurity gas detection and analysis within SNF canisters, thus ensuring safe storage and management of SNFs.

47 OTHER INSTRUMENTATION↗

DOE OSTI2026

Enhancing segmentation fairness through curriculum learning and progressive loss: a centralized and federated perspective on radiograph analysis

Bias in medical image segmentation can lead to unequal performance across demographic subgroups, raising concerns about fairness and reliability in clinical AI systems. While deep learning models have achieved high segmentation accuracy, ensuring equitable performance across race and gender remains a significant challenge, particularly in privacy-sensitive healthcare environments. This study investigates fairness-aware medical image segmentation for hip and knee radiographs using deep learning models evaluated in both centralized and Federated Learning (FL) settings. We introduce Curriculum Learning (CL) strategies and Progressive Loss (PL) functions to regulate sample difficulty during training. In addition, we propose two novel fairness-oriented federated learning algorithms, Federated Intersection over Union (FedIoU) and Federated Intersection over Union with Outlier Analysis (FedIoUoutlier). Experiments are conducted using multiple segmentation backbones and simulated multi-site data partitions derived from the Osteoarthritis Initiative dataset. Model performance is evaluated using Intersection over Union (IoU), IoU standard deviation, Skewed Error Ratio (SER), and Min-Max Disparity across race and gender subgroups. Statistical significance was verified using paired t-tests to compare per-sample IoU performance against baseline configurations. Across both hip and knee segmentation tasks, curriculum learning and progressive loss strategies consistently improved segmentation accuracy and reduced demographic performance disparities in centralized training. In federated settings, fairness-aware aggregation further enhanced performance. Notably, FedIoUoutlier combined with balanced curriculum learning and tiered progressive loss achieved the highest mean IoU while yielding the lowest SER and Min-Max Disparity, indicating improved fairness without sacrificing accuracy. In several configurations, federated models matched or exceeded the performance of optimized centralized models, with statistically significant improvements in per-sample IoU over baseline configurations. The results demonstrate that structured training strategies and fairness-aware federated aggregation can jointly improve accuracy, stability, and demographic fairness in medical image segmentation. By integrating curriculum learning, progressive loss, and novel FL algorithms, this work provides a practical pathway toward equitable and privacy-preserving AI systems for medical imaging.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1990

Recent developments in learning control and system identification for robots and structures

This paper reviews recent results in learning control and learning system identification, with particular emphasis on discrete-time formulation, and their relation to adaptive theory. Related continuous-time results are also discussed. Among the topics presented are proportional, derivative, and integral learning controllers, time-domain formulation of discrete learning algorithms. Newly developed techniques are described including the concept of the repetition domain, and the repetition domain formulation of learning control by linear feedback, model reference learning control, indirect learning control with parameter estimation, as well as related basic concepts, recursive and non-recursive methods for learning identification.

Phan, M.↗

NASA NTRSDate not supplied

Mapping National Forest Aboveground Biomass in Mexico By Integrating GEDI and Landsat Times Series Data

Mexico is one of the countries with great potential for the UN's Reducing Emissions from Deforestation and Forest Degradation (REDD+) program, a key nature-based solution for the forest sector. To monitor carbon stock changes, there is a growing demand for unbiased Monitoring Reporting Verification (MRV) systems to facilitate effective forest management and climate change mitigation strategies. Remote sensing-based national aboveground biomass density (AGBD) estimation over Mexico is scarce and often limited to one-time static mapping, leading to spatiotemporal inconsistency in inputs. As an effort under NASA's Carbon Monitoring System (CMS) program, we have developed a remote sensing-based approach to create consistent historical AGBD maps of Mexico using multi-stream remote sensing data, including spaceborne lidar GEDI and long-term Landsat time series, as well as topographic information. We employ the continuous change detection and classification (CCDC) algorithm for temporal modeling of Landsat surface reflectance, followed by the inference of forest AGBD using a random forest machine learning algorithm with the temporal information of land surface dynamics extracted by the CCDC as input. GEDI provides unprecedented forest structure and AGBD sampling datasets for model training and validation practices. In this presentation, we share the progress made in developing a spatially explicit mapping of historical AGBD changes associated with land surface changes and post-disturbance landscapes.

Taejin Park↗

NASA NTRSDate not supplied

Machine Learning Application for Improving Cloud Detection and Phase Determination Over Sunglint Regions for Geostationary Satellites

Cloud detection and phase determination over sunglint regions has been a challenge, especially for geostationary (GEO) satellites. Sunglint is observed when the sunlight specular reflection is at the same viewing angle of the satellite sensor. This intense reflection in the visible channels (VIS) is often comparable to that from optically thick clouds. It also contaminates the shortwave infrared channels (SWIR). Consequently, VIS and SWIR channels become less useful - or not useful- when they are saturated, hampering the detection of cloudy and clear-sky pixels. Sunglint contamination happens frequently and exists nearly in every daytime GEO full disk satellite images. However, sunglint intensity and region are difficult to model due to variable viewing geometry and ocean surface conditions. Moreover, existing physical models do not meet the accuracy required for operational GEO satellite cloud detection. We developed a machine learning algorithm to improve cloud detection in sunglint conditions for the NASA Langley’s Satellite ClOud and radiation Property retrieval System (SatCORPS). This poster presents our recent progress in the algorithm development, validation and applications. The algorithm is validated using collocated SatCORPS GOES-East and GOES-West cloud products. We demonstrate that the machine learning cloud detection in sunglint regions is superior to the traditional approach by improving temporal consistency between sunglint and non-sunglint conditions.

Machine Learning, Cloud detection, Sunglint, SatCO↗

NASA NTRSDate not supplied

Environment Adversarial Reinforcement Learning

This paper presents a training method for increasing performance of reinforcement learning agents. The method is named Environment Adversarial Reinforcement Learning. The method requires the reinforcement learning environment to be parameterizeable. Over the course of training, environment parameters are updated in a direction of increasing difficulty for the agent. The direction for these updates is found using a performance prediction network trained on data from tests of the agent under varying environment parameters. The method was tested on a CartPole environment. A 28-58\% improvement in mean return was found when comparing performance to a baseline reinforcement learning algorithm on both easy and hard versions of the task.

machine learning↗

NASA NTRS2013

SIM_EXPLORE: Software for Directed Exploration of Complex Systems

Physics-based numerical simulation codes are widely used in science and engineering to model complex systems that would be infeasible to study otherwise. While such codes may provide the highest- fidelity representation of system behavior, they are often so slow to run that insight into the system is limited. Trying to understand the effects of inputs on outputs by conducting an exhaustive grid-based sweep over the input parameter space is simply too time-consuming. An alternative approach called "directed exploration" has been developed to harvest information from numerical simulators more efficiently. The basic idea is to employ active learning and supervised machine learning to choose cleverly at each step which simulation trials to run next based on the results of previous trials. SIM_EXPLORE is a new computer program that uses directed exploration to explore efficiently complex systems represented by numerical simulations. The software sequentially identifies and runs simulation trials that it believes will be most informative given the results of previous trials. The results of new trials are incorporated into the software's model of the system behavior. The updated model is then used to pick the next round of new trials. This process, implemented as a closed-loop system wrapped around existing simulation code, provides a means to improve the speed and efficiency with which a set of simulations can yield scientifically useful results. The software focuses on the case in which the feedback from the simulation trials is binary-valued, i.e., the learner is only informed of the success or failure of the simulation trial to produce a desired output. The software offers a number of choices for the supervised learning algorithm (the method used to model the system behavior given the results so far) and a number of choices for the active learning strategy (the method used to choose which new simulation trials to run given the current behavior model). The software also makes use of the LEGION distributed computing framework to leverage the power of a set of compute nodes. The approach has been demonstrated on a planetary science application in which numerical simulations are used to study the formation of asteroid families.

Burl, Michael↗

NASA NTRS2020

Efficient Active Learning for New Domains

The promise of active learning is to reduce the number of labeled examples required by supervised machine learning algorithms. The largest potential benefits lie in entirely new domains, for which no labeled examples yet exist. Yet to date, most active learning studies are retroactive and demonstrate the benefits that could have been gained if active learning had been used. What are the barriers to true adoption and utilization of active learning? We focus on two: (1) the cold start or class discovery problem, in which active learning methods may struggle to make progress with zero labeled examples, and (2) the cost of having the classifier in the loop to select the next example to be labeled. We assess different active learning approaches in the context of these two barriers and conclude with recommendations for how to employ active learning in new domains. As an example, we report on the use of active learning on a large, novel data set of Mars surface images.

Lu, Steven↗

NASA NTRS2025

Improving Adhesive Bondline Time of Flight Predictions During Autoclave Cure Utilizing Machine Learning

Composite materials are increasingly being used in aerospace applications due to their superior strength-to-weight ratio compared to commonly used metals. A current limitation to widespread adoption is the certification of adhesively bonded joints. One approach to improving adhesive bonding in composites is accurately measuring the thickness of adhesive bondlines in composite laminates. Precise bondline thickness control is essential for aerospace applications where adhesive layer thickness directly affects joint fracture properties and structural performance. This study focused on implementing machine learning techniques to determine the ultrasonic time of flight (directly correlated to thickness) in adhesive bondlines throughout autoclave cure cycles. A high-temperature (use up to 180°C) ultrasonic scanning system was deployed in an autoclave to provide time of flight data through composite panels. Three experiments were conducted on the curing of 305 mm × 305 mm unidirectional composite panels. In the first experiment, a piecewise function was fit for the temperature correction factor to account for changing autoclave temperatures. Due to deficiencies in the first calibration experiment, a second experiment was run, and the results were used to train a machine learning model. The revised experiment, in combination with the machine learning model, significantly increased the accuracy of the bondline time of flight predictions (~14% error reduced to <1%). Data was processed using the Regression Learner Application in MATLAB®, with a Support Vector Machine selected for the model. The result was a machine learning algorithm capable of reliably quantifying ultrasonic time of flight through adhesive bondlines. The third experiment provided independent test data for the machine learning model, demonstrating that the model produces accurate predictions from data beyond its training set.

Machine Learning↗

NASA NTRS2020

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-of-the-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Deep-learning↗

NASA NTRS2020

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-ofthe-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Hesham ElAbd↗

NASA NTRS1995

Linear and nonlinear trending and prediction for AVHRR time series data

The variability of AVHRR calibration coefficient in time was analyzed using algorithms of linear and non-linear time series analysis. Specifically we have used the spline trend modeling, autoregressive process analysis, incremental neural network learning algorithm and redundancy functional testing. The analysis performed on available AVHRR data sets revealed that (1) the calibration data have nonlinear dependencies, (2) the calibration data depend strongly on the target temperature, (3) both calibration coefficients and the temperature time series can be modeled, in the first approximation, as autonomous dynamical systems, (4) the high frequency residuals of the analyzed data sets can be best modeled as an autoregressive process of the 10th degree. We have dealt with a nonlinear identification problem and the problem of noise filtering (data smoothing). The system identification and filtering are significant problems for AVHRR data sets. The algorithms outlined in this study can be used for the future EOS missions. Prediction and smoothing algorithms for time series of calibration data provide a functional characterization of the data. Those algorithms can be particularly useful when calibration data are incomplete or sparse.

Smid, J.↗