Results for “foundation model” — Page 2

NASA NTRS2016

Canonical Statistical Model for Maximum Expected Immission of Wire Conductor in an Aperture Enclosure

Prediction of the maximum expected electromagnetic pick-up of conductors inside a realistic shielding enclosure is an important canonical problem for system-level EMC design of space craft, launch vehicles, aircraft and automobiles. This paper introduces a simple statistical power balance model for prediction of the maximum expected current in a wire conductor inside an aperture enclosure. It calculates both the statistical mean and variance of the immission from the physical design parameters of the problem. Familiar probability density functions can then be used to predict the maximum expected immission for deign purposes. The statistical power balance model requires minimal EMC design information and solves orders of magnitude faster than existing numerical models, making it ultimately viable for scaled-up, full system-level modeling. Both experimental test results and full wave simulation results are used to validate the foundational model.

Computational electromagnetics↗

NASA NTRSDate not supplied

Improving Sim-to-Real Transfer in Vision-Based Robot Navigation Via Instance-Level GAN-Based Data Augmentation

Achieving robust vision-based robotic tasks requires large amounts of data, which are often difficult to obtain in real-world scenarios. Simulators and synthetic data offer a cost-effective alternative, but the visual gap between simulation and reality hinders the performance of models when deployed in real-world environments. In this paper, we present a data augmentation pipeline that integrates a foundation model (Segment Anything Model) with an unsupervised image-to-image translation model (CycleGAN) for instance-level domain transfer from simulation to reality. This pipeline enables the generation of realistic labeled data from synthetic images for training supervised machine learning models in vision-based navigation tasks. We evaluate our approach on real-world data for ego-vehicle pose estimation, a critical autonomous navigation task involving the prediction of cross-track position and heading angle relative to road center line markings. The results of our tests show that our GAN-based data augmentation pipeline significantly outperforms models trained solely on simulation data or on data processed with standard image augmentation methods for sim-to-real transfer, enhancing model robustness and generalizability in real-world scenarios. Our method provides a scalable and flexible data augmentation tool for leveraging large synthetic datasets to enhance vision-based robotic navigation tasks.

artificial intelligence↗

NASA NTRS1995

Report on twisted nematic and supertwisted nematic device characterization program

In this study we measured the optical characteristics of normally white twisted nematic (NWTN) and super twisted nematic (STN ) cells. Though no dynamic computer model was available, the static observations were compared with computer simulated behavior. The measurements were taken as a function of both viewing angle and applied voltage and included in the static case not only luminance but also contrast ratio and chromaticity . We employed the computer model Twist Cell Optics, developed at Kent State in conjunction with this study, and whose optical modeling foundation, Iike the ViDEOS program, is the 4 x 4 matrix method of Berreman. In order to resolve discrepancies between the experimental and modeled data the optical parameters of the individual cell components, where not known, were determined using refractometry, profilometry, and various forms of ellipsometry. The resulting agreement between experiment and model is quite good due primarily to a better understanding of the structure and optics of dichroic sheet polarizers. A description of the model and test cells employed are given in section 2. Section 3 contains the experimental data gathered and section 4 gives examples of the fit between model and experiment. Also included with this report are a pair of papers which resulted from the research and which detail the polarizer properties and some of the cell characterization methods.

Source record↗

NASA NTRSDate not supplied

Implementation of a SysML Model-Based Concept of Operations for Level of Care IV: Long-Duration Lunar Orbital and Surface Operations

One goal of the Human Research Program (HRP) Exploration Medical Capability (ExMC) Element Systems Engineering (SE) team is to define the technical system needed to support crew medical system capabilities for future exploration missions, including orbital and surface operations for long duration lunar missions. This is accomplished through the development of a Medical System Foundation, which communicates medical system requirements, capabilities, conditions, and resources that define a starting point for a medical system that meets the specification for design reference missions and associated Levels of Care (as defined by NASA-STD-3001). The starting point for creating this Medical System Foundation Model is the development of a Concept of Operations (ConOps) that describes the operation of the system from the point of view of the users. It includes a comprehensive and thoroughly vetted list of the users, their specific needs, the goals of the system, key assumptions about the system, and definitions of the system’s operational environments. The use case scenarios included in the ConOps illustrate required medical system capabilities for Level of Care IV and enable the ExMC SE team to develop integrated medical system requirements and identify capabilities required to meet those requirements. For this “Long Duration” Medical System Foundation development effort, ExMC replaced the traditional document-based ConOps with a model-based ConOps using model-based systems engineering. This approach has several advantages, including the facilitation of more efficient understanding of the material through information-dense images with less opportunity for misinterpretation than text alone, the roll-out of changes to the ConOps to stakeholders in real-time as they are approved, and the consolidation of all salient information into one centralized location. This discussion will focus on both how the model represents the ConOps content and how the SE team utilized lessons learned in an agile environment to improve the way this information was created and presented to stakeholders.

M Kaetzer↗

NASA NTRSDate not supplied

Implementation of a SysML Model-Based Concept of Operations for Level of Care IV: Long-Duration Lunar Orbital and Surface Operations

One goal of the Human Research Program (HRP) Exploration Medical Capability (ExMC) Element Systems Engineering (SE) team is to define the technical system needed to support crew medical system capabilities for future exploration missions, including orbital and surface operations for long duration lunar missions. This is accomplished through the development of a Medical System Foundation, which communicates medical system requirements, capabilities, conditions, and resources that define a starting point for a medical system that meets the specification for design reference missions and associated Levels of Care (as defined by NASA-STD-3001). The starting point for creating this Medical System Foundation Model is the development of a Concept of Operations (ConOps) that describes the operation of the system from the point of view of the users. It includes a comprehensive and thoroughly vetted list of the users, their specific needs, the goals of the system, key assumptions about the system, and definitions of the system’s operational environments. The use case scenarios included in the ConOps illustrate required medical system capabilities for Level of Care IV and enable the ExMC SE team to develop integrated medical system requirements and identify capabilities required to meet those requirements. For this “Long Duration” Medical System Foundation development effort, ExMC replaced the traditional document-based ConOps with a model-based ConOps using model-based systems engineering. This approach has several advantages, including the facilitation of more efficient understanding of the material through information-dense images with less opportunity for misinterpretation than text alone, the roll-out of changes to the ConOps to stakeholders in real-time as they are approved, and the consolidation of all salient information into one centralized location. This discussion will focus on both how the model represents the ConOps content and how the SE team utilized lessons learned in an agile environment to improve the way this information was created and presented to stakeholders.

Mary Susan Kaetzer↗

NASA NTRSDate not supplied

Concept-based Analysis of Neural Networks via Vision-Language Models

The analysis of vision-based deep neural networks (DNNs) is highly desirable but it is very challenging due to the difficulty of expressing formal specifications for vision tasks and the lack of efficient verification procedures. In this paper, we propose to leverage emerging multimodal, vision-language, foundation models (VLMs) as a lens through which we can reason about vision models. VLMs have been trained on a large body of images accompanied by their textual description, and are thus implicitly aware of high-level, human-understandable concepts describing the images. We describe a logical specification language Con spec designed to facilitate writing specifications in terms of these concepts. To define and formally check Con spec specifications, we build a map between the internal representations of a given vision model and a VLM, leading to an efficient verification procedure of natural-language properties for vision models. We demonstrate our techniques on a ResNet-based classifier trained on the RIVAL-10 dataset using CLIP as the multimodal model.

Large Vision Language Models↗

NASA NTRS1994

TEXCAD: Textile Composite Analysis for Design. Version 1.0: User's manual

The Textile Composite Analysis for Design (TEXCAD) code provides the materials/design engineer with a user-friendly desktop computer (IBM PC compatible or Apple Macintosh) tool for the analysis of a wide variety of fabric reinforced woven and braided composites. It can be used to calculate overall thermal and mechanical properties along with engineering estimates of damage progression and strength. TEXCAD also calculates laminate properties for stacked, oriented fabric constructions. It discretely models the yarn centerline paths within the textile repeating unit cell (RUC) by assuming sinusoidal undulations at yarn cross-over points and uses a yarn discretization scheme (which subdivides each yarn not smaller, piecewise straight yarn slices) together with a 3-D stress averaging procedure to compute overall stiffness properties. In the calculations for strength, it uses a curved beam-on-elastic foundation model for yarn undulating regions together with an incremental approach in which stiffness properties for the failed yarn slices are reduced based on the predicted yarn slice failure mode. Nonlinear shear effects and nonlinear geometric effects can be simulated. Input to TEXCAD consists of: (1) materials parameters like impregnated yarn and resin properties such moduli, Poisson's ratios, coefficients of thermal expansion, nonlinear parameters, axial failure strains and in-plane failure stresses; and (2) fabric parameters like yarn sizes, braid angle, yarn packing density, filament diameter and overall fiber volume fraction. Output consists of overall thermoelastic constants, yarn slice strains/stresses, yarn slice failure history, in-plane stress-strain response and ultimate failure strength. Strength can be computed under the combined action of thermal and mechanical loading (tension, compression and shear).

Naik, Rajiv A.↗

NASA NTRS1995

TEXCAD: TEXile Composite Analysis for Design

The Textile Composite Analysis for Design (TEXCAD) code provides the materials/design engineer with a user-friendly, desktop computer based tool for the analysis of a wide variety of fabric reinforced woven and braided composites. It can be used to calculate overall thermal and mechanical properties along with engineering estimates of damage progression and strength. TEXCAD also calculates laminate properties for stacked, oriented fabric constructions. It discretely models the yarn centerline paths within the textile repeating unit cell (RUC) by assuming sinusoidal undulations at yarn cross-over points and uses a year discretization scheme (which subdivides each yarn into smaller, piecewise straight yarn slices) together with a 3-D stress averaging procedure to compute overall stiffness properties. In the calculations for strength, it uses a curved beam-on-elastic foundation model for yarn undulating regions together with incremental approach in which stiffness properties for the failed yarn slices are reduced based on the predicted yarn slice failure mode. Nonlinear shear effects and nonlinear geometric effects can be stimulated. Input to TEXCAD consists of: (1) material parameters like impregnated yarn and resin properties such as moduli, Poisson's ratios, coefficients of thermal expansion, nonlinear shear parameters, axial failure strains, and in-plane failure stresses; and (2) fabric parameters like yarn sizes, braid angle, yarn packing density, filament diameter, and overall fiber volume fraction. Output consists of overall thermoelastic constants, yarn slice strains/stresses, yarn slice failure history, in-plane strain response, and ultimate failure strength. Strength can be computed under the combined action of thermal and mechanical loading (tension, compression, and shear). A brief overview of the analytical capabilities, program organization, and modules, input and output parameters, computer platforms, distribution, and modifications/extensions of the TEXCAD code are presented here.

Naik, Rajiv A.↗

NASA NTRSDate not supplied

Fast Machine Learning Lidar Surrogate Simulator: Pristine Clear Sky

The simulations of lidar signals and retrievals rely on a range of optic-physical models, such as radiative transfer models, particle scattering and absorption models, along with the output data from atmospheric physical models. Integrating these different models to represent signals of a lidar system is computationally expensive, and performing backward retrievals can be complex and ambiguous. However, with the advantages of Machine Learning, there is a new potential for building effective lidar signal database linked to corresponding atmospheric profiles. For this project, we are developing a fast pre-trained neural network as the lidar surrogate simulator using simulated data for a CALIPSO-like lidar (355 nm, 532nm, and 1064nm), and a CO2 differential absorption lidar (DIAL) near 1571nm. Specifically, we utilize a long short-term memory (LSTM) model to map the relationships between atmospheric profiles (pressure, temperature, air density and CO2 mixing ratio) and lidar signals. This approach allows us to build machine learning based simulators that can reconstruct lidar signals at specific bands from MERRA reanalysis data, and perform retrievals of atmospheric profiles using lidar signals at various wavelengths. As a first step, the results show the potential of this method to establish a foundational model for sensor signals. This model offers the promise of enabling both accurate predictions and rapid retrievals, providing a more efficient approach to signal processing and analysis.

Shan Zeng↗

NASA NTRS1993

Nonlinear probabilistic finite element models of laminated composite shells

A probabilistic finite element analysis procedure for laminated composite shells has been developed. A total Lagrangian finite element formulation, employing a degenerated 3-D laminated composite shell with the full Green-Lagrange strains and first-order shear deformable kinematics, forms the modeling foundation. The first-order second-moment technique for probabilistic finite element analysis of random fields is employed and results are presented in the form of mean and variance of the structural response. The effects of material nonlinearity are included through the use of a rate-independent anisotropic plasticity formulation with the macroscopic point of view. Both ply-level and micromechanics-level random variables can be selected, the latter by means of the Aboudi micromechanics model. A number of sample problems are solved to verify the accuracy of the procedures developed and to quantify the variability of certain material type/structure combinations. Experimental data is compared in many cases, and the Monte Carlo simulation method is used to check the probabilistic results. In general, the procedure is quite effective in modeling the mean and variance response of the linear and nonlinear behavior of laminated composite shells.

Engelstad, S. P.↗

NASA NTRSDate not supplied

Implementing Artificial Thinking Autonomy with Model-Based System Engineering

Complex autonomous systems capable of successfully operating independently under ‘known unknowns’ and harsh conditions require paradigm innovation in modern development strategies. In the field of autonomy, developing a system-of-systems which can ostensibly think for itself in the face of ‘unknown unknowns’ is still a field of ongoing research. Maturing the systems architecting and modeling methodologies for developing henceforth named Thinking Autonomous Systems, which are verified with digital mission simulation, can potentially usher in the next generation of artificial intelligence for space exploration. The concept presented in this paper incorporates multiple Model-Based Systems Engineering and simulation methodologies combined as a new paradigm to design a novel, biomimetic thinking autonomy strategy. Anachronistic concepts from classical Kantian philosophy will be leveraged to inspire architectural designs that could be used for complex distributed systems in deep space. To accomplish this, digital transformation of a document-based implementation plan for Thinking Autonomous Systems, generated by experienced NASA software engineers, is implemented for NASA’s Platform for Autonomous Systems by creating descriptive and executable software models in SysML to prototype real-time operating capabilities. This conceptual implementation has been developed by incorporating model-based digital simulations to theorize how a cyberphysical thinking system would achieve specific strategies without crew reliance, while simultaneously being resilient to all operating conditions and remaining functional when devoid of ground communication. Additionally, ensuring that an autonomous system framework is an ethical Artificial Intelligence requires careful consideration of system behavior and accountability, human factors for teaming with a thinking autonomous system, and comparison to other modern approaches used for implementing true autonomy. This paper presents the first steps in formalizing the metacognition required for instantiating a truly Thinking Autonomous System; the approach described symphonizes autonomy characteristics from classical philosophical into a unified software architecture describing human thought. In the future, the foundational models described in this paper can be further leveraged to help advance research into thinking autonomy requirements for future deep space missions as well as for current near-term applications, i.e., living aboard crewed spacecraft like a NASA Gateway cislunar habitat.

Artificial Thought↗

NASA NTRSDate not supplied

GeoAI advances in specific landform mapping

Landform mapping (also referred to as geomorphology or geomorphometry) can be divided into two domains: general and specific (Evans 2012). Whereas general landform mapping categorizes all elements of the study area into landform classes, such as ridges, valleys, peaks, and depressions, the mapping of specific landforms requires the delineation (even if fuzzy) of individual landforms. The former is mainly driven by physical properties such as elevation, slope, and curvature. The latter, however, must consider the cognitive (human) reasoning that discriminates individual landforms in addition to these physical properties (Arundel and Sinha 2018). Both mapping forms are important. General geomorphometry is needed to understand geological and ecological processes and as boundary layer input to climate and environmental models. Specific geomorphometry supports such activities as disaster management and recovery, emergency response, transportation, and navigation. In the United States, individual landforms of interest are named in the U.S. Geological Survey (USGS) Geographic Names Information System, a point dataset captured specifically to digitize geographic names from the USGS Historical Topographic Map Collection (HTMC). Named landform extent is represented only by the name placement in the HTMC. Recent work has investigated CNN-based deep learning methods to capture these extents in machine-readable form. These studies first relied on physical properties (Arundel et al. 2020) and then included the HTMC as a band in RGB images in limited testing (Arundel et al. 2023). Results from the HTMC dataset surpassed those using just physical properties and using the HTMC alone performed best due to the hillshading and elevation (contour) data incorporated into the topographic maps. However, results fell short of an operational capacity to map all named landforms in the United States. Thus, our current work expands upon past research by focusing on the HTMC and physical information as inputs and the named landform label extents. Specifically, we propose to leverage pre-trained foundation models for segmentation and optical character recognition (OCR) models to jointly map landforms in the United States. Our approach aims to bridge the disparities among the independent information sources to facilitate informed decision-making. The modeling pipeline performs (1) segmentation using the physical information and (2) information extraction using OCR, in parallel. Then a computer vision approach merges the two branches into a labeled segmentation. References: Arundel, Samantha T., Wenwen Li, and Sizhe Wang. 2020. “GeoNat v1.0: A Dataset for Natural Feature Mapping with Artificial Intelligence and Supervised Learning.” Transactions in GIS 24 (3): 556–72. https://doi.org/10.1111/tgis.12633. Arundel, Samantha T, and Gaurav Sinha. 2018. “Validating GEOBIA Based Terrain Segmentation and Classification for Automated Delineation of Cognitively Salient Landforms BT - Proceedings of Workshops and Posters at the 13th International Conference on Spatial Information Theory (COSIT 2017).” In Proceedings of Workshops and Posters at the 13th International Conference on Spatial Information Theory (COSIT 2017), Lecture Notes in Geoinformation and Cartography, edited by Paolo Fogliaroni, Andrea Ballatore, and Eliseo Clementini, 9–14. Cham: Springer International Publishing. Arundel, Samantha T., Gaurav Sinha, Wenwen Li, David P. Martin, Kevin G. McKeehan, and Philip T. Thiem. 2023. “Historical Maps Inform Landform Cognition in Machine Learning.” Abstracts of the ICA 6 (August): 1–2. https://doi.org/10.5194/ica-abs-6-10-2023. Evans, Ian S. 2012. “Geomorphometry and Landform Mapping: What Is a Landform?” Geomorphology 137 (1): 94–106. https://doi.org/10.1016/j.geomorph.2010.09.029.

machine learning↗

NASA NTRSDate not supplied

GeoAI Advances in Specific Landform Mapping

Landform mapping (also referred to as geomorphology or geomorphometry) can be divided into two domains: general and specific (Evans 2012). Whereas general landform mapping categorizes all elements of the study area into landform classes, such as ridges, valleys, peaks, and depressions, the mapping of specific landforms requires the delineation (even if fuzzy) of individual landforms. The former is mainly driven by physical properties such as elevation, slope, and curvature. The latter, however, must consider the cognitive (human) reasoning that discriminates individual landforms in addition to these physical properties (Arundel and Sinha 2018). Both mapping forms are important. General geomorphometry is needed to understand geological and ecological processes and as boundary layer input to climate and environmental models. Specific geomorphometry supports such activities as disaster management and recovery, emergency response, transportation, and navigation. In the United States, individual landforms of interest are named in the U.S. Geological Survey (USGS) Geographic Names Information System, a point dataset captured specifically to digitize geographic names from the USGS Historical Topographic Map Collection (HTMC). Named landform extent is represented only by the name placement in the HTMC. Recent work has investigated CNN-based deep learning methods to capture these extents in machine-readable form. These studies first relied on physical properties (Arundel et al. 2020) and then included the HTMC as a band in RGB images in limited testing (Arundel et al. 2023). Results from the HTMC dataset surpassed those using just physical properties and using the HTMC alone performed best due to the hillshading and elevation (contour) data incorporated into the topographic maps. However, results fell short of an operational capacity to map all named landforms in the United States. Thus, our current work expands upon past research by focusing on the HTMC and physical information as inputs and the named landform label extents. Specifically, we propose to leverage pre-trained foundation models for segmentation and optical character recognition (OCR) models to jointly map landforms in the United States. Our approach aims to bridge the disparities among the independent information sources to facilitate informed decision-making. The modeling pipeline performs (1) segmentation using the physical information and (2) information extraction using OCR, in parallel. Then a computer vision approach merges the two branches into a labeled segmentation.

machine learning↗

NASA NTRSDate not supplied

Advancing Open Source Science Initiatives Through Public-Private Partnerships

Collaboration is fundamental to advancing open science within the science community. With the recent developments in technology and research, the establishment of formal partnerships between the private sector and government agencies are needed to bridge the knowledge gaps and expedite the time to actionable science. NASA’s Interagency Implementation and Advanced Concepts Team (IMPACT) seeks to address this challenge by establishing non-reimbursable Space Act Agreements with industry leaders in cloud computing, artificial intelligence (AI) and machine learning. The purpose of these agreements is to advance open source science initiatives in the areas of data discovery, access and use of high value NASA science data sets on the cloud. As well as, jointly work on common research problems to accelerate the development and adoption of new AI technologies. Current success stories include co-locating NASA datasets from multiple science disciplines on one platform using Amazon Web Services Open Data Registry, developing AI Foundation Models for Science with IBM and co-hosting training workshops and tutorials for the science community aimed at providing hands-on experience with using NASA data and AI models on the cloud. In summary, we will present an overview of our partnerships supporting open source science initiatives, describe current activities and lessons learned that may be useful to others considering similar partnerships with the private sector.

Elizabeth Fancher↗

NASA NTRS2024

Artificial Intelligence Workshop Report

The 4th NASA Science Mission Directorate (SMD) Artificial Intelligence (AI) Workshop, held during March 25-27, 2024, in Huntsville, AL, highlighted the significant potential of AI and machine learning (ML) in scientific research and processes. The workshop, supported by the NASA Office of Chief Science Data Officer (OCSDO), emphasized the critical role of foundation models (FMs) and large language models (LLMs) in advancing scientific disciplines. The event brought together domain scientists, computer scientists, AI experts, program managers, program scientists, and industry partners to address key challenges and explore opportunities in applying these advanced technologies.

Manil Maskey↗

NASA NTRSDate not supplied

Advancing Wildfire Monitoring with TEMPO and ML tools: Hourly Smoke and Fire‑Front Mapping and Near‑Surface NO₂ Predictions

Wildfires impose substantial impacts on communities and regions downwind of wildfire smoke. We present a TEMPO‑enabled workflow that generates value‑added Level‑3 smoke‑plume masks and fire‑front maps for large wildfires, such as 2024 Park Fire, using the self‑supervised deep learning system SIT‑FUSE, along with near‑surface NO₂ predictions produced by a foundation model (Microsoft Aurora). We conclude by outlining a roadmap for expanding these capabilities to additional Western U.S. wildfire events and for delivering actionable tools to stakeholders. This open-source, reproducible workflow provides a scalable framework for cross-agency wildfire monitoring to overcome traditional limitations in smoke-cloud discrimination and air-quality forecasting by incorporating TEMPO data and beyond.

Xiaohua Pan↗

NASA NTRS1986

Theoretical Foundation for Weld Modeling

Differential equations describe physics of tungsten/inert-gas and plasma-arc welding in aluminum. Report collects and describes necessary theoretical foundation upon which numerical welding model is constructed for tungsten/inert gas or plasma-arc welding in aluminum without keyhole. Governing partial differential equations for flow of heat, metal, and current given, together with boundary conditions relevant to welding process. Numerical estimates for relative importance of various phenomena and required properties of 2219 aluminum included

Traugott, S.↗

NASA NTRS2024

Out of Plane Stresses Using a Beam-Based Elastic Foundation Bonded Joint Model

Efficient, semi-analytical models of bonded joints are utilized for initial joint design and sizing due to their convenience and speed. Many of these models are based on beam and plate theories, which are most applicable when in-plane stresses/strains dominate. While this can be a sensible assumption for obtaining the deformations of the joint, out of plane (transverse) stresses are often of critical importance for failure in bonded joints composed of fiber reinforced composite materials. Since through-thickness strains are neglected in these semi-analytical theories, the present study shows a method to back-calculate the out of plane stresses based on stress balance equations, which use the in-plane stresses and displacements that have already been obtained. The method was implemented within the Joint Element Designer (JED) software, and verification and validation examples have been shown. As is typical for these types of models, corner stress singularities are not captured by the JED model, but the general trends of the stress and deformation fields agreed with a higher fidelity finite element model. As such, this method for obtaining out of plane stresses, when coupled with the JED code, is very useful for the complete stress analysis of bonded composite joints.

composites↗