Search NASA⌕ Search

SEARCH · Search NASA

Results for “VISION”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Self-assembly of wood-based shape memory composites triggered by solar-thermal energy

Transporting and assembling large, complex structures poses significant challenges due to their size, geometry, and cost. Additionally, the installation sites are often inaccessible or hazardous for humans, necessitating self-assembling capabilities in these structures. To mitigate these challenges, we propose using 3D printing materials with shape memory effect (SME) for both transport and construction. This approach involves developing 3D modular components into flat sheets for easier transportation, and then self-assembling into 3D structures on-site using solar energy. To gain a deeper understanding of the factors influencing material memory performance, we have chosen a composite PLA/WF, which is polylactic acid (PLA) with 20 wt% wood flour (WF) for this purpose, leveraging its high tensile modulus at 0.966 GPa, low cost, and sustainability. Printed shapes with this material can maintain a recovery ratio over 90% after 3 cycles. While traditional composites fillers (e.g. glass or carbon fiber) are added to enhance mechanical and thermal properties, the addition of bio-based fillers like WF accomplish similar goals without compromising sustainability. We conducted multiple experiments to demonstrate how environmental conditions (i.e. temperature) maximize the material’s SME. Although still at an early stage, this study provides initial insights into bridging the gap between the small-scale nature of shape memory polymers (SMPs) and their potential for large-scale additive manufacturing, addressing a critical need for efficient and sustainable construction. In the long term, we hope our study contributes to the design vision of utilizing SMPs for transportation, assembly, and deployment of complex structures, providing a new pathway for sustainable construction and transportation of large-scale structures to hard-to-access locations such as disaster-affected areas and remote deserts, etc.

4D printing↗
DOE OSTI2024

Offshore Wind Energy Validation Experiment Hierarchy

This paper provides a summary of planning work for experiments that will be necessary to address the long-term model validation needs required to meet offshore wind energy deployment goals. Conceptual experiments are identified and laid out in a validation hierarchy for both wind turbine and wind plant applications. Instrumentation needs that will be required for the offshore validation experiments to be impactful are then listed. The document concludes with a nominal vision for how these experiments can be accomplished.

17 WIND ENERGY↗
DOE OSTI2025

Climate Nowcasting

The climate is changing so rapidly that climatologies based on historical statistics cannot reliably capture the current risk of extreme weather events hazardous to society. Decision relevant projections of weather extreme probability over the next 10–15 years are needed to enable adaptation and resilience in the face of this evolving risk. Current weather forecasts/predictions and long-term climate projections for decades into the future are inadequate for providing this information to stakeholders that need it, targeting forecast horizons either too short or too far into the future. We argue that a new approach is needed: climate nowcasting. Climate nowcasting would focus on user-inspired extreme metrics over the next 10–15 year time frame, targeting specific impacts and locations down to a local scale, by engaging with stakeholders to understand their needs and provide information in a format relevant for decision making. Importantly, climate nowcasting will not consist of a single approach or data set, involving rather data fusion from different sources of information: simulations, observations and data driven methods, likely through different weights depending on the metric of interest. Predictions must be accompanied by serious engagement with stakeholders facing climate risks, and clearly present uncertainties and limitations of any prediction. Such a vision is very different from how typical climate or weather forecasts are applied today.

Gettelman, Andrew↗
DOE OSTI2025

Data imbalance in drug response prediction: multi-objective optimization approach in deep learning setting

Abstract Drug response prediction (DRP) methods tackle the complex task of associating the effectiveness of small molecules with the specific genetic makeup of the patient. Anti-cancer DRP is a particularly challenging task requiring costly experiments as underlying pathogenic mechanisms are broad and associated with multiple genomic pathways. The scientific community has exerted significant efforts to generate public drug screening datasets, giving a path to various machine learning models that attempt to reason over complex data space of small compounds and biological characteristics of tumors. However, the data depth is still lacking compared to application domains like computer vision or natural language processing domains, limiting current learning capabilities. To combat this issue and improves the generalizability of the DRP models, we are exploring strategies that explicitly address the imbalance in the DRP datasets. We reframe the problem as a multi-objective optimization across multiple drugs to maximize deep learning model performance. We implement this approach by constructing Multi-Objective Optimization Regularized by Loss Entropy loss function and plugging it into a Deep Learning model. We demonstrate the utility of proposed drug discovery methods and make suggestions for further potential application of the work to achieve desirable outcomes in the healthcare field.

Biochemistry & Molecular Biology↗
DOE OSTI2024

Current and future directions in network biology

Network biology is an interdisciplinary field bridging computational and biological sciences that has proved pivotal in advancing the understanding of cellular functions and diseases across biological systems and scales. Although the field has been around for two decades, it remains nascent. It has witnessed rapid evolution, accompanied by emerging challenges. These stem from various factors, notably the growing complexity and volume of data together with the increased diversity of data types describing different tiers of biological organization. We discuss prevailing research directions in network biology, focusing on molecular/cellular networks but also on other biological network types such as biomedical knowledge graphs, patient similarity networks, brain networks, and social/contact networks relevant to disease spread. In more detail, we highlight areas of inference and comparison of biological networks, multimodal data integration and heterogeneous networks, higher-order network analysis, machine learning on networks, and network-based personalized medicine. Following the overview of recent breakthroughs across these five areas, we offer a perspective on future directions of network biology. Additionally, we discuss scientific communities, educational initiatives, and the importance of fostering diversity within the field. This article establishes a roadmap for an immediate and long-term vision for network biology.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Direct Reactions and Spectroscopy with Hydrogen Targets at the RIBF

The implementation of MINOS, a liquid-hydrogen target and vertex-tracking system, at the Radioactive Isotope Beam Factory (RIBF) of the RIKEN Nishina Center has led to numerous nuclear physics discoveries. In experiments involving rare-isotope beams at intermediate energies, MINOS was integrated with the DALI2 + γ-ray spectrometer and other auxiliary devices. These were positioned in front of both the ZeroDegree Spectrometer and the SAMURAI magnetic spectrometers for the identification of ejectiles following quasi-free scattering direct reactions. Notable outcomes include the first spectroscopy of unstable doubly magic nucleus, investigations into multi-neutron decays, and the study of shell evolution along isotopic chains. These studies have significantly advanced our understanding of nuclear structure, as well as the mechanism of direct reactions with fast beams. Simultaneously, these direct reactions enabled raising new physics questions and paved the way for next-generation experimental setups to address them. Through the combination of particle- and γ-spectroscopy with a broad range of nuclear reactions, new frontiers will be established in our understanding of nuclear structure and reaction mechanisms. Here, we review the past ten years of studies on direct reactions and spectroscopy of exotic nuclei with liquid-hydrogen targets at the RIBF and define a vision for the future.

Bertulani, Carlos A [East Texas A & M Univ., Comme↗
DOE OSTI2025

Bats flying through a Y-maze are visually attracted to wind turbine surfaces

Wind energy’s rapid expansion has led to unintended consequences for wildlife, with migratory bats among the species most at risk. The behavioural mechanisms underlying collisions remain poorly understood, but one hypothesis is that bats are attracted to wind turbine structures. Vision is important to bat orientation and obstacle avoidance, yet it has been relatively understudied in the context of bat–turbine interactions. We hypothesize that light reflected off turbine surfaces could attract bats, acting as a sensory pollutant that may increase collision risk. To test whether reflective turbine surfaces elicit attraction, we flew 242 Lasiurus cinereus and 154 Lasionycteris noctivagans through Y-maze assays. Bats were at least twice as likely to fly towards white turbine blade sections compared to less reflective black ones. This attraction intensified when the alternative exit was a dark, empty flyway, with 74% of L. cinereus and 97% of L. noctivagans flying towards the white turbine blade. These findings provide evidence that visual sensory pollutants could underlie bat–turbine interactions, and if so, wind turbines could be ecological traps.

17 WIND ENERGY↗
DOE OSTI2024

Essay: Quantum Sensing with Atomic, Molecular, and Optical Platforms for Fundamental Physics

Atomic, molecular, and optical (AMO) physics has been at the forefront of the development of quantum science while laying the foundation for modern technology. With the growing capabilities of quantum control of many atoms for engineered many-body states and quantum entanglement, a key question emerges: what critical impact will the second quantum revolution with ubiquitous applications of entanglement bring to bear on fundamental physics? In this Essay, we argue that a compelling long-term vision for fundamental physics and novel applications is to harness the rapid development of quantum information science to define and advance the frontiers of measurement physics, with strong potential for fundamental discoveries. As quantum technologies, such as fault-tolerant quantum computing and entangled quantum sensor networks, become much more advanced than today’s realization, we wonder what doors of basic science can these tools unlock. We anticipate that some of the most intriguing and challenging problems, such as quantum aspects of gravity, fundamental symmetries, or new physics beyond the minimal standard model, will be tackled at the emerging quantum measurement frontier. . Published by the American Physical Society 2024

Physics↗
DOE OSTI2025

Discorpy : algorithms and software for camera calibration and correction

Camera or lens-based detector calibration is essential for spatial accuracy in applications like dimensional tomography, optical metrology, and computer vision. Many methods and software exist yet there is still a lack of approaches that achieve both high accuracy and robustness while being easy to use and capable of handling a wide range of distortions. Radial lens distortion is common in high-resolution X-ray detector optics used in parallel-beam tomography at synchrotrons. Achieving sub-pixel accuracy requires calibrating with an optical target image. Although methods for characterizing radial distortion are well established, acquired images often also include perspective distortion and optical center offset. Here, we present our approaches to individually characterize and correct both types of distortion using a single calibration image, implemented in the Discorpy software.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Tula: Optimizing Time, Cost, and Generalization in Distributed Large-Batch Training

Distributed training increases the number of batches processed per iteration either by scaling-out (adding more nodes) or scaling-up (increasing the batch-size). However, the largest configuration does not necessarily yield the best performance. Horizontal scaling introduces additional communication overhead, while vertical scaling is constrained by computation cost and device memory limits. Thus, simply increasing the batch-size leads to diminishing returns: training time and cost decrease initially but eventually plateaus, creating a knee-point in the time/cost vs. batch-size pareto curve. The optimal batch-size therefore depends on the underlying model, data and available compute resources. Large batches also suffer from worse model quality due to the well-known “generalization gap”. In this paper, we present Tula, an online service that automatically optimizes time, cost, and convergence quality for large-batch training of convolutional models. It combines parallel-systems modeling with statistical performance prediction to identify the optimal batchsize. Tula predicts training time and cost within 7.5−14% error across multiple models, and achieves up to 20× overall speedup and improves test accuracy by ≈9% on average over standard large-batch training on various vision tasks, thus successfully mitigating the generalization gap and accelerating training at the same time.

Tyagi, Sahil [ORNL] (ORCID:0009000783144745)↗
DOE OSTI2024

Workflow Provenance in the Computing Continuum for Responsible, Trustworthy, and Energy-Efficient AI

As Artificial Intelligence (AI) becomes more pervasive in our society, it is crucial to develop, deploy, and assess Responsible and Trustworthy AI (RTAI) models, i.e., those that consider not only accuracy but also other aspects, such as explainability, fairness, and energy efficiency. Workflow provenance data have historically enabled critical capabilities towards RTAI. Provenance data derivation paths contribute to responsible workflows through transparency in tracking artifacts and resource consumption. Provenance data are well-known for their trustworthiness helping explainability, reproducibility, and accountability. However, there are complex challenges to achieve RTAI, which are further complicated by the heterogeneous infrastructure in the computing continuum (Edge-Cloud-HPC) used to develop and deploy models. As a result, a significant research and development gap remains between workflow provenance data management and RTAI. In this paper, we present a vision of the pivotal role of workflow provenance in supporting RTAI and discuss related challenges. We present a schematic view between RTAI and provenance, and highlight open research directions.

Santos Souza, Renan↗
DOE OSTI2025

Rapid Event Detection via Synchro-Waveform Based Temporal Attention Network in Distributed Grid

Compared with the information collected from phasor measurement units, synchro-waveforms contain high-fidelity disturbances of the grid, which can be a granular and authentic representation of measurements in the modern power system. However, the dynamic changing morphology makes it challenging to effectively capture various disturbance information from the synchro-waveforms. To tackle this issue, this paper proposes a Synchro-waveform based Temporal Attention (STA) network to achieve rapid event detection. First, a multi-scenario distributed model with renewable integration is established to generate synchro-waveforms under various uncertainties. Then, three typical temporal features are extracted directly from the synchro-waveform measurements. Additionally, the lightweight STA network is deployed to identify the most common event types in renewable energy systems via the self-attention based vision transformer module. The results from simulated experiments demonstrate that the proposed approach can achieve rapid and real-time detection within 0.81 ms and over 96.27 % accuracy.

Dong, Yuqing [University of Tennessee (UT)]↗
DOE OSTI2025

Constrained GAN-Generated X-Ray CT Data For Self-Supervised And Foundation-Model Segmentation Of Concrete Microstructures

Three-dimensional characterization of materials using X-ray computed tomography (XCT) is challenging due to the complexity of internal structures, noise, and variations in resolution. Traditional computer vision models often struggle to accurately segment these images, particularly in domain-specific applications like materials science. While supervised deep learning approaches have been developed to address the limitations of conventional algorithms, they typically require large amounts of labeled training data and often fail to generalize across different datasets. Self-supervised, few-and zero-shot learning methods have gained prominence in natural image processing and segmentation tasks, but their application to scientific imaging remains limited due to the unique structural complexity, noise, and textural artifacts present in materials science data. In this work, we investigate how domain adaptation, leveraging physics-based and GAN-generated synthetic data, impacts segmentation performance. We introduce a modified Contrastive Unpaired Translation (CUT) model designed to generate realistic labeled data, which can be used for training, pre-training, and fine-tuning segmentation models for real XCT microstructure data. We evaluate the performance of two segmentation approaches: a self-supervised network (SSL-ALPNet) and a foundation model (Segment Anything Model), assessing their improvements when pre-trained and/or fine-tuned on the synthesized data. Our results demonstrate that leveraging synthetic data significantly enhances segmentation performance, particularly in challenging materials science applications.

Ziabari, Amir [ORNL] (ORCID:000000034776457X)↗
DOE OSTI2024

Towards Diverse and Representative Global Pretraining Datasets for Remote Sensing Foundation Models

The design of a pretraining dataset is emerging as a critical component for the generality of foundation models. In the remote sensing realm, large volumes of imagery and benchmark datasets exist that can be leveraged to pretrain foundation models, however using this imagery in absence of a well-crafted sampling strategy is inefficient and has the potential to create biased and less generalizable models. Here, we provide a discussion and vision for the curation and assessment of pretraining datasets for remote sensing geospatial foundation models. We highlight the importance of geographic, temporal, and image acquisition diversity and review possible strategies to enable such diversity at global scale. In addition to these characteristics, support for various spatial-temporal pretext tasks within the dataset is also critical. Ultimately, our primary objective is to place emphasis on and draw attention to the data curation stage of the foundation model development pipeline. By doing so, we think it is possible to reduce biases of geospatial foundation models, as well as enable broader generalization to downstream remote sensing tasks and applications.

Arndt, Jacob↗
DOE OSTI2026

Pixel-Resolved Long-Context Learning for Turbulence at Exascale: Resolving Small-scale Eddies Toward the Viscous Limit

Turbulence plays a crucial role in multiphysics applications, including aerodynamics, fusion, and combustion. Accurately capturing turbulence's multiscale characteristics is essential for reliable predictions of multiphysics interactions, but remains a grand challenge even for exascale supercomputers and advanced deep learning models. The extreme-resolution data required to represent turbulence, ranging from billions to trillions of grid points, pose prohibitive computational costs for models based on architectures like vision transformers. To address this challenge, we introduce a multiscale hierarchical Turbulence Transformer that reduces sequence length from billions to a few millions and a novel RingX sequence parallelism approach that enables scalable long-context learning. We perform scaling and science runs on the Frontier supercomputer. Our approach demonstrates excellent performance up to 1.1 EFLOPS on 32,768 AMD GPUs, with a scaling efficiency of 94\%. To our knowledge, this is the first AI model for turbulence that can capture small-scale eddies down to the dissipative range in three-dimensional turbulence at high Reynolds numbers.

Yin, Junqi [ORNL] (ORCID:0000000338435520)↗
DOE OSTI2026

Supervised Learning-Based Spatial Position Estimation with Vertical Displacement for Hovering UAV Wireless Power Transfer

This study presents a supervised learning-based spatial position estimation approach for wireless power transfer (WPT) systems supporting hovering unmanned aerial vehicle (UAV) charging. Unlike stationary charging scenarios, hovering UAVs introduce continuous lateral misalignment and vertical displacement, leading to variations in magnetic coupling and reduced power transfer efficiency. To address this challenge, the proposed method estimates the relative spatial position of the receiver coil using only electrical measurements obtained at the secondary side. A supervised learning model is trained to map output voltage and current features to spatial coordinates, enabling position awareness without requiring external sensors, vision systems, or communication links. The sensing functionality is inherently integrated into the WPT system, allowing simultaneous power transfer and localization through the same magnetic interface. Experimental validation is conducted on a laboratory-scale prototype under varying lateral offsets and air-gap conditions. In addition, spline-based interpolation is employed to increase spatial data density for training. The results demonstrate that the proposed framework can capture spatial variations associated with both lateral and vertical displacement, providing reliable position estimation under hovering conditions. This work establishes a hardware-efficient, sensorless solution for UAV wireless charging and serves as a baseline for advanced data-driven position estimation methods in dynamic WPT systems.

Asa, Erdem [ORNL] (ORCID:0000000190884812)↗
DOE OSTI2025

Designing Future Energy Systems with Generative AI

Energy systems are experiencing various changes that impact the distribution, use, and reliability of energy. Local utilities and municipalities must respond and adapt to these changes, moving towards a future energy system with modernized infrastructure and other targeted investments and policy decisions. However, planning for and enacting these advancements requires significant effort from experts and engineers to develop strategies that ensure a reliable and secure energy future. This includes characterizing the current energy infrastructure, identifying areas for development, and engaging with local community members. Emerging generative artificial intelligence techniques can alleviate pain points and help support the development of the next generation of energy systems. Here, in this article, we highlight on-going generative AI work in the areas of atmospheric modeling, building energy management, and distribution network design, and we propose a vision for the role of generative AI that considers opportunities and identifies challenges inherent to this technology.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2024

Transformer Masked Autoencoders for RF Device Fingerprinting

Machine learning methods for RF device fingerprinting typically rely on CNN-based models. Transformer-based models have outperformed CNNs for modulation classification tasks, but there are few implementations for device fingerprinting. We train a transformer for device fingerprinting with the largest device count to date and explore several variations of the architecture. Additionally, we demonstrate that pre-training an RF transformer as a Masked Autoencoder improves classification accuracy, as has been observed for CNN fingerprinting models and vision transformers.

artificial intelligence↗