SEARCH · Search NASA

Results for “COMPUTER 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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 523 records · Page 29

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Backpropagation-based learning with local derivative approximation and memory replay in biologically plausible neural systems

When learning, the brain modifies individual synaptic connections to reach a desired behavior. Animal and human brains have been shown to be incredibly capable of learning complex and varied functions across a wide variety of tasks. In recent years, artificial neural networks, inspired by human and animal brains, have shown great capabilities in learning a wide variety of difficult tasks. However, artificial neural networks primarily teach themselves through the use of backpropagation, a learning method which has no clear analogue within the brain. Additionally, Artificial Neural Networks primarily use continuous activation functions, which differ significantly from the spiking neuronal behavior present in the brain. In this paper, we discuss and demonstrate a biologically plausible learning method that approximates backpropagation through two techniques on Spiking Neural Networks. First, we show that the local temporal derivatives that are necessary for backpropagation can be approximately recovered through reconstruction using spike timings. Second, we show that through learning during a sleep phase, inspired by neuroscience research into memory replay, the localized parallel feedback path can learn to approximate the derivative through the forward path weight matrix, thus solving the weight transport problem. Lastly, we demonstrate that the combination of these two methods can approach or exceed the accuracy of backpropagation-based methods for a variety of neuromorphic vision tasks while maintaining biological plausibility.

42 ENGINEERING↗

DOE OSTI2025

ROOT RNTuple and EOS: The Next Generation of Event Data I/O

For several years, the ROOT team is developing the new RNTuple I/O subsystem in preparation of the next generation of collider experiments. Both HL-LHC and DUNE are expected to start data taking by the end of this decade. They pose unprecedented challenges to event data I/O in terms of data rates, event sizes, and event complexity. At the same time, the I/O landscape is becoming more diverse. HPC cluster file systems and object stores, NVMe disk cache layers in analysis facilities, and S3 storage on cloud resources are mixing with traditional XRootD-managed spinning disk pools.The ROOT team will finalize a first production version of the RNTuple binary format by the end of 2024. After this point, ROOT will provide backward compatibility for RNTuple data. This contribution provides an overview of the RNTuple feature set, the related R&D activities and the long-term vision for RNTuple. We report on performance, interface design, tooling, robustness, integration with experiment frameworks, and validation results, as well as recent R&D on parallel reading and writing and exploitation of modern hardware and storage systems. We will give an outlook on possible future features after a first production release.Collaboratively, the IT and EP departments at CERN have launched a formal project within the Research and Computing sector to evaluate the novel data format for physics analysis data utilized in LHC experiments and other fields. This part of the project focuses on validating the scalability of the EOS storage backend during the transition from the over 25 years old TTree production format to the newly developed RNTuple format, using both replicated and erasure-coded storage profiles.

Blomer, Jakob [CERN]↗

DOE OSTI2020

Management of Risk and Uncertainty Through Optimized Co-Operation of Transmission Systems and Microgrids With Responsive Loads (Final Report)

The evolution of the power system to the reliable, efficient and sustainable system of the future will involve development of both demand- and supply-side technology and operations. Ambitious national and state-level goals around the decarbonization of electricity relies on the integration of very high levels of renewable resources, most of which are variable and intermittent. The use of demand response is an ideal approach to counterbalance the intermittency of renewable generation and brings the consumer into the spotlight. Though individual consumers are interconnected at the low-voltage distribution system, these resources are typically modeled as variables at the transmission network level. Demand-side participation cannot be leveraged effectively without explicitly including the distribution system dynamics in the optimization-based wholesale market operations. This project grew from a vision for co-optimized interaction of distribution systems, or microgrids, with the high-voltage transmission system. In this framework, microgrids encompass consumers, distributed renewables and storage. The energy management system of the lower voltage system (distribution or microgrid) can also sell (buy) excess (necessary) energy from the transmission system. Until recently, very little research had been conducted on the co-optimization of these two systems due to computational limitations. However, advances in computational capabilities, and the judicious use of decomposition methods and innovative approximation methods for high-dimension dynamic programming made this goal a viable objective for this project, leading to a fundamental shift in the ability to integrate and fully utilize demand-side resources. To this end, the modeling framework developed introduces a novel co-optimization framework, to include the operations of both the transmission and distribution systems (or microgrids) in operational decision making. This framework was used to analyze renewable and distributed generation along with responsive demand and to compare the capability of co-optimized systems to perform with higher levels of variable renewables. An ideal microgrid is defined as an electric entity capable of operating in both interconnected (with the high-voltage grid) and islanded mode. As such, the microgrid should incorporate generating units (traditional units and intermittent) and if needed, exchange power with the high-voltage grid. The interplay between the microgrid and high-voltage grid motivated the development of the co-optimization approach to ensure efficient performance of the interconnected network. Results show that the use of a bi-level optimization approach is an appropriate structure, capable of co-optimizing a transmission system with multiple distribution systems and microgrids. While increasing the number of connected systems provides increasing flexibility for renewables integration this can also the economic benefits to the low-voltage subsystems with each additional system connected. Comparison of a traditional single-level decision structure with the co-optimization approach illustrates a reduction in overall system cost under co-optimization, while specific cost allocations to transmission and distribution systems are changed.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2025

Accelerated data-driven materials science with the Materials Project

The Materials Project was launched formally in 2011 to drive materials discovery forwards through high-throughput computation and open data. More than a decade later, the Materials Project has become an indispensable tool used by more than 600,000 materials researchers around the world. This Perspective describes how the Materials Project, as a data platform and a software ecosystem, has helped to shape research in data-driven materials science. We cover how sustainable software and computational methods have accelerated materials design while becoming more open source and collaborative in nature. Next, we present cases where the Materials Project was used to understand and discover functional materials. We then describe our efforts to meet the needs of an expanding user base, through technical infrastructure updates ranging from data architecture and cloud resources to interactive web applications. Finally, we discuss opportunities to better aid the research community, with the vision that more accessible and easy-to-understand materials data will result in democratized materials knowledge and an increasingly collaborative community.

Horton, Matthew K↗

DOE OSTI2024

Transformer Neural Networks with Spatiotemporal Attention for Predictive Control and Optimization of Industrial Processes

In the context of real-time optimization and model predictive control of industrial systems, machine learning, and neural networks represent cutting-edge tools that hold promise for enhancing dynamic modeling. This work presents a novel transformer neural network architecture for real-time optimization and model predictive control. This network design includes a modified attention mechanism inspired by positional embedding attention from vision transformers and task-specific modifications to the input-output structure of the transformer’s decoder stack. Experiments were conducted using data from a 450 MW coal-fired power plant to evaluate this approach's effectiveness. The transformer neural network was compared with conventional recurrent models, including GRU and LSTM. The transformer exhibited a 6% increase in the R-squared (R2) value of predictions and an 83% reduction in mean squared error (MSE). Computation time was also reduced by 84% compared to conventional recurrent models.

Gallup, Ethan R.↗

DOE OSTI2023

LDRD 2022 Annual Report: Laboratory Directed Research and Development Program Activities

Each year, Brookhaven National Laboratory (BNL) is required to provide a report of its completed Laboratory Directed Research and Development Program (LDRD) projects to the Department of Energy (DOE) Office of Scientific and Technical Information in accordance with DOE Order 413.2C Chg1 (MinChg) dated August 2, 2018. This report provides a detailed look at the scientific and technical activities for each of the LDRD projects funded by BNL in FY 2022, in fulfillment of that requirement. In FY 2022, the BNL LDRD Program funded 70 projects, 30 of which were new starts, at a total cost of $17.2M. The investments that BNL makes in its LDRD program support the Laboratory’s strategic goals. BNL has identified seven scientific initiatives that define the Laboratory’s scientific future and that will enable it to realize its overall vision. This requires simultaneous excellence in all aspects of BNL’s work – from science and operations, to external partnerships with the local, state, and national communities, and beyond. This is enabled by safe, efficient, and secure operations; by an unwavering commitment to a diverse, equitable, and inclusive environment, including workforce development, both with staff and reaching out to the community; and by a strong focus on renewed infrastructure. The seven scientific initiatives are: 1) Nuclear Physics: uncover the structure of visible matter by constructing and operating the Electron-Ion Collider at BNL to maintain international leadership in nuclear physics for decades; 2) Clean Energy and Climate: support a net-zero U.S. economy through fundamental research in basic energy and climate sciences to revolutionize grid-scale storage, renewable integration, and the study of atmospheric processes with a new facility to improve climate predictability; 3) Quantum Information Science and Technology: discover new quantum materials to enhance quantum computers and develop an entanglement sharing quantum network as a prototype for the first quantum internet; 4) Discovery Science Driven by the Human-AI Facility Integration: revolutionize the operation of experiments across the sciences at user facilities and in core programs; 5) High Energy Physics: understand the origin of space and time with the ATLAS high luminosity upgrade at CERN and the future Long Baseline Neutrino Facility/Deep Underground Neutrino Experiment; 6) Isotope Production: accelerate and expand isotope production to ensure the security of the Nation’s supply; 7) Accelerator Science and Technology: harness the cross-cutting accelerator science expertise at BNL to develop new facilities, improve and expand its user facilities, and promote the use of accelerators in industry. The funded projects support BNL’s seven scientific initiatives and priority programs as well as new areas of research and competencies at the Laboratory that are consistent with the Laboratory’s vision and mission. In total, these LDRD investments supported 43 postdoctoral researchers in whole or in part and resulted in 138 publications and 7 awards. This Program Activities Report represents the future of BNL science; it is an impressive body of exploratory work that investigates many scientific and technical directions in support of the DOE and BNL missions.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2025

Foveal vision reduces neural resources in agent-based game learning

Efficient processing of information is crucial for the optimization of neural resources in both biological and artificial visual systems. In this paper, we study the efficiency that may be obtained via the use of a fovea. Using biologically-motivated agents, we study visual information processing, learning, and decision making in a controlled artificial environment, namely the Atari Pong video game. We compare the resources necessary to play Pong between agents with and without a fovea. Our study shows that a fovea can significantly reduce the neural resources, in the form of number of neurons, number of synapses, and number of computations, while at the same time maintaining performance at playing Pong. To our knowledge, this is the first study in which an agent must simultaneously optimize its visual system, along with its decision making and action generation capabilities. That is, the visual system is integral to a complete agent.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2025

ORBIT-2 Dataset for Scaling Exascale Vision Foundation Models for Weather and Climate Downscaling

This dataset release corresponds to the work conducted in ORBIT-2: Scaling Exascale Vision Foundation Models for Weather and Climate Downscaling, where large-scale AI methods were applied to improve climate and weather resolution. The collection integrates four widely used, publicly available datasets: ERA5, PRISM, DAYMET, and IMERG. To prepare the data for ORBIT-2 model training and evaluation, we applied a preprocessing pipeline that generates paired low-resolution and high-resolution samples, enabling supervised downscaling experiments. The transformation from coarse to fine scales was performed using bilinear regridding, consistent with the procedures described in WeatherBench2, a community benchmark for weather and climate AI models. This dataset supports the development and evaluation of foundation models designed for weather and climate downscaling at exascale. Additional details on methodology and applications can be found in Wang et al., ORBIT-2 (arXiv:2505.04802, 2025).

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Running Ensemble Workflows at Extreme Scale: Lessons Learned and Path Forward

The ever-increasing volumes of scientific data combined with sophisticated techniques for extracting information from them have led to the increasing popularity of ensemble workflows which are a collection of runs of individual workflows. A traditional approach followed by scientists to run ensembles is to rely on simple scripts to execute different runs and manage resources. This approach is not scalable and is error-prone, thereby motivating the development of workflow management systems that specialize in executing ensembles on HPC clusters. However, when the size of both the ensemble and the target system reach extreme scales, existing workflow management systems face new challenges that hamper their efficient execution. In this paper, we describe our experience scaling an ensemble workflow from the computational biology domain from the early design stages to the execution at extreme scale on Summit, a leadership class supercomputer at the Oak Ridge National Laboratory. We discuss challenges that arise when scaling ensembles to several million runs on thousands of HPC nodes. We identify challenges with composition of the ensemble itself, its execution at large scale, post-processing of the generated data, and scalability of the file system. Based on the experience acquired, we develop a generic vision of the capabilities and abstractions to add to existing workflow management systems to enable the execution of ensemble workflows at extreme scales. We believe that the understanding of these fundamental challenges will help application teams along with workflow system developers with designing the next generation of infrastructure for composing and executing extreme-scale ensemble workflows.

Mehta, Kshitij↗

DOE OSTI2024

Spinning metasurface stack for spectro-polarimetric thermal imaging

Spectro-polarimetric imaging in the long-wave infrared (LWIR) region plays a crucial role in applications from night vision and machine perception to trace gas sensing and thermography. However, the current generation of spectro-polarimetric LWIR imagers suffers from limitations in size, spectral resolution, and field of view (FOV). While meta-optics-based strategies for spectro-polarimetric imaging have been explored in the visible spectrum, their potential for thermal imaging remains largely unexplored. In this work, we introduce an approach for spectro-polarimetric decomposition by combining large-area stacked meta-optical devices with advanced computational imaging algorithms. The co-design of a stack of spinning dispersive metasurfaces along with compressive sensing and dictionary learning algorithms allows simultaneous spectral and polarimetric resolution without the need for bulky filter wheels or interferometers. Our spinning-metasurface-based spectro-polarimetric stack is compact (<10×10×10cm) and robust, and it offers a wide field of view (20.5°). We show that the spectral resolving power of our system substantially enhances performance in machine learning tasks such as material classification, a challenge for conventional panchromatic thermal cameras. Our approach represents a significant advance in the field of thermal imaging for a wide range of applications including heat-assisted detection and ranging (HADAR).

Optics↗

DOE OSTI2023

Advancing the Theory of Nuclear Data Evaluations [Slides]

This presentation covers a few key topics in relation to advancing the theory of nuclear data evaluations. First it targets and highlights the NCSP mission and vision along with the five-year plan. This lecture focuses upon the uncertainty quantification and SAMMY modernization. This presentation covers theory, and a framework is provided. The presentation concludes with goals, outlooks, and accomplishments.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2025

A Data-Agnostic, Continuous Machine Learning Framework for Application in High Energy Physics and Beyond: Phase 1 Final Scientific/Technical Report

This Phase 1 effort has focused on the development of continual learning frameworks for use in machine learning, specifically in the applied context of High Energy Physics (HEP). Machine learning (ML) is a transformative technology by which computers, typically through the use of neural networks, are able to perform tasks with proficiency that rivals or surpasses that of human users. Model Degradation & Catastrophic Forgetting are two undesired phenomena which can occur in ML where the performance of a model degrades when either deployed on novel data streams, or trained on novel data which are sufficiently different than the data the models were initially trained on. A natural example where these sorts of effects can be observed is in the performance of detectors in harsh environments, where the detector signature may change over the lifetime of the detector as it ages and deteriorates — precisely what occurs in the experiments conducted in HEP. Real world HEP data is therefore an excellent test-ground and use-case for Continual Learning paradigms, which are techniques used in ML to counteract these problems. Ensemble learning is one such technique, where multiple smaller models are trained on subsets of the overall data and are ensembled together during inference. The intuition behind this technique is that, although there are shifts in the distributions which govern the incoming data streams, these shifts are not expected to be homogeneous or global. If a sufficient diversity in solutions within the various sub-models has been achieved, then at least one sub-model is expected to retain its performance within the overall ensemble. One further strength of this approach is that the architectures of the various models do not need to be identical, and in fact even different modalities of data can naturally be combined in this way. This work focused on applying ensemble learning techniques to derive results using two main datasets, anomaly detection in HEP data & time-series forecasting in semiconductor manufacturing data. Semiconductor manufacturing involves data with surprising similarity to that of HEP (e.g. wafer maps look very similar to digi-occupancy maps) and Cerium Lab’s prominence within the semiconductor industry makes semiconductor manufacturing a natural opportunity for commercialization of this work. Our efforts have led to two strong results. The first is that we evaluated the proposed ensembling techniques using previously proposed machine learning architectures for use in anomaly detection, namely AutoEncoder based models and their derivatives. We also developed new architectures which have not been evaluated in this context before. In fact, this work marks the first use of Vision Transformers for anomaly detection in HEP. Second, we demonstrated that ensemble learning significantly improves model performance in scenarios prone to degradation, validating its effectiveness across both HEP and semiconductor datasets. These results further support ensemble learning as a powerful strategy for mitigating catastrophic forgetting and maintaining robust performance in evolving data environments.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

LCLS Big Data Handling – How I Learned to Stop Worrying and Love the Data Deluge

Advanced data and computing systems are vital to Linac Coherent Light Source (LCLS) operations, data interpretation and overall scientific productivity. The transition to MHz-era operation marks a fundamental change in scale that requires new infrastructure and architectures to link LCLS to the required scale of computing needed for scientific interpretation. The LCLS-II Data System meets big data challenges by implementing configurable data reduction that can adapt to multiple science areas, real-time analysis frameworks to provide visualization and fast feedback, and the ability to transfer data to local and remote computational facilities for near real time analysis at the appropriate scale. Feature extracted information generated in the data analysis pipeline - at the edge, local compute, or remote High-Performance Computing (HPC) resources - can be used to steer experiments and inform user decisions during beam time. Artificial Intelligence and Machine Learning (AI/ML) techniques present new opportunities to rapidly analyse large datasets and direct experiments, but create new challenges in scaling, adaptability, complexity, and trustworthiness. We describe how the LCLS-II Data System architecture addresses its data-driven challenges in the areas of data acquisition, data processing, data management, and workflow orchestration to decrease the overall time-to-science and provide a vision for future developments.

artificial intelligence↗

DOE OSTI2025

Resolving femtosecond photoinduced energy flow: capture of nonadiabatic reaction pathway topography and wavepacket dynamics from photoexcitation through the conical intersection seam (Final Technical Report)

The dynamics that take place within just tens to hundreds of femtoseconds following the absorption of light by a molecule can play a critical role in how the absorbed energy is directed, allowing it to be used for a specific function or dissipated harmlessly. The form of chemical change that occurs rapidly in these molecules is called a “nonadiabatic electronic transition.” Such transitions are known to mediate energy flow in natural biological systems such as the ultraviolet photoprotection mechanism of DNA and the first step of the human vision response. Understanding how these mechanisms work precisely may help scientists achieve controlled manipulation of solar energy or optical control of a wide range of energy management functions in artificial systems. Experimental methods, however, have not yet allowed a precisely resolved and complete measurement of nonadiabatic electronic transitions. This constitutes a major obstacle to progress in the field. For progress to occur that would inform a wide body of research aiming to efficiently harness the energy of light for practical purposes, it is especially important to benchmark computational models of the molecules undergoing these rapid changes with experimental measurements, in order to learn which models are accurate. With Dept. of Energy funding, we have made strong progress towards establishing a new optical method for experimentally detecting the full nonadiabatic electronic transition. This requires having coordinated pulses of light covering the visible through the mid-infrared range of the electromagnetic spectrum that last only ten femtoseconds. We have developed a new, relatively simple approach for generating such pulses of laser light, and have incorporated them into a time-resolved spectrometer for measuring rapid changes in molecules. These tools can provide the greater precision and new types of data that are needed to benchmark computational models of molecular change and thus to make progress in the field. Our tools were tested on graphene, an excellent solid-state sample for verifying the capabilities and limitations of our instrumentation. The investment made in these tools by the Dept. of Energy Office of Science will allow new fundamental scientific understanding of energy dynamics in molecules in future studies.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Machine learning materials properties with accurate predictions, uncertainty estimates, domain guidance, and persistent online accessibility

One compelling vision of the future of materials discovery and design involves the use of machine learning (ML) models to predict materials properties and then rapidly find materials tailored for specific applications. However, realizing this vision requires both providing detailed uncertainty quantification (model prediction errors and domain of applicability) and making models readily usable. At present, it is common practice in the community to assess ML model performance only in terms of prediction accuracy (e.g. mean absolute error), while neglecting detailed uncertainty quantification and robust model accessibility and usability. Here, we demonstrate a practical method for realizing both uncertainty and accessibility features with a large set of models. We develop random forest ML models for 33 materials properties spanning an array of data sources (computational and experimental) and property types (electrical, mechanical, thermodynamic, etc). All models have calibrated ensemble error bars to quantify prediction uncertainty and domain of applicability guidance enabled by kernel-density-estimate-based feature distance measures. All data and models are publicly hosted on the Garden-AI infrastructure, which provides an easy-to-use, persistent interface for model dissemination that permits models to be invoked with only a few lines of Python code. We demonstrate the power of this approach by using our models to conduct a fully ML-based materials discovery exercise to search for new stable, highly active perovskite oxide catalyst materials.

domain of applicability↗

DOE OSTI2025

Basic Research Needs for Inverse Methods for Complex Systems under Uncertainty [Brochure]

The four priority research directions outlined in this brochure represent a cohesive vision for advancing the science of inverse problems for complex systems under uncertainty. Together, they address the critical challenges of: discovering, exploiting, and preserving physical and problem structure; overcoming model limitations; integrating disparate, multimodal, and/or dynamic data; and tailoring the solution of inverse problems to downstream tasks. While each PRD focuses on a distinct aspect of inverse-problem research, their interconnected nature highlights the importance of a holistic approach that leverages progress across all areas to achieve transformative solutions. This agenda calls for research across mathematics, statistics, and computer science disciplines, which are guided and complemented by rapid advances in artificial intelligence, high-performance computing, and experimental facilities, to unlock new capabilities, maximize scientific impact, and meet the growing demands of inverse problems that arise across applications that are critical to DOE's mission.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Vision Foundation Models in Remote Sensing: A survey

Artificial intelligence (AI) technologies have profoundly transformed the field of remote sensing (RS), revolutionizing data collection, processing, and analysis. Traditionally reliant on manual interpretation and task-specific models, RS research has been significantly enhanced by the advent of foundation models (FMs)—large-scale pretrained AI models capable of performing a wide array of tasks with unprecedented accuracy and efficiency. This article provides a comprehensive survey of FMs in the RS domain. We categorize these models based on their architectures, pretraining datasets, and methodologies. Through detailed performance comparisons, we highlight emerging trends and the significant advancements achieved by those FMs. Additionally, we discuss technical challenges, practical implications, and future research directions, addressing the need for high-quality data, computational resources, and improved model generalization. Our research also finds that pretraining methods, particularly self-supervised learning (SSL) techniques like contrastive learning (CL) and masked autoencoders (MAEs), remarkably enhance the performance and robustness of FMs. This survey aims to serve as a resource for researchers and practitioners by providing a panorama of advances and promising pathways for the continued development and application of FMs in RS.

data models↗

DOE OSTI2020

A systems engineering vision for floating offshore wind cost optimization

The U.S. offshore wind resource potential is immense, and often in close proximity to densely-populated coastal load centers. In many U.S. coastal areas, water depths favor the deployment of floating over fixed-bottom offshore wind technology. Floating offshore wind plants have the potential to be cost-competitive with fixed-bottom installations, but because the technology has not yet been deployed at commercial scale, it is not clear when and with what configurations this potential cost parity can be achieved. This article first reviews the state of floating offshore wind technology and deployments to identify key gaps that must be addressed to bring down the overall cost of energy produced. The article then puts forth a long-term vision for a research program and design methodology that may be able to push floating wind plants toward a lower levelized cost of energy than fixed-bottom offshore wind. The method involves a fully integrated systems-engineering and techno-economic design approach to capture the complex interactions between the physics, manufacturing, installation, and operation of floating wind turbines to achieve transformational cost reductions. The approach also envisions multifidelity and uncertainty management strategies to examine the most robust and viable concepts in the design trade-space. To better focus the computational resources, engineering lessons learned from existing offshore wind systems and concept studies are used to develop a set of criteria that can be applied to prefilter candidate technology building blocks that have the greatest cost reduction potential.

17 WIND ENERGY↗