Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Deep Operator Networks”

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 1,009 records · Page 56
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Runway Sign Classifier: A DAL C Certifiable Machine Learning System

In recent years, the remarkable progress of Machine Learning (ML) technologies within the domain of Artificial Intelligence (AI) systems has presented unprecedented opportunities for the aviation industry, paving the way for further advancements in automation, including the potential for single pilot or fully autonomous operation of large commercial airplanes. However, ML technology faces major incompatibilities with existing airborne certification standards, such as ML model traceability and explainability issues or the inadequacy of traditional coverage metrics. Certification of ML-based airborne systems using current standards is problematic due to these challenges. This paper presents a case study of an airborne system utilizing a Deep Neural Network (DNN) for airport sign detection and classification. Building upon our previous work, which demonstrates compliance with Design Assurance Level (DAL) ”D”, we upgrade the system to meet the more stringent requirements of Design Assurance Level ”C”. To achieve DAL C, we employ an established architectural mitigation technique involving two redundant and dissimilar Deep Neural Networks. The application of novel ML-specific data management techniques further enhances this approach. This work is intended to illustrate how the certification challenges of ML-based systems can be addressed for medium criticality airborne applications.

Flight Software↗
DOE OSTI2021

IoT Intrusion Detection Taxonomy, Reference Architecture, and Analyses

This paper surveys the deep learning (DL) approaches for intrusion-detection systems (IDSs) in Internet of Things (IoT) and the associated datasets toward identifying gaps, weaknesses, and a neutral reference architecture. A comparative study of IDSs is provided, with a review of anomaly-based IDSs on DL approaches, which include supervised, unsupervised, and hybrid methods. All techniques in these three categories have essentially been used in IoT environments. To date, only a few have been used in the anomaly-based IDS for IoT. For each of these anomaly-based IDSs, the implementation of the four categories of feature(s) extraction, classification, prediction, and regression were evaluated. We studied important performance metrics and benchmark detection rates, including the requisite efficiency of the various methods. Four machine learning algorithms were evaluated for classification purposes: Logistic Regression (LR), Support Vector Machine (SVM), Decision Tree (DT), and an Artificial Neural Network (ANN). Therefore, we compared each via the Receiver Operating Characteristic (ROC) curve. The study model exhibits promising outcomes for all classes of attacks. The scope of our analysis examines attacks targeting the IoT ecosystem using empirically based, simulation-generated datasets (namely the Bot-IoT and the IoTID20 datasets).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Universal linear intensity transformations using spatially incoherent diffractive processors

Abstract Under spatially coherent light, a diffractive optical network composed of structured surfaces can be designed to perform any arbitrary complex-valued linear transformation between its input and output fields-of-view (FOVs) if the total number ( N ) of optimizable phase-only diffractive features is ≥~2 N i N o , where N i and N o refer to the number of useful pixels at the input and the output FOVs, respectively. Here we report the design of a spatially incoherent diffractive optical processor that can approximate any arbitrary linear transformation in time-averaged intensity between its input and output FOVs. Under spatially incoherent monochromatic light, the spatially varying intensity point spread function ( H ) of a diffractive network, corresponding to a given, arbitrarily-selected linear intensity transformation, can be written as H ( m , n ; m ′, n ′) = | h ( m , n ; m ′, n ′)| 2 , where h is the spatially coherent point spread function of the same diffractive network, and ( m , n ) and ( m ′, n ′) define the coordinates of the output and input FOVs, respectively. Using numerical simulations and deep learning, supervised through examples of input-output profiles, we demonstrate that a spatially incoherent diffractive network can be trained to all-optically perform any arbitrary linear intensity transformation between its input and output if N ≥ ~2 N i N o . We also report the design of spatially incoherent diffractive networks for linear processing of intensity information at multiple illumination wavelengths, operating simultaneously. Finally, we numerically demonstrate a diffractive network design that performs all-optical classification of handwritten digits under spatially incoherent illumination, achieving a test accuracy of >95%. Spatially incoherent diffractive networks will be broadly useful for designing all-optical visual processors that can work under natural light.

36 MATERIALS SCIENCE↗
NASA NTRS1972

Precision time distribution within a deep space communications complex

The Precision Time Distribution System (PTDS) at the Golstone Deep Space Communications Complex is a practical application of existing technology to the solution of a local problem. The problem was to synchronize four station timing systems to a master source with a relative accuracy consistently and significantly better than 10 microseconds. The solution involved combining a precision timing source, an automatic error detection assembly and a microwave distribution network into an operational system. Upon activation of the completed PTDS two years ago, synchronization accuracy at Goldstone (two station relative) was improved by an order of magnitude. It is felt that the validation of the PTDS mechanization is now completed. Other facilities which have site dispersion and synchronization accuracy requirements similar to Goldstone may find the PTDS mechanization useful in solving their problem. At present, the two station relative synchronization accuracy at Goldstone is better than one microsecond.

Curtright, J. B.↗
NASA NTRS2003

Recent results in Mars Relay Network planning and scheduing

At different time periods in the future, Mars missions will overlap and previous studies indicate that during such periods existing deep space communication infrastructure will not be able to handle all Mars communication needs. A plausible solution is to perform optimal resource allocation for the Mars relay communication network; a network consisting of multiple surface units and orbiters on Mars and the Deep Space Stations. Unlike direct-to-earth, a relay communication, either in real-time or store-and-forward, can increase network science data return, reduce surface unit's direct-to-earth communication demands, and enable communication even when the surface unit is not facing Earth. It is the objective of this paper to take advantage of the relay operation to efficiently plan and schedule the network communications.

Relay Network planning and scheduling constrained ↗
NASA NTRS2008

Rapid Turnaround of Costing/Designing of Space Missions Operations

The Ground Segment Team (GST), at NASA's Jet Propulsion Laboratory in Pasadena, California, provides high-level mission operations concepts and cost estimates for projects that are in the formulation phase. GST has developed a tool to track costs, assumptions, and mission requirements, and to rapidly turnaround estimates for mission operations, ground data systems, and tracking for deep space and near Earth missions. Estimates that would often take several weeks to generate are now generated in minutes through the use of an integrated suite of cost models. The models were developed through interviews with domain experts in areas of Mission Operations, including but not limited to: systems engineering, payload operations, tracking resources, mission planning, navigation, telemetry and command, and ground network infrastructure. Data collected during interviews were converted into parametric cost models and integrated into one tool suite. The tool has been used on a wide range of missions from small Earth orbiters, to flagship missions like Cassini. The tool is an aid to project managers and mission planners as they consider different scenarios during the proposal and early development stages of their missions. The tool is also used for gathering cost related requirements and assumptions and for conducting integrated analysis of multiple missions.

Cost estimation↗
DOE OSTI2024

Leveraging generative AI for urban digital twins: a scoping review on the autonomous generation of urban data, scenarios, designs, and 3D city models for smart city advancement

The digital transformation of modern cities by integrating advanced information, communication, and computing technologies has marked the epoch of data-driven smart city applications for efficient and sustainable urban management. Despite their effectiveness, these applications often rely on massive amounts of high-dimensional and multi-domain data for monitoring and characterizing different urban sub-systems, presenting challenges in application areas that are limited by data quality and availability, as well as costly efforts for generating urban scenarios and design alternatives. As an emerging research area in deep learning, Generative Artificial Intelligence (GenAI) models have demonstrated their unique values in content generation. This paper aims to explore the innovative integration of GenAI techniques and urban digital twins to address challenges in the planning and management of built environments with focuses on various urban sub-systems, such as transportation, energy, water, and building and infrastructure. The survey starts with the introduction of cutting-edge generative AI models, such as the Generative Adversarial Networks (GAN), Variational Autoencoders (VAEs), Generative Pre-trained Transformer (GPT), followed by a scoping review of the existing urban science applications that leverage the intelligent and autonomous capability of these techniques to facilitate the research, operations, and management of critical urban subsystems, as well as the holistic planning and design of the built environment. Based on the review, we discuss potential opportunities and technical strategies that integrate GenAI models into the next-generation urban digital twins for more intelligent, scalable, and automated smart city development and management.

3D city modeling↗
DOE OSTI2022

Variational encoder geostatistical analysis (VEGAS) with an application to large scale riverine bathymetry

Estimation of riverbed profiles, also known as bathymetry, plays a vital role in many applications, such as safe and efficient inland navigation, prediction of bank erosion, land subsidence, and flood risk management. The high cost and complex logistics of direct bathymetry surveys, i.e, depth imaging, have encouraged the use of indirect measurements such as surface flow velocities. However, estimating high-resolution bathymetry from indirect measurements is an inverse problem that can be computationally challenging. Here, we propose a reduced-order model (ROM) based approach that utilizes a variational autoencoder (VAE), a type of deep neural network with a narrow layer in the middle, to compress bathymetry and flow velocity information and accelerate bathymetry inverse problems from flow velocity measurements. In our application, the shallow-water equations (SWE) with appropriate boundary conditions (BCs), e.g., the discharge and/or the free surface elevation, constitute the forward problem, to predict flow velocity. Then, ROMs of the SWEs are constructed on a nonlinear manifold of low dimensionality through a variational encoder and the bathymetry inversion problem is derived on the low-dimensional latent space in a Hierarchical Bayesian setting. Further, the reformulation allows variational inference with a small number (e.g., $\mathscr{O}$ (100) of ROM runs and efficient uncertainty quantification. We have tested our inversion approach on a one-mile reach of the Savannah River, GA, USA. Once the neural network is trained (offline stage), the proposed technique can perform the inversion operation orders of magnitude faster than traditional inversion methods that are commonly based on linear projections, such as principal component analysis (PCA), or the principal component geostatistical approach (PCGA). Furthermore, tests show that the algorithm can estimate the bathymetry with good accuracy even with sparse flow velocity measurements.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

PANTHER: A Programmable Architecture for Neural Network Training Harnessing Energy-Efficient ReRAM

The wide adoption of deep neural networks has been accompanied by ever-increasing energy and performance demands due to the expensive nature of training them. Additionally, numerous special-purpose architectures have been proposed to accelerate training: both digital and hybrid digital-analog using resistive RAM (ReRAM) crossbars. ReRAM-based accelerators have demonstrated the effectiveness of ReRAM crossbars at performing matrix-vector multiplication operations that are prevalent in training. However, they still suffer from inefficiency due to the use of serial reads and writes for performing the weight gradient and update step. A few works have demonstrated the possibility of performing outer products in crossbars, which can be used to realize the weight gradient and update step without the use of serial reads and writes. However, these works have been limited to low precision operations which are not sufficient for typical training workloads. Moreover, they have been confined to a limited set of training algorithms for fully-connected layers only. To address these limitations, we propose a bit-slicing technique for enhancing the precision of ReRAM-based outer products, which is substantially different from bit-slicing for matrix-vector multiplication only. We incorporate this technique into a crossbar architecture with three variants catered to different training algorithms. To evaluate our design on different types of layers in neural networks (fully-connected, convolutional, etc.) and training algorithms, we develop PANTHER, an ISA-programmable training accelerator with compiler support. Our design can also be integrated into other accelerators in the literature to enhance their efficiency. Our evaluation shows that PANTHER achieves up to 8.02×, 54.21×, and 103× energy reductions as well as 7.16×, 4.02×, and 16× execution time reductions compared to digital accelerators, ReRAM-based accelerators, and GPUs, respectively.

42 ENGINEERING↗
NASA NTRS1991

Calorimetric output power measurements on a CW 20 kW 7.16 GHz microwave transmitter

A calorimetric measurement technique developed for NASA's Deep Space Network (DSN) transmitters that does not require data on the coolant's thermal parameters is described. Calibration of the measurement system is achieved by measuring the DC input power to the klystron and relating coolant temperature increases to this known power dissipation. Agreement between calorimetric and electrical measurements of total system power was good, the difference being less than 2 percent. The operation of the system was not greatly affected by the composition of the coolant, which was varied from pure water to 40 percent ethylene glycol by mass. Good accuracy was also shown at output power levels, which varied over a 3.6:1 range.

Perez, Raul M.↗
NASA NTRS1999

Architectural Design for a Mars Communications and Navigation Orbital Infrastructure

The planet Mars has become the focus of an intensive series of missions that span decades of time, a wide array of international agencies and an evolution from robotics to humans. The number of missions to Mars at any one time, and over a period of time, is unprecedented in the annals of space exploration. To meet the operational needs of this exploratory fleet will require the implementation of new architectural concepts for communications and navigation. To this end, NASA's Jet Propulsion Laboratory has begun to define and develop a Mars communications and navigation orbital infrastructure. This architecture will make extensive use of assets at Mars, as well as use of traditional Earth-based assets, such as the Deep Space Network, DSN. Indeed, the total system can be thought of as an extension of DSN nodes and services to the Mars in-situ region. The concept has been likened to the beginnings of an interplanetary Internet that will bring the exploration of Mars right into our living rooms. The paper will begin with a high-level overview of the concept for the Mars communications and navigation infrastructure. Next, the mission requirements will be presented. These will include the relatively near-term needs of robotic landers, rovers, ascent vehicles, balloons, airplanes, and possibly orbiting, arriving and departing spacecraft. Requirements envisioned for the human exploration of Mars will also be described. The important Mars orbit design trades on telecommunications and navigation capabilities will be summarized, and the baseline infrastructure will be described. A roadmap of NASA's plan to evolve this infrastructure over time will be shown. Finally, launch considerations and delivery to Mars will be briefly treated.

Ceasrone R. J.↗
NASA NTRS2015

Orbit Determination for the Lunar Reconnaissance Orbiter Using an Extended Kalman Filter

Since launch, the FDF has performed daily OD for LRO using the Goddard Trajectory Determination System (GTDS). GTDS is a batch least-squares (BLS) estimator. The tracking data arc for OD is 36 hours. Current operational OD uses 200 x 200 lunar gravity, solid lunar tides, solar radiation pressure (SRP) using a spherical spacecraft area model, and point mass gravity for the Earth, Sun, and Jupiter. LRO tracking data consists of range and range-rate measurements from: Universal Space Network (USN) stations in Sweden, Germany, Australia, and Hawaii. A NASA antenna at White Sands, New Mexico (WS1S). NASA Deep Space Network (DSN) stations. DSN data was sparse and not included in this study. Tracking is predominantly (50) from WS1S. The OD accuracy requirements are: Definitive ephemeris accuracy of 500 meters total position root-mean-squared (RMS) and18 meters radial RMS. Predicted orbit accuracy less than 800 meters root sum squared (RSS) over an 84-hour prediction span.

LRO↗
NASA NTRS2018

Enabling Technologies for Deep Space Imaging

From the beginning of the Space Age, imagery, particularly motion imagery, has been a part of crewed and un-crewed missions. As technologies have evolved the imagery gets better, more compelling, and more useful for operations and monitoring of systems, crew, and spacecraft. As we look forward now to crewed missions beyond low-Earth orbit, such as the Lunar Orbiting Platform-Gateway being considered as a pre-cursor to future crewed Mars missions, there are both opportunities and challenges in implementing a multi-faceted imaging system that advances mission capabilities and technology. This paper will present a vision for an imaging system that is relevant for operations of the ISS and future crewed missions in deep space, with a detailed look at some of the key innovative technologies required to enable such a system. Specific enabling technologies included are: Innovative camera systems capable of providing a 360deg field-of-view without moving parts; Ultra-high Definition (or higher) resolution; High Efficiency Video Coding compression; Compatibility with Delay Tolerant Network protocols; and Intelligent systems capable of monitoring the field-of-view for un-crewed missions. Opportunities where Standardization can enable interoperability are also identified.

Grubbs, Rodney↗
NASA NTRS2022

The Sun Radio Interferometer Space Experiment (SunRISE) Mission

The Sun Radio Interferometer Space Experiment (SunRISE) will provide an entirely new view on particle acceleration and transport in the inner heliosphere by creating the first low radio frequency interferometer in space to localize heliospheric radio emissions. By imaging and determining the location of decametric-hectometric (DH, 0.1 MHz–23 MHz) solar radio bursts, SunRISE will provide key information on particle acceleration mechanisms associated with coronal mass ejections (CMEs) and the magnetic field topology from active regions into interplanetary space. The SunRISE Observatory will consist of six space vehicles in a passive formation, in orbits designed to keep them within approximately 10 km of each other, and flying in a supersynchronous geosynchronous Earth orbit (GEO). Each space vehicle consists of a Solar DH-GNSS payload and a 6U form factor spacecraft. The SunRISE Observatory together with significant ground-based processing, will enable imaging of the Sun in a portion of the spectrum that is blocked by the ionosphere and cannot be observed from Earth. Key aspects that enable this mission are that only position knowledge of the space vehicles is required, not active control, and that the architecture involves a modest amount of on-board processing coupled with significant ground-based processing for navigation, position determination, and science operations. Mission-enabling advances in software-defined radios, GPS navigation and timing, and small spacecraft technologies, developed and flown on the DARPA High Frequency Research (DHFR) and the Community Initiative for Continuing Earth Radio Occultation (CICERO) have made this mission affordable and low-risk. The SunRISE mission will exploit the multiple spacecraft per aperture (MSPA) capability of NASA’s Deep Space Network (DSN), for more efficient data transfers of larger data volumes, and utilize commercial access to space, in which the SunRISE space vehicles will be carried to their target orbit as secondary payloads in conjunction with a larger host spacecraft intended for GEO.

Neilson, Tim↗
DOE OSTI2024

Learning Only on Boundaries: A Physics-Informed Neural Operator for Solving Parametric Partial Differential Equations in Complex Geometries

Recently, deep learning surrogates and neural operators have shown promise in solving partial differential equations (PDEs). However, they often require a large amount of training data and are limited to bounded domains. In this work, we present a novel physics-informed neural operator method to solve parameterized boundary value problems without labeled data. By reformulating the PDEs into boundary integral equations (BIEs), we can train the operator network solely on the boundary of the domain. This approach reduces the number of required sample points from $O(N^d)$ to $O(N^{d-1}$), where $d$ is the domain’s dimension, leading to a significant acceleration of the training process. Additionally, our method can handle unbounded problems, which are unattainable for existing physics-informed neural networks (PINNs) and neural operators. Finally, our numerical experiments show the effectiveness of parameterized complex geometries and unbounded problems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

ExtremeMETA: High-speed Lightweight Image Segmentation Model by Remodeling Multi-channel Metamaterial Imagers

Deep neural networks (DNNs) have heavily relied on traditional computational units, such as CPUs and GPUs. However, this conventional approach brings significant computational burden, latency issues, and high power consumption, limiting their effectiveness. This has sparked the need for lightweight networks such as ExtremeC3Net. Meanwhile, there have been notable advancements in optical computational units, particularly with metamaterials, offering the exciting prospect of energy-efficient neural networks operating at the speed of light. Yet, the digital design of metamaterial neural networks (MNNs) faces precision, noise, and bandwidth challenges, limiting their application to intuitive tasks and low-resolution images. In this study, we proposed a large kernel lightweight segmentation model, ExtremeMETA. Based on ExtremeC3Net, our proposed model, ExtremeMETA maximized the ability of the first convolution layer by exploring a larger convolution kernel and multiple processing paths. With the large kernel convolution model, we extended the optic neural network application boundary to the segmentation task. To further lighten the computation burden of the digital processing part, a set of model compression methods was applied to improve model efficiency in the inference stage. The experimental results on three publicly available datasets demonstrated that the optimized efficient design improved segmentation performance from 92.45 to 95.97 on mIoU while reducing computational FLOPs from 461.07 MMacs to 166.03 MMacs. The large kernel lightweight model ExtremeMETA showcased the hybrid design’s ability on complex tasks.

large convolution kernel↗