Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep 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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
NASA NTRS2020

Improving Trust in Deep Neural Networks with Nearest Neighbors

Deep neural networks are used increasingly for perception and decision-making in UAVs. For example, they can be used to recognize objects from images and decide what actions the vehicle should take. While deep neural networks can perform very well at complex tasks, their decisions may be unintuitive to a human operator. When a human disagrees with a neural network prediction, due to the black box nature of deep neural networks, it can be unclear whether the system knows something the human does not or whether the system is malfunctioning. This uncertainty is problematic when it comes to ensuring safety. As a result, it is important to develop technologies for explaining neural network decisions for trust and safety. This paper explores a modification to the deep neural network classification layer to produce both a predicted label and an explanation to support its prediction. Specifically, at test time, we replace the final output layer of the neural network classifier by a k-nearest neighbor classifier. The nearest neighbor classifier produces 1) a predicted label through voting and 2) the nearest neighbors involved in the prediction, which represent the most similar examples from the training dataset. Because prediction and explanation are derived from the same underlying process, this approach guarantees that the explanations are always relevant to the predictions. We demonstrate the approach on a convolutional neural network for a UAV image classification task. We perform experiments using a forest trail image dataset and show empirically that the hybrid classifier can produce intuitive explanations without loss of predictive performance compared to the original neural network. We also show how the approach can be used to help identify potential issues in the network and training process.

Lee, Ritchie↗
DOE OSTI2021

Enhanced Data Efficiency Using Deep Neural Networks and Gaussian Processes for Aerodynamic Design Optimization

Adjoint-based optimization methods are attractive for aerodynamic shape design primarily due to their computational costs being independent of the dimensionality of the input space and their ability to generate high-fidelity gradients that can then be used in a gradient-based optimizer. This makes them very well suited for high-fidelity simulation based aerodynamic shape optimization of highly parametrized geometries such as aircraft wings. However, the development of adjoint-based solvers involve careful mathematical treatment and their implementation require detailed software development. Furthermore, they can become prohibitively expensive when multiple optimization problems are being solved, each requiring multiple restarts to circumvent local optima. In this work, we propose a machine learning enabled, surrogate-based framework that replaces the expensive adjoint solver, without compromising on predicting predictive accuracy. Specifically, we first train a deep neural network (DNN) from training data generated from evaluating the high-fidelity simulation model on a model-agnostic design of experiments on the geometry shape parameters. The optimum shape may then be computed by using a gradient-based optimizer coupled with the trained DNN. Subsequently, we also perform a gradient-free Bayesian optimization, where the trained DNN is used as the prior mean. We observe that the latter framework (DNN-BO) improves upon the DNN-only based optimization strategy for the same computational cost. Overall, this framework predicts the true optimum with very high accuracy, while requiring far fewer high-fidelity function calls compared to the adjoint-based method. Furthermore, we show that multiple optimization problems can be solved with the same machine learning model with high accuracy, to amortize the offline costs associated with constructing our models. Our methodology finds applications in the early stages of aerospace design. (C) 2021 Published by Elsevier Masson SAS.

Renganathan, S. Ashwin↗
NASA NTRS2010

Coherent Frequency Reference System for the NASA Deep Space Network

The NASA Deep Space Network (DSN) requires state-of-the-art frequency references that are derived and distributed from very stable atomic frequency standards. A new Frequency Reference System (FRS) and Frequency Reference Distribution System (FRD) have been developed, which together replace the previous Coherent Reference Generator System (CRG). The FRS and FRD each provide new capabilities that significantly improve operability and reliability. The FRS allows for selection and switching between frequency standards, a flywheel capability (to avoid interruptions when switching frequency standards), and a frequency synthesis system (to generate standardized 5-, 10-, and 100-MHz reference signals). The FRS is powered by redundant, specially filtered, and sustainable power systems and includes a monitor and control capability for station operations to interact and control the frequency-standard selection process. The FRD receives the standardized 5-, 10-, and 100-MHz reference signals and distributes signals to distribution amplifiers in a fan out fashion to dozens of DSN users that require the highly stable reference signals. The FRD is also powered by redundant, specially filtered, and sustainable power systems. The new DSN Frequency Distribution System, which consists of the FRS and FRD systems described here, is central to all operational activities of the NASA DSN. The frequency generation and distribution system provides ultra-stable, coherent, and very low phase-noise references at 5, l0, and 100 MHz to between 60 and 100 separate users at each Deep Space Communications Complex.

Tucker, Blake C.↗
NASA NTRS1971

The Deep Space Network, volume 4

Deep Space Network progress report, including mission support, development and implementation, and operations and facilities

Source record↗
NASA NTRS1972

The deep space network, Volume 11

Deep Space Network progress in flight project support, Tracking and Data Acquisition research and technology, network engineering, hardware and software implementation, and operations are presented. Material is presented in each of the following categories: description of DSN; mission support; radio science; support research and technology; network engineering and implementation; and operations and facilities.

Source record↗
NASA NTRS1975

Single- and dual-carrier microwave noise abatement in the deep space network

The NASA/JPL Deep Space Network (DSN) microwave ground antenna systems are presented which simultaneously uplink very high power S-band signals while receiving very low level S- and X-band downlinks. Tertiary mechanisms associated with elements give rise to self-interference in the forms of broadband noise burst and coherent intermodulation products. A long-term program to reduce or eliminate both forms of interference is described in detail. Two DSN antennas were subjected to extensive interference testing and practical cleanup program; the initial performance, modification details, and final performance achieved at several planned stages are discussed. Test equipment and field procedures found useful in locating interference sources are discussed. Practices deemed necessary for interference-free operations in the DSN are described. Much of the specific information given is expected to be easily generalized for application in a variety of similar installations. Recommendations for future investigations and individual element design are given.

Bathker, D. A.↗
NASA NTRS1977

The deep space network

Presented is Deep Space Network (DSN) progress in flight project support, tracking and data acquisition (TDA) research and technology, network engineering, hardware and software implementation, and operations.

Source record↗
NASA NTRS2009

Ka-Band High-Rate Telemetry System Upgrade for the NASA Deep Space Network

The NASA Deep Space Network (DSN) has a new requirement to support high-data-rate Category A (Cat A) missions (within 2 million kilometers of Earth) with simultaneous S-band uplink, S-band downlink and Ka-band downlink. The S-band links are required for traditional TT&C (Telemetry, Tracking, and Command) support to the spacecraft, while the Ka-band link is intended for high-data-rate science returns. The new Ka-band system combines the use of proven DSN cryogenic designs, for low system temperature, and high data rate capability using commercial telemetry receivers. The initial Cat A support is required for the James Webb Space Telescope (JWST) in 2013 and possibly other missions. The upgrade has been implemented into 3 different 34-meter Beam Waveguide (BWG) antennas in the DSN, one at each of the complexes in Canberra (Australia), Goldstone (California) and Madrid (Spain). System test data is presented to show that the requirements were met and the DSN is ready for Cat A Ka-band operational support.

space communications↗
NASA NTRS2003

Master clock and time distribution system for the NASA Deep Space Network

NASA's Deep Space Network (DSN) consists of more than 20 antennas located at three globally spaced ground communications facilities. Local generation and distribution of precise time and frequency reference signals comprise an essential and central component of each complex. Within each complex synchronized timing references are required by approximately 100 users located at distances up to 30 kilometers from the central control center and station Master Clock. In this paper a highly modular, hot-swappable, and expandable system design for generation, delivery, and synchronization of highly precise and stable timing signals over fiber optic cables is described.

Tjoelker, R. L.↗
NASA NTRS2021

User Preference Optimization for Oversubscribed Scheduling of NASA’s Deep Space Network

NASA’s Deep Space Network (DSN) is the primary resource for communications and navigation for interplanetary space missions, for both NASA and partner agencies. Growth in mission demand, both in number of spacecraft and in data return, has led to increased loading levels on the network, and actual demand frequently exceeds network capacity. The DSN scheduling process involves peer-to-peer collaborative negotiation, which consumes significant time and resources in order to reach a baseline version of the schedule, and then to manage and agree to changes. Process delays are exacerbated by the high level of oversubscription experienced by the DSN: it is not unusual for the scheduling process to start with 20-40% more requested time can be accommodated on the available antennas. The other NASA networks make use of a static mission priority list to address a similar problem: missions are ranked in priority order, then the schedule is populated by priority from highest to lowest. Such a mechanism would not work for DSN due to the heterogeneity of the mission set, and to the time-varying mission requirements with mission phase. This paper describes an alternative approach for the DSN that addresses key problems inherent in the current process — oversubsubscription and how to “fairly” reduce it to a manageable level. The main characteristics of the new approach are the use of loading-based limits based on balancing requested time, along with priorities and user preferences as the basis for optimization criteria that can be used by new algorithms.

Johnston, Mark D↗
NASA NTRS2021

A Demand Access Paradigm for NASA’s Deep Space Network

NASA’s Deep Space Network (DSN) is the primary resource for communications and navigation for interplanetary space missions, for both NASA and partner agencies. As part of an investigation into improved efficiency and responsiveness, we have been exploring and prototyping the infusion of a ”demand access” model into the DSN scheduling process. Today, DSN is fully pre-scheduled in advance, and many users rely on a stable schedule to plan their own spacecraft activities, weeks in advance of execution. However, a new class of missions is emerging that may not be scheduled as far in advance, and may be event-driven in coming across science targets at unpredictable times. These users could take advantage of an on-demand mechanism to download data. Simulations have shown that such a mechanism could improve latency (time from data collection to download) by 2x, as well as more efficiently utilize the available DSN antennas. In this paper, we describe a prototype of a demand access process and how it addresses the challenges of co-existing with a staticallyscheduled body of missions, while providing the benefits of lower latency science data return.

Lucena, Girly↗
DOE OSTI2022

Nonlocal Kernel Network (NKN): a Stable and Resolution-Independent Deep Neural Network.

Neural operators have recently become popular tools for designing solution maps between function spaces in the form of neural networks. Differently from classical scientific machine learning approaches that learn parameters of a known partial differential equation (PDE) for a single instance of the input parameters at a fixed resolution, neural operators approximate the solution map of a family of PDEs [6, 7]. Despite their success, the uses of neural operators are so far restricted to relatively shallow neural networks and confined to learning hidden governing laws. In this work, we propose a novel nonlocal neural operator, which we refer to as nonlocal kernel network (NKN), that is resolution independent, characterized by deep neural networks, and capable of handling a variety of tasks such as learning governing equations and classifying images. Our NKN stems from the interpretation of the neural network as a discrete nonlocal diffusion reaction equation that, in the limit of infinite layers, is equivalent to a parabolic nonlocal equation, whose stability is analyzed via nonlocal vector calculus. The resemblance with integral forms of neural operators allows NKNs to capture long-range dependencies in the feature space, while the continuous treatment of node-to-node interactions makes NKNs resolution independent. The resemblance with neural ODEs, reinterpreted in a nonlocal sense, and the stable network dynamics between layers allow for generalization of NKN’s optimal parameters from shallow to deep networks. This fact enables the use of shallow-to-deep initialization techniques [8]. Our tests show that NKNs outperform baseline methods in both learning governing equations and image classification tasks and generalize well to different resolutions and depths.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1999

The Future of the Deep Space Network: Technology Development for K2-Band Deep Space Communications

Projections indicate that in the future the number of NASA's robotic deep space missions is likely to increase significantly. A launch rate of up to 4-6 launches per year is projected with up to 25 simultaneous missions active [I]. Future high resolution mapping missions to other planetary bodies as well as other experiments are likely to require increased downlink capacity. These future deep space communications requirements will, according to baseline loading analysis, exceed the capacity of NASA's Deep Space Network in its present form. There are essentially two approaches for increasing the channel capacity of the Deep Space Network. Given the near-optimum performance of the network at the two deep space communications bands, S-Band (uplink 2.025-2.120 GHz, downlink 2.2-2.3 GHz), and X-Band (uplink 7.145-7.19 GHz, downlink 8.48.5 GHz), additional improvements bring only marginal return for the investment. Thus the only way to increase channel capacity is simply to construct more antennas, receivers, transmitters and other hardware. This approach is relatively low-risk but involves increasing both the number of assets in the network and operational costs.

Bhanji, Alaudin M.↗
NASA NTRS1983

The Deep Space Network: A Radio Communications Instrument for Deep Space Exploration

The primary purpose of the Deep Space Network (DSN) is to serve as a communications instrument for deep space exploration, providing communications between the spacecraft and the ground facilities. The uplink communications channel provides instructions or commands to the spacecraft. The downlink communications channel provides command verification and spacecraft engineering and science instrument payload data.

Renzetti, N. A.↗
DOE OSTI2022

Deep Neural Network High Spatiotemporal Resolution Precipitation Estimation (Deep-STEP) Using Passive Microwave and Infrared Data

Recent developments in “headline-making” deep neural networks (DNNs), specifically convolutional neural networks (CNNs), along with advancements in computational power, open great opportunities to integrate massive amounts of real-time observations to characterize spatiotemporal structures of surface precipitation. This study aims to develop a CNN algorithm, named Deep Neural Network High Spatiotemporal Resolution Precipitation Estimation (Deep-STEP), that ingests direct satellite passive microwave (PMW) brightness temperatures (Tbs) at emission and scattering frequencies combined with infrared (IR) Tbs from geostationary satellites and surface information to automatically extract geospatial features related to the precipitable clouds. These features allow the end-to-end Deep-STEP algorithm to instantaneously map surface precipitation intensities with a spatial resolution of 4 km. The main advantages of Deep-STEP, as compared to current state-of-the-art techniques, are 1) it learns and estimates complex precipitation systems directly from raw measurements in near–real time, 2) it uses the automatic spatial neighborhood feature extraction approach, and 3) it fuses coarse-resolution PMW footprints with IR images to reliably retrieve surface precipitation at a high spatial resolution. We anticipate our proposed DNN algorithm to be a starting point for more sophisticated and efficient precipitation retrieval systems in terms of accuracy, fine spatial pattern detection skills, and computational costs.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2006

The Deep Space Network Array

This document is a viewgraph presentation that reviews the costs, and technological processing required to replace the current network of Deep Space Antennas. The concept of using an array for space communications is much less of a concern than the cost of implementing and operating such an array. Within the cost question, the cost uncertainty of the front-end components (repeated n-times) is of most importance. The activities at JPL have focused on both these aspects of the cost. A breadboard array of three antennas at JPL has been the vehicle to perform many investigations into the development of the new DSN. Several pictures of the antennas at JPL are shown.

Deep Space Network (DSN)↗