Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “learned priors”

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 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
NASA NTRS2016

Using Multimodal Input for Autonomous Decision Making for Unmanned Systems

Autonomous decision making in the presence of uncertainly is a deeply studied problem space particularly in the area of autonomous systems operations for land, air, sea, and space vehicles. Various techniques ranging from single algorithm solutions to complex ensemble classifier systems have been utilized in a research context in solving mission critical flight decisions. Realized systems on actual autonomous hardware, however, is a difficult systems integration problem, constituting a majority of applied robotics development timelines. The ability to reliably and repeatedly classify objects during a vehicles mission execution is vital for the vehicle to mitigate both static and dynamic environmental concerns such that the mission may be completed successfully and have the vehicle operate and return safely. In this paper, the Autonomy Incubator proposes and discusses an ensemble learning and recognition system planned for our autonomous framework, AEON, in selected domains, which fuse decision criteria, using prior experience on both the individual classifier layer and the ensemble layer to mitigate environmental uncertainty during operation.

Neilan, James H.↗
NASA NTRS2010

Fast Solution in Sparse LDA for Binary Classification

An algorithm that performs sparse linear discriminant analysis (Sparse-LDA) finds near-optimal solutions in far less time than the prior art when specialized to binary classification (of 2 classes). Sparse-LDA is a type of feature- or variable- selection problem with numerous applications in statistics, machine learning, computer vision, computational finance, operations research, and bio-informatics. Because of its combinatorial nature, feature- or variable-selection problems are NP-hard or computationally intractable in cases involving more than 30 variables or features. Therefore, one typically seeks approximate solutions by means of greedy search algorithms. The prior Sparse-LDA algorithm was a greedy algorithm that considered the best variable or feature to add/ delete to/ from its subsets in order to maximally discriminate between multiple classes of data. The present algorithm is designed for the special but prevalent case of 2-class or binary classification (e.g. 1 vs. 0, functioning vs. malfunctioning, or change versus no change). The present algorithm provides near-optimal solutions on large real-world datasets having hundreds or even thousands of variables or features (e.g. selecting the fewest wavelength bands in a hyperspectral sensor to do terrain classification) and does so in typical computation times of minutes as compared to days or weeks as taken by the prior art. Sparse LDA requires solving generalized eigenvalue problems for a large number of variable subsets (represented by the submatrices of the input within-class and between-class covariance matrices). In the general (fullrank) case, the amount of computation scales at least cubically with the number of variables and thus the size of the problems that can be solved is limited accordingly. However, in binary classification, the principal eigenvalues can be found using a special analytic formula, without resorting to costly iterative techniques. The present algorithm exploits this analytic form along with the inherent sequential nature of greedy search itself. Together this enables the use of highly-efficient partitioned-matrix-inverse techniques that result in large speedups of computation in both the forward-selection and backward-elimination stages of greedy algorithms in general.

Moghaddam, Baback↗
DOE OSTI2023

DOE STTR Phase I Final Report Report: Machine-learning Based Prediction of Thermal Limits for Conjugated Organic Materials

The SCANN-DT Phase I STTR project led by NLM Photonics and partnering with the National Renewable Energy Laboratory (NREL) sought to apply machine learning techniques based on graph neural networks (GNNs) towards the prediction of decomposition temperatures (Td) of organic semiconductors, based on prior work on bond dissociation energy (BDE) prediction as implemented in NREL’s ALFABET prediction tool. Using a curated set of experimental decomposition energies, the project examined GNN-based, classical quantitative structure-property relationship (QSPR) based on DFT calculations, and combinations of both methods to predict Td. While the best MAEs in Td achieved were near 40°C, below project targets, the project led to improvements in the ALFABET model, improvements in cloud-based implementations of NWChem software, and a substantial dataset of calculations on medium-sized conjugated organic molecules.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Deep Learning with Quantized Neural Networks for Gravitational-wave Forecasting of Eccentric Compact Binary Coalescence

We present the first application of deep learning forecasting for binary neutron stars, neutron star–black hole systems, and binary black hole mergers that span an eccentricity range e ≤ 0.9. In this work, we train neural networks that describe these astrophysical populations, and then test their performance by injecting simulated eccentric signals in advanced Laser Interferometer Gravitational-Wave Observatory (LIGO) noise available at the Gravitational Wave Open Science Center to (1) quantify how fast neural networks identify these signals before the binary components merge; (2) quantify how accurately neural networks estimate the time to merger once gravitational waves are identified; and (3) estimate the time-dependent sky localization of these events from early detection to merger. Our findings show that deep learning can identify eccentric signals from a few seconds (for binary black holes) up to tens of seconds (for binary neutron stars) prior to merger. A quantized version of our neural networks achieves 4× reduction in model size, and up to 2.5× inference speedup. These novel algorithms may be used to facilitate time-sensitive multimessenger astrophysics observations of compact binaries in dense stellar environments.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

MIONet: Learning Multiple-Input Operators via Tensor Product

As an emerging paradigm in scientific machine learning, neural operators aim to learn operators, via neural networks, that map between infinite-dimensional function spaces. Several neural operators have been recently developed. However, all the existing neural operators are only designed to learn operators defined on a single Banach space; i.e., the input of the operator is a single function. Here, for the first time, we study the operator regression via neural networks for multiple-input operators defined on the product of Banach spaces. We first prove a universal approximation theorem of continuous multiple-input operators. We also provide a detailed theoretical analysis including the approximation error, which provides guidance for the design of the network architecture. Based on our theory and a low-rank approximation, we propose a novel neural operator, MIONet, to learn multiple-input operators. MIONet consists of several branch nets for encoding the input functions and a trunk net for encoding the domain of the output function. Here, we demonstrate that MIONet can learn solution operators involving systems governed by ordinary and partial differential equations. In our computational examples, we also show that we can endow MIONet with prior knowledge of the underlying system, such as linearity and periodicity, to further improve accuracy.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Accelerating Random Forest Classification on GPU and FPGA

Random Forests (RFs) are a commonly used machine learning method for classification and regression tasks spanning a variety of application domains, including bioinformatics, business analytics, and software optimization. While prior work has focused primarily on improving performance of the training of RFs, many applications, such as malware identification, cancer prediction, and banking fraud detection, require fast RF classification. In this work, we accelerate RF classification on GPU and FPGA. In order to provide efficient support for large datasets, we propose a hierarchical memory layout suitable to the GPU/FPGA memory hierarchy. We design three RF classification code variants based on that layout, and we investigate GPU- and FPGA-specific considerations for these kernels. Our experimental evaluation, performed on an Nvidia Xp GPU and on a Xilinx Alveo U250 FPGA accelerator card using publicly available datasets on the scale of millions of samples and tens of features, covers various aspects. First, we evaluate the performance benefits of our hierarchical data structure over the standard compressed sparse row (CSR) format. Second, we compare our GPU implementation with cuML, a machine learning library targeting Nvidia GPUs. Third, we explore the performance/accuracy tradeoff resulting from the use of different tree depths in the RF. Finally, we perform a comparative performance analysis of our GPU and FPGA implementations. Our evaluation shows that for high accuracy targets, our GPU implementation yields 5-9x speedup over CSR, and up to a 2x speedup over cuML.

FPGA, Xilinx FPGA, GPU, Random Forest classificati↗
DOE OSTI2020

Deblending galaxies with variational autoencoders: A joint multiband, multi-instrument approach

ABSTRACT Blending of galaxies has a major contribution in the systematic error budget of weak-lensing studies, affecting photometric and shape measurements, particularly for ground-based, deep, photometric galaxy surveys, such as the Rubin Observatory Legacy Survey of Space and Time (LSST). Existing deblenders mostly rely on analytic modelling of galaxy profiles and suffer from the lack of flexible yet accurate models. We propose to use generative models based on deep neural networks, namely variational autoencoders (VAE), to learn probabilistic models directly from data. We train a VAE on images of centred, isolated galaxies, which we reuse, as a prior, in a second VAE-like neural network in charge of deblending galaxies. We train our networks on simulated images including six LSST bandpass filters and the visible and near-infrared bands of the Euclid satellite, as our method naturally generalizes to multiple bands and can incorporate data from multiple instruments. We obtain median reconstruction errors on ellipticities and r-band magnitude between ±0.01 and ±0.05, respectively, in most cases, and ellipticity multiplicative bias of 1.6 per cent for blended objects in the optimal configuration. We also study the impact of decentring and prove the method to be robust. This method only requires the approximate centre of each target galaxy, but no assumptions about the number of surrounding objects, pointing to an iterative detection/deblending procedure we leave for future work. Finally, we discuss future challenges about training on real data and obtain encouraging results when applying transfer learning.

Arcelin, Bastien↗
NASA NTRS2014

The Double Flybys of the Cassini Mission: Navigation Challenges and Lessons Learned

Since 2004, the Cassini spacecraft has flown by Titan and other Saturn moons numerous times, successfully accomplishing its 100th targeted encounter of Titan in March 2014. The navigation of Cassini is challenging, even more so with "double flybys," two encounters separated by at most a few days. Because of this tight spacing, there is not enough time for a maneuver in between. Additionally, maneuvers prior to a double flyby only target one of the two encounters. This paper discusses the challenges faced by the Cassini Navigation Team with each double flyby, as well as lessons learned during operational support of each dual encounter. The strengths and weaknesses of the targeting strategies considered for each double flyby are also detailed, by comparing downstream ?V costs and changes to the non-targeted flyby conditions.

maneuver↗
NASA NTRS2014

Full Scale Advanced Systems Testbed (FAST): Capabilities and Recent Flight Research

At the NASA Armstrong Flight Research Center research is being conducted into flight control technologies that will enable the next generation of air and space vehicles. The Full Scale Advanced Systems Testbed (FAST) aircraft provides a laboratory for flight exploration of these technologies. In recent years novel but simple adaptive architectures for aircraft and rockets have been researched along with control technologies for improving aircraft fuel efficiency and control structural interaction. This presentation outlines the FAST capabilities and provides a snapshot of the research accomplishments to date. Flight experimentation allows a researcher to substantiate or invalidate their assumptions and intuition about a new technology or innovative approach Data early in a development cycle is invaluable for determining which technology barriers are real and which ones are imagined Data for a technology at a low TRL can be used to steer and focus the exploration and fuel rapid advances based on real world lessons learned It is important to identify technologies that are mature enough to benefit from flight research data and not be tempted to wait until we have solved all the potential issues prior to getting some data Sometimes a stagnated technology just needs a little real world data to get it going One trick to getting data for low TRL technologies is finding an environment where it is okay to take risks, where occasional failure is an expected outcome Learning how things fail is often as valuable as showing that they work FAST has been architected to facilitate this type of testing for control system technologies, specifically novel algorithms and sensors Rapid prototyping with a quick turnaround in a fly-fix-fly paradigm Sometimes it's easier and cheaper to just go fly it than to analyze the problem to death The goal is to find and test control technologies that would benefit from flight data and find solutions to the real barriers to innovation. The FAST vehicle is a flexible laboratory for nascent technologies that would benefit from early life cycle flight research data It provides a robust and safe environment where innovative techniques can be explored in a fly-fix-fly rapid prototyping paradigm IRAC Simple adaptive control technologies can provide real benefits without undo complexity Adverse pilot/adaptive system interactions can be mitigated and tools have been developed to evaluate those interactions ICP Substantial fuel savings can be achieved over a broad range of vehicles and configurations with intelligent control solutions LVAC The AAC design is robust and effective for the SLS mission, and promises to provide benefits to other platforms as well OCLA Hopefully will show that structural feedback can be seamlessly integrated with performance and stability objectives All of these control technologies have been implemented into the same baseline control law and could be combined into one control solution that answers many pressing questions for modern vehicle configurations

flight controls↗
NASA NTRS2011

Lessons Learned from the Space Shuttle Engine Cutoff System (ECO) Anomalies

The Space Shuttle Orbiter's main engine cutoff (ECO) system first failed ground checkout in April, 2005 during a first tanking test prior to Return-to-Flight. Despite significant troubleshooting and investigative efforts that followed, the root cause could not be found and intermittent anomalies continued to plague the Program. By implementing hardware upgrades, enhancing monitoring capability, and relaxing the launch rules, the Shuttle fleet was allowed to continue flying in spite of these unexplained failures. Root cause was finally determined following the launch attempts of STS-122 in December, 2007 when the anomalies repeated, which allowed drag-on instrumentation to pinpoint the fault (the ET feedthrough connector). The suspect hardware was removed and provided additional evidence towards root cause determination. Corrective action was implemented and the system has performed successfully since then. This white paper presents the lessons learned from the entire experience, beginning with the anomalies since Return-to-Flight through discovery and correction of the problem. To put these lessons in better perspective for the reader, an overview of the ECO system is presented first. Next, a chronological account of the failures and associated investigation activities is discussed. Root cause and corrective action are summarized, followed by the lessons learned.

Martinez, Hugo E.↗
DOE OSTI2024

Targeted Biomining and Machine Learning Approaches in Critical Minerals Revealed by a Biogeochemical Survey of a Coal Mine Drainage Remediation System

Abandoned coal mine drainage (AMD) remediation systems in Pennsylvania can concentrate critical minerals and materials (CMM) at levels comparable to mining-grade ores. Remediation systems have varying engineering features and are open to the environment, resulting in diverse microbial colonization and seasonal climate influences that may impact CMM speciation. The location of CMMs, the types of bacterial communities tolerant of these pollutant conditions, and the influence of localized climate on CMM rich remediation systems are not well characterized. Through a one-year spatiotemporal survey of biogeochemistry at a remediation system, we have initiated the process to address these questions. Rare Earth Elements (REE) ranged 180-1,200 ppm and greater than 1,500 bacterial ASVs were classified via 16S sequencing. Analyses indicate biogeochemical differences are heavily influenced by engineering features. Additionally, REE precipitants correlate strongly with the elements Al, Cu, Zn, Be, and U. Unearthing these trends has refined our line of inquiry to explore biological mining opportunities more closely with these metals. Furthermore, we created a Machine Learning Model for predicting AMD REE content, with 89% accuracy, using the data from this study and several others. Further training data is required to create a more reputable model. Recently, global research efforts have prioritized modeling work or the use of the few historical surveys to design experiments. Through our data, we challenge this approach, emphasizing the importance of expanding fundamental survey efforts prior to advanced product design and experimentation.

critical minerals↗
DOE OSTI2024

Haar-Like Wavelets on Hierarchical Trees

Here, discrete wavelet methods, originally formulated in the setting of regularly sampled signals, can be adapted to data defined on a point cloud if some multiresolution structure is imposed on the cloud. A wide variety of hierarchical clustering algorithms can be used for this purpose, and the multiresolution structure obtained can be encoded by a hierarchical tree of subsets of the cloud. Prior work introduced the use of Haar-like bases defined with respect to such trees for approximation and learning tasks on unstructured data. This paper builds on that work in two directions. First, we present an algorithm for constructing Haar-like bases on general discrete hierarchical trees. Second, with an eye towards data compression, we present thresholding techniques for data defined on a point cloud with error controlled in the $L$ $\infty$ norm and in a Hölder-type norm. In a concluding trio of numerical examples, we apply our methods to compress a point cloud dataset, study the tightness of the $L$ $\infty$ error bound, and use thresholding to identify MNIST classifiers with good generalizability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Probing the evolution of fault properties during the seismic cycle with deep learning

We use seismic waves that pass through the hypocentral region of the 2016 M6.5 Norcia earthquake together with Deep Learning (DL) to distinguish between foreshocks, aftershocks and time-to-failure (TTF). Binary and N-class models defined by TTF correctly identify seismograms in test with > 90% accuracy. We use raw seismic records as input to a 7 layer CNN model to perform the classification. Here we show that DL models successfully distinguish seismic waves pre/post mainshock in accord with lab and theoretical expectations of progressive changes in crack density prior to abrupt change at failure and gradual postseismic recovery. Performance is lower for band-pass filtered seismograms (below 10 Hz) suggesting that DL models learn from the evolution of subtle changes in elastic wave attenuation. Tests to verify that our results indeed provide a proxy for fault properties included DL models trained with the wrong mainshock time and those using seismic waves far from the Norcia mainshock; both show degraded performance. Our results demonstrate that DL models have the potential to track the evolution of fault zone properties during the seismic cycle. If this result is generalizable it could improve earthquake early warning and seismic hazard analysis.

58 GEOSCIENCES↗
DOE OSTI2024

Reinforcement learning pulses for transmon qubit entangling gates

The utility of a quantum computer is highly dependent on the ability to reliably perform accurate quantum logic operations. For finding optimal control solutions, it is of particular interest to explore model-free approaches, since their quality is not constrained by the limited accuracy of theoretical models for the quantum processor—in contrast to many established gate implementation strategies. In this work, we utilize a continuous control reinforcement learning algorithm to design entangling two-qubit gates for superconducting qubits; specifically, our agent constructs cross-resonance and CNOT gates without any prior information about the physical system. Using a simulated environment of fixed-frequency fixed-coupling transmon qubits, we demonstrate the capability to generate novel pulse sequences that outperform the standard cross-resonance gates in both fidelity and gate duration, while maintaining a comparable susceptibility to stochastic unitary noise. We further showcase an augmentation in training and input information that allows our agent to adapt its pulse design abilities to drifting hardware characteristics, importantly, with little to no additional optimization. Our results exhibit clearly the advantages of unbiased adaptive-feedback learning-based optimization methods for transmon gate design.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

Identification of Non-Fermi Liquid Physics in a Quantum Critical Metal via Quantum Loop Topography

Non-Fermi liquid physics is ubiquitous in strongly correlated metals, manifesting itself in anomalous transport properties, such as a $\textit{T}$-linear resistivity in experiments. However, its theoretical understanding in terms of microscopic models is lacking, despite decades of conceptual work and attempted numerical simulations. In this work, we demonstrate that a combination of sign-problem-free quantum Monte Carlo sampling and quantum loop topography, a physics-inspired machine-learning approach, can map out the emergence of non-Fermi liquid physics in the vicinity of a quantum critical point (QCP) with little prior knowledge. Using only three parameter points for training the underlying neural network, we are able to robustly identify a stable non-Fermi liquid regime tracing the fans of metallic QCPs at the onset of both spin-density wave and nematic order. In particular, we establish for the first time that a spin-density wave QCP commands a wide fan of non-Fermi liquid region that funnels into the quantum critical point. Our study thereby provides an important proof-of-principle example that new physics can be detected via unbiased machine-learning approaches.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
NASA NTRS1993

Towards a machine learning framework for acquiring and exploiting monitoring and diagnostic knowledge

In this paper we address the problem of detecting and diagnosing faults in physical systems, for which neither prior expertise for the task nor suitable system models are available. We propose an architecture that integrates the on-line acquisition and exploitation of monitoring and diagnostic knowledge. The focus of the paper is on the component of the architecture that discovers classes of behaviors with similar characteristics by observing a system in operation. We investigate a characterization of behaviors based on best fitting approximation models. An experimental prototype has been implemented to test it. We present preliminary results in diagnosing faults of the Reaction Control System of the Space Shuttle. The merits and limitations of the approach are identified and directions for future work are set.

Manganaris, Stefanos↗
NASA NTRSDate not supplied

OSIRIS-REx Screening Experience & Lessons Learned

On September 24, 2023, the NASA OSIRIS-REx spacecraft returned a sample of asteroid Bennu, approaching on a hyperbolic impact trajectory relative to Earth. At approximately 108,000 km altitude and four hours prior to hitting the top of the atmosphere, the Sample Return Capsule (SRC) separated from the spacecraft bus, targeting the Utah Test and Training Range. Twenty minutes later, the bus performed a large divert maneuver to fly by at a perigee altitude of ~780 km. Given the immense scientific value of the sample, substantial efforts were undertaken by NASA CARA and the OSIRIS-REx mission to ensure safe passage of the objects from GEO through LEO. This presentation provides an overview of collision avoidance maneuver (CAM) options, results of screenings, coordination with DoD for contingency observations, and lessons learned for screening future Earth flyby and entry trajectories. Two particular challenges included screening two objects using a much larger than normal 250 km spherical monitor volume and decision planning for a single CAM while the SRC was still attached that would alter both trajectories. The larger, non-standard screening volume resulted in custom procedures that CARA personnel had to run manually for each screening, and the initial long propagation times while the objects were more than a week out required significant computing resources. With a single CAM opportunity at 13 hours out, the decision to execute a maneuver had to consider the risk to both objects, where one could imagine a situation where the maneuver would clear a high-risk conjunction for one object and introduce one for the other – begging the question of relative risk posture between the SRC and bus. These among other unique aspects of Earth flyby and entry missions led to novel planning and implementation procedures to support future missions of this type.

Dolan E. Highsmith↗
DOE OSTI2023

An exploration of machine learning models for the determination of reaction coordinates associated with conformational transitions

Determining collective variables (CVs) for conformational transitions is crucial to understanding their dynamics and targeting them in enhanced sampling simulations. Often, CVs are proposed based on intuition or prior knowledge of a system. However, the problem of systematically determining a proper reaction coordinate (RC) for a specific process in terms of a set of putative CVs can be achieved using committor analysis (CA). Identifying essential degrees of freedom that govern such transitions using CA remains elusive because of the high dimensionality of the conformational space. Various schemes exist to leverage the power of machine learning (ML) to extract an RC from CA. Here, we extend these studies and compare the ability of 17 different ML schemes to identify accurate RCs associated with conformational transitions. We tested these methods on an alanine dipeptide in vacuum and on a sarcosine dipeptoid in an implicit solvent. Our comparison revealed that the light gradient boosting machine method outperforms other methods. In order to extract key features from the models, we employed Shapley Additive exPlanations analysis and compared its interpretation with the “feature importance” approach. For the alanine dipeptide, our methodology identifies ϕ and θ dihedrals as essential degrees of freedom in the C7ax to C7eq transition. For the sarcosine dipeptoid system, the dihedrals ψ and ω are the most important for the cisαD to transαD transition. We further argue that analysis of the full dynamical pathway, and not just endpoint states, is essential for identifying key degrees of freedom governing transitions.

Chemistry↗