Search NASA⌕ Search

SEARCH · Search NASA

Results for “Loss function”

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 415 records · Page 23

Kalman filter estimation of attitude and gyro bias with the QUEST observation model

The loss function for the Wahba attitude estimation problem employs unit vector observations with scalar weights. It is usually associated with the QUEST observation model, where the actual sensor noise is assumed the same for all components of the observed vector, regardless of the position in the field of view. The QUEST model has the great advantage of being sensor-independent except for the scalar parameter that characterizes the sensor errors. Although efficient algorithms for solving the Wahba problem exist, extending these algorithms to estimate gyro biases or sensor alignments has had mixed success. However, it is straightforward to estimate bias and alignment parameters with a Kalman filter. This paper investigates the use of an extended Kalman filter for the attitude and gyro bias that incorporates the QUEST observation model, to be referred to as the Unit Vector Filter (UVF). The UVF results are compared with those from a more conventional filter, the Real-Time Sequential Filter (RTSF), for which the residual is the two-dimensional projection of the unit vector onto the plane perpendicular to the sensor boresight. The RTSF is similar in design to that used by Multimission Modular Spacecraft for onboard attitude determination. An apparent obstacle to the use of unit vectors as measurements is their singular noise covariance matrix. Shuster has shown that this problem should not affect filter performance. The UVF and RTSF are tested using actual flight data from the Extreme Ultraviolet Explorer (EUVE). It is found that these filters generate nearly identical attitude and gyro bias estimates, thus validating the use of unit vectors and the QUEST noise model.

Sedlak, J.↗

Turbines For Optomechanical Energy Conversion

Class of proposed optomechanical energy-conversion machines exploit multiple reflections from precise, high-reflectivity mirrors, some of which fixed to turbinelike rotors that spin at high speeds. Basic idea to increase frequencies of photons (and thus energy of light) by repeated Doppler-shifting reflections from moving mirrors. Gain in optical energy in such reflection proportional to speed of mirror, while loss function of imperfections of mirror.

Dick, G. John↗

Fast Integer Ambiguity Resolution for GPS Attitude Determination

In this paper, a new algorithm for GPS (Global Positioning System) integer ambiguity resolution is shown. The algorithm first incorporates an instantaneous (static) integer search to significantly reduce the search space using a geometric inequality. Then a batch-type loss function is used to check the remaining integers in order to determine the optimal integer. This batch function represents the GPS sightline vectors in the body frame as the sum of two vectors, one depending on the phase measurements and the other on the unknown integers. The new algorithm has several advantages: it does not require an a-priori estimate of the vehicle's attitude; it provides an inherent integrity check using a covariance-type expression; and it can resolve the integers even when coplanar baselines exist. The performance of the new algorithm is tested on a dynamic hardware simulator.

Lightsey, E. Glenn↗

30 Years of Wahba's Problem

This article discusses a problem proposed by Grace Wahba: Find the orthogonal matrix A with determinant +1 that minimizes the loss function L(A) is identical with the 1/2(summation from i( a(sub i) * absolute value of (b(sub i)- Ar(sub i))squared). where {b(sub i)} is a set of unit vectors measured in a spacecraft's body frame, {r(sub i)} e the corresponding unit vectors in a reference frame and {a(sub i) are non-negative weights). The orthogonal Procrustes problem is to find the orthogonal matrix A that is closest to B in the sense of the Fobenius norm, so that Wahba's Problem is equivalent to the orthogonal Procrustes problem with the provision that the determinant of A must be +1. The article reviews various solutions to this problem and the implementations of the various solutions.

Markley, F. Landis↗

How to Estimate Attitude from Vector Observations

In many spacecraft attitude systems, the attitude observations are naturally represented as unit vectors. Typical examples are the unit vectors giving the direction to the sun or a star and the unit vector in the direction of the Earth's magnetic field. In 1965, Grace Wahba, proposed the following problem: Find the orthogonal matrix A with determinant +1 that minimizes the loss function L(A) is identity with 1/2(Sum from i a(sub i) (absolute value of b(sub i - A(r(sub i))(exp 2))) where the set of b(sub i) is a set of unit vectors measured in a spacecraft's body frame, the set of r(sub i) are the corresponding unit vectors in a reference frame, and the set of a(sub i) are non-negative weights. Wahba's problem can be related to Maximum Likelihood Estimation if the weights are chosen to be inverse variances, a(sub i) = sigma((sub -2). Wahba didn't assume this, but it will be convenient to assume it in this paper. Wahba'soptimality condition has provided the basis for many attitude determination algorithms. The purpose of this paper is to give an overview of the most popular and most promising algorithm and to provide accuracy and speed comparisons.

Markley,F. Landis↗

How to Estimate Attitude from Vector Observations

The most robust estimators minimizing Wahba's loss function are Davenport's q method and the Singular Value Decomposition (SVD) method. The q method is faster than the SVD method with three or more measurements. The other algorithms are less robust since they solve the characteristic polynomial equation to find the maximum eigenvalue of Davenport's K matrix. They are only preferable when speed or processor power is an important consideration. Of these, Fast Optimal Attitude Matrix (FOAM) is the most robust and faster than the q method. Robustness is only an issue for measurements with widely differing accuracies, so the fastest algorithms, Quaternion ESTimator (QUEST), EStimator of the Optimal Quaternion (ESOQ), and ESOQ2, are well suited to star sensor applications.

Markley, F. Landis↗

New Developments in Quaternion Estimation from Vector Observations

This paper contains a critical comparison of estimators minimizing Wahba's loss function. Some new results are presented for the QUaternion ESTimator (QUEST) and EStimators of the Optimal Quaternion (ESOQ and ESOQ2) to avoid the computational burden of sequential rotations in these algorithms. None of these methods is as robust in principle as Davenport's q method or the Singular Value Decomposition (SVD) method, which are significantly slower. Robustness is only an issue for measurements with widely differing accuracies, so the fastest estimators, the modified ESOQ and ESOQ2, are well suited to sensors that track multiple stars with comparable accuracies. More robust forms of ESOQ and ESOQ2 are developed that are intermediate in speed.

Markley, F. Landis↗

New Attitude Sensor Alignment Calibration Algorithms

Accurate spacecraft attitudes may only be obtained if the primary attitude sensors are well calibrated. Launch shock, relaxation of gravitational stresses and similar effects often produce large enough alignment shifts so that on-orbit alignment calibration is necessary if attitude accuracy requirements are to be met. A variety of attitude sensor alignment algorithms have been developed to meet the need for on-orbit calibration. Two new algorithms are presented here: ALICAL and ALIQUEST. Each of these has advantages in particular circumstances. ALICAL is an attitude independent algorithm that uses near simultaneous measurements from two or more sensors to produce accurate sensor alignments. For each set of simultaneous observations the attitude is overdetermined. The information content of the extra degrees of freedom can be combined over numerous sets to provide the sensor alignments. ALIQUEST is an attitude dependent algorithm that combines sensor and attitude data into a loss function that has the same mathematical form as the Wahba problem. Alignments can then be determined using any of the algorithms (such as the QUEST quaternion estimator) that have been developed to solve the Wahba problem for attitude. Results from the use of these methods on active missions are presented.

Hashmall, Joseph A.↗

Surrogate-based Analysis and Optimization

A major challenge to the successful full-scale development of modem aerospace systems is to address competing objectives such as improved performance, reduced costs, and enhanced safety. Accurate, high-fidelity models are typically time consuming and computationally expensive. Furthermore, informed decisions should be made with an understanding of the impact (global sensitivity) of the design variables on the different objectives. In this context, the so-called surrogate-based approach for analysis and optimization can play a very valuable role. The surrogates are constructed using data drawn from high-fidelity models, and provide fast approximations of the objectives and constraints at new design points, thereby making sensitivity and optimization studies feasible. This paper provides a comprehensive discussion of the fundamental issues that arise in surrogate-based analysis and optimization (SBAO), highlighting concepts, methods, techniques, as well as practical implications. The issues addressed include the selection of the loss function and regularization criteria for constructing the surrogates, design of experiments, surrogate selection and construction, sensitivity analysis, convergence, and optimization. The multi-objective optimal design of a liquid rocket injector is presented to highlight the state of the art and to help guide future efforts.

Queipo, Nestor V.↗

Predictive Game Theory

Probability theory governs the outcome of a game; there is a distribution over mixed strat.'s, not a single "equilibrium". To predict a single mixed strategy must use our loss function (external to the game's players. Provides a quantification of any strategy's rationality. Prove rationality falls as cost of computation rises (for players who have not previously interacted). All extends to games with varying numbers of players.

Wolpert, David H.↗

Translational Cellular Research on the International Space Station

The emerging field of Translational Research aims to coalesce interdisciplinary findings from basic science for biomedical applications. To complement spaceflight research using human subjects, translational studies can be designed to address aspects of space-related human health risks and help develop countermeasures to prevent or mitigate them, with therapeutical benefits for analogous conditions experienced on Earth. Translational research with cells and model organisms is being conducted onboard the International Space Station (ISS) in connection with various human systems impacted by spaceflight, such as the cardiovascular, musculoskeletal, and immune systems. Examples of recent cell-based translational investigations on the ISS include the following. The JAXA investigation Cell Mechanosensing seeks to identify gravity sensors in skeletal muscle cells to develop muscle atrophy countermeasures by analyzing tension fluctuations in the plasma membrane, which changes the expression of key proteins and genes. Earth applications of this study include therapeutic approaches for some forms of muscular dystrophy, which appear to parallel aspects of muscle wasting in space. Spheroids is an ESA investigation examining the system of endothelial cells lining the inner surface of all blood vessels in terms of vessel formation, cellular proliferation, and programmed cell death, because injury to the endothelium has been implicated as underpinning various cardiovascular and musculoskeletal problems arising during spaceflight. Since endothelial cells are involved in the functional integrity of the vascular wall, this research has applications to Earth diseases such as atherosclerosis, diabetes, and hypertension. The goal of the T-Cell Activation in Aging NASA investigation is to understand human immune system depression in microgravity by identifying gene expression patterns of candidate molecular regulators, which will provide further insight into factors that may play a critical role in immune function loss during aging. In addition, Omics investigations with cells have synergistic applications ranging from the evaluation of pharmacological countermeasures to drug discovery. Thus, cell-based translational research onboard the ISS is bidirectionally bridging cutting-edge cellular and molecular approaches with space bioastronautics and human health methodologies on Earth.

Love, John↗

Quasi-Periodic Fluctuations and Chromospheric Evaporation in a Solar Flare Ribbon Observed by Hinode/EIS, IRIS, and RHESSI

The Hinode/Extreme-ultraviolet Imaging Spectrometer (EIS) obtained rapid cadence (11.2 s) EUV stare spectra of an M7.3 flare ribbon in AR 12036 on 2014 April 18. Quasi-periodic (P approx. = 75.6 +/- 9.2 s) intensity fluctuations occurred in emission lines of O IV, Mg VI, Mg VII, Si VII, Fe XIV, and Fe XVI during the flare's impulsive rise, and ended when the maximum intensity in Fe XXIII was reached. The profiles of the O IV- Fe XVI lines reveal that they were all redshifted during most of the interval of quasi-periodic intensity fluctuations, while the Fe XXIII profile revealed multiple components including one or two highly blueshifted ones. This indicates that the flare underwent explosive chromospheric evaporation during its impulsive rise. Fluctuations in the relative Doppler velocities were seen, but their amplitudes were too subtle to extract significant quasi-periodicities. RHESSI detected 25-100 keV hard-X-ray sources in the ribbon near the EIS slit's pointing position during the peaks in the EIS intensity fluctuations. The observations are consistent with a series of energy injections into the chromosphere by nonthermal particle beams. Electron densities derived with Fe XIV (4.6 x 10(exp 10) per cu cm) and Mg VII (7.8 x 10(exp 9) per cu cm) average line intensity ratios during the interval of quasi-periodic intensity fluctuations, combined with the radiative loss function of an optically thin plasma, yield radiative cooling times of 32 s at 2.0 x 10(exp 6) K, and 46 s at 6.3 x 10(exp 5) K (about half the quasi-period); assuming Fe XIV's density for Fe XXIII yields a radiative cooling time of 10(exp 3) s (13 times the quasi-period) at 1.4 x 10(exp 7) K.

Brosius, Jeffrey W.↗

Predictive Modeling for Differential Diagnosis and Mortality Risk Assessment

The prevalence of electronic health record (EHR) systems has brought prodigious biomedical informatics opportunity. Automated machine learning methods can effectively utilize such data and have become common tools for healthcare predictive modeling. Researches in medical informatics have explored the potential of deep learning and classical models in emergent care scenarios. In particular, predicting differential diagnoses for admissions have proven useful in decreasing unnecessary lab tests and improving inpatient triage decision-making. Moreover, identification of high-risk patients for in-hospital mortality is vitally important to maximize allocation of medical resources.The Medical Information Mart for Intensive Care (MIMIC-III) database, containing de-identified critical care inpatient was used in our study. This data set captures hospital patient laboratory measurements, pharmacologic prescriptions, diagnostic data and procedure event recordings. When considering adult patients and discounting admissions with ICU length of stay less than 24 hours, there were 37,787 unique admissions and 30,414 total patients. We examined the top 25 most prevalent ICD-9 group-level disease specificities in MIMIC-III using a multi-label classification model. In-hospital mortality was modeled as binary classification with 4,155 (13%) adult patients that expired, of which 3,138 (75.5%) were in the ICU setting. The metrics AUC, F1 score, sensitivity and specificity values calculated for each disease label measured prediction performance.The usage of ICD-9 group codes reduced feature dimension from 14,567 to 942 and greatly improved distribution of patient diagnostic categories. Disease temporal patterns were captured by considering the most frequently sampled 6 vital signs and 13 laboratory values. Missing data were imputed at each time-stamp. Time-series raw hourly average values were converted into 5 summary features (mean, standard deviation, number of observations, min & max values). Patient demographic variables such as age, gender, marital status and ethnicity were also factored into the modeling. Choi et al showed that contextual embedding of medical data, diagnostic and procedural codes alone can predict future diagnoses with sensitivity as high as 0.79. We utilized an embedding technique called word2vec which allowed sparse representations of medical history to be transformed into dense word vectors. The mappings captured contextual information by treating each admission as a sentence and learning the most likely neighboring words in a sliding window fashion. Binary and multi-label classification was achieved via collapse models, which do not consider temporal information, as well as recurrent neural networks with regularization, Softmax output layer activation together with categorical cross-entropy as the loss function.

US Army collaboration↗

An Ensemble of Bayesian Neural Networks for Exoplanetary Atmospheric Retrieval

Machine learning (ML) is now used in many areas of astrophysics, from detecting exoplanets in Kepler transit signals to removing telescope systematics. Recent work demonstrated the potential of using ML algorithms for atmospheric retrieval by implementing a random forest (RF) to perform retrievals in seconds that are consistent with the traditional, computationally expensive nested-sampling retrieval method. We expand upon their approach by presenting a new ML model, plan-net, based on an ensemble of Bayesian neural networks (BNNs) that yields more accurate inferences than the RF for the same data set of synthetic transmission spectra. We demonstrate that an ensemble provides greater accuracy and more robust uncertainties than a single model. In addition to being the first to use BNNs for atmospheric retrieval, we also introduce a new loss function for BNNs that learns correlations between the model outputs. Importantly, we show that designing ML models to explicitly incorporate domain-specific knowledge both improves performance and provides additional insight by inferring the covariance of the retrieved atmospheric parameters. We apply plan-net to the Hubble Space Telescope Wide Field Camera 3 transmission spectrum for WASP-12b and retrieve an isothermal temperature and water abundance consistent with the literature. We highlight that our method is flexible and can be expanded to higher resolution spectra and a larger number of atmospheric parameters.

Adam D. Cobb↗

Anomaly Detection in Flight Operational Data Using Deep Learning

In this session, we demonstrate two recently developed deep learning models for anomaly detection in flight operational data by the Data Sciences Group at NASA Ames Research Center. The first model is Convolutional Variational Auto-Encoder (CVAE) [1], which is an unsupervised deep encoder-decoder model, designed specifically for finding anomalies in heterogeneous multivariate time series data. We will demonstrate its application to finding anomalies in streaming data from NASA’s Digital Information Platform’s Fuser source. CVAE identifies data instances that are not representative of expected nominal behavior as anomalous. Since it is an unsupervised approach, the flagged anomalies will need to be reviewed by the subject matter experts (SMEs) for validation and labeling and is designed to assist with vulnerability discovery within Safety Monitoring System programs. The second model is Robust and Explainable Semi-supervised Anomaly Detection (RESAD) model [2], which builds on CVAE to allow learning from both minimally labeled data (previously reviewed by the SMEs) as well as majority unlabeled data. RESAD takes advantage of graph theoretic techniques to propagate the labels from the labeled data to the unlabeled data based on a pre-defined similarity metric and structures the learned feature space from flight time-series so that data of the same class would cluster tightly together. This model characteristic is enabled by training with an augmented loss function and allows learning of a more informative feature space for down-stream tasks such as search and active learning. We demonstrate RESAD using data from the NASA DASHlink project [3].

anomaly detection↗

Differential Equation Approximation Using Gradient-Boosted Quantile Regression

The operation of cyber-physical-human (CPH) systems is subject to various epistemic and aleatory uncertainties. Overall trustworthiness of CPH systems relies on the trustworthiness of its components and their interactions. It is important that computational models comprising the cyber component of CPH provide predictions accompanied by a measure of confidence in model outcomes. Uncertainty quantification (UQ) and propagation are especially important in safety critical CPH systems. Gradient-boosted trees is a modeling approach capable both of learning the dynamics of a system and performing UQ. In this paper, we devise a method for using gradient boosting to learn the dynamics of a second order differential equation and estimate uncertainty at the same time. We do this by creating a custom loss function that trains the model to approximate the second derivative of a noisy time series, and to penalize based on a parameter that corresponds to the desired quantile. The resulting gradient boosting model can simulate stochastic trajectories of the system given a single starting point, that is, it can estimate both the expected trajectory and its uncertainty. We show that the uncertainty estimation is well calibrated and that the model can learn the dynamics even in the presence of noise. We demonstrate the approach on a simple cartpole system.

Autonomous systems↗

Deep Interacting Multiple Model Filtering

In this paper, a deep learning-based multiple model estimation framework is presented for the state estimation of hybrid dynamical systems from high dimensional observations such as camera images. A low dimensional vector which represents the measurement of the latent dynamical system and its corresponding variance are learned using a deep encoder neural network. An Interacting Multiple Model (IMM) filter is used to generate the latent state estimates and covariances using multiple dynamical models, which can be learned using backpropagation through time. The state estimates of the dynamical system and the corresponding covariance matrix are generated from the latent state estimates and covariance using a deep decoder neural network. The whole network is trained in an end-to-end manner using a loss function which minimizes the negative log-likelihood of the neural network parameters. Simulation results are presented using a 2D bouncing ball example and estimation error statistics are computed which demonstrates the accuracy and consistency of the estimation.

Ghananeel Rotithor↗

A Neural Network Parametrization of Volumetric Cloud Fraction Profiles Using Satellite Observations and MERRA-2 Reanalysis Meteorological Data

Clouds play a crucial role in regulating the hydrologic cycle and Earth's radiative energy budget, yet they are often poorly represented in global climate models (GCMs). This study applies deep machine learning techniques to develop a physical parameterization of volumetric cloud fraction (VCF), the fraction of a 3-D grid volume occupied by clouds using satellite lidar-radar measurements. The neural network (NN) captures the complicated relationships between observed VCF profiles and collocated meteorological variables from MERRA-2 reanalysis data. Our results show that the NN model, particularly a sequence-to-sequence long short-term memory (LSTM) network with a sixfactor loss function, effectively learns the underlying cloud physical processes. The NN model outperforms MERRA-2 reanalysis in representing low-level clouds in tropical and subtropical regions and low- and middle-level clouds over midlatitude storm-track regions, and also improves VCF histograms. These improvements are reflected in the vertical distributions of zonally, meridionally, and globally averaged VCFs, geographic distributions of low-, middle-, and high-level clouds, and seasonal variations in monthly-mean VCF. Furthermore, the NN predictions effectively capture the El Niño-Southern Oscillation (ENSO) effects and other interannual variations. The NN parameterization is further evaluated through a sensitivity analysis, in which a single predictor is perturbed at a time. This reveals that relative humidity (RH) is the dominant factor influencing variations in globally averaged VCF at low and middle altitudes, followed by temperature. At higher altitudes, temperature becomes the primary driver of VCF through its effect on RH. Changes in wind components had minimal impact on globally averaged VCF.

Shan Zeng↗