Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Machine learning”

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 1,405 records · Page 78

Tradeoffs When Considering Deep Reinforcement Learning for Contingency Management in Advanced Air Mobility

Air transportation is undergoing a rapid evolution globally with the introduction of Advanced Air Mobility (AAM) and with it comes novel challenges and opportunities for transforming aviation. As AAM operations introduce increasing heterogeneity in vehicle capabilities and density, increased levels of automation are likely necessary to achieve operational safety and efficiency goals. This paper focuses on one example where increased automation has been suggested. Autonomous operations will need contingency management systems that can monitor evolving risk across a span of interrelated (or interdependent) hazards and, if necessary, execute appropriate control interventions via supervised or automated decision making. Accommodating this complex environment may require automated functions (autonomy) that apply artificial intelligence (AI) techniques that can adapt and respond to a quickly changing environment. This paper explores the use of Deep Reinforcement Learning (DRL) which has shown promising performance in complex and high-dimensional environments where the objective can be constructed as a sequential decision-making problem. An extension of a prior formulation of the contingency management problem as a Markov Decision Process (MDP) is presented and uses a DRL framework to train agents that mitigate hazards present in the simulation environment. A comparison of these learning-based agents and classical techniques is presented in terms of their performance, verification difficulties, and development process.

machine learningautonomous systems; flight simulat↗

Improving Learning Performance Through Rational Resource Allocation

This article shows how rational analysis can be used to minimize learning cost for a general class of statistical learning problems. We discuss the factors that influence learning cost and show that the problem of efficient learning can be cast as a resource optimization problem. Solutions found in this way can be significantly more efficient than the best solutions that do not account for these factors. We introduce a heuristic learning algorithm that approximately solves this optimization problem and document its performance improvements on synthetic and real-world problems.

resource optimization↗

ICE-RASSOR: Intelligent Capabilities Enhanced Regolith Advanced Surface Systems Operations Robot

NASA’s Regolith Advanced Surface Systems Operations Robot (RASSOR) is principally designed to mine and deliver regolith for In-Situ Resource Utilization (ISRU) processing. RASSOR’s design enables it to efficiently collect and deposit regolith, return collected material for processing, and myriad related ISRU activities. To reliably perform these operations on the lunar surface, RASSOR software and sensory systems need to be robust and maximize the information extracted from on-board sensory. Herein, we present preliminary findings from the Intelligent Capabilities Enhanced RASSOR project. We apply supervised learning using real data to estimate the soil mass collected without the need for mass flow rate monitors or other explicate sensing techniques. We also create a reduced-order simulation environment to develop autonomous trenching controllers via reinforcement learning and proto-type state estimation architectures. Our initial results suggest that excavated regolith mass can be inferred within 2.9% RMS error of full scale, and reinforcement learning for autonomous operations has learned viable trenching strategies and helped identify desirable sensing capabilities, arrangements, and considerations. Future work includes regolith mass estimation during dynamic operation, expanding our simulation to more complex environments, and transfer learning from simulation to hardware.

machine learning↗

Microstructure Segmentation with Deep Learning Encoders Pre-Trained on a Large Microscopy Dataset

This study examined the improvement of microscopy segmentation accuracy by transfer learning from a large dataset of microscopy images called MicroNet. Many neural network encoder architectures, including VGG, Inception, and ResNet, were trained on over 100,000 labelled microscopy images from 54 classes. These pre-trained encoders were then embedded into multiple segmentation architectures including U-Net and DeepLabV3+ to evaluate segmentation performance on newly created benchmark microscopy datasets. Compared to ImageNet pre-training, models pre-trained on MicroNet generalized better to out-of-distribution micrographs taken under different imaging and sample conditions and were more accurate with less training data. When training with only a single Ni-superalloy image, pre-training on MicroNet produced a 72.2 percent reduction in relative segmentation error. These results suggest that transfer learning from large in-domain datasets generate models with learned feature representations that are more useful for downstream tasks and will likely improve any microscopy image analysis technique that can leverage pre-trained encoders.

machine learning↗

Microstructure Segmentation With Deep Learning Encoders Pre-Trained on a Large Microscopy Dataset

This study examined the improvement of microscopy segmentation intersection over union accuracy by transfer learning from a large dataset of microscopy images called MicroNet. Many neural network encoder architectures were trained on over 100,000 labeled microscopy images from 54 material classes. These pre-trained encoders were then embedded into multiple segmentation architectures including UNet and DeepLabV3+ to evaluate segmentation performance on created benchmark microscopy datasets. Compared to ImageNet pre-training, models pre-trained on MicroNet generalized better to out-of-distribution micrographs taken under different imaging and sample conditions and were more accurate with less training data. When training with only a single Ni-superalloy image, pre-training on MicroNet produced a 72.2% reduction in relative intersection over union error. These results suggest that transfer learning from large in-domain datasets generate models with learned feature representations that are more useful for downstream tasks and will likely improve any microscopy image analysis technique that can leverage pre-trained encoders.

machine learning↗

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-of-the-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Deep-learning↗

Image Database Exploration: Progress and Challenges

In this paper we discuss the general problem of automated image database exploration, the particular aspects of image databases which distinguish them from other databases, and how this impacts the application of off-the-shelf learning algorithms to problems of this nature. Current progress will be illustrated using two large-scale image exploration projects at JPL. The paper concludes with a discussion of current and future challenges.

image database exploration pattern recognition tec↗

Prediction of Weather Impacted Airport Capacity using Ensemble Learning

Ensemble learning with the Bagging Decision Tree (BDT) model was used to assess the impact of weather on airport capacities at selected high-demand airports in the United States. The ensemble bagging decision tree models were developed and validated using the Federal Aviation Administration (FAA) Aviation System Performance Metrics (ASPM) data and weather forecast at these airports. The study examines the performance of BDT, along with traditional single Support Vector Machines (SVM), for airport runway configuration selection and airport arrival rates (AAR) prediction during weather impacts. Testing of these models was accomplished using observed weather, weather forecast, and airport operation information at the chosen airports. The experimental results show that ensemble methods are more accurate than a single SVM classifier. The airport capacity ensemble method presented here can be used as a decision support model that supports air traffic flow management to meet the weather impacted airport capacity in order to reduce costs and increase safety.

Weather impact↗

Anomaly Detection, Active Learning, Precursor Identification,and Human Knowledge for Autonomous System Safety

The project Autonomy Teaming and TRajectories for ComplexTrusted Operational Reliability (ATTRACTOR) researched and developed Artificial Intelligence with application to multi-Unmanned Aerial Systems (UAS) missions. Such missions, like other complex systems-of-systems, are likely to have previously-unknown, safety relevant anomalies occur due to many possible factors including system failures or degradations, emergent behavior, changes in the environment in which the systems operate, changes in the way the systems are operated. We discuss the application of anomaly detection, active learning, and precursor identification to identify such anomalies and the conditions under which they are more likely to appear. We demonstrate results on simulated multi-UAS missions that show promise to be applied to real missions.

machine learning↗

Prognostics for Systems Health Management - Model and Hybrid Based Approaches. Where are We Heading?

To facilitate and solve the prediction problem, awareness of the current state and health of the system is key, since it is necessary to perform condition-based system health predictions. To accurately predict the future state of any system, it is required to possess knowledge of its current health state and future operational conditional. In case of next generation electric aircrafts, computing remaining flying time is safety-critical, since an aircraft that runs out of power (battery charge) while in the air will eventually lose control leading to catastrophe. In order to tackle and solve the prediction problem, it is essential to have awareness of the current health state of the system, especially since it is necessary to perform condition-based predictions. To be able to predict the future state of the system, it is also required to possess knowledge of the current and future operational conditions and flight profiles for accurate estimation of end-of-discharge (EOD) for the batteries. Similar framework can be implemented to other complex systems and subsystems. Our research approach is to develop a system level health monitoring safety indicator which runs estimation and prediction algorithms to estimate remaining useful life predictions at system, subsystem swell as component levels. Given models of the current and future system behavior, a general approach of model-based prognostics is discussed as a solution to the prediction problem and further for decision making. Data driven prognostics approaches have been equally used with good results in the past, where respective approaches have their own challenges to tackle. This limits their applicability to complex real-world domains: (a) high complexity or incompleteness of physics-based models and (b) limited representativeness of the training dataset for data-driven models. With the advent of internet of things for data collection and increased use of ML algorithms, hybrid approaches are the next avenue to reduce the challenges and achieve better results. An hybrid framework for fusing information from physics-based performance models along with deep learning algorithms for prognostics of complex safety critical systems is presented. In this framework, we use physics-based performance models to infer unobservable model parameters related to the system's components health solving a calibration problem.

Prognostics↗

Sampling Functions from Gaussian Processes and Structured Covariance Gaussian Networks

When learning aerodynamic models from data, it is critical to incorporate estimates of model uncertainty. This motivates the design of probabilistic aerodynamic databases which can be sampled to generate physically and statistically plausible aerodynamic models. In this talk we discuss how to sample deterministic functions from two different kinds of probabilistic models and demonstrate their use. First, Gaussian Process Regressors (GPRs) are a widely used probabilistic kernel-based model which can be thought of as Gaussian distributions over functions. GPRs are generally trained by maximizing the marginal likelihood of seeing the training data over the kernel parameter space. Sample functions are easily generated by drawing points from the Gaussian distribution at desired input points. However, when the points are not known ahead of time, the classical sampling approach is not possible since successive function samples will generate different function realizations. We present an approach for sampling consistent function evaluations from a GPR over multiple samples. Second, we describe a neural network architecture which learns a conditional Gaussian distribution by maximizing the marginal likelihood at each point in the input space. We then discuss and compare several options for generating sample functions which match this distribution. Finally, we demonstrate the use of these probabilistic aerodynamic models in an atmospheric reentry simulation.

Gaussian process regression↗

A Deep Learning Approach to Fast Radiative Transfer

Due to the sheer volume of data, leveraging satellite instrument observations effectively in a data assimilation context for numerical weather prediction or for remote sensing requires a radiative transfer model as an observation operator that is both fast and accurate at the same time. Physics-based line-by-line radiative transfer (RT) models fulfil the requirement for accuracy, but are too slow and too costly in computational terms for operational applications. Therefore, fast methods were developed to be able to perform fast RT calculations using techniques such as spectral sampling or pre-computed look-up tables. The operational fast models currently calculate the absorption and scattering coefficients from the pre-computed regression coefficients and atmospheric state and cloud profiles. As a novel solution to this problem, this work investigates a deep learning approach to replace the regression coefficients in the fast RT models. A selection of hidden-layer neural network configurations is trained against atmospheric transmittance profile data computed by an accurate line-by-line model and their performance is evaluated and their advantages and disadvantages are discussed.

Machine learning↗

Ensemble-Based Deep Learning for Estimating PM2.5 over California with Multisource Big Data Including Wildfire Smoke

Estimating PM2.5 concentrations and their prediction uncertainties at a high spatiotemporal resolution is important for air pollution health effect studies. This is particularly challenging for California, which has high variability in natural (e.g. wildfires, dust) and anthropogenic emissions, meteorology, topography (e.g. desert surfaces, mountains, snow cover) and land use.

air quality↗

Structured Covariance Gaussian Networks for Orion Crew Module Aerodynamic Uncertainty Quantification

In this paper we propose a new approach for nonlinear regression and uncertainty quantification. The method is based on a pair of neural networks which parameterize mean and dense covariance functions of a multivariate Gaussian process, trained together to maximize the log-likelihood of observing the given data. The covariance matrix is made positive definite at every input by construction. We also propose a sampling approach that produces viable surrogate function realizations from the Gaussian process. We call the proposed model a Structured Covariance Gaussian Network (SCGN). We illustrate the use of SCGNs for learning an aerodynamic response surface with built-in uncertainty for the Orion crew module. We find that SCGN provides an efficient and systematic way to learn nonlinear functional relationships and dense covariances. We compare results to a baseline Gaussian process regressor and observe that the SCGN provides comparable uncertainty descriptions with improved scalability to dataset size. The sample functions generated by SCGN are fast to evaluate online and are therefore convenient for use in trajectory simulations. These results suggest that SCGN may be a viable computational method for aerodynamic uncertainty quantification.

machine learning↗

Structured Covariance Gaussian Networks for Orion Crew Module Aerodynamic Uncertainty Quantification

In this paper we propose a new approach for nonlinear regression and uncertainty quantification. The method is based on a pair of neural networks which parameterize mean and dense covariance functions of a multivariate Gaussian process, trained together to maximize the log-likelihood of observing the given data. The covariance matrix is made positive definite at every input by construction. We also propose a sampling approach that produces viable surrogate function realizations from the Gaussian process. We call the proposed model a Structured Covariance Gaussian Network (SCGN). We illustrate the use of SCGNs for learning an aerodynamic response surface with built-in uncertainty for the Orion crew module. We find that SCGN provides an efficient and systematic way to learn nonlinear functional relationships and dense covariances. We compare results to a baseline Gaussian process regressor and observe that the SCGN provides comparable uncertainty descriptions with improved scalability to dataset size. The sample functions generated by SCGN are fast to evaluate online and are therefore convenient for use in trajectory simulations. These results suggest that SCGN may be a viable computational method for aerodynamic uncertainty quantification.

machine learning↗

Predicting Airport Runway Configuration for Decision-Support Using Supervised Learning

One of the most challenging tasks for air traffic controllers is runway configuration management (RCM). It deals with the optimal selection of runways to operate on (for arrivals and departures) based on current and forecast of traffic, surface wind speed, wind direction, other environmental variables, noise constraints, and several other airport-specific factors. In this paper, a methodology using supervised learning is developed to build a predictive model for RCM decision-support from large volumes of historical data. Data from two full years (2018 and 2019) related to current and forecast weather, demand/capacity, etc. is collected, analyzed, and fused together. A variety of supervised learning algorithms are tested for predicting runway configuration and hyperparameter tuning is carried out to select the best performing model. The validation process involves two airports of low (Charlotte Douglas International Airport, CLT) and high (Denver International Airport, DEN) complexity of configuration decision-making. The results show significant promise for the two airports with test accuracy of 93% (CLT) and 73% (DEN). The methodology is scalable and generalizable to other airports across the U.S. National Airspace System.

air traffic management↗