Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “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 487 records · Page 27

Real-time Unimpeded Taxi Out Machine Learning Service

This paper describes a study on the estimation of the unimpeded taxi out time using Machine Learning (ML) tools and proposes an implementation that can be used to make real-time predictions at any airport in the National Airspace System. Kedro, an open-source pipeline framework, is used to develop the model definition and training. Models are stored in scikit-learn containers on a MLFlow server where they can be retrieved and served to make predictions in the live system. These open source frameworks provide common structures between ML services, allow for easier maintenance and updates, and overall deliver an easier CI/CD (Continuous Integration/Continuous Deployment) process. The current models were trained on data acquired at KCLT and KDFW from June 1st to December 31st, 2019 and compute taxi time in the ramp, airport movement area (AMA) and total (from gates to runways). The current versions of the models achieve relatively low uncertainties of about 10 to 15% for the total and AMA taxi times and about 20% for the ramp taxi time at both KCLT and KDFW. Initial tests on offline data from 2020 and 2021 show a small degradation (10 to 15%) in accuracy performance indicating the model’s resilience to operational changes over time.

machine learning↗

Predicting Arrival and Departure Runway Assignments with Machine Learning

Runway assignments at major airports are made by air traffic controllers subject to various constraints, and to achieve various objectives. In this research, we describe our efforts training machine learning (ML) models to predict both departure and arrival runway assignments using an entirely data-driven approach. This approach is compared to existing rule-based approaches developed in previous research using input from Subject Matter Experts. The models have features derived from various FAA data feeds, and leverage multiple machine learning algorithms. Results for models trained for nine major U.S. airports are described and compared to one another across various important dimensions. Particular attention was paid to developing a repeatable framework for training these models so the approach could be scaled to other airports, and to developing models that are useful in a real-time environment. In addition, the models were designed to be functional in a real-time environment to support NASA’s ATD-2 project, as part of an ML-powered shadow system to compare against the performance of the fielded system.

machine learning↗

A Machine Learning Approach to Predict Aircraft Landing Times using Mediated Predictions from Existing Systems

We developed a novel approach for predicting the landing time of airborne flights in real-time operations. The first step predicts a landing time by using mediation rules to select from among physics-based predictions (relying on the expected flight trajectory) already available in real time in the Federal Aviation Administration System Wide Information Management system data feeds. The second step uses a machine learning model built upon the mediated predictions. The model is trained to predict the error in the mediated prediction, using features describing the current state of an airborne flight. These features are calculated in real time from a relatively small number of data elements that are readily available for airborne flights. Initial results based on five months of data at six large airports demonstrate that incorporating a machine learning model on top of the mediated physics-based prediction can lead to substantial additional improvements in prediction quality.

Machine learning↗

Real-time Unimpeded Taxi Out Machine Learning Service

This presentation describes a study on the estimation of the unimpeded taxi out time using Machine Learning (ML) tools and proposes an implementation that can be used to make real-time predictions at any airport in the National Airspace System. Kedro, an open-source pipeline framework, is used to develop the model definition and training. Models are stored in scikit-learn containers on a MLFlow server where they can be retrieved and served to make predictions in the live system. These open source frameworks provide common structures between ML services, allow for easier maintenance and updates, and overall deliver an easier CI/CD (Continuous Integration/Continuous Deployment) process. The current models were trained on data acquired at KCLT and KDFW from June 1st to December 31st, 2019 and compute taxi time in the ramp, airport movement area (AMA) and total (from gates to runways). The current versions of the models achieve relatively low uncertainties of about 10 to 15% for the total and AMA taxi times and about 20% for the ramp taxi time at both KCLT and KDFW. Initial tests on offline data from 2020 and 2021 show a small degradation (10 to 15%) in accuracy performance indicating the model’s resilience to operational changes over time.

Machine Learning↗

A Machine Learning Approach to Predict Aircraft Landing Times using Mediated Predictions from Existing Systems

We developed a novel approach for predicting the landing time of airborne flights in real-time operations. The first step predicts a landing time by using mediation rules to select from among physics-based predictions (relying on the expected flight trajectory) already available in real time in the Federal Aviation Administration System Wide Information Management system data feeds. The second step uses a machine learning model built upon the mediated predictions. The model is trained to predict the error in the mediated prediction, using features describing the current state of an airborne flight. These features are calculated in real time from a relatively small number of data elements that are readily available for airborne flights. Initial results based on five months of data at six large airports demonstrate that incorporating a machine learning model on top of the mediated physics-based prediction can lead to substantial additional improvements in prediction quality.

Machine learning↗

Predicting Arrival and Departure Runway Assignments with Machine Learning

Runway assignments at major airports are made by air traffic controllers subject to various constraints, and to achieve various objectives. In this research, we describe our efforts training machine learning (ML) models to predict both departure and arrival runway assignments using an entirely data-driven approach. This approach is compared to existing rule-based approaches developed in previous research using input from Subject Matter Experts. The models have features derived from various FAA data feeds, and leverage multiple machine learning algorithms. Results for models trained for nine major U.S. airports are described and compared to one another across various important dimensions. Particular attention was paid to developing a repeatable framework for training these models so the approach could be scaled to other airports, and to developing models that are useful in a real-time environment. In addition, the models were designed to be functional in a real-time environment to support NASA’s ATD-2 project, as part of an ML-powered shadow system to compare against the performance of the fielded system.

machine learning↗

Machine Learning Models to Predict Cognitive Impairment of Rodents Subjected to Space Radiation

This research uses machine-learned computational analyses to predict the cognitive performance impairment of rats induced by irradiation. The experimental data in the analyses is from a rodent model exposed to ≤ 15 cGy of individual Galactic Cosmic Radiation (GCR) ions: 4He, 16O, 28Si, 48Ti, or 56Fe, expected for a Lunar or Mars mission. This work investigates rats at a subject-based level and uses performance scores taken before irradiation to predict impairment in Attentional Set-shifting (ATSET) data post-irradiation. Here, the worst performing rats of the control group define the impairment thresholds based on population analyses via cumulative distribution functions, leading to the labeling of impairment for each subject. A significant finding is the exhibition of a dose-dependent increasing probability of impairment for 1 to 10 cGy of 28Si or 56Fe in the Simple Discrimination (SD) stage of the ATSET, and for 1 to 10 cGy of 56Fe in the Compound Discrimination (CD) stage. On a subject-based level, implementing Machine Learning (ML) classifiers such as the Gaussian Naïve Bayes, Support Vector Machine, and Artificial Neural Networks identifies rats that have a higher tendency for impairment after GCR exposure. The algorithms employ the experimental prescreenperformance scores as multidimensional input features to predict each rodent’s susceptibility to cognitive impairment due to space radiation exposure. The receiver operating characteristic and the precision-recall curves of the ML models show a better prediction of impairment when 56Feis the ion in question in both SD and CD stages. They, however, do not depict impairment due to 4Hein SD and 28Siin CD, suggesting no dose-dependent impairment response in these cases. One key finding of our study is that prescreen performance scores can be used to predict the ATSET performance impairments. This result is significant to crewed space missions as it supports the potential of predicting an astronaut’s impairment in a specific task before spaceflight through the implementation of appropriately trained ML tools. Future research can focus on constructing ML ensemble methods to integrate the findings from the methodologies implemented in this study for morerobust predictionsof cognitive decrements due to space radiation exposure.

space radiation↗

A Framework for Deep Learning Emulation of Numerical Models With a Case Study in Satellite Remote Sensing

Numerical models based on physics represent the state of the art in Earth system modeling and comprise our best tools for generating insights and predictions. Despite rapid growth in computational power, the perceived need for higher model resolutions overwhelms the latest generation computers, reducing the ability of modelers to generate simulations for understanding parameter sensitivities and characterizing variability and uncertainty. Thus, surrogate models are often developed to capture the essential attributes of the full-blown numerical models. Recent successes of machine learning methods, especially deep learning (DL), across many disciplines offer the possibility that complex nonlinear connectionist representations may be able to capture the underlying complex structures and nonlinear processes in Earth systems. A difficult test for DL-based emulation, which refers to function approximation of numerical models, is to understand whether they can be comparable to traditional forms of surrogate models in terms of computational efficiency while simultaneously reproducing model results in a credible manner. A DL emulation that passes this test may be expected to perform even better than simple models with respect to capturing complex processes and spatiotemporal dependencies. Here, we examine, with a case study in satellite-based remote sensing, the hypothesis that DL approaches can credibly represent the simulations from a surrogate model with comparable computational efficiency. Our results are encouraging in that the DL emulation reproduces the results with acceptable accuracy and often even faster performance. We discuss the broader implications of our results in light of the pace of improvements in high-performance implementations of DL and the growing desire for higher resolution simulations in the Earth sciences.

Bayesian Deep Learning↗

Learning from Past Missions for Today’s Case Studies

As the small spacecraft industry continues to grow, the extensive selection of subsystem parts and service providers can present unique challenges in mission implementation. These challenges include inadequate trade studies, missing lessons learned, and unknown solutions to common pitfalls. Trade studies frequently fall short because products and services ultimately do not meet customer expectations; the actual outcomes of a particular service or technology are not always shared with others seeking similar practices. It is difficult to: quantify the viability and robustness of the desired products and services, truly understand what solutions exist, and account for unique failures from previous smallsat missions. One reason for the lack of information as input for trade studies is that it is problematic to capture and disseminate the relevant lessons learned, experienced anomalies, and programmatic issues in both the laboratory and on-orbit setting as these types of information can be extremely sensitive and are specific to each unique mission. Therefore, they are not always made publicly available. The utilization and sharing of existing tools, databases, and literature can help distribute useful information for future smallsat missions to hopefully avoid these common pitfalls. This paper will provide the required framework for current and future smallsat mission implementation by providing best practices and identifying helpful resources to assist with subsystem parts and services trade study selection.

Small Spacecraft↗

Assuring Safety-Critical Machine Learning Enabled Systems: Challenges and Promise

Machine learning is increasingly being used in safety-critical systems, where the public safety requires a rigorous assurance process. We shall outline how assurance processes work for conventional systems and identify the primary difficulty in applying them to machine learning enabled systems. We will then outline a path forward including identifying where considerable basic research remains

Safety-critical systems↗

Deep Reinforcement Learning based Routing in an Air-to-Air Ad-hoc Network

This paper studies the Multiple Sources and Multiple Destinations (MSMD) routing problem in a dynamic Air-to-Air Ad-hoc Network (AAAN). We consider a spectrum limited scenario where multiple links have to share the same frequency channel so that co-channel interference becomes inevitable. As a result, routing decisions and spectrum access are coupled and must be jointly considered. This paper proposes a deep Q-learning based algorithm to find an optimal routing and channel selection strategy that minimizes the end-to-end communication delay. Specifically, under the assumption that only local information is available to every node, the Deep Q-Network (DQN) is trained offline to learn the optimal routing and channel selection strategy. After the trained DQN is implemented in every node, multiple relay nodes can simultaneously determine their next-hop relay and channel selections in real-time. Simulation results demonstrate the efficacy of our proposed algorithm.

AAAN↗

Use of Machine Learning to Reduce Uncertainties in Particle Number Concentration and Aerosol Indirect Radiative Forcing Predicted by Climate Models

The radiative forcing of anthropogenic aerosols associated with aerosol–cloud interactions (RF(sub aci)) remains the largest source of uncertainty in climate prediction. The calculation of particle number concentration (PNC), one of the critical parameters affecting RF(sub aci), is generally simplified in climate models. Here we employ outputs from long-term (30-years) simulations of a global size-resolved (sectional) aerosol microphysics model and a machine-learning tool to develop a Random Forest Regression Model (RFRM) for PNC. We have implemented the PNC RFRM in GISS-ModelE2.1 with a mass-based One-Moment Aerosol module, which is one of CMIP6 models. Compared to the default setting, the GISS-ModelE2.1 simulation based on RFRM reduces the changes of cloud droplet number concentration associated with anthropogenic emissions, and decreases the RF(sub aci) from −1.46 W⋅m(exp −2) to −1.11 W⋅m(exp −2). This work highlights a promising approach based on machine learning to reduce uncertainties of climate models in predicting PNC and RF(sub aci) without compromising their computing efficiency.

Radiative forcing↗

Deep Reinforcement Learning based Routing in an Air-to-Air Ad-hoc Network

This paper studies the Multiple Sources and Multiple Destinations (MSMD) routing problem in a dynamic Air-to-Air Ad-hoc Network (AAAN). We consider a spectrum limited scenario where multiple links have to share the same frequency channel so that co-channel interference becomes inevitable. As a result, routing decisions and spectrum access are coupled and must be jointly considered. This paper proposes a deep Q-learning based algorithm to find an optimal routing and channel selection strategy that minimizes the end-to-end communication delay. Specifically, under the assumption that only local information is available to every node, the Deep Q-Network (DQN) is trained offline to learn the optimal routing and channel selection strategy. After the trained DQN is implemented in every node, multiple relay nodes can simultaneously determine their next-hop relay and channel selections in real-time. Simulation results demonstrate the efficacy of our proposed algorithm.

AAAN↗

Scheduling For Urban Air Mobility Using Safe Learning

This work considers the scheduling problem for Urban Air Mobility (UAM) vehicles travelling between origin-destination pairs with both hard and soft trip deadlines. Each route is described by a discrete probability distribution over trip completion times (or delay) and over interarrival times of requests (or demand) for the route along with a fixed hard or soft deadline. Soft deadlines carry a cost that is incurred when the deadline is missed. An online, safe scheduler is developed that ensures that hard deadlines are never missed and that average cost of missing soft deadlines is minimized. The system is modelled as a Markov Decision Process (MDP) and safe model based learning is used to find the probabilistic distributions over route delays and demand. Monte Carlo Tree Search (MCTS) Earliest Deadline First (EDF) is used to safely explore the learned models in an online fashion and develop a near-optimal non-preemptive scheduling policy. These results are compared with Value Iteration (VI) and MCTS (Random) scheduling solutions.

Urban Air Mobility↗

Scheduling for Urban Air Mobility using Safe Learning

This work considers the scheduling problem for Urban Air Mobility (UAM) vehicles travelling between origin-destination pairs with both hard and soft trip deadlines. Each route is described by a discrete probability distribution over trip completion times (or delay) and over interarrival times of requests (or demand) for the route along with a fixed hard or soft deadline. Soft deadlines carry a cost that is incurred when the deadline is missed. An online, safe scheduler is developed that ensures that hard deadlines are never missed and that average cost of missing soft deadlines is minimized. The system is modelled as a Markov Decision Process (MDP) and safe model based learning is used to find the probabilistic distributions over route delays and demand. Monte Carlo Tree Search (MCTS) Earliest Deadline First (EDF) is used to safely explore the learned models in an online fashion and develop a near-optimal non-preemptive scheduling policy. These results are compared with Value Iteration (VI) and MCTS (Random) scheduling solutions.

Urban Air Mobility↗

NASA Symposium on Turbulence Modeling: Roadblocks, and the Potential for Machine Learning

A three-day symposium sponsored by NASA was held in July 2022 in Suffolk, Virginia on the subject of Turbulence Modeling: Roadblocks, and the Potential for Machine Learning. This meeting brought together over 80 experts from academia, government, and industry to discuss critical issues for Reynolds-averaged Navier-Stokes turbulence and transition models, as well as to evaluate the results from a collaborative testing challenge based on data-driven methods and machine learning technology. This report puts this symposium in context with an earlier similar meeting and summarizes many of the questions, discussions, and conclusions that arose from it. Next steps are suggested.

machine learning↗

Assuring Safety-Critical Machine Learning Enabled Systems: Challenges and Promise

Machine learning is increasingly being used in safety-critical systems, where the public safety requires a rigorous assurance process. We shall outline how assurance processes work for conventional systems and identify the primary difficulty in applying them to machine learning enabled systems. We will then outline a path forward including identifying where considerable basic research remains.

machine learning↗

NASA's Composite Overwrap Lessons Learned on 40k Thrust Chamber Assemblies

Filament wound carbon fiber/polymer composite overwraps combined with additively manufactured (AM) copper alloy combustion chambers are a critical part of NASA’s Rapid Analysis and Manufacturing Propulsion Technology (RAMPT) project. The composite overwrap is a high strength-to-weight material and allows for significant weight reduction of the high-pressure thrust chamber assembly, reduction of overall cost, and reduced fabrication time while offering structural robustness. Integrated AM/composite overwrap chamber assemblies using 2,000 lbf (2k) and 7,000 lbf (7k) thrust were successfully hot fire tested at NASA Marshall Space Flight Center (MSFC). These assemblies allowed the team to learn lessons related to the design, analysis, and manufacturing of these integrated assemblies. A final goal of the RAMPT project is to design and test a 40,000 lbf (40k) thrust chamber assembly. The40k chamber presented the team several new challenges including an integral inlet port in the main body of the composite overwrap. Through multiple iterations of manufacturing techniques, varied materials, and Non-Destructive Evaluations (NDE), many lessons were learned to obtain a final design solution that would not only allow for the interrupted port inlet but also allow for sufficient material margins for the most critically stressed part of the chamber. This paper will cover the manufacturing techniques exploring fabric designs to help subsidize the coverage in the areas left bare by the inlet port during the filament winding process and custom tooling needed for the final full chamber processing. The original material choice of bismaleimide (BMI) 5250-4 from the 7k chamber performed well for the extreme environment of hot fire testing but proved to be difficult with the increased scale and interrupted port design of the 40kchamber. Discussion in this paper will cover these changes, including evaluation of the design, comparison to the prior 2k and 7k designs, and show the final hardware that will be tested. One primary goal of this work and RAMPT project is to provide this development process and data to industry for infusion into future engine designs.

Filament Wound↗