Results for “Deep Learning” — Page 31

NASA NTRS1994

Re-engineering the Multimission Command System at the Jet Propulsion Laboratory

The Operations Engineering Lab (OEL) at JPL has developed the multimission command system as part of JPL's Advanced Multimission Operations System. The command system provides an advanced multimission environment for secure, concurrent commanding of multiple spacecraft. The command functions include real-time command generation, command translation and radiation, status reporting, some remote control of Deep Space Network antenna functions, and command file management. The mission-independent architecture has allowed easy adaptation to new flight projects and the system currently supports all JPL planetary missions (Voyager, Galileo, Magellan, Ulysses, Mars Pathfinder, and CASSINI). This paper will discuss the design and implementation of the command software, especially trade-offs and lessons learned from practical operational use. The lessons learned have resulted in a re-engineering of the command system, especially in its user interface and new automation capabilities. The redesign has allowed streamlining of command operations with significant improvements in productivity and ease of use. In addition, the new system has provided a command capability that works equally well for real-time operations and within a spacecraft testbed. This paper will also discuss new development work including a multimission command database toolkit, a universal command translator for sequencing and real-time commands, and incorporation of telecommand capabilities for new missions.

Alexander, Scott↗

NASA NTRS2007

Near Earth Object (NEO) Mitigation Options Using Exploration Technologies

This work documents the advancements in MSFC threat modeling and mitigation technology research completed since our last major publication in this field. Most of the work enclosed here are refinements of our work documented in NASA TP-2004-213089. Very long development times from start of funding (10-20 years) can be expected for any mitigation system which suggests that delaying consideration of mitigation technologies could leave the Earth in an unprotected state for a significant period of time. Fortunately there is the potential for strong synergy between architecture requirements for some threat mitigators and crewed deep space exploration. Thus planetary defense has the potential to be integrated into the current U.S. space exploration effort. The number of possible options available for protection against the NEO threat was too numerous for them to all be addressed within the study; instead, a representative selection were modeled and evaluated. A summary of the major lessons learned during this study is presented, as are recommendations for future work.

Arnold William↗

NASA NTRSDate not supplied

QuantifyML: How good is my machine learning model?

This paper presents an approach, QuantifyML, which employs model counting to assess the learnability and robustness of machine learning models. Typically the efficacy of machine learning models is determined by computing their accuracy statistically on test data sets. However, this may be misleading, if the test data is not representative of the problem that is being studied. Further, two different models may have the same accuracy on a given data set, measured statistically, but may be very different in their behavior on unseen data. Also, models with high accuracy could have poor adversarial robustness. In QuantifyML, our goal is to precisely quantify the extent to which machine learning models have learned and generalized from the given data. In QuantifyML, a trained model is translated into a C program, which is fed to the CBMC model checking tool to produce a formula in Conjunctive Normal Form (CNF), which in turn is analyzed with state-of-the-art model counters to efficiently obtain precise counts w.r.t different outputs. QuantifyML enables i) evaluating the learnability of models by comparing the counts for the outputs to ground truth, expressed as logical predicates (if available), ii) comparing the performance of different models that may be built with different machine learning algorithms (e.g., decision-trees vs. neural networks), and iii) quantifying the robustness of trained models around given inputs. Our evaluation demonstrates these applications of QuantifyML on decision trees and neural networks trained to learn relational properties of graphs, for which we know the ground truth, and to perform image classification, for which we do not have the ground truth, but we can quantify local robustness.

Deep Neural Networks↗

NASA NTRS2017

Future Food Production System Development Pulling from Space Biology Crop Growth Testing in Veggie

Preliminary crop testing using Veggie indicates the environmental conditions provided by the ISS are generally suitable for food crop production. When plant samples were returned to Earth for analysis, their levels of nutrients were comparable to Earth-grown ground controls. Veggie-grown produce food safety microbiology analysis indicated that space-grown crops are safe to consume. Produce sanitizing wipes were used on-orbit to further reduce risk of foodborne illness. Validation growth tests indicated abiotic challenges of insufficient or excess fluid delivery, potentially reduced air flow leading to excess water, elevated CO2 leading to physiological responses, and microorganisms that became opportunistic pathogens. As NASA works to develop future space food production, several areas of research to define these systems pull from the Veggie technology validation tests. Research into effective, reusable water delivery and water recovery methods for future food production systems arises from abiotic challenges observed. Additionally, impacts of elevated CO2 and refinement of fertilizer and light recipes for crops needs to be assessed. Biotic pulls include methods or technologies to effectively sanitize produce with few consumables and low inputs; work to understand the phytomicrobiome and potentially use it to protect crops or enhance growth; selection of crops with high harvest index and desirable flavors for supplemental nutrition; crops that provide psychosocial benefits, and custom space crop development. Planning for future food production in a deep space gateway or a deep space transit vehicle requires methods of handling and storing seeds, and ensuring space seeds are free of contaminants and long-lived. Space food production systems may require mechanization and autonomous operation, with preliminary testing initiated to identify operations and capabilities that are candidates for automation. Food production design is also pulling from Veggie logistics lessons, as we learn about growing at different scales and move toward developing systems that require less launch mass. Veggie will be used as a test bed for novel food production technologies. Veggie is a relatively simple precursor food production system but the knowledge gained from space biology validation tests in Veggie will have far reaching repercussions on future exploration food production.

Veggie↗

NASA NTRS2017

Future Food Production System Development Pulling From Space Biology Crop Growth Testing in Veggie

Preliminary crop testing using Veggie indicates the environmental conditions provided by the ISS are generally suitable for food crop production. When plant samples were returned to Earth for analysis, their levels of nutrients were comparable to Earth-grown ground controls. Veggie-grown produce food safety microbiology analysis indicated that space-grown crops are safe to consume. Produce sanitizing wipes were used on-orbit to further reduce risk of foodborne illness. Validation growth tests indicated abiotic challenges of insufficient or excess fluid delivery, potentially reduced air flow leading to excess water, elevated CO2 leading to physiological responses, and microorganisms that became opportunistic pathogens. As NASA works to develop future space food production, several areas of research to define these systems pull from the Veggie technology validation tests. Research into effective, reusable water delivery and water recovery methods for future food production systems arises from abiotic challenges observed. Additionally, impacts of elevated CO2 and refinement of fertilizer and light recipes for crops needs to be assessed. Biotic pulls include methods or technologies to effectively sanitize produce with few consumables and low inputs; work to understand the phytomicrobiome and potentially use it to protect crops or enhance growth; selection of crops with high harvest index and desirable flavors for supplemental nutrition; crops that provide psychosocial benefits, and custom space crop development. Planning for future food production in a deep space gateway or a deep space transit vehicle requires methods of handling and storing seeds, and ensuring space seeds are free of contaminants and long-lived. Space food production systems may require mechanization and autonomous operation, with preliminary testing initiated to identify operations and capabilities that are candidates for automation. Food production design is also pulling from Veggie logistics lessons, as we learn about growing at different scales and move toward developing systems that require less launch mass. Veggie will be used as a test bed for novel food production technologies. Veggie is a relatively simple precursor food production system but the knowledge gained from space biology validation tests in Veggie will have far reaching repercussions on future exploration food production. This work is supported by NASA.

Plants↗

NASA NTRS2019

Orion Heat Shield Manufacturing Producibility Improvements for the EM-1 Flight Test Program

This paper describes how the Orion program is incorporating improvements in the heat shield design and manufacturing processes reducing programmatic risk and ensuring crew safety in support of NASA’s Exploration Missions. The Orion program successfully completed the EFT-1 flight test in 2014 and is currently developing the EM-1 spacecraft to meet the test objectives of an orbital mission to the moon and return to earth in 2019. Lessons learned from the EFT-1 manufacturing and flight test experience are being incorporated into a wide variety of vehicle systems and manufacturing processes to reduce risk to the Orion missions and flight crew. A critical contributor to crew safety is the heat shield that protects the crew capsule during re-entry through the earth’s atmosphere for return from deep space. The first flight test vehicle, EFT-1, was manufactured and tested in the Neil Armstrong Operations and Checkout (O&C) facility at KSC to demonstrate early risk reduction including the functionality of the Thermal Protection Systems (TPS) for capsule safe return to earth. The approach for the EFT-1 heat shield utilized a low risk approach using Apollo heritage design and manufacturing processes using an Avcoat TPS ablator with a honeycomb substrate to provide a one piece heat shield to meet the mission re-entry heating environments. The manufacturing processes used honeycomb cell injection guns which were redeveloped from the Apollo Lunar Program processes to build the EFT-1 heat shield. The completed heat shield was transported across the country by aircraft to the O&C at KSC for installation onto the capsule. The EFT-1 heat shield successfully performed its mission and experienced ~80% of the re-entry velocity (50% heating rate) for a lunar return for an Exploration Mission. The second flight test vehicle is the EM-1 mission which will have additional flight systems installed to fly to the moon and return. Heat shield design and producibility improvements have been incorporated in the EM-1 vehicle to meet deep space mission and programmatic requirements. The design continues to use the Avcoat material, but in a “block” configuration to enable improvements in the application processes as well as additional improvements in the carrier structure design and manufacturing operations. Incorporating flight test results and producibility improvements from EFT-1 for the heat shield system design and processes have improved the thermal protection capability, improved the producibility, and cost for EM-1 flight test.

Producibility↗

NASA NTRS1978

Economic evaluation of DSS 13 unattended operations demonstration

The goals and data collection requirements to be used for the economic and performance evaluation indexes and life cycle cost parameters for the upcoming operations demonstration of an automated Deep Space Station (DSS) run unattended and controlled remotely from JPL are presented. These evaluation indexes compare the remote operation of telemetry at DSS 13 with the cost and performance of a comparable manned operation at DSS 11. A description is presented of the data that needs to be collected, how the data will be analyzed, and what can and cannot be learned from this operations demonstration.

Remer, D. S.↗

NASA NTRS2016

Applications and Innovations for Use of High Definition and High Resolution Digital Motion Imagery in Space Operations

The first live High Definition Television (HDTV) from a spacecraft was in November, 2006, nearly ten years before the 2016 SpaceOps Conference. Much has changed since then. Now, live HDTV from the International Space Station (ISS) is routine. HDTV cameras stream live video views of the Earth from the exterior of the ISS every day on UStream, and HDTV has even flown around the Moon on a Japanese Space Agency spacecraft. A great deal has been learned about the operations applicability of HDTV and high resolution imagery since that first live broadcast. This paper will discuss the current state of real-time and file based HDTV and higher resolution video for space operations. A potential roadmap will be provided for further development and innovations of high-resolution digital motion imagery, including gaps in technology enablers, especially for deep space and unmanned missions. Specific topics to be covered in the paper will include: An update on radiation tolerance and performance of various camera types and sensors and ramifications on the future applicability of these types of cameras for space operations; Practical experience with downlinking very large imagery files with breaks in link coverage; Ramifications of larger camera resolutions like Ultra-High Definition, 6,000 [pixels] and 8,000 [pixels] in space applications; Enabling technologies such as the High Efficiency Video Codec, Bundle Streaming Delay Tolerant Networking, Optical Communications and Bayer Pattern Sensors and other similar innovations; Likely future operations scenarios for deep space missions with extreme latency and intermittent communications links.

Grubbs, Rodney↗

NASA NTRS1977

Learning to build large structures in space

The paper examines some of the key technologies and forms of construction know-how that will have to be developed and tested for eventual application to building large structures in space. Construction of a shuttle-tended space construction/demonstration platform would comprehensively demonstrate large structure technology, develop construction capability, and furnish a construction platform for a variety of operational large structures. Completion of this platform would lead to demonstrations of the Satellite Power System (SPS) concept, including microwave transmission, fabrication of 20-m-deep beams, conductor installation, rotary joint installation, and solar blanket installation.

Hagler, T.↗

NASA NTRS2014

Evolution of Software-Only-Simulation at NASA IV and V

Software-Only-Simulations have been an emerging but quickly developing field of study throughout NASA. The NASA Independent Verification Validation (IVV) Independent Test Capability (ITC) team has been rapidly building a collection of simulators for a wide range of NASA missions. ITC specializes in full end-to-end simulations that enable developers, VV personnel, and operators to test-as-you-fly. In four years, the team has delivered a wide variety of spacecraft simulations that have ranged from low complexity science missions such as the Global Precipitation Management (GPM) satellite and the Deep Space Climate Observatory (DSCOVR), to the extremely complex missions such as the James Webb Space Telescope (JWST) and Space Launch System (SLS).This paper describes the evolution of ITCs technologies and processes that have been utilized to design, implement, and deploy end-to-end simulation environments for various NASA missions. A comparison of mission simulators are discussed with focus on technology and lessons learned in complexity, hardware modeling, and continuous integration. The paper also describes the methods for executing the missions unmodified flight software binaries (not cross-compiled) for verification and validation activities.

Embedded↗

NASA NTRS2017

From Prototype to Flight: Qualifying a Ka-Band Parabolic Deployable Antenna (KaPDA) for CubeSats

CubeSats have experienced a number of exciting technological advancements in the past several years. However, until recently, there has been very limited development in the area of high gain CubeSat antennas, which are critical for both high data rate communications and radar science. A Ka-band high gain antenna would provide a 10,000 times increase in data communication rates over an X-band patch antenna and a 100 times increase over state-of-the-art S-band parabolic antennas. Because of this, three years ago the Jet Propulsion Laboratory (JPL) initiated a research and technology development effort to advance CubeSat communication capabilities, with one of the key thrusts being the Ka-band parabolic deployable antenna (KaPDA). This antenna started with the ambitious goal of fitting a 42 dB, 0.5 meter, 35 Ghz antenna in a 1.5U canister. This paper discusses the process of taking the antenna from a first prototype to the flight design, how the design successfully met its goals, and lessons learned. A prototype antenna was constructed in early 2015, and then upgraded to an engineering model at the end of 2016. KaPDA will be flying on the RainCube mission, and earth science CubeSat. KaPDA is the second deployable parabolic antenna to fly on a CubeSat, and the first of its kind to operate at Ka-band enabling a number of opportunities for high rate deep space antenna communications and radar science.

Sauder, Jonathan↗

NASA NTRS2012

Principles to Products: Toward Realizing MOS 2.0

This is a report on the Operations Revitalization Initiative, part of the ongoing NASA-funded Advanced Multi-Mission Operations Systems (AMMOS) program. We are implementing products that significantly improve efficiency and effectiveness of Mission Operations Systems (MOS) for deep-space missions. We take a multi-mission approach, in keeping with our organization's charter to "provide multi-mission tools and services that enable mission customers to operate at a lower total cost to NASA." Focusing first on architectural fundamentals of the MOS, we review the effort's progress. In particular, we note the use of stakeholder interactions and consideration of past lessons learned to motivate a set of Principles that guide the evolution of the AMMOS. Thus guided, we have created essential patterns and connections (detailed in companion papers) that are explicitly modeled and support elaboration at multiple levels of detail (system, sub-system, element...) throughout a MOS. This architecture is realized in design and implementation products that provide lifecycle support to a Mission at the system and subsystem level. The products include adaptable multi-mission engineering documentation that describes essentials such as operational concepts and scenarios, requirements, interfaces and agreements, information models, and mission operations processes. Because we have adopted a model-based system engineering method, these documents and their contents are meaningfully related to one another and to the system model. This means they are both more rigorous and reusable (from mission to mission) than standard system engineering products. The use of models also enables detailed, early (e.g., formulation phase) insight into the impact of changes (e.g., to interfaces or to software) that is rigorous and complete, allowing better decisions on cost or technical trades. Finally, our work provides clear and rigorous specification of operations needs to software developers, further enabling significant gains in productivity.

AMMOS↗

NASA NTRSDate not supplied

Safety Assessment of a Machine Learning-Based Aircraft Emergency Braking System: A Case Study

Machine Learning (ML) is revolutionizing many technological fields, but its use in aviation remains restricted due to stringent certification requirements. Efforts by the aviation community to establish standards for certifying ML-based systems are progressing, yet challenges persist, particularly with safety assessment methods for ML-based systems. This research addresses these challenges through a case study of an autonomous emergency braking system utilizing a computer vision deep neural network (DNN). We demonstrate a safety assessment process tailored to ML-specific concerns, such as low integrity and performance variability in quantitative safety analysis. This study can serve as an illustrative example to facilitate the discussion and convergence on certification aspects for ML-based systems within the aviation community.

Safety certification↗

NASA NTRS2011

Lessons Learned for Planning and Estimating Operations Support Requirements

Operations (phase E) costs are typically small compared to the spacecraft development and test costs. This, combined with the long lead time for realizing operations costs, can lead projects to focus on hardware development schedules and costs, de-emphasizing estimation of operations support requirements during proposal, early design, and replan cost exercises. The Discovery and New Frontiers (D&NF) programs comprise small, cost-capped missions supporting scientific exploration of the solar system. Even moderate yearly underestimates of the operations costs can present significant LCC impacts for deep space missions with long operational durations, and any LCC growth can directly impact the programs ability to fund new missions. The D&NF Program Office at Marshall Space Flight Center recently studied cost overruns for 7 D&NF missions related to phase C/D development of operational capabilities and phase E mission operations. The goal was to identify the underlying causes for the overruns and develop practical mitigations to assist the D&NF projects in identifying potential operations risks and controlling the associated impacts to operations development and execution costs. The study found that the drivers behind these overruns include overly optimistic assumptions regarding the savings resulting from the use of heritage technology, late development of operations requirements, inadequate planning for sustaining engineering and the special requirements of long duration missions (e.g., knowledge retention and hardware/software refresh), and delayed completion of ground system development work. This presentation summarizes the study and the results, providing a set of lessons NASA can use to improve early estimation and validation of operations costs.

Newhouse, Marilyn↗

NASA NTRS2015

Exploring Life Support Architectures for Evolution of Deep Space Human Exploration

Life support system architectures for long duration space missions are often explored analytically in the human spaceflight community to find optimum solutions for mass, performance, and reliability. But in reality, many other constraints can guide the design when the life support system is examined within the context of an overall vehicle, as well as specific programmatic goals and needs. Between the end of the Constellation program and the development of the "Evolvable Mars Campaign", NASA explored a broad range of mission possibilities. Most of these missions will never be implemented but the lessons learned during these concept development phases may color and guide future analytical studies and eventual life support system architectures. This paper discusses several iterations of design studies from the life support system perspective to examine which requirements and assumptions, programmatic needs, or interfaces drive design. When doing early concept studies, many assumptions have to be made about technology and operations. Data can be pulled from a variety of sources depending on the study needs, including parametric models, historical data, new technologies, and even predictive analysis. In the end, assumptions must be made in the face of uncertainty. Some of these may introduce more risk as to whether the solution for the conceptual design study will still work when designs mature and data becomes available.

Anderson, Molly↗

NASA NTRS2015

Exploring Life Support Architectures for Evolution of Deep Space Human Exploration

Life support system architectures for long duration space missions are often explored analytically in the human spaceflight community to find optimum solutions for mass, performance, and reliability. But in reality, many other constraints can guide the design when the life support system is examined within the context of an overall vehicle, as well as specific programmatic goals and needs. Between the end of the Constellation program and the development of the "Evolvable Mars Campaign", NASA explored a broad range of mission possibilities. Most of these missions will never be implemented but the lessons learned during these concept development phases may color and guide future analytical studies and eventual life support system architectures. This paper discusses several iterations of design studies from the life support system perspective to examine which requirements and assumptions, programmatic needs, or interfaces drive design. When doing early concept studies, many assumptions have to be made about technology and operations. Data can be pulled from a variety of sources depending on the study needs, including parametric models, historical data, new technologies, and even predictive analysis. In the end, assumptions must be made in the face of uncertainty. Some of these may introduce more risk as to whether the solution for the conceptual design study will still work when designs mature and data becomes available.

Anderson, Molly S.↗

NASA NTRS1993

Melt spinning study

Containerless processing of materials provides an excellent opportunity to study nucleation phenomena and produce unique materials, primarily through the formation of metastable phases and deep undercoolings. Deep undercoolings can be readily achieved in falling drops of molten material. Extended solute solubilities and greatly refined microstructures can also be obtained in containerless processing experiments. The Drop Tube Facility at Marshall Space Flight Center has played an important role in enhancing that area of research. Previous experiments performed in the Drop Tube with refractory metals has shown very interesting microstructural changes associated with deep undercoolings. It is apparent also that the microstructure of the deep undercooled species may be changing due to the release of the latent heat of fusion during recalescence. For scientific purposes, it is important to be able to differentiate between the microstructures of the two types of metallic species. A review of the literature shows that although significant advances have been made with respect to the engineering aspects of rapid solidification phenomena, there is still much to be learned in terms of understanding the basic phenomena. The two major ways in which rapid solidification processing provides improved structures and hence improved properties are: (1) production of refined structures such as fine dendrites and eutectics, and (2) production of new alloy compositions, microstructures, and phases through extended solid solubility, new phase reaction sequences, and the formation of metallic-glass microstructures. The objective of this work has been to determine the optimal methodology required to extract this excess energy without affecting the thermo-physical parameters of the under-cooled melt. In normal containerless processing experiments recalescence occurs as the melt returns toward the melting point in order to solidify. A new type of experiment is sought in which the resultant microstructure of the undercooled species is frozen in without going through the melting point regime and subsequent near equilibrium solidification of the remaining liquid. This experimental approach entails the design of an appropriate melt spinning system which is compatible with Drop Tube operations and processing constraints. That work is the goal of this study.

Workman, Gary L.↗

NASA NTRS2021

Applying the Cognitive Space Gateway to Swarm Topologies

NASA's future vision for interplanetary networking includes a lunar network, Cube Satellite (CubeSat) constellations, and deep space robotic missions, comprising what could be viewed as a network of networks. Delay-tolerant networking (DTN) architecture and protocols provide a standard network layer among these varying scenarios and mitigate many challenges of the space environment, such as long delays, unplanned service interruptions, and asymmetric links. The Cognitive Space Gateway (CSG) is a routing method in a DTN architecture that uses spiking neural networks as the learning element to optimize routing decisions in a complex environment. This work aims to further develop cognitive networking technologies in several critical areas, including DTN, the CSG algorithm, CubeSat swarm topologies, and cloud services. To test the algorithm in a realistic scenario, the emulated network topology is based on a CubeSat swarm. The swarm may function as a mesh of nodes or as a hub-and-spoke network. An emulation environment will be built upon a commercial cloud service, such as Amazon Web Services (AWS) Elastic Compute Cloud. The cloud environment may enable a flexible, lower maintenance approach versus a multi-hop network based in a physical laboratory. The cloud platform will provide a secure environment allowing for collaboration among government and academic entities.

Ricardo Lent↗