Search NASA⌕ Search

SEARCH · Search NASA

Results for “continual 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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
NASA NTRS2005

Multi Agent Reward Analysis for Learning in Noisy Domains

In many multi agent learning problems, it is difficult to determine, a priori, the agent reward structure that will lead to good performance. This problem is particularly pronounced in continuous, noisy domains ill-suited to simple table backup schemes commonly used in TD(lambda)/Q-learning. In this paper, we present a new reward evaluation method that allows the tradeoff between coordination among the agents and the difficulty of the learning problem each agent faces to be visualized. This method is independent of the learning algorithm and is only a function of the problem domain and the agents reward structure. We then use this reward efficiency visualization method to determine an effective reward without performing extensive simulations. We test this method in both a static and a dynamic multi-rover learning domain where the agents have continuous state spaces and where their actions are noisy (e.g., the agents movement decisions are not always carried out properly). Our results show that in the more difficult dynamic domain, the reward efficiency visualization method provides a two order of magnitude speedup in selecting a good reward. Most importantly it allows one to quickly create and verify rewards tailored to the observational limitations of the domain.

Tumer, Kagan↗
NASA NTRSDate not supplied

Cyber Resiliency and the Implementation of a Host-Based Intrusion Detection System in an Urban Air Mobility Environment

With the growth in Urban Air Mobility systems and the increasing reliance on interconnected technologies, ensuring the security of these complex components has become critical. As cities evolve into smart urban centers, the vulnerability to cyber threats escalates, possibly endangering citizens safety and the efficiency of transportation networks.In response to these challenges, this paper presents a study on the need for cyber resilient techniques within future air traffic environments. It will pay specific attention to the implementation of a Host-Based Intrusion Detection System (HIDS) utilizing Atomic OSSEC software, tailored specifically to a NASA simulation of an UrbanAirMobility environments’ unique demands. Further, this study seeks to outline the rational for NASA’s recommendation for a HIDS in such environments. It explores the design, development, and deployment of the proposed HIDS, focusing on its adaptability to monitor the hybrid nature of the Urban Air Mobility environment. Leveraging machine learning algorithms and anomaly detection techniques, the HIDS is equipped to continuously monitor and analyze the behavior of individual host systems, vehicles, and devices, thereby providing a proactive approach to threat detection. Implementing a HIDS is a pivotal strategy for enhancing cyber resiliency, as it gives an organization granular visibility into internal system activities, enables rapid detection and response to anomalous behavior and cyber threats, and fortifies the organizations overall cybersecurity posture. Finally, this study aims to provide recommendations and include learned takeaways that the Urban Air Mobility industry should consider. In brief, this paper highlights the significance of host-based intrusion detection in UrbanAirMobility environments and underscores the necessity of tailored security solutions to safeguard against emerging cyber threats.

UAM↗
DOE OSTI2025

Multi-Agent Hierarchical Deep Reinforcement Learning for HVAC Control With Flexible DERs

As electricity consumption in commercial and residential buildings continues to rise, reducing energy costs presents an increasing challenge. Heating, ventilating, and air-conditioning (HVAC) systems, which typically account for 40%-50% of a building's energy use, are prime targets for energy savings. Intelligent control of HVAC temperature through the exploitation of HVAC load flexibility brings significant potential to reduce energy consumption and electricity expenses. The nonlinear models of HVAC systems challenge traditional control methods, while the uncertainty introduced by HVAC load flexibility complicates distributed energy resource (DER) management using conventional optimal dispatch techniques. In response to these challenges, we propose a hierarchical multi-agent deep reinforcement learning (DRL) approach. The lower-level agents focus on balancing comfort and energy conservation, while the upper-level DRL agents optimize the use of DERs to reduce peak demand based on the control outcomes of the HVAC by the lower-level agents. Here, in the upper-level agents, we incorporate a multi-agent structure based on ensemble learning, which acts based on historical and current data without relying on precise load forecasting to address the delayed rewarding issue in DRL. This allows for the effective reduction of energy costs. The proposed method is tested using a real-world microgrid comprising 413 buildings in Southern California, and the results demonstrate that our approach can significantly reduce overall electricity bills while ensuring the comfort of consumers and residents.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Backpropagation-based learning with local derivative approximation and memory replay in biologically plausible neural systems

When learning, the brain modifies individual synaptic connections to reach a desired behavior. Animal and human brains have been shown to be incredibly capable of learning complex and varied functions across a wide variety of tasks. In recent years, artificial neural networks, inspired by human and animal brains, have shown great capabilities in learning a wide variety of difficult tasks. However, artificial neural networks primarily teach themselves through the use of backpropagation, a learning method which has no clear analogue within the brain. Additionally, Artificial Neural Networks primarily use continuous activation functions, which differ significantly from the spiking neuronal behavior present in the brain. In this paper, we discuss and demonstrate a biologically plausible learning method that approximates backpropagation through two techniques on Spiking Neural Networks. First, we show that the local temporal derivatives that are necessary for backpropagation can be approximately recovered through reconstruction using spike timings. Second, we show that through learning during a sleep phase, inspired by neuroscience research into memory replay, the localized parallel feedback path can learn to approximate the derivative through the forward path weight matrix, thus solving the weight transport problem. Lastly, we demonstrate that the combination of these two methods can approach or exceed the accuracy of backpropagation-based methods for a variety of neuromorphic vision tasks while maintaining biological plausibility.

42 ENGINEERING↗
DOE OSTI2024

Variational data augmentation for a learning-based granular predictive model of power outages

As the trend in climate change continues, extreme weather events are expected to occur with increasing frequency and severity and pose a significant threat to the electric power infrastructure. Regardless of the efforts a utility puts towards hardening the grid, storm-induced damage to the utility assets such as cables and distributed energy resources (DERs) that are particularly vulnerable to such events is unavoidable. Access to a highly granular, in space and time, outage forecasting tool with long lead times (i.e., days ahead) will enhance the efficiency of service restoration efforts. Here, in this study, we propose to develop and implement a multi-model framework as an operational tool based on a granular and multi-day outage forecasting model using operational numerical weather prediction model forecasts and detailed component outage information. An innovative two-layered recurrent neural network, i.e., a long-short-term-memory (LSTM)-based variational autoencoder (VAE) framework and a sliding window are used to address the uneven distribution of different types of weather events and make better use of the time-series data. Case studies are performed to demonstrate the performance of the new framework.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2011

ISS Solar Array Alpha Rotary Joint (SARJ) Bearing Failure and Recovery: Technical and Project Management Lessons Learned

The photovoltaic solar panels on the International Space Station (ISS) track the Sun through continuous rotating motion enabled by large bearings on the main truss called solar array alpha rotary joints (SARJs). In late 2007, shortly after installation, the starboard SARJ had become hard to turn and had to be shut down after exceeding drive current safety limits. The port SARJ, of the same design, had been working well for over 2 years. An exhaustive failure investigation ensued that included multiple extravehicular activities to collect information and samples for engineering forensics, detailed structural and thermal analyses, and a careful review of the build records. The ultimate root cause was determined to be kinematic design vulnerability coupled with inadequate lubrication, and manufacturing flaws; this was corroborated through ground tests, metallurgical studies, and modeling. A highly successful recovery plan was developed and implemented that included replacing worn and damaged components in orbit and applying space-compatible grease to improve lubrication. Beyond the technical aspects, however, lie several key programmatic lessons learned. These lessons, such as running ground tests to intentional failure to experimentally verify failure modes, are reviewed and discussed so they can be applied to future projects to avoid such problems.

DellaCorte, Christopher↗
DOE OSTI2024

NNSA MSIIP Intern Biweekly Report - #3

These couple of weeks I continued doing solvent tests with the 3D printed actuators. I made measurements and organized the data obtained from the tests. I was also able to get a logic gate design that worked according to the logic and was able to start printing and modifying the design. I've been understanding more about what parts of the design it would be best to modify and why certain ideas do not work versus others. I've been communicating with my mentor when I make edits to the design to ensure I keep moving forward with it and make it look closer and closer to what we need it to. I've continued to attend the seminars held by the department and learned more about the different methods of 3d printing and the different projects that are being worked on here.

36 MATERIALS SCIENCE↗
NASA NTRS1993

TimeSet: A computer program that accesses five atomic time services on two continents

TimeSet is a shareware program for accessing digital time services by telephone. At its initial release, it was capable of capturing time signals only from the U.S. Naval Observatory to set a computer's clock. Later the ability to synchronize with the National Institute of Standards and Technology was added. Now, in Version 7.10, TimeSet is able to access three additional telephone time services in Europe - in Sweden, Austria, and Italy - making a total of five official services addressable by the program. A companion program, TimeGen, allows yet another source of telephone time data strings for callers equipped with TimeSet version 7.10. TimeGen synthesizes UTC time data strings in the Naval Observatory's format from an accurately set and maintained DOS computer clock, and transmits them to callers. This allows an unlimited number of 'freelance' time generating stations to be created. Timesetting from TimeGen is made feasible by the advent of Becker's RighTime, a shareware program that learns the drift characteristics of a computer's clock and continuously applies a correction to keep it accurate, and also brings .01 second resolution to the DOS clock. With clock regulation by RighTime and periodic update calls by the TimeGen station to an official time source via TimeSet, TimeGen offers the same degree of accuracy within the resolution of the computer clock as any official atomic time source.

Petrakis, P. L.↗
NASA NTRS1996

Lewis Research Center: Commercialization Success Stories

The NASA Lewis Research Center, located in Cleveland, Ohio, has a portfolio of research and technology capabilities and facilities that afford opportunities for productive partnerships with industry in a broad range of industry sectors. In response to the President's agenda in the area of technology for economic growth (Clinton/Gore 1993), the National Performance Review (1993), NASA's Agenda for Change (1994), and the needs of its customers, NASA Lewis Research Center has sought and achieved significant successes in technology transfer and commercialization. This paper discusses a sampling of Lewis Research Center's successes in this area, and lessons learned that Lewis Research Center is applying in pursuit of continuous improvement and excellence in technology transfer and commercialization.

Heyward, Ann O.↗
NASA NTRS2003

An Old Dog and New Tricks

As I approach my 55th birthday, the old adage 'you can't teach an old dog new tricks' keeps coming to mind. I'm not sure why, because I don't feel old and I'm still interested in taking on new challenges and learning new tricks. However, as I mentor new project managers, I am also aware that others may consider me an old dog unable to learn new tricks. To the contrary, the people I mentor continue to teach me new tricks and challenge my assumptions about project management.

Cameron, W. Scott↗
NASA NTRS2010

Space Shuttle Program Primary Avionics Software System (PASS) Success Legacy -Major Accomplishments and Lessons Learned

This presentation has shown the accomplishments of the PASS project over three decades and highlighted the lessons learned. Over the entire time, our goal has been to continuously improve our process, implement automation for both quality and increased productivity, and identify and remove all defects due to prior execution of a flawed process in addition to improving our processes following identification of significant process escapes. Morale and workforce instability have been issues, most significantly during 1993 to 1998 (period of consolidation in aerospace industry). The PASS project has also consulted with others, including the Software Engineering Institute, so as to be an early evaluator, adopter, and adapter of state-of-the-art software engineering innovations.

Orr, James K.↗
NASA NTRS2012

MODIS On-Orbit Calibration and Lessons Learned

The Moderate Resolution Imaging Spectroradiometer (MODIS) is a key instrument for NASA's Earth Observing System (EOS) Terra and Aqua missions. Since launch, Terra and Aqua MODIS have successfully operated for more than 12 and 10 years, respectively, and generated an unprecedented amount of data products for the science and user community over a wide range of applications. MODIS was developed with improved design and stringent calibration requirements over its heritage sensors in order . to extend and enhance their long-term data records. Its follow-on instrument, the Visible/Infrared Imager Radiometer Suite (VIIRS), was launched on-board the Suomi National Polar-orbiting Partnership (NPP) spacecraft October 28, 2011. MODIS collects data in 36 spectral bands, covering wavelengths from 0.41 to 14.S!Jlll, and at 250m, SOOm, and lkm spatial resolutions (nadir). MODIS on-orbit calibration is provided by a set of onboard calibrators (OBC), including a solar diffuser (SO), a solar diffuser stability monitor (SDSM), a blackbody (BB), and a spectroradiometric calibration assembly (SRCA). In addition to the onboard calibrators, regular lunar observations are made by both Terra and Aqua MODIS to track their calibration stability in the reflective solar region. This tutorial session provides an overview of MODIS on-orbit calibration and characterization methodologies. It discusses challenging issues and lessons learned from sensor design, operation, calibration, and inter-comparisons. Examples of instrument on-orbit performance are illustrated with a focus on the improvements made based on various lessons learned. It is expected that MODIS experience and lessons will continue to provide valuable information for future earth observing missions/sensors.

Xiong, Jack↗
NASA NTRS2015

Asteroid Redirect Crewed Mission Space Suit and EVA System Maturation

The Asteroid Redirect Crewed Mission (ARCM) requires a Launch/Entry/Abort (LEA) suit capability and short duration Extra Vehicular Activity (EVA) capability from the Orion spacecraft. For this mission, the pressure garment that was selected, for both functions, is the Modified Advanced Crew Escape Suit (MACES) with EVA enhancements and the life support option that was selected is the Exploration Portable Life Support System (PLSS). The proposed architecture was found to meet the mission constraints, but much more work is required to determine the details of the required suit upgrades, the integration with the PLSS, and the rest of the tools and equipment required to accomplish the mission. This work has continued over the last year to better define the operations and hardware maturation of these systems. EVA simulations have been completed in the NBL and interfacing options have been prototyped and analyzed with testing planned for late 2014. For NBL EVA simulations, in 2013, components were procured to allow in-house build up for four new suits with mobility enhancements built into the arms. Boots outfitted with clips that fit into foot restraints have also been added to the suit and analyzed for possible loads. Major suit objectives accomplished this year in testing include: evaluation of mobility enhancements, ingress/egress of foot restraint, use of foot restraint for worksite stability, ingress/egress of Orion hatch with PLSS mockup, and testing with two crew members in the water at one time to evaluate the crew's ability to help one another. Major tool objectives accomplished this year include using various other methods for worksite stability, testing new methods for asteroid geologic sampling and improving the fidelity of the mockups and crew equipment. These tests were completed on a medium fidelity capsule mockup, asteroid vehicle mockup, and asteroid mockups that were more accurate for an asteroid type EVA than previous tests. Another focus was the design and fabrication of the interface between the MACES and the PLSS. The MACES was not designed to interface with a PLSS, hence an interface kit must accommodate the unique design qualities of the MACES and provide the necessary life support function connections to the PLSS. A prototype interface kit for MACES to PLSS has been designed and fabricated. Unmanned and manned testing of the interface will show the usability of the kit while wearing a MACES. The testing shows viability of the kit approach as well as the operations concept. The design will be vetted through suit and PLSS experts and, with the findings from the testing, the best path forward will be determined. As the Asteroid Redirect Mission matures, the suit/life support portion of the mission will mature along with it and EVA Tools & Equipment can be iterated to accommodate the overall mission objectives and compromises inherent in EVA Suit optimization. The goal of the EVA architecture for ARCM is to continue to build on the previously developed technologies and lessons learned, and accomplish the ARCM EVAs while providing a stepping stone to future missions and destinations.

Bowie, Jonathan T.↗
NASA NTRS2020

PHALANX: Expendable Projectile Sensor Networks for Planetary Exploration

Technologies enabling long-term, wide-ranging measurement in hard-to-reach areas are a critical need for planetary science inquiry. Phenomena of interest include flows or variations in volatiles, gas composition or concentration, particulate density, or even simply temperature. Improved measurement of these processes enables understanding of exotic geologies and distributions or correlating indicators of trapped water or biological activity. However, such data is often needed in unsafe areas such as caves, lava tubes, or steep ravines not easily reached by current spacecraft and planetary robots. To address this capability gap, we have developed miniaturized, expendable sensors which can be ballistically lobbed from a robotic rover or static lander - or even dropped during a flyover. These projectiles can perform sensing during flight and after anchoring to terrain features. By augmenting exploration systems with these sensors, we can extend situational awareness, perform long-duration monitoring, and reduce utilization of primary mobility resources, all of which are crucial in surface missions. We call the integrated payload that includes a cold gas launcher, smart projectiles, planning software, network discovery, and science sensing: PHALANX. In this paper, we introduce the mission architecture for PHALANX and describe an exploration concept that pairs projectile sensors with a rover “mothership.” Science use cases explored include reconnaissance using ballistic cameras, volatiles detection, and building timelapse maps of temperature and illumination conditions. Strategies to autonomously coordinate constellations of deployed sensors to self-discover and localize with peer ranging (i.e. a “local GPS”) are summarized, thus providing communications infrastructure beyond-line-of-sight (BLOS) of the rover. Capabilities were demonstrated through both simulation and physical testing with a terrestrial prototype. The approach to developing a terrestrial prototype is discussed, including design of the launching mechanism, projectile optimization, micro-electronics fabrication, and sensor selection. Results from early testing and characterization of commercial-off-the-shelf (COTS) components are reported. Nodes were subjected to successful burn-in tests over 48 hours at full logging duty cycle. Integrated field tests were conducted in the Roverscape, a half-acre planetary analog environment at NASA Ames, where we tested up to 10 sensor nodes simultaneously coordinating with an exploration rover. Ranging accuracy has been demonstrated to be within +/-10cm over 20m using commodity radios when compared to high-resolution laser scanner ground truthing. Evolution of the design, including progressive miniaturization of the electronics and iterated modifications of the enclosure housing for streamlining and optimized radio performance are described. Finally, lessons learned to date, gaps toward eventual flight mission implementation, and continuing future development plans are discussed.

Dille, Michael↗
NASA NTRS2022

Flight Software Dictionary Development for the Mars2020 Rover

The Mars2020 project, developed and operated by the Jet Propulsion Laboratory (JPL), successfully landed the Perseverance rover and its flying companion Ingenuity on the surface of Mars on February 18th 2021. Perseverance combines heritage and cutting-edge flight software and hardware to accomplish crucial mission requirements related to Martian surface sampling. The design, development, and operation of NASA’s large strategic science missions require the ability to communicate spacecraft capabilities to hundreds of engineers across multiple disciplines. The interaction between flight and ground software development, Verification and Validation (V&V), Assembly, Test, and Launch Operations (ATLO), and management each demand quick understanding of unique slices of information for each discipline. This information includes the current capabilities of the flight system as well as future capabilities and their status as they are developed and tested. Despite the fundamental and critical nature of this information, the flight software dictionaries used to track it are a stumbling block for many projects. These dictionaries provide the cornerstone for the interpretation of data sent from the spacecraft, allowing for quick comprehension by engineers on the ground. During both spacecraft development and operations, flight software dictionary management includes significant challenges due to the large number of interfacing systems and the subtle yet distinct needs of each.The engineering of flight software dictionaries for Mars2020 had numerous challenges, most-notably: parallel dictionary development to support simultaneous separate flight software build campaigns for each mission phase (cruise and surface), managing requests for operations-enabling information without perturbing the heritage interface with the rover, and the introduction of new tools by the dictionary stakeholders that forced the dictionary team to innovate and redesign the heritage tool chain. These challenges generated guiding principles for the dictionary development effort: emphasize coding best practices and unit testing in the dictionary code development tool chain, use institutionally provided COTS (commercial-off-the-shelf) tools whenever possible, and maintain the heritage flight-ground interface all while advancing operations-enabling information via a loosely coupled interface.Throughout development and operations, the Mars2020 dictionary toolchain included IBM DOORS Next Generation, GitHub, Microsoft Excel, Docker, Jenkins, and a significant custom-built Python codebase. Significant interfaces included JPL’s command and control software, heritage flight software team tools and processes, and the many cloud-based ground tools developed for the mission.This paper will discuss the requirements for the Mars2020 dictionary development, the development team’s response to those requirements, lessons learned throughout the process, steps taken towards automated deliveries and continuous integration of stakeholder inputs, potential toolchain improvements for Mars2020, and key takeaways that could be applied to future missions.

Pyrzak, Guy↗
NASA NTRSDate not supplied

Testing a Run-Time Assurance Framework Coupled with Integrated Risk Mitigation Capabilities for Autonomous Urban UAS Flights

The In-Time Aviation Safety Management System (IASMS) Concept of Operations (ConOps) envisions new capabilities to monitor, assess, and mitigate flight safety risks. Systems will be tailored to mission type, vehicle/equipage type, operational environment, and safety risk tolerance. Within an IASMS framework, several capabilities may be implemented spanning three operational phases (pre-flight, in-flight, and post-flight/off-line); and consisting of lower level functions and information services which may reside onboard the aircraft, on third-party server(s), and/or on ground/operator station(s). Each capability will be designed to produce and disseminate safety-relevant information; perform detection, diagnosis, and prediction of unsafe situations; and/or execute mitigation actions when hazardous events warrant such changes. This paper focuses on recent testing of airborne capabilities that demonstrate inflight aspects of the overarching concept for autonomous unmanned aircraft systems (UAS) operations in urban environments. A flight test architecture is described that applies run-time assurance principles (e.g., executes independent of the unassured autopilot), real-time risk assessment, and a technique to execute contingencies if necessary either automatically or via pilot intervention. Several tests using small UAS were conducted to verify the assured in-flight risk mitigation capability. The paper draws significantly from a larger NASA technical report and recent prior conference papers, providing additional details. Data is analyzed for two representative flights to illustrate the performance for various sequential and simultaneous hazards used during testing. During each automated flight, several hazards are encountered at various points along the flight path. At each point, the hazard is mitigated by the system, with the vehicle then continuing to subsequent points. The paper concludes with lessons-learned regarding relevant aspects of the overarching IASMS concept and how it may be updated and further advanced in the future.

population activity↗
NASA NTRS1999

Designing Specification Languages for Process Control Systems: Lessons Learned and Steps to the Future

Previously, we defined a blackbox formal system modeling language called RSML (Requirements State Machine Language). The language was developed over several years while specifying the system requirements for a collision avoidance system for commercial passenger aircraft. During the language development, we received continual feedback and evaluation by FAA employees and industry representatives, which helped us to produce a specification language that is easily learned and used by application experts. Since the completion of the PSML project, we have continued our research on specification languages. This research is part of a larger effort to investigate the more general problem of providing tools to assist in developing embedded systems. Our latest experimental toolset is called SpecTRM (Specification Tools and Requirements Methodology), and the formal specification language is SpecTRM-RL (SpecTRM Requirements Language). This paper describes what we have learned from our use of RSML and how those lessons were applied to the design of SpecTRM-RL. We discuss our goals for SpecTRM-RL and the design features that support each of these goals.

Leveson, Nancy G.↗