Results for “increasing automation” — Page 3

NASA NTRSDate not supplied

Where is the Human in the Loop? Human Factors Analysis of Extended Visual Line of Sight Unmanned Aerial System Operations within a Remote Operations Environment

Many envisioned technological and conceptual innovations focus on allocating more functions to automation, relegating the human as an afterthought if not a nuisance. Yet, until complete autonomy is realized, the human will remain “in the loop”. The National Aeronautics and Space Administration is supporting research for the development and maturation of automated technologies and architectures for the future of advanced air mobility. Standing up a remote operations center used to control, manage, and monitor multiple highly automated vehicles is an important step towards realizing the advanced air mobility vision. At the National Aeronautics and Space Administration’s Langley Research Center, a remote operation center exists and has been tested using humans piloting simulated vehicles. In this paper, we explore the human element within live flight operations that rely on increasingly automated technologies. Four ground control station operators performed multiple live flight operations. We employed a naturalistic approach and relied on qualitative data such as interviews and discussions with subject matter experts to help facilitate discovery. The present work evaluates the functions that the human and the automation had during the live operations, lists psychological constructs that may have promoted or reduced task performance, and provides recommendations on design of the remote operations environment and training of future ground control station operators.

Advanced Air Mobility↗

NASA NTRSDate not supplied

Where is the Human in the Loop? Human Factors Analysis of Extended Visual Line of Sight Unmanned Aerial System Operations within a Remote Operations Environment

Many envisioned technological and conceptual innovations focus on allocating more functions to automation, relegating the human as an afterthought if not a nuisance. Yet, until complete autonomy is realized, the human will remain “in the loop”. The National Aeronautics and Space Administration is supporting research for the development and maturation of automated technologies and architectures for the future of advanced air mobility. Standing up a remote operations center used to control, manage, and monitor multiple highly automated vehicles is an important step towards realizing the advanced air mobility vision. At the National Aeronautics and Space Administration’s Langley Research Center, a remote operation center exists and has been tested using humans piloting simulated vehicles. In this paper, we explore the human element within live flight operations that rely on increasingly automated technologies. Four ground control station operators performed multiple live flight operations. We employed a naturalistic approach and relied on qualitative data such as interviews and discussions with subject matter experts to help facilitate discovery. The present work evaluates the functions that the human and the automation had during the live operations, lists psychological constructs that may have promoted or reduced task performance, and provides recommendations on design of the remote operations environment and training of future ground control station operators.

Advanced Air Mobility↗

NASA NTRS2006

Mission Control Technologies: A New Way of Designing and Evolving Mission Systems

Current mission operations systems are built as a collection of monolithic software applications. Each application serves the needs of a specific user base associated with a discipline or functional role. Built to accomplish specific tasks, each application embodies specialized functional knowledge and has its own data storage, data models, programmatic interfaces, user interfaces, and customized business logic. In effect, each application creates its own walled-off environment. While individual applications are sometimes reused across multiple missions, it is expensive and time consuming to maintain these systems, and both costly and risky to upgrade them in the light of new requirements or modify them for new purposes. It is even more expensive to achieve new integrated activities across a set of monolithic applications. These problems impact the lifecycle cost (especially design, development, testing, training, maintenance, and integration) of each new mission operations system. They also inhibit system innovation and evolution. This in turn hinders NASA's ability to adopt new operations paradigms, including increasingly automated space systems, such as autonomous rovers, autonomous onboard crew systems, and integrated control of human and robotic missions. Hence, in order to achieve NASA's vision affordably and reliably, we need to consider and mature new ways to build mission control systems that overcome the problems inherent in systems of monolithic applications. The keys to the solution are modularity and interoperability. Modularity will increase extensibility (evolution), reusability, and maintainability. Interoperability will enable composition of larger systems out of smaller parts, and enable the construction of new integrated activities that tie together, at a deep level, the capabilities of many of the components. Modularity and interoperability together contribute to flexibility. The Mission Control Technologies (MCT) Project, a collaboration of multiple NASA Centers, led by NASA Ames Research Center, is building a framework to enable software to be assembled from flexible collections of components and services.

Trimble, Jay↗

NASA NTRSDate not supplied

Flight Deck Design of a Hybrid Turbine/Electric Passenger Aircraft

NASA is exploring the development of a 180-passenger subsonic single engine aft turbine aircraft, The aft turbine provides electric power in a hybrid design to wing mounted electric engines, creating a highly efficient, high-bypass-ratio fan equivalent. The SUbsonic Single Aft eNgine (SUSAN) aircraft is being developed as a sustainable subsonic regional aircraft that seeks to reduce emission levels by 50% in the next few decades. Pilot-in-the-loop studies were conducted at the NASA Langley Research Center in Hampton, Virginia, to explore the flight deck design for the hybrid electric aircraft. Following modern trends in commercial aircraft flight decks with full time augmented controls and a quiet and dark philosophy, single throttle and simplified engine displays were developed for the SUSAN aircraft. The aircraft includes a single aft mounted turbine engine and 16 wing mounted electric fans. The final design was developed from feedback received during an earlier pilot-in-the-loop study where one, two, and three throttles were tested in standard airline operations, including various failures of the turbine and electric engines. Current flight deck designs normally provide control inceptors for each propulsion engine and an engine display for all primary aircraft engine parameters. With full time augmentation expected, a single throttle control with autothrottle always engaged, even during failures, is desired. Augmentation of flight controls using distributed thrust also requires full time control of the electric engines using automation. Additionally, electric engine thrust is augmented during climb based on battery state of charge. Thrust augmentation changes faster than human reaction time and therefore requires full-time automation. A pilot-in-the-loop study was conducted at the NASA Langley Research Center in Hampton, Virginia, to test the final design of the single throttle with simplified engine displays. Fourteen airline pilots evaluated the single throttle and engine display concept. Electric engine failures included one, four symmetric, and eight non-symmetric electric engine failures. The turbine engine was evaluated for complete and partial failure during critical phases of flight to include takeoff as well as enroute. Unexpected go-arounds increase workload and require significant throttle manipulation. Go-arounds were included to ensure the single throttle was usable for all phases of flight. Failures during takeoff required a return to the departure field and failures enroute required a diversion except for one and four electric engine failures as these failures did not affect aircraft flyability or range. There are currently no Part 25 aircraft certified with hybrid systems or electric engines with batteries as emergency propulsion. For turbine engine failures in the SUSAN aircraft design, range is limited to 30 minutes at full power. Battery state of charge and battery health displays were developed and tested for usability and to determine how well they supported pilot decisions for alternate airports during emergency diversions. Novel displays using shape and color were developed to provide immediate feedback when state of charge became critical. This paper details the pilot study including pilot feedback supporting the potential for increased automation and a single throttle control. Detailed recommendations are provided for a novel single throttle control and additional pilot controls to support selection of engines during start, shutdown, and engine troubleshooting procedures. This design deviates significantly from current practice of providing throttles for each propulsion engine. Engine display recommendations are provided based on pilot feedback during a guided post-evaluation interview. Battery state of charge and battery health display recommendations were collected from all airline crews. The simplified engine displays design was rated excellent as measured with a usability scale. Quantitative metrics include airspeed tracking, time to complete checklists, time to make diversion decisions and the quality of the diversion decision. Recommendations for future studies are documented with supporting research and current observations about upcoming flight deck certifications.

autothrottle↗

NASA NTRSDate not supplied

Flight Deck Design of a Hybrid Turbine/Electric Passenger Aircraft

NASA is exploring the development of a 180-passenger subsonic single engine aft turbine aircraft, The aft turbine provides electric power in a hybrid design to wing mounted electric engines, creating a highly efficient, high-bypass-ratio fan equivalent. The SUbsonic Single Aft eNgine (SUSAN) aircraft is being developed as a sustainable subsonic regional aircraft that seeks to reduce emission levels by 50% in the next few decades. Pilot-in-the-loop studies were conducted at the NASA Langley Research Center in Hampton, Virginia, to explore the flight deck design for the hybrid electric aircraft. Following modern trends in commercial aircraft flight decks with full time augmented controls and a quiet and dark philosophy, single throttle and simplified engine displays were developed for the SUSAN aircraft. The aircraft includes a single aft mounted turbine engine and 16 wing mounted electric fans. The final design was developed from feedback received during an earlier pilot-in-the-loop study where one, two, and three throttles were tested in standard airline operations, including various failures of the turbine and electric engines. Current flight deck designs normally provide control inceptors for each propulsion engine and an engine display for all primary aircraft engine parameters. With full time augmentation expected, a single throttle control with autothrottle always engaged, even during failures, is desired. Augmentation of flight controls using distributed thrust also requires full time control of the electric engines using automation. Additionally, electric engine thrust is augmented during climb based on battery state of charge. Thrust augmentation changes faster than human reaction time and therefore requires full-time automation. A pilot-in-the-loop study was conducted at the NASA Langley Research Center in Hampton, Virginia, to test the final design of the single throttle with simplified engine displays. Fourteen airline pilots evaluated the single throttle and engine display concept. Electric engine failures included one, four symmetric, and eight non-symmetric electric engine failures. The turbine engine was evaluated for complete and partial failure during critical phases of flight to include takeoff as well as enroute. Unexpected go-arounds increase workload and require significant throttle manipulation. Go-arounds were included to ensure the single throttle was usable for all phases of flight. Failures during takeoff required a return to the departure field and failures enroute required a diversion except for one and four electric engine failures as these failures did not affect aircraft flyability or range. There are currently no Part 25 aircraft certified with hybrid systems or electric engines with batteries as emergency propulsion. For turbine engine failures in the SUSAN aircraft design, range is limited to 30 minutes at full power. Battery state of charge and battery health displays were developed and tested for usability and to determine how well they supported pilot decisions for alternate airports during emergency diversions. Novel displays using shape and color were developed to provide immediate feedback when state of charge became critical. This paper details the pilot study including pilot feedback supporting the potential for increased automation and a single throttle control. Detailed recommendations are provided for a novel single throttle control and additional pilot controls to support selection of engines during start, shutdown, and engine troubleshooting procedures. This design deviates significantly from current practice of providing throttles for each propulsion engine. Engine display recommendations are provided based on pilot feedback during a guided post-evaluation interview. Battery state of charge and battery health display recommendations were collected from all airline crews. The simplified engine displays design was rated excellent as measured with a usability scale. Quantitative metrics include airspeed tracking, time to complete checklists, time to make diversion decisions and the quality of the diversion decision. Recommendations for future studies are documented with supporting research and current observations about upcoming flight deck certifications.

autothrottle↗

NASA NTRS2017

Robotically Assembled Aerospace Structures: Digital Material Assembly using a Gantry-Type Assembler

This paper evaluates the development of automated assembly techniques for discrete lattice structures using a multi-axis gantry type CNC machine. These lattices are made of discrete components called digital materials. We present the development of a specialized end effector that works in conjunction with the CNC machine to assemble these lattices. With this configuration we are able to place voxels at a rate of 1.5 per minute. The scalability of digital material structures due to the incremental modular assembly is one of its key traits and an important metric of interest. We investigate the build times of a 5x5 beam structure on the scale of 1 meter (325 parts), 10 meters (3,250 parts), and 30 meters (9,750 parts). Utilizing the current configuration with a single end effector, performing serial assembly with a globally fixed feed station at the edge of the build volume, the build time increases according to a scaling law of n4, where n is the build scale. Build times can be reduced significantly by integrating feed systems into the gantry itself, resulting in a scaling law of n3. A completely serial assembly process will encounter time limitations as build scale increases. Automated assembly for digital materials can assemble high performance structures from discrete parts, and techniques such as built in feed systems, parallelization, and optimization of the fastening process will yield much higher throughput.

Mechanical Structure↗

NASA NTRS2017

Robotically Assembled Aerospace Structures: Digital Material Assembly using a Gantry-Type Assembler

This paper evaluates the development of automated assembly techniques for discrete lattice structures using a multi-axis gantry type CNC machine. These lattices are made of discrete components called "digital materials." We present the development of a specialized end effector that works in conjunction with the CNC machine to assemble these lattices. With this configuration we are able to place voxels at a rate of 1.5 per minute. The scalability of digital material structures due to the incremental modular assembly is one of its key traits and an important metric of interest. We investigate the build times of a 5x5 beam structure on the scale of 1 meter (325 parts), 10 meters (3,250 parts), and 30 meters (9,750 parts). Utilizing the current configuration with a single end effector, performing serial assembly with a globally fixed feed station at the edge of the build volume, the build time increases according to a scaling law of n4, where n is the build scale. Build times can be reduced significantly by integrating feed systems into the gantry itself, resulting in a scaling law of n3. A completely serial assembly process will encounter time limitations as build scale increases. Automated assembly for digital materials can assemble high performance structures from discrete parts, and techniques such as built in feed systems, parallelization, and optimization of the fastening process will yield much higher throughput.

Manufacturing↗

NASA NTRSDate not supplied

Informing New Concepts for UAS and Autonomous System Safety Management using Disaster Management and First Responder Scenarios

As emerging flight operations become more prevalent and increasingly automated and distributed, the capabilities for managing safety of vehicles and operations will also need to evolve. To address this challenge, the National Academies has envisioned an In-Time Aviation Safety Management System (IASMS) capability for a wide range of aviation operations including current commercial operations as well as new entrants envisioned with advanced air mobility (AAM). The suite of IASMS services, functions, and capabilities (SFCs) would be implemented in a federated approach and would address trends as well as individual operations. Through predictive modeling and data analysis, IASMS is envisioned to identify arising risks so that they can be mitigated, in-time, before a safety incident occurs. IASMS and its requisite set of SFCs must leverage a wide range of information to perform. To better understand these new needs, FSF worked with the aviation and humanitarian communities to develop and validate scenarios that include traditional aviation operations and UAS operations intermingled as they are deployed for disaster management and first responder (DMFR) situations. The three scenarios developed include: • Post Natural disaster response, such as a hurricane, involving multiple parties utilizing traditional aviation and UAS to support rescue operations, surveil damage, and locate survivors needing assistance. • Wildfire fighting in remote locations with traditional aircraft for transport and fire-retardant delivery combined with UAS for surveillance of fire locations as well as to track individual firefighter locations. • Medical Operations and AAM in Urban Environments including passenger-carrying helicopters and AAM vehicles, medical missions (such as transport of radio-pharmaceuticals), and other UAS delivery operations (such as the delivery of defibrillators). Each scenario was developed and validated by representatives with expertise in humanitarian operations, urban and rural emergency response, air traffic management, UAS operations, and traditional flight operations. The scenario definitions address roles and responsibilities of individual actors, the appropriate utilization of UAS, and the actions taken by those actors to appropriately manage risks associated with the mission and environment. The risks to aviation traffic and to people on the ground explored included potential risks arising from incompatibilities in calculating reference altitudes (eg, differing uses of AGL, MSL, barometric, or GPS-derived values), loss of command and control (C2) communications, rapid changes in weather and winds, and physical interference. For each risk, IASMS SFCs were postulated in the context of monitoring services, risk assessment capabilities, and identifying appropriate mitigation strategies. The identified SFC capabilities were envisioned from known services postulated for IASMS and for UTM. For these unique environments, IASMS SFCs are needed to address conditions such as hazardous payloads, micro-climates and urban canyons, and the need to keep uninvolved air traffic out of the area where DMFR operations are being conducted. The second phase of analysis focused on inferring the specific information needs and the SFCs for IASMS, utilizing a structure of 16 information classes to organize requirements. For each of the risks identified in the workshops, it was postulated what data sources would be necessary to monitor critical aspects of the risk (eg, surrounding air traffic, ground population, terrain, etc). to be directly measured as well as data that would be derived, which implies additional SFCs for different actors to understand what information would likely be exchanged between parties. For an IASMS to be effective, additional research is needed to develop the advanced algorithms that can address the increasingly autonomous and complex operations in differing environments and to develop means of identifying unknown risks. Looking at these scenarios highlighted a number of research issues. These include the ability to quickly "cordon off" airspace thru temporary flight restrictions (TFRs) or other means, developing clear definitions to enable automation-based algorithms for prioritizing operations, defining airspace density metrics, standardization of altitude reporting, and establishing a basis for safety data metrics definition and collection. This paper seeks to outline the development of an IASMS in the context of the DMFR scenarios and resulting demonstrations. Utilizing this contextual approach, NASA will generate recommendations for an assured safety framework for AAM operations that enables AAM operations to safely access the NAS.

In Time Aviation Safety Management System↗

NASA NTRS1994

Mini All-purpose Satellite Control Center (MASCC)

A new generation of Mini All-purpose Satellite Control Centers (MASCC) has been developed by CNES (F). They turn out to be easily adaptable to different kinds of satellites, both Low Earth Orbital or Geostationary. The features of MASCC allow both standard satellite control activities, and checking of passengers experiments hosted on a space platform. In the different environments in which it may be used, MASCC provides standard broadcasting of telemetry parameters on animated synoptics (curves, bar graphs, alphanumeric displays, ...), which turns out to be a very useful and ergonomic medium for operational teams or satellite specialists. Special care has been taken during the MASCC development about two points: - automation of all routine tasks, allowing automated operation, and limiting human commitment to system supervision and decision making, - software adaptability. To reach these two main objectives, the MASCC design provides:(1) a simple, robust and flexible hardware architecture, based on powerful distributed workstations; and (2) a table-driven software architecture, easily adapted to various operational needs. Satellite characteristics are described in a central Data Base. Hence, the processing of telemetry and commands is largely independent from the satellite itself. In order to validate these capabilities, the MASCC has been customized to several types of satellites and orbital platforms: (1) SPOT4, the French new generation of remote sensing satellites; (2) TELECOM2, the French geostationary TV and telecommunication satellite; and (3) MIR, the Russian orbital platform. MASCC development has been completed by the third quarter of 1993. This paper will provide first a description of the MASCC basic functions, of its hardware and software design. It will then detail the increased automation capability, along with the easy adaptation of the MASCC to new satellites with minimal software modifications.

Zaouche, Gerard↗

NASA NTRS2024

Integration of Automated Systems Test Campaign NC-IAS

NASA’s 2022 strategic plan included the goal to catalyze economic growth and drive innovation in the aviation industry, and to address challenges in air transportation and airspace management within the National Airspace System (NAS). The Aeronautics Research Mission Directorate (ARMD) specifically was tasked with leading aviation innovation to enable safe and sustainable air transportation through revolutionary vehicle advances and efficient flight operations. As a part of meeting these objectives and advancing air mobility concepts, NASA initiated the National Campaign (NC) in the Advanced Air Mobility (AAM) Project within the Airspace Operations and Safety Program (AOSP). The NC was designed to support operational demonstrations with industry as well as the research and development needed to support NASA-led research flight demonstrations. Within NC, the Integration of Automated Systems (IAS), an NC activity, tested and evaluated flight deck automation and airspace operations management functions needed to enable Urban Air Mobility (UAM) operations. This was accomplished through a partnership with Sikorsky Aircraft (specifically Sikorsky Innovations), a Lockheed Martin company, and DARPA (Defense Advanced Research Projects Agency), by leveraging two automation-enabled helicopters equipped with unique capabilities that enabled NASA to develop and test two-ship conflict encounters to demonstrate flight path management and hazard avoidance technologies. The enabler in this testing was NASA-developed “Middleware” (MW) software (also known as Expandable Variable Autonomy Architecture, or EVAA), which among other things allowed multiple algorithms to be incorporated into one software build that was hosted on the dissimilar-type Sikorsky helicopters. The IAS test campaign period of performance was from March 2022 through October 2023 and was structured as a phased, or spiral, approach that ultimately led to the first-ever demonstration of two-ship UAM/AAM operations designed to safely choreograph specific conflict encounters and mission scenarios to test the research algorithms for strategic and tactical aircraft deconfliction. Lessons learned are included in the body of the report. Data collected will be used to inform FAA and industry standards groups on the increasingly automated systems needed for future AAM operations. The test encounters developed for these flight tests were proven to be highly predictable, repeatable, and safely exercised flight path planning and Detect and Avoid (DAA) algorithms. These same test encounters should be leveraged by future flight test campaigns to verify that operational safety is not compromised as the AAM architecture matures. Next steps include repeating similar encounters using unmanned aircraft carrying DAA sensors in the National Airspace.

AAM↗

NASA NTRS2019

Human Performance Contributions to Safety in Commercial Aviation

Every day in aviation, pilots, air traffic controllers, and other front-line personnel perform countless correct judgments and actions in a variety of operational environments. These judgments and actions are often the difference between an accident and a non-event. Ironically, data on these behaviors are rarely collected or analyzed. Data-driven decisions about safety management and design of safety-critical systems are limited by the available data, which influence how decision makers characterize problems and identify solutions. Large volumes of data are collected on the failures and errors that result in infrequent incidents and accidents, but in the absence of data on behaviors that result in routine successful outcomes, safety management and system design decisions are based on a small sample of nonrepresentative safety data. This assessment aimed to find and document “safety successes” made possible by human operators. With many Aeronautics Research Mission Directorate (ARMD) Programs and Projects focusing on increased automation and autonomy and decreased human involvement, failure to fully consider the human contributions to successful system performance in civil aviation represents a significant risk — a risk that has not been recognized to date. Without understanding how humans contribute to safety, any estimate of predicted safety of autonomous capabilities is incomplete and inherently suspect. Furthermore, understanding the ways in which humans contribute to safety can promote strategic interactions among safety technologies, functions, procedures and the people using them. Without this understanding, the full benefits of an integrated, optimized human/technology or autonomous system will not be realized. Historically, safety has been consistently defined in terms of the occurrence of accidents or recognized risks (i.e., in terms of things that go wrong). These adverse outcomes are explained by identifying their causes, and safety is restored by eliminating or mitigating these causes. An alternative to this approach is to focus on what goes right and identify how to replicate that process. Focusing on the rare cases of failures attributed to “human error” provides little information about why human performance routinely prevents adverse events. Hollnagel has proposed that things go right because people continuously adjust their work to match their operating conditions. These adjustments become increasingly important as systems continue to grow in complexity. Thus, the definition of safety should reflect not only “avoiding things that go wrong” but “ensuring that things go right.” The basis for safety management requires developing an understanding of everyday activities. However, few mechanisms to monitor everyday work exist in the aviation domain, which limits opportunities to learn how designs function in reality. This concept of safety thinking and safety management is reflected in the emerging field of resilience engineering. According to Hollnagel, a system is resilient if it can sustain required operations under expected and unexpected conditions by adjusting its functioning prior to, during, or following changes, disturbances, and opportunities. To explore “positive” behaviors that contribute to resilient performance in commercial aviation, the assessment team examined a range of existing sources of data about pilot and air traffic control (ATC) tower controller performance, including subjective interviews with domain experts and objective aircraft flight data records. These data were used to identify strategies that support resilient performance, methods for exploring and refining those strategies in existing data, and proposed methods for capturing and analyzing new data.

Null, Cynthia H.↗

DOE OSTI2024

Formally Verified ZTA Requirements for OT/ICS Environments with Isabelle/HOL

The clean energy transformation includes the integration of distributed energy resources with the power grid, which has led to a substantial increase in the complexity of power grids infrastructure and the underlying operational technology environment. Power grids infrastructure represents an operational technology environment that has become a system of systems, integrating heterogeneous devices which are both software-and hardware-intensive; as a result, there are increasing demands to exploit advances in the commodity of software-hardware infrastructures to improve energy systems requirements such as cybersecurity and resilience. In such a setting, system requirements at different levels mix, which leads to vulnerabilities and undesirable outcomes. The use of formal methods to characterize and prove system requirements removes ambiguity, increases automation, and provides high levels of assurance and reliability. In this paper, we contribute a methodology and a framework for the system-level verification of zero trust architecture requirements in operational technology environments. We define a formal specification for the core functionalities of operational technology environments, the corresponding invariants, and security proofs. Of particular note is our modular approach for the formal verification of asynchronous interactions in operational technology environments. The formal specification and the proofs have been mechanized using the interactive theorem proving environment Isabelle/HOL.

formal methods↗

DOE OSTI2021

Model predictive control for integrated control of air-conditioning and mechanical ventilation, lighting and shading systems

Modern buildings are increasingly automated and often equipped with multiple building services (e.g., air-conditioning and mechanical ventilation (ACMV), dynamic shading, dimmable lighting). These systems are conventionally controlled individually without considering their interactions, affecting the building’s overall energy inefficiency and occupant comfort. A model predictive control (MPC) system that features a multi-objective MPC scheme to enable coordinated control of multiple building services for overall optimized energy efficiency, indoor thermal and visual comfort, as well as a hybrid model for predicting indoor visual comfort and lighting power is proposed. The MPC system was implemented in a test facility having two identical, side-by-side experimental cells to facilitate comparison with a building management system (BMS) employing conventional reactive feedback control. The MPC system coordinated the control of the ACMV, dynamic façade and automated dimmable lighting systems in one cell while the BMS controlled the building services in the other cell in a conventional manner. The MPC side achieved 15.1–20.7% electricity consumption reduction, as compared to the BMS side. Simultaneously, the MPC system improved indoor thermal comfort by maintaining the room within the comfortable range (-0.5 < predicted mean vote < 0.5) for 98.3% of the time, up from 91.8% of the time on the BMS side. Visual comfort, measured by indoor daylight glare probability (DGP) and horizontal illuminance level at work plane height, was maintained for the entire test period on the MPC side, improving from having visual comfort for 94.5% and 85.7% of the time, respectively, on the BMS side.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2022

Using artificial intelligence to detect human errors in nuclear power plants: A case in operation and maintenance

Human error (HE) is an important concern in safety-critical systems such as nuclear power plants (NPPs). HE has played a role in many accidents and outage incidents in NPPs. Despite the increased automation in NPPs, HE remains unavoidable. Hence, the need for HE detection is as important as HE prevention efforts. In NPPs, HE is rather rare. Hence, anomaly detection, a widely used machine learning technique for detecting rare anomalous instances, can be repurposed to detect potential HE. In this study, we develop an unsupervised anomaly detection technique based on generative adversarial networks (GANs) to detect anomalies in manually collected surveillance data in NPPs. More specifically, our GAN is trained to detect mismatches between automatically recorded sensor data and manually collected surveillance data, and hence, identify anomalous instances that can be attributed to HE. We test our GAN on both a real-world dataset and an external dataset obtained from a testbed, and we benchmark our results against state-of-the-art unsupervised anomaly detection algorithms, including one-class support vector machine and isolation forest. Our results show that the proposed GAN provides improved anomaly detection performance. Our study is promising for the future development of artificial intelligence based HE detection systems.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

Introduction: Neuromorphic Materials

The explosive growth in data collection and the need to process it efficiently, as well as the desire to automate increasingly complex tasks in transportation, medical care, manufacturing, security and many other fields have motivated a growing interest in neuromorphic computing. Unlike the binary, transistorbased ON/OFF logic gates and separate logic and memory functionalities employed in digital computing, neuromorphic computing is inspired by animal brains that use interconnected synapses and neurons to perform processing, storage and transmission of information at the same location, while only consuming ~20 W or less of power. Motivated by the brain’s efficiency, adaptability, self-learning and resiliency qualities, neuromorphic computing can be broadly defined as an approach to processing and storing information using hardware and algorithms inspired by models of biological neural systems. Present research in neuromorphic computing encompasses approaches that vary significantly in their degree of neuro-inspiration, from systems that only incorporate features such as asynchronous, event-driven operation or use crossbar arrays of non-volatile memory (NVM) elements to accelerate deep neural networks (DNNs), to designs that embrace the extreme parallelism, sparsity, reconfigurability, adaptability, complexity and stochasticity observed in nervous systems. The term ‘neuromorphic’ computing is often credited to Carver Mead, who in the 1980s investigated Si-based analog electronics to replicate functions of the animal retina. Earlier important advances in this field include the work of Frank Rosenblatt, who proposed the concept of the perceptron, Bernard Widrow, who used this concept to build one of the first analog neural networks, the Adaline and many other researchers (see ref. 6 for an historical perspective on neuromorphic computing). With the recent increase in the use of artificial intelligence and large language models, and rising concerns over the associated energy costs, interest in neuromorphic hardware has expanded rapidly. According to some estimates, driven largely by the drastic growth in the training use of artificial intelligence (AI) models using the current computing architectures, the energy cost of computing is projected to reach the energy supply worldwide by 2045. Furthermore, while this is not a realistic outcome, it means that, if more efficient computing technologies are not developed -- soon -- the world will soon become one where demand for energy and market constraints limit the continued increase of societal access to AI and cloud services from data centers. Data centers used for training and use of these models consume hundreds of terawatt hours of electricity, already past 4% of the US electricity demand.

Circuits↗

DOE OSTI2020

Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels

Abstract A majority of experimental disciplines face the challenge of exploring large and high-dimensional parameter spaces in search of new scientific discoveries. Materials science is no exception; the wide variety of synthesis, processing, and environmental conditions that influence material properties gives rise to particularly vast parameter spaces. Recent advances have led to an increase in the efficiency of materials discovery by increasingly automating the exploration processes. Methods for autonomous experimentation have become more sophisticated recently, allowing for multi-dimensional parameter spaces to be explored efficiently and with minimal human intervention, thereby liberating the scientists to focus on interpretations and big-picture decisions. Gaussian process regression (GPR) techniques have emerged as the method of choice for steering many classes of experiments. We have recently demonstrated the positive impact of GPR-driven decision-making algorithms on autonomously-steered experiments at a synchrotron beamline. However, due to the complexity of the experiments, GPR often cannot be used in its most basic form, but rather has to be tuned to account for the special requirements of the experiments. Two requirements seem to be of particular importance, namely inhomogeneous measurement noise (input-dependent or non-i.i.d.) and anisotropic kernel functions, which are the two concepts that we tackle in this paper. Our synthetic and experimental tests demonstrate the importance of both concepts for experiments in materials science and the benefits that result from including them in the autonomous decision-making process.

36 MATERIALS SCIENCE↗

DOE OSTI2021

Use of computer vision for analysis of image datasets from high temperature plasma experiments

Great strides have been made in improving the quality of x-ray radiographs in high energy density plasma experiments, enabled in part by innovations in engineering and manufacturing of integrated circuits and materials. As a consequence, the radiographs of today are filled with a great deal of detail, but few of these features are extracted in a systematic way. Analysis techniques familiar to plasma physicists tend toward brittle 1D lineout or Fourier transform type analyses. The techniques applied to process our data have not kept pace with improvements in the quality of our data. Fortunately, the field of computer vision has a wealth of tools to offer, which have been widely used in industrial imaging and, more recently, adopted in biological imaging. We demonstrate the application of computer vision techniques to the analysis of x-ray radiographs from high energy density plasma experiments, as well as give a brief tutorial on the computer vision techniques themselves. These tools robustly extract 2D contours of shocks, boundaries of inhomogeneities, and secondary flows, thereby allowing for increased automation of analysis, as well as direct and quantitative comparisons with simulations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2025

Detection of Diversion in a Realistic Heat Pipe Microreactor Using Supervised Machine Learning

Microreactors (MRs) pose new challenges for international safeguards. Here, their small size and mass reproducibility make them ideal for deployment in greater numbers and in remote locations, making the job of safeguards inspectors more challenging. Machine learning (ML) is currently being applied to many fields to augment human performance and increase automation; in particular, ML could be used to provide insight for international inspectors to help detect the diversion of nuclear fuel from MR cores. Four ML model types (k-nearest neighbors, decision tree, random forest, and histogram-based gradient boosted ensemble) were trained on integrated flux and critical control drum angle data generated with Serpent 2 for a realistic heat pipe MR design, achieving nearly 100% binary classification accuracy of nominal and diversion core configurations by the end of 1 full power year for three of the four model types. Regression model variants were also trained, using the same input data, for predicting the number of fuel pins diverted. Root-mean-square errors below 5% of the total number of fuel pins were achieved by the 1 full power year mark for all models.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗