Engineering topics

Iverson, David

Publications and source records attributed to Iverson, David.

12 records

NASA NTRS2018

Toward Justifiable Trust in Autonomous Systems Incorporating Human Knowledge in Autonomous Systems through Machine Learning

Trust in Autonomous Systems is largely about humans trusting the decisions made by autonomous systems. This trust can be increased through learning from domain experts. In particular, autonomous systems can learn offline from past mission operations before conducting any operations of its own. Additionally, autonomous systems can learn online by obtaining human feedback during operations. We will discuss several classes of machine learning methods and our application of them to autonomous systems. The first class of methods is anomaly detection, which uses operations data to identify examples of anomalous operations. The second class of methods is inverse reinforcement learning, also known as apprenticeship learning, that takes past operations data as input and yields a controller that is able to duplicate the operations described by the data. The third class is active learning, which identifies examples on which the model is most uncertain and requests domain expert feedback.

Oza, Nikunj C.↗

NASA NTRS2016

Real-Time Early Warning Techniques

Contemporary approaches to aerospace vehicle system monitoring rely heavily on thresholds that represent a compromise between providing warning early enough to avert a mishap while simultaneously minimizing false alarms. While this reliance on thresholds has been in place for decades and has permeated both cockpits and control rooms, we often find it insufficient when retrospectively analyzing data from an accident. To modernize and enhance flight test safety, we introduce new methods of monitoring for anomalous patterns of interaction rather than for thresholds exceedance. For systems with a well characterized baseline we show how the Inductive Monitoring System (IMS), utilized by NASA in the aftermath of the Columbia accident, might be implemented in real-time to provide earlier warning than currently employed techniques. For systems without such a baseline, we introduce new developments in statistical methods relating to critical slowing down, first applied in medicine and physics, which show promise for adaptation to flight test. Finally, the familiar resource constrained environment leads to a reliance on increased instrumentation that is challenging the limits of the current one-sensor, one-indicator threshold paradigm. Existing methods thus fail to accurately reflect the true complexity of a vehicle rich with interdependent interacting systems. We highlight these concepts in a brief summary of the 2001 Air Transat Flight 236 deadstick landing in the Azores. We then suggest control room and cockpit modifications to better display the information gleaned using these novel monitoring methods.

Javorsek, Dan↗

NASA NTRS2016

Enhancing Flight Test Safety with Real-Time Early Warning Techniques

Contemporary approaches to aerospace vehicle system monitoring rely heavily on thresholds that represent a compromise between providing warning early enough to avert a mishap while simultaneously minimizing false alarms. While this reliance on thresholds has been in place for decades and has permeated both cockpits and control rooms, we often find it insufficient when retrospectively analyzing data from an accident. To modernize and enhance flight test safety, we introduce new methods of monitoring for anomalous patterns of interaction rather than for thresholds exceedance. For systems with a well characterized baseline we show how the Inductive Monitoring System (IMS), utilized by NASA in the aftermath of the Columbia accident, might be implemented in real-time to provide earlier warning than currently employed techniques. For systems without such a baseline, we introduce new developments in statistical methods relating to critical slowing down, first applied in medicine and physics, which show promise for adaptation to flight test. Finally, the familiar resource constrained environment leads to a reliance on increased instrumentation that is challenging the limits of the current "one-sensor, one-indicator" threshold paradigm. Existing methods thus fail to accurately reflect the true complexity of a vehicle rich with interdependent interacting systems. We highlight these concepts in a brief summary of the 2001 Air Transat Flight 236 deadstick landing in the Azores. We then suggest control room and cockpit modifications to better display the information gleaned using these novel monitoring methods.

Javorsek, Daniel, II↗

NASA NTRS2014

Exploration Technologies for Operations

Although the International Space Station (ISS) assembly has been completed, the Operations support teams continue to seek more efficient and effective ways to prepare for and conduct the ISS operations and future exploration missions beyond low earth orbit. This search for improvement has led to a significant collaboration between the NASA research and advanced software development community at NASA Ames Research Center and the Mission Operations community at NASA Johnson Space Center. Since 2001, NASA Ames Research Center has been developing and applying its advanced intelligent systems and human systems integration research to mission operations tools for several of the unmanned Mars missions operations. Since 2006, NASA Ames Research Center has also been developing and applying its advanced intelligent systems and human systems integration research to mission operations tools for manned operations support with the Mission Operations Directorate at NASA Johnson Space Center. This paper discusses the completion of the development and deployment of a variety of intelligent and human systems technologies adopted for manned mission operations. The technologies associated with the projects include advanced software systems for operations and human-centered computing. Human-centered computing looks to the processes and procedures that people do to perform any given job, then attempts to identify opportunities to improve these processes and procedures. In particular, for mission operations, improvements are quantified by specifically identifying how a tool can increase a persons efficiency, enhance a persons functional capability, andor improve the assurance of a persons decisions. The Ames development team has collaborated with the Mission Operations team to identify areas of efficiencies through technology infusion applications in support of the Plan, Train, and Fly activities of human-spaceflight mission operations. The specific applications discussed in this paper are in the areas of mission planning systems, mission operations design modeling and workflow automation, advanced systems monitoring, mission control technologies, search tools, training management tools, spacecraft solar array management, spacecraft power management, and spacecraft attitude planning. We discuss these specific projects between the Ames Research Center and the Johnson Space Centers Mission Operations Directorate, and how these technologies and projects are enhancing the mission operations support for the International Space Station. We also discuss the challenges, problems, and successes associated with long-distance and multi-year development projects between the research team at Ames and the Mission Operations customers at Johnson Space center. Finally, we discuss how these technology infusion applications and underlying technologies might be used in the future to support on-board operations of the crew and spacecraft systems as human exploration expands beyond low earth orbit to destinations in the solar system where communications delays will require more on-board autonomy and planning by the crew. Longer communications delays will require that the ground mission operations support will be primarily strategic in nature, while the tactical level of planning, systems monitoring and control, and failure analysisisolationrecovery will be the responsibility of both the spacecraft autonomous systems and the crew. Our expectation is that the technologies

mission operations↗

NASA NTRS2013

Advanced Caution and Warning System, Final Report - 2011

The work described in this report is a continuation of the ACAWS work funded in fiscal year (FY) 2010 under the Exploration Technology Development Program (ETDP), Integrated Systems Health Management (ISHM) project. In FY 2010, we developed requirements for an ACAWS system and vetted the requirements with potential users via a concept demonstration system. In FY 2011, we developed a working prototype of aspects of that concept, with placeholders for technologies to be fully developed in future phases of the project. The objective is to develop general capability to assist operators with system health monitoring and failure diagnosis. Moreover, ACAWS was integrated with the Discrete Controls (DC) task of the Autonomous Systems and Avionics (ASA) project. The primary objective of DC is to demonstrate an electronic and interactive procedure display environment and multiple levels of automation (automatic execution by computer, execution by computer if the operator consents, and manual execution by the operator).

Spirkovska, Lilly↗

NASA NTRS2009

On Organization of Information: Approach and Early Work

In this report we describe an approach for organizing information for presentation and display. "e approach stems from the observation that there is a stepwise progression in the way signals (from the environment and the system under consideration) are extracted and transformed into data, and then analyzed and abstracted to form representations (e.g., indications and icons) on the user interface. In physical environments such as aerospace and process control, many system components and their corresponding data and information are interrelated (e.g., an increase in a chamber s temperature results in an increase in its pressure). "ese interrelationships, when presented clearly, allow users to understand linkages among system components and how they may affect one another. Organization of these interrelationships by means of an orderly structure provides for the so-called "big picture" that pilots, astronauts, and operators strive for.

Degani, Asaf↗

NASA NTRS2006

Integrated System Health Management (ISHM) Technology Demonstration Project Final Report

Integrated System Health Management (ISHM) is an essential capability that will be required to enable upcoming explorations mission systems such as the Crew Exploration Vehicle (CEV) and Crew Launch Vehicle (CLV), as well as NASA aeronautics missions. However, the lack of flight experience and available test platforms have held back the infusion by NASA Ames Research Center (ARC) and the Jet Propulsion Laboratory (JPL) of ISHM technologies into future space and aeronautical missions. To address this problem, a pioneer project was conceived to use a high-performance aircraft as a low-cost proxy to develop, mature, and verify the effectiveness of candidate ISHM technologies. Given the similarities between spacecraft and aircraft, an F/A-18 currently stationed at Dryden Flight Research Center (DFRC) was chosen as a suitable host platform for the test bed. This report describes how the test bed was conceived, how the technologies were integrated on to the aircraft, and how these technologies were matured during the project. It also describes the lessons learned during the project and a forward path for continued work.

Mackey, Ryan↗

NASA NTRS2005

Modeling and Diagnostic Software for Liquefying-Fuel Rockets

A report presents a study of five modeling and diagnostic computer programs considered for use in an integrated vehicle health management (IVHM) system during testing of liquefying-fuel hybrid rocket engines in the Hybrid Combustion Facility (HCF) at NASA Ames Research Center. Three of the programs -- TEAMS, L2, and RODON -- are model-based reasoning (or diagnostic) programs. The other two programs -- ICS and IMS -- do not attempt to isolate the causes of failures but can be used for detecting faults. In the study, qualitative models (in TEAMS and L2) and quantitative models (in RODON) having varying scope and completeness were created. Each of the models captured the structure and behavior of the HCF as a physical system. It was noted that in the cases of the qualitative models, the temporal aspects of the behavior of the HCF and the abstraction of sensor data are handled outside of the models, and it is necessary to develop additional code for this purpose. A need for additional code was also noted in the case of the quantitative model, though the amount of development effort needed was found to be less than that for the qualitative models.

Poll, Scott↗

NASA NTRS2003

System Modeling and Diagnostics for Liquefying-Fuel Hybrid Rockets

A Hybrid Combustion Facility (HCF) was recently built at NASA Ames Research Center to study the combustion properties of a new fuel formulation that burns approximately three times faster than conventional hybrid fuels. Researchers at Ames working in the area of Integrated Vehicle Health Management recognized a good opportunity to apply IVHM techniques to a candidate technology for next generation launch systems. Five tools were selected to examine various IVHM techniques for the HCF. Three of the tools, TEAMS (Testability Engineering and Maintenance System), L2 (Livingstone2), and RODON, are model-based reasoning (or diagnostic) systems. Two other tools in this study, ICS (Interval Constraint Simulator) and IMS (Inductive Monitoring System) do not attempt to isolate the cause of the failure but may be used for fault detection. Models of varying scope and completeness were created, both qualitative and quantitative. In each of the models, the structure and behavior of the physical system are captured. In the qualitative models, the temporal aspects of the system behavior and the abstraction of sensor data are handled outside of the model and require the development of additional code. In the quantitative model, less extensive processing code is also necessary. Examples of fault diagnoses are given.

Poll, Scott↗

NASA NTRS2003

Characterization of Model-Based Reasoning Strategies for Use in IVHM Architectures

Open architectures are gaining popularity for Integrated Vehicle Health Management (IVHM) applications due to the diversity of subsystem health monitoring strategies in use and the need to integrate a variety of techniques at the system health management level. The basic concept of an open architecture suggests that whatever monitoring or reasoning strategy a subsystem wishes to deploy, the system architecture will support the needs of that subsystem and will be capable of transmitting subsystem health status across subsystem boundaries and up to the system level for system-wide fault identification and diagnosis. There is a need to understand the capabilities of various reasoning engines and how they, coupled with intelligent monitoring techniques, can support fault detection and system level fault management. Researchers in IVHM at NASA Ames Research Center are supporting the development of an IVHM system for liquefying-fuel hybrid rockets. In the initial stage of this project, a few readily available reasoning engines were studied to assess candidate technologies for application in next generation launch systems. Three tools representing the spectrum of model-based reasoning approaches, from a quantitative simulation based approach to a graph-based fault propagation technique, were applied to model the behavior of the Hybrid Combustion Facility testbed at Ames. This paper summarizes the characterization of the modeling process for each of the techniques.

Poll, Scott↗

NASA NTRS1997

Multiple Fault Isolation in Redundant Systems

Fault diagnosis in large-scale systems that are products of modern technology present formidable challenges to manufacturers and users. This is due to large number of failure sources in such systems and the need to quickly isolate and rectify failures with minimal down time. In addition, for fault-tolerant systems and systems with infrequent opportunity for maintenance (e.g., Hubble telescope, space station), the assumption of at most a single fault in the system is unrealistic. In this project, we have developed novel block and sequential diagnostic strategies to isolate multiple faults in the shortest possible time without making the unrealistic single fault assumption.

Pattipati, Krishna R.↗

NASA NTRS1994

Matrix Failure Modes and Effects Analysis as a Knowledge Base for a Real Time Automated Diagnosis Expert System

Failure Modes and Effects Analysis contain a wealth of information that can be used to create the knowledge base required for building automated diagnostic Expert systems. A real time monitoring and diagnosis expert system based on an actual NASA project's matrix failure modes and effects analysis was developed. This Expert system Was developed at NASA Ames Research Center. This system was first used as a case study to monitor the Research Animal Holding Facility (RAHF), a Space Shuttle payload that is used to house and monitor animals in orbit so the effects of space flight and microgravity can be studied. The techniques developed for the RAHF monitoring and diagnosis Expert system are general enough to be used for monitoring and diagnosis of a variety of other systems that undergo a Matrix FMEA. This automated diagnosis system was successfully used on-line and validated on the Space Shuttle flight STS-58, mission SLS-2 in October 1993.

Herrin, Stephanie↗