SEARCH · Engineering Papers
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.
This paper discusses the combination of software robots and expert systems to automate everyday business tasks. Tasks which require people to repetitively interact with multiple systems screens as well as multiple systems.
The importance of the incorporation of various graphics-oriented features into CLIPS is discussed. These popular features, which have been implemented in a version of CLIPS developed for a popular workstation, are described and their usefulness in the development of expert systems is examined.
We are attempting to show how independent but cooperating expert systems, executing within a parallel production system (PCLIPS), can operate and control a completely automated, fault tolerant prototype of a factory of the future (The Lincoln Logs Factory of the Future). The factory consists of a CAD system for designing the Lincoln Log Houses, two workcells, and a materials handling system. A workcell consists of two robots, part feeders, and a frame mounted vision system.
This paper describes an ongoing program to augment such an expert system gas turbine startup advisor, known as the EPRI SA VANT System, by including an intelligent training package. It will give a brief background on the SA VANT development and an overview of its evolution into a full-blown Gas Turbine Information System (GTIS) for rapid access of on-line documentation, diagnostics, and training. In particular, the paper will address: (1) the conversion of the knowledge base used by the SA VANT startup advisor so that it can be used for both training and job aiding; and (2) the hypertext-oriented user manuals being incorporated into the system for rapidly accessing on-line documentation at the job site.
Whether optimization techniques or expert systems technologies are used, the underlying inference processes and the model or knowledge base for a computerized problem-solving system are likely to be incomplete for any given complex, real-world task. To deal with the resultant brittleness, it has been suggested that 'cooperative' rather than 'automated' problem-solving systems be designed. Such cooperative systems are proposed to explicitly enhance the collaboration of people and the computer system when working in partnership to solve problems. This study evaluates the impact of alternative design concepts on the performance of airline pilots interacting with such a cooperative system designed to support enroute flight planning. Thirty pilots were studied using three different versions of the system. The results clearly demonstrate that different system design concepts can strongly influence the cognitive processes of users. Indeed, one of the designs studied caused four times as many pilots to accept a poor flight amendment. Based on think-aloud protocols, cognitive models are proposed to account for how features of the computer system interacted with specific types of scenarios to influence exploration and decision-making by the pilots. The results are then used to develop recommendations for guiding the design of cooperative systems.
An effort is underway at the University of Tennessee Space Institute to develop diagnostic expert system methodologies based on the analysis of patterns of behavior of physical mechanisms. In this approach, fault diagnosis is conceptualized as the mapping or association of patterns of sensor data to patterns representing fault conditions. Neural networks are being investigated as a means of storing and retrieving fault scenarios. Neural networks offer several powerful features in fault diagnosis, including (1) general pattern matching capabilities, (2) resistance to noisy input data, (3) the ability to be trained by example, and (4) the potential for implementation on parallel computer architectures. This paper presents (1) an autoassociative neural network topology, i.e. the network input and output is identical when properly trained, and hence learning is unsupervised; (2) the training regimen used; and (3) the response of the system to inputs representing both previously observed and unkown fault scenarios. The effects of noise on the integrity of the diagnosis are also evaluated.
Many important applications can be formalized as constrained optimization tasks. For example, we are studying the engineering domain of two-dimensional (2-D) structural design. In this task, the goal is to design a structure of minimum weight that bears a set of loads. A solution to a design problem in which there is a single load (L) and two stationary support points (S1 and S2) consists of four members, E1, E2, E3, and E4 that connect the load to the support points is discussed. In principle, optimal solutions to problems of this kind can be found by numerical optimization techniques. However, in practice [Vanderplaats, 1984] these methods are slow and they can produce different local solutions whose quality (ratio to the global optimum) varies with the choice of starting points. Hence, their applicability to real-world problems is severely restricted. To overcome these limitations, we propose to augment numerical optimization by first performing a symbolic compilation stage to produce: (a) objective functions that are faster to evaluate and that depend less on the choice of the starting point and (b) selection rules that associate problem instances to a set of recommended solutions. These goals are accomplished by successive specializations of the problem class and of the associated objective functions. In the end, this process reduces the problem to a collection of independent functions that are fast to evaluate, that can be differentiated symbolically, and that represent smaller regions of the overall search space. However, the specialization process can produce a large number of sub-problems. This is overcome by deriving inductively selection rules which associate problems to small sets of specialized independent sub-problems. Each set of candidate solutions is chosen to minimize a cost function which expresses the tradeoff between the quality of the solution that can be obtained from the sub-problem and the time it takes to produce it. The overall solution to the problem, is then obtained by solving in parallel each of the sub-problems in the set and computing the one with the minimum cost. In addition to speeding up the optimization process, our use of learning methods also relieves the expert from the burden of identifying rules that exactly pinpoint optimal candidate sub-problems. In real engineering tasks it is usually too costly to the engineers to derive such rules. Therefore, this paper also contributes to a further step towards the solution of the knowledge acquisition bottleneck [Feigenbaum, 1977] which has somewhat impaired the construction of rulebased expert systems.
This Project has two main thrusts-preservation of special knowledge and its useful representation via computers. NASA is losing the expertise of its engineers and scientists who put together the great missions of the past. We no longer are landing men on the moon. Some of the equipment still used today (such as the RL-10 rocket) was designed decades ago by people who are now retiring. Furthermore, there has been a lack, in some areas of technology, of new projects that overlap with the old and that would have provided opportunities for monitoring by senior engineers of the young ones. We are studying this problem and trying out a couple of methods of soliciting and recording rare knowledge from experts. One method is that of Concept Maps which produces a graphical interface to knowledge even as it helps solicit that knowledge. We arranged for experienced help in this method from John Coffey of the Institute of Human and Machine Technology at the University of West Florida. A second method which we plan to try out in May, is a video-taped review of selected failed missions (e.g., the craft tumbled and blew up). Five senior engineers (most already retired from NASA) will, as a team, analyze available data, illustrating their thought processes as they try to solve the problem of why a space craft failed to complete its mission. The session will be captured in high quality audio and with at least two video cameras. The video can later be used to plan future concept mapping interviews and, in edited form, be a product in itself. Our computer representations of the amassed knowledge may eventually, via the methods of expert systems, be joined with other software being prepared as a suite of tools to aid future engineers designing rocket engines. In addition to representation by multimedia concept maps, we plan to consider linking vast bodies of text (and other media) by hypertexting methods.
The increasing requirement to dramatically reduce the cost of mission operations led to increased emphasis on automation technology. The expert system technology used at the Goddard Space Flight Center (MD) is currently being applied to the automation of spacecraft control center activities. The generic inferential executor (GENIE) is a tool which allows pass automation applications to be constructed. The pass script templates constructed encode the tasks necessary to mimic flight operations team interactions with the spacecraft during a pass. These templates can be configured with data specific to a particular pass. Animated graphical displays illustrate the progress during the pass. The first GENIE application automates passes of the solar, anomalous and magnetospheric particle explorer (SAMPEX) spacecraft.
The application of artificial neural networks to capture structural design expertise is demonstrated. The principal advantage of a trained neural network is that it requires a trivial computational effort to produce an acceptable new design. For the class of problems addressed, the development of a conventional expert system would be extremely difficult. In the present effort, a structural optimization code with multiple nonlinear programming algorithms and an artificial neural network code NETS were used. A set of optimum designs for a ring and two aircraft wings for static and dynamic constraints were generated using the optimization codes. The optimum design data were processed to obtain input and output pairs, which were used to develop a trained artificial neural network using the code NETS. Optimum designs for new design conditions were predicted using the trained network. Neural net prediction of optimum designs was found to be satisfactory for the majority of the output design parameters. However, results from the present study indicate that caution must be exercised to ensure that all design variables are within selected error bounds.
Phase 2 of the Lunar-Mars Life Support Test Project was conducted in June and July of 1996 at the NASA Johnson Space Center. The primary objective of Phase 2 was to demonstrate and evaluate an integrated physicochemical air revitalization and regenerative water recovery system capable of sustaining a human crew of four for 30 days inside a closed chamber. The crew (3 males and 1 female) was continuously present inside a chamber throughout the 30-day test. The objective of this paper was to describe crew interactions and human factors for the test. Crew preparations for the test included training and familiarization of chamber systems and accommodations, and medical and psychological evaluations. During the test, crew members provided metabolic loads for the life support systems, performed maintenance on chamber systems, and evaluated human factors inside the chamber. Overall, the four crew members found the chamber to be comfortable for the 30-day test. The crew performed well together and this was attributed in part to team dynamics, skill mix (one commander, two system experts, and one logistics lead), and a complementary mix of personalities. Communication with and support by family, friends, and colleagues were identified as important contributors to the high morale of the crew during the test. Lessons learned and recommendations for future testing are presented by the crew in this paper.
The General Aviation (GA) industry has suffered a ten-year decline in the number of airplanes sold. This decline is due mainly to the increase cost associated with purchasing, insuring, maintaining, operating, and pilot training a GA airplane. In response to this decline the Federal Aviation Administration (FAA) and the National Aeronautics and Space Administration (NASA) developed a program (Advanced General Aviation Transport Experiments - AGATE) to address these issues. The purpose of AGATE focused within this report is to reduce the costs to acquire and maintain instrument-flight-proficiency. The AGATE program defined four elements necessary to accomplish these goals: (1) new and intuitive cockpit displays and controls, (2) situation technologies for weather, traffic, and navigation, (3) expert systems for system monitoring, and (4) reduced cost training methods. One recognized need for the GA pilot and airplane is to provide cockpit displays and systems already available to transport category airplane. These displays such as Electronic Flight and Instrument System (EFIS), graphic weather and traffic displays, and flight management systems. The goal of this grant was to develop the AGATE GA Display Evaluation Workstation as a tool to test these existing and emerging technologies in the GA environment.
Contract efforts are focused on refining the Robust Design Methodology for Conceptual Aircraft Design. Robust Design Simulation (RDS) was developed earlier as a potential solution to the need to do rapid trade-offs while accounting for risk, conflict, and uncertainty. The core of the simulation revolved around Response Surface Equations as approximations of bounded design spaces. An ongoing investigation is concerned with the advantages of using Neural Networks in conceptual design. Thought was also given to the development of systematic way to choose or create a baseline configuration based on specific mission requirements. Expert system was developed, which selects aerodynamics, performance and weights model from several configurations based on the user's mission requirements for subsonic civil transport. The research has also resulted in a step-by-step illustration on how to use the AMV method for distribution generation and the search for robust design solutions to multivariate constrained problems.
In this study, we proposed an Advanced Health Management System (AHMS) functional architecture and conducted a technology assessment for liquid propellant rocket engine lifecycle health management. The purpose of the AHMS is to improve reusable rocket engine safety and to reduce between-flight maintenance. During the study, past and current reusable rocket engine health management-related projects were reviewed, data structures and health management processes of current rocket engine programs were assessed, and in-depth interviews with rocket engine lifecycle and system experts were conducted. A generic AHMS functional architecture, with primary focus on real-time health monitoring, was developed. Fourteen categories of technology tasks and development needs for implementation of the AHMS were identified, based on the functional architecture and our assessment of current rocket engine programs. Five key technology areas were recommended for immediate development, which (1) would provide immediate benefits to current engine programs, and (2) could be implemented with minimal impact on the current Space Shuttle Main Engine (SSME) and Reusable Launch Vehicle (RLV) engine controllers.
Jet Engines may experience severe vibration due to the sudden imbalance caused by blade failure. This research investigates employment of on board magnetic bearings or piezoelectric actuators to cancel these forces in flight. This operation requires identification of the source of the vibrations via an expert system, determination of the required phase angles and amplitudes for the correction forces, and application of the desired control signals to the magnetic bearings or piezo electric actuators. This paper will show the architecture of the software system, details of the control algorithm used for the sudden imbalance correction project described above, and the laboratory test results.
Probabilistic graphical models are being used widely in artificial intelligence and statistics, for instance, in diagnosis and expert systems, as a framework for representing and reasoning with probabilities and independencies. They come with corresponding algorithms for performing statistical inference. This offers a unifying framework for prototyping and/or generating data analysis algorithms from graphical specifications. This paper illustrates the framework with an example and then presents some basic techniques for the task: problem decomposition and the calculation of exact Bayes factors. Other tools already developed, such as automatic differentiation, Gibbs sampling, and use of the EM algorithm, make this a broad basis for the generation of data analysis software.
Probabilistic graphical models (directed and undirected Markov fields, and combined in chain graphs) are used widely in expert systems, image processing and other areas as a framework for representing and reasoning with probabilities. They come with corresponding algorithms for performing probabilistic inference. This paper discusses an extension to these models by Spiegelhalter and Gilks, plates, used to graphically model the notion of a sample. This offers a graphical specification language for representing data analysis problems. When combined with general methods for statistical inference, this also offers a unifying framework for prototyping and/or generating data analysis algorithms from graphical specifications. This paper outlines the framework and then presents some basic tools for the task: a graphical version of the Pitman-Koopman Theorem for the exponential family, problem decomposition, and the calculation of exact Bayes factors. Other tools already developed, such as automatic differentiation, Gibbs sampling, and use of the EM algorithm, make this a broad basis for the generation of data analysis software.
On February 23, 1999, thermal imagery of Marshall Space Flight Center, Alabama was collected using an airborne thermal camera. Ground resolution was I in. Approximately 40 km 2 of thermal imagery in and around Marshall Space Flight Center (MSFC) was analyzed to determine the location of springs for groundwater monitoring. Subsequently, forty-five springs were located ranging in flow from a few ml/sec to approximately 280 liter/sec. Groundwater temperatures are usually near the mean annual surface air temperature. On thermography collected during the winter, springs show up as very warm spots. Many of the new springs were submerged in lakes, streams, or swamps; consequently, flow measurements were difficult. Without estimates of discharge, the impacts of contaminated discharge on surface streams would be difficult to evaluate. An approach to obtaining an estimate was developed using the Environmental Protection Agency (EPA) Cornell Mixing Zone Expert System (CORMIX). The thermography was queried to obtain a temperature profile down the center of the surface plume. The spring discharge was modeled with CORMIX, and the flow adjusted until the surface temperature profile was matched. The presence of volatile compounds in some of the new springs also allowed MSFC to unravel the natural system of solution cavities of the karst aquifer. Sampling results also showed that two springs on either side of a large creek had the same water source so that groundwater was able to pass beneath the creek.