Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “learned”

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 937 records · Page 52

Application of fuzzy logic-neural network based reinforcement learning to proximity and docking operations: Translational controller results

The reinforcement learning techniques developed at Ames Research Center are being applied to proximity and docking operations using the Shuttle and Solar Maximum Mission (SMM) satellite simulation. In utilizing these fuzzy learning techniques, we also use the Approximate Reasoning based Intelligent Control (ARIC) architecture, and so we use two terms interchangeable to imply the same. This activity is carried out in the Software Technology Laboratory utilizing the Orbital Operations Simulator (OOS). This report is the deliverable D3 in our project activity and provides the test results of the fuzzy learning translational controller. This report is organized in six sections. Based on our experience and analysis with the attitude controller, we have modified the basic configuration of the reinforcement learning algorithm in ARIC as described in section 2. The shuttle translational controller and its implementation in fuzzy learning architecture is described in section 3. Two test cases that we have performed are described in section 4. Our results and conclusions are discussed in section 5, and section 6 provides future plans and summary for the project.

Jani, Yashvant↗

A self-learning rule base for command following in dynamical systems

In this paper, a self-learning Rule Base for command following in dynamical systems is presented. The learning is accomplished though reinforcement learning using an associative memory called SAM. The main advantage of SAM is that it is a function approximator with explicit storage of training samples. A learning algorithm patterned after the dynamic programming is proposed. Two artificially created, unstable dynamical systems are used for testing, and the Rule Base was used to generate a feedback control to improve the command following ability of the otherwise uncontrolled systems. The numerical results are very encouraging. The controlled systems exhibit a more stable behavior and a better capability to follow reference commands. The rules resulting from the reinforcement learning are explicitly stored and they can be modified or augmented by human experts. Due to overlapping storage scheme of SAM, the stored rules are similar to fuzzy rules.

Tsai, Wei K.↗

Fuzzy learning under and about an unfamiliar fuzzy teacher

This study addresses the problem of optimal parametric learning in unfamiliar fuzzy environments. Prior studies in the domain of unfamiliar environments, which employed either crisp or fuzzy approaches to model the uncertainty or imperfectness of the learning environment, assumed that the training sample labels provided by the unfamiliar teacher were crisp, even if not perfect. Here, the more realistic problem of fuzzy learning under an unfamiliar teacher who provides only fuzzy (instead of crisp) labels, is tackled by expanding the previously defined fuzzy membership concepts to include an additional component representative of the fuzziness of the teacher. The previously studied scenarios, namely, crisp and fuzzy learning under (crisp) unfamiliar teacher, can be looked upon as special cases of this new methodology. As under the earlier studies, the estimated membership functions can then be deployed during the ensuing classification decision phase to judiciously take into account the imperfectness of the learning environment. The study also offers some insight into the properties of several of these fuzzy membership function estimators by examining their behavior under certain specific scenarios.

Dasarathy, Belur V.↗

On the integration of reinforcement learning and approximate reasoning for control

The author discusses the importance of strengthening the knowledge representation characteristic of reinforcement learning techniques using methods such as approximate reasoning. The ARIC (approximate reasoning-based intelligent control) architecture is an example of such a hybrid approach in which the fuzzy control rules are modified (fine-tuned) using reinforcement learning. ARIC also demonstrates that it is possible to start with an approximately correct control knowledge base and learn to refine this knowledge through further experience. On the other hand, techniques such as the TD (temporal difference) algorithm and Q-learning establish stronger theoretical foundations for their use in adaptive control and also in stability analysis of hybrid reinforcement learning and approximate reasoning-based controllers.

Berenji, Hamid R.↗

Learning and tuning fuzzy logic controllers through reinforcements

This paper presents a new method for learning and tuning a fuzzy logic controller based on reinforcements from a dynamic system. In particular, our generalized approximate reasoning-based intelligent control (GARIC) architecture (1) learns and tunes a fuzzy logic controller even when only weak reinforcement, such as a binary failure signal, is available; (2) introduces a new conjunction operator in computing the rule strengths of fuzzy control rules; (3) introduces a new localized mean of maximum (LMOM) method in combining the conclusions of several firing control rules; and (4) learns to produce real-valued control actions. Learning is achieved by integrating fuzzy inference into a feedforward neural network, which can then adaptively improve performance by using gradient descent methods. We extend the AHC algorithm of Barto et al. (1983) to include the prior control knowledge of human operators. The GARIC architecture is applied to a cart-pole balancing system and demonstrates significant improvements in terms of the speed of learning and robustness to changes in the dynamic system's parameters over previous schemes for cart-pole balancing.

Berenji, Hamid R.↗

Finding accurate frontiers: A knowledge-intensive approach to relational learning

An approach to analytic learning is described that searches for accurate entailments of a Horn Clause domain theory. A hill-climbing search, guided by an information based evaluation function, is performed by applying a set of operators that derive frontiers from domain theories. The analytic learning system is one component of a multi-strategy relational learning system. We compare the accuracy of concepts learned with this analytic strategy to concepts learned with an analytic strategy that operationalizes the domain theory.

Pazzani, Michael↗

Locomotion training of legged robots using hybrid machine learning techniques

In this study artificial neural networks and fuzzy logic are used to control the jumping behavior of a three-link uniped robot. The biped locomotion control problem is an increment of the uniped locomotion control. Study of legged locomotion dynamics indicates that a hierarchical controller is required to control the behavior of a legged robot. A structured control strategy is suggested which includes navigator, motion planner, biped coordinator and uniped controllers. A three-link uniped robot simulation is developed to be used as the plant. Neurocontrollers were trained both online and offline. In the case of on-line training, a reinforcement learning technique was used to train the neurocontroller to make the robot jump to a specified height. After several hundred iterations of training, the plant output achieved an accuracy of 7.4%. However, when jump distance and body angular momentum were also included in the control objectives, training time became impractically long. In the case of off-line training, a three-layered backpropagation (BP) network was first used with three inputs, three outputs and 15 to 40 hidden nodes. Pre-generated data were presented to the network with a learning rate as low as 0.003 in order to reach convergence. The low learning rate required for convergence resulted in a very slow training process which took weeks to learn 460 examples. After training, performance of the neurocontroller was rather poor. Consequently, the BP network was replaced by a Cerebeller Model Articulation Controller (CMAC) network. Subsequent experiments described in this document show that the CMAC network is more suitable to the solution of uniped locomotion control problems in terms of both learning efficiency and performance. A new approach is introduced in this report, viz., a self-organizing multiagent cerebeller model for fuzzy-neural control of uniped locomotion is suggested to improve training efficiency. This is currently being evaluated for a possible patent by NASA, Johnson Space Center. An alternative modular approach is also developed which uses separate controllers for each stage of the running stride. A self-organizing fuzzy-neural controller controls the height, distance and angular momentum of the stride. A CMAC-based controller controls the movement of the leg from the time the foot leaves the ground to the time of landing. Because the leg joints are controlled at each time step during flight, movement is smooth and obstacles can be avoided. Initial results indicate that this approach can yield fast, accurate results.

Simon, William E.↗

There is No Free Lunch: Tradeoffs in the Utility of Learned Knowledge

With the recent introduction of learning in integrated systems, there is a need to measure the utility of learned knowledge for these more complex systems. A difficulty arrises when there are multiple, possibly conflicting, utility metrics to be measured. In this paper, we present schemes which trade off conflicting utility metrics in order to achieve some global performance objectives. In particular, we present a case study of a multi-strategy machine learning system, mutual theory refinement, which refines world models for an integrated reactive system, the Entropy Reduction Engine. We provide experimental results on the utility of learned knowledge in two conflicting metrics - improved accuracy and degraded efficiency. We then demonstrate two ways to trade off these metrics. In each, some learned knowledge is either approximated or dynamically 'forgotten' so as to improve efficiency while degrading accuracy only slightly.

Kedar, Smadar T.↗

Learning from Inconsistency

This position paper argues that inconsistencies that occur during the development of a software specification offer an excellent way of learning more about the development process. We base this argument on our work on inconsistency management. Much attention has been devoted recently to the need to allow inconsistencies to occur during software development, to facilitate flexible development strategies, especially for collaborative work. Recent work has concentrated on reasoning in the presence of inconsistency, tracing inconsistencies with 'pollution markers' and supporting resolution. We argue here that one of the most important aspects of inconsistency is the learning opportunity it provides. We are therefore concerned with how to capture this learning outcome so that its significance is not lost. We present a small example of how apprentice software engineers learn from their mistakes, and outline how an inconsistency management tool could support this learning. We then argue that the approach can be used more generally as part of continuous process improvement.

Easterbrook, Steve↗

NASA/Mir Phase 1: A Lesson Learned in Long Duration Mission Planning and Operations

The NASA/Mir Phase I Program is a precursor to the assembly and operations of the International Space Station (ISS). The Phase I Program covers a four-year period and encompasses over ten shuttle flights to the Mir, culminating with the flight of STS-91 in June 1998. The Mir-Orbiting Station (Mir-OS) provides an excellent environment, both onboard and on the ground, for learning and studying the characteristics associated with crew-tended operations on long duration missions. By the end of the Phase I program, NASA astronauts will have accumulated over two years of continuous operations experience working and living onboard the Mir-OS. Ground controllers, working in Russia and at remote sites in the United States, will have gained valuable experience in the generation, maintenance, and coordination of the information and products required to support the long duration crew members onboard the Mir-OS. The experience gained by the crew and ground support personnel in the area of long duration mission operations has resulted in a number of factors that need to be considered in the definition and implementation of the processes and products for the ISS. While there are many areas in which experience has been gained, this paper focuses on the lessons learned regarding the level of detail and flexibility of crew timelines. Provided in this paper is an overview of the characteristics associated with long duration mission timeline development and execution, a summary of applicable lessons learned identified and documented through participation in the NASA/Mir Phase I Program, a detailed discussion of "Gross Timelines" as implementation which addresses the lessons learned. Also discussed are the challenges that must be overcome in getting the ISS program to recognize, evaluate, and learn from the crew and ground controller experience gained through participation in the NASA/Mir Phase I Program.

Hagopian, Jeff↗

Operations for Learning with Graphical Models

This paper is a multidisciplinary review of empirical, statistical learning from a graphical model perspective. Well-known examples of graphical models include Bayesian net- works, directed graphs representing a Markov chain, and undirected networks representing a Markov field. These graphical models are extended to model data analysis and empirical learning using the notation of plates. Graphical operations for simplifying and manipulating a problem are provided including decomposition, differentiation, and the manipulation of probability models from the exponential family. These operations adapt existing techniques from statistics and automatic differentiation to graphs. Two standard algorithm schemes for learning are reviewed in a graphical framework: Gibbs sampling and the expectation maximization algorithm. Some algorithms are developed in this graphical framework including a generalized version of linear regression, techniques for feed-forward networks, and learning Gaussian and discrete Bayesian networks from data. The paper concludes by sketching some implications for data analysis and summarizing some popular algorithms that fall within the framework presented. The main original contributions here are the decomposition techniques and the demonstration that graphical models provide a framework for understanding and developing complex learning algorithms.

Buntine, Wray L.↗

Technology Development, Implementation and Assessment: K-16 Pre-Service, In-Service and Distance Learning Initiatives

The technologies associated with distance learning are evolving rapidly, giving to educators a potential tool for enhancing the educational experiences of large numbers of students simultaneously. This enhancement, in order to be effective, must take into account the various agendas of teachers, administrators, state systems, and of course students. It must also make use of the latest research on effective pedagogy. This combination, effective pedagogy and robust information technology, is a powerful vehicle for communicating, to a large audience of school children the excitement of mathematics and science--an excitement that for the most part is now well-hidden. This project,"Technology Development, Implementation and Assessment," proposed to bring to bear on the education of learners in grades 3 - 8 in science and mathematics both advances in information technology and in effective pedagogy. Specifically, the project developed components NASA CONNECT video series--problem-based learning modules that focus on the scientific method and that incorporate problem-based learning scenarios tied to national mathematics and science standards. These videos serve two purposes; they engage students in the excitement of hands-on learning and they model for the teachers of these students the problem-based learning practices that are proving to be excellent ways to teach science and mathematics to school students. Another component of NASA CONNECT is the accompanying web-site.

Williams, William B., Jr.↗

Learning and Tuning of Fuzzy Rules

In this chapter, we review some of the current techniques for learning and tuning fuzzy rules. For clarity, we refer to the process of generating rules from data as the learning problem and distinguish it from tuning an already existing set of fuzzy rules. For learning, we touch on unsupervised learning techniques such as fuzzy c-means, fuzzy decision tree systems, fuzzy genetic algorithms, and linear fuzzy rules generation methods. For tuning, we discuss Jang's ANFIS architecture, Berenji-Khedkar's GARIC architecture and its extensions in GARIC-Q. We show that the hybrid techniques capable of learning and tuning fuzzy rules, such as CART-ANFIS, RNN-FLCS, and GARIC-RB, are desirable in development of a number of future intelligent systems.

Berenji, Hamid R.↗

The Relationship Between Fidelity and Learning in Aviation Training and Assessment

Flight simulators can be designed to train pilots or assess their flight performance. Low-Fidelity simulators maximize the initial learning rate of novice pilots and minimize initial costs; whereas, expensive, high-fidelity simulators predict the realworld in-flight performance of expert pilots (Fink & Shriver, 1978 Hays & Singer 1989; Kinkade & Wheaton. 1972). Although intuitively appealing and intellectually convenient to generalize concepts of learning and assessment, what holds true for the role of fidelity in assessment may not always hold true for learning, and vice versa. To bring clarity to this issue, the author distinguishes the role of fidelity in learning from its role in assessment as a function of skill level by applying the hypothesis of Alessi (1988) and reviewing the Laughery, Ditzian, and Houtman (1982) study on simulator validity. Alessi hypothesized that there is it point beyond which one additional unit of flight-simulator fidelity results in a diminished rate of learning. The author of this current paper also suggests the existence of an optimal point beyond which one additional unit of flight-simulator fidelity results in a diminished rate of practical assessment of nonexpert pilot performance.

Noble, Cliff↗

Lessons Learned and Technical Standards: A Logical Marriage for Future Space Systems Design

A comprehensive database of engineering lessons learned that corresponds with relevant technical standards will be a valuable asset to those engaged in studies on future space vehicle developments, especially for structures, materials, propulsion, control, operations and associated elements. In addition, this will enable the capturing of technology developments applicable to the design, development, and operation of future space vehicles as planned in the Space Launch Initiative. Using the time-honored tradition of passing on lessons learned while utilizing the newest information technology, NASA has launched an intensive effort to link lessons learned acquired through various Internet databases with applicable technical standards. This paper will discuss the importance of lessons learned, the difficulty in finding relevant lessons learned while engaged in a space vehicle development, and the new NASA effort to relate them to technical standards that can help alleviate this difficulty.

Gill, Paul S.↗

Multiple subclasses of Purkinje cells in the primate floccular complex provide similar signals to guide learning in the vestibulo-ocular reflex

The neural "learning rules" governing the induction of plasticity in the cerebellum were analyzed by recording the patterns of neural activity in awake, behaving animals during stimuli that induce a form of cerebellum-dependent learning. We recorded the simple- and complex-spike responses of a broad sample of Purkinje cells in the floccular complex during a number of stimulus conditions that induce motor learning in the vestibulo-ocular reflex (VOR). Each subclass of Purkinje cells carried essentially the same information about required changes in the gain of the VOR. The correlation of simple-spike activity in Purkinje cells with activity in vestibular pathways could guide learning during low-frequency but not high-frequency stimuli. Climbing fiber activity could guide learning during all stimuli tested but only if compared with the activity present approximately 100 msec earlier in either vestibular pathways or Purkinje cells.

Non-NASA Center↗

The APPL "Learning Map"

Root Learning, a learning consulting organization with a background in strategic planning, recognizes the knowledge gap that frequently exists between a leadership team and the rest of an organization. Team members supposedly working toward the same goal don't always have the same vision as to where the organization is headed, and they may not understand how the piece they are accountable for fits into the big picture. To address these complex problems, Root Learning utilizes the age-old tools of sarcasm, metaphor and graphics (much in the same way that ASK uses a traditional storytelling format.) The company is best known for creating "Learning Maps" like this one: humorous drawings based on the inner workings of an organization. Their purpose is to put complex topics on the table, to stimulate discussion, and to ultimately give team members a common vision of where the organization is going and what role they personally play in getting there. APPL knows how effective it is to incorporate new and engaging techniques into its knowledge sharing programs. By collaborating with Root Learning, we were able to expand the knowledge of the organization and add one more of these techniques to our repertoire.

Source record↗

Development of Advanced Verification and Validation Procedures and Tools for the Certification of Learning Systems in Aerospace Applications

Adaptive control technologies that incorporate learning algorithms have been proposed to enable automatic flight control and vehicle recovery, autonomous flight, and to maintain vehicle performance in the face of unknown, changing, or poorly defined operating environments. In order for adaptive control systems to be used in safety-critical aerospace applications, they must be proven to be highly safe and reliable. Rigorous methods for adaptive software verification and validation must be developed to ensure that control system software failures will not occur. Of central importance in this regard is the need to establish reliable methods that guarantee convergent learning, rapid convergence (learning) rate, and algorithm stability. This paper presents the major problems of adaptive control systems that use learning to improve performance. The paper then presents the major procedures and tools presently developed or currently being developed to enable the verification, validation, and ultimate certification of these adaptive control systems. These technologies include the application of automated program analysis methods, techniques to improve the learning process, analytical methods to verify stability, methods to automatically synthesize code, simulation and test methods, and tools to provide on-line software assurance.

Jacklin, Stephen↗