Search NASA⌕ Search

SEARCH · Search NASA

Results for “Learning with errors”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
NASA NTRS2016

Human Error and the International Space Station: Challenges and Triumphs in Science Operations

Any system with a human component is inherently risky. Studies in human factors and psychology have repeatedly shown that human operators will inevitably make errors, regardless of how well they are trained. Onboard the International Space Station (ISS) where crew time is arguably the most valuable resource, errors by the crew or ground operators can be costly to critical science objectives. Operations experts at the ISS Payload Operations Integration Center (POIC), located at NASA's Marshall Space Flight Center in Huntsville, Alabama, have learned that from payload concept development through execution, there are countless opportunities to introduce errors that can potentially result in costly losses of crew time and science. To effectively address this challenge, we must approach the design, testing, and operation processes with two specific goals in mind. First, a systematic approach to error and human centered design methodology should be implemented to minimize opportunities for user error. Second, we must assume that human errors will be made and enable rapid identification and recoverability when they occur. While a systematic approach and human centered development process can go a long way toward eliminating error, the complete exclusion of operator error is not a reasonable expectation. The ISS environment in particular poses challenging conditions, especially for flight controllers and astronauts. Operating a scientific laboratory 250 miles above the Earth is a complicated and dangerous task with high stakes and a steep learning curve. While human error is a reality that may never be fully eliminated, smart implementation of carefully chosen tools and techniques can go a long way toward minimizing risk and increasing the efficiency of NASA's space science operations.

Harris, Samantha S.↗
NASA NTRS1992

Discrete time learning control in nonlinear systems

In this paper digital learning control methods are developed primarily for use in single-input, single-output nonlinear dynamic systems. Conditions for convergence of the basic form of learning control based on integral control concepts are given, and shown to be satisfied by a large class of nonlinear problems. It is shown that it is not the gross nonlinearities of the differential equations that matter in the convergence, but rather the much smaller nonlinearities that can manifest themselves during the short time interval of one sample time. New algorithms are developed that eliminate restrictions on the size of the learning gain, and on knowledge of the appropriate sign of the learning gain, for convergence to zero error in tracking a feasible desired output trajectory. It is shown that one of the new algorithms can give guaranteed convergence in the presence of actuator saturation constraints, and indicate when the requested trajectory is beyond the actuator capabilities.

Longman, Richard W.↗
NASA NTRS1991

Search-based model identification of smart-structure damage

This paper describes the use of a combined model and parameter identification approach, based on modal analysis and artificial intelligence (AI) techniques, for identifying damage or flaws in a rotating truss structure incorporating embedded piezoceramic sensors. This smart structure example is representative of a class of structures commonly found in aerospace systems and next generation space structures. Artificial intelligence techniques of classification, heuristic search, and an object-oriented knowledge base are used in an AI-based model identification approach. A finite model space is classified into a search tree, over which a variant of best-first search is used to identify the model whose stored response most closely matches that of the input. Newly-encountered models can be incorporated into the model space. This adaptativeness demonstrates the potential for learning control. Following this output-error model identification, numerical parameter identification is used to further refine the identified model. Given the rotating truss example in this paper, noisy data corresponding to various damage configurations are input to both this approach and a conventional parameter identification method. The combination of the AI-based model identification with parameter identification is shown to lead to smaller parameter corrections than required by the use of parameter identification alone.

Glass, B. J.↗
NASA NTRS1987

Learning control for robotic manipulators with sparse data

Learning control algorithms have been proposed for error compensation in repetitive robotic manipulator tasks. It is shown that the performance of such control algorithms can be seriously degraded when the feedback data they use is relatively sparse in time, such as might be provided by vision systems. It is also shown that learning control algorithms can be modified to compensate for the effects of sparse data and thereby yield performance which approaches that of systems without limitations on the sensory information available for control.

Morita, Atsushi↗
NASA NTRS1998

Preliminary Work for Examining the Scalability of Reinforcement Learning

Researchers began studying automated agents that learn to perform multiple-step tasks early in the history of artificial intelligence (Samuel, 1963; Samuel, 1967; Waterman, 1970; Fikes, Hart & Nilsonn, 1972). Multiple-step tasks are tasks that can only be solved via a sequence of decisions, such as control problems, robotics problems, classic problem-solving, and game-playing. The objective of agents attempting to learn such tasks is to use the resources they have available in order to become more proficient at the tasks. In particular, each agent attempts to develop a good policy, a mapping from states to actions, that allows it to select actions that optimize a measure of its performance on the task; for example, reducing the number of steps necessary to complete the task successfully. Our study focuses on reinforcement learning, a set of learning techniques where the learner performs trial-and-error experiments in the task and adapts its policy based on the outcome of those experiments. Much of the work in reinforcement learning has focused on a particular, simple representation, where every problem state is represented explicitly in a table, and associated with each state are the actions that can be chosen in that state. A major advantage of this table lookup representation is that one can prove that certain reinforcement learning techniques will develop an optimal policy for the current task. The drawback is that the representation limits the application of reinforcement learning to multiple-step tasks with relatively small state-spaces. There has been a little theoretical work that proves that convergence to optimal solutions can be obtained when using generalization structures, but the structures are quite simple. The theory says little about complex structures, such as multi-layer, feedforward artificial neural networks (Rumelhart & McClelland, 1986), but empirical results indicate that the use of reinforcement learning with such structures is promising. These empirical results make no theoretical claims, nor compare the policies produced to optimal policies. A goal of our work is to be able to make the comparison between an optimal policy and one stored in an artificial neural network. A difficulty of performing such a study is finding a multiple-step task that is small enough that one can find an optimal policy using table lookup, yet large enough that, for practical purposes, an artificial neural network is really required. We have identified a limited form of the game OTHELLO as satisfying these requirements. The work we report here is in the very preliminary stages of research, but this paper provides background for the problem being studied and a description of our initial approach to examining the problem. In the remainder of this paper, we first describe reinforcement learning in more detail. Next, we present the game OTHELLO. Finally we argue that a restricted form of the game meets the requirements of our study, and describe our preliminary approach to finding an optimal solution to the problem.

Clouse, Jeff↗
NASA NTRSDate not supplied

Avoiding Selection Bias in Generating Examples of Plans in the Presence of Heuristic Error

It is generally understood that heuristic error hurts the performance of search algorithms, measured in terms of search effort. Hence there is an interest in understanding how to reduce heuristic error. One way to do this is to learn a heuristic from a set of examples of plans generated offline, e.g. bootstrapping methods. In this paper, we consider how some methods for generating examples of plans may skew the training set in the presence of heuristic errors. Initial theoretical results show that duplicate detection is one source of selection bias in the canonical A* algorithm. We introduce a duplicate selection scheme for A* that avoids selection bias in generating cost-optimal examples, without compromising memory efficiency, and develop ideas in the satisficing setting. We evaluate our approach on n x m grids with multiple cost-optimal solutions and synthetic heuristic error. Finally, we attempt to extend these ideas to the problem of generating extreme examples of plans.

Alison S Paredes↗
NASA NTRS1992

Multilayer perceptron, fuzzy sets, and classification

A fuzzy neural network model based on the multilayer perceptron, using the back-propagation algorithm, and capable of fuzzy classification of patterns is described. The input vector consists of membership values to linguistic properties while the output vector is defined in terms of fuzzy class membership values. This allows efficient modeling of fuzzy or uncertain patterns with appropriate weights being assigned to the backpropagated errors depending upon the membership values at the corresponding outputs. During training, the learning rate is gradually decreased in discrete steps until the network converges to a minimum error solution. The effectiveness of the algorithm is demonstrated on a speech recognition problem. The results are compared with those of the conventional MLP, the Bayes classifier, and the other related models.

Pal, Sankar K.↗
NASA NTRS1988

Pitch-Learning Algorithm For Speech Encoders

Adaptive algorithm detects and corrects errors in sequence of estimates of pitch period of speech. Algorithm operates in conjunction with techniques used to estimate pitch period. Used in such parametric and hybrid speech coders as linear predictive coders and adaptive predictive coders.

Bhaskar, B. R. Udaya↗
NASA NTRS1993

A Software Package for Neural Network Applications Development

Original Backprop (Version 1.2) is an MS-DOS package of four stand-alone C-language programs that enable users to develop neural network solutions to a variety of practical problems. Original Backprop generates three-layer, feed-forward (series-coupled) networks which map fixed-length input vectors into fixed length output vectors through an intermediate (hidden) layer of binary threshold units. Version 1.2 can handle up to 200 input vectors at a time, each having up to 128 real-valued components. The first subprogram, TSET, appends a number (up to 16) of classification bits to each input, thus creating a training set of input output pairs. The second subprogram, BACKPROP, creates a trilayer network to do the prescribed mapping and modifies the weights of its connections incrementally until the training set is leaned. The learning algorithm is the 'back-propagating error correction procedures first described by F. Rosenblatt in 1961. The third subprogram, VIEWNET, lets the trained network be examined, tested, and 'pruned' (by the deletion of unnecessary hidden units). The fourth subprogram, DONET, makes a TSR routine by which the finished product of the neural net design-and-training exercise can be consulted under other MS-DOS applications.

Baran, Robert H.↗
NASA NTRS1993

Recent SEL experiments and studies

The studies discussed in this paper are all examples of activities that are performed as part of the Software Engineering Laboratory's (SEL's) process improvement model. Using this model, the SEL starts by understanding the product and process, then assesses the impact of new technologies, and finally packages what was learned. The preliminary examination of maintenance effort, error, and change profiles to establish a maintenance baseline exemplifies understanding-phase activities. The ongoing testing study that is examining the effects of various testing approaches on process and product measures is an example of typical assessing efforts. Finally, the derivation of cost and schedule estimation models from locally driven factors such as reuse level, application type, and language is an example of experience packaging. In the SEL, no study is ever really completed. Studies will be repeated and iterated upon in the future as part of the ongoing software improvement process.

Pajerski, Rose↗
NASA NTRS2013

Scatter-Reducing Sounding Filtration Using a Genetic Algorithm and Mean Monthly Standard Deviation

Retrieval algorithms like that used by the Orbiting Carbon Observatory (OCO)-2 mission generate massive quantities of data of varying quality and reliability. A computationally efficient, simple method of labeling problematic datapoints or predicting soundings that will fail is required for basic operation, given that only 6% of the retrieved data may be operationally processed. This method automatically obtains a filter designed to reduce scatter based on a small number of input features. Most machine-learning filter construction algorithms attempt to predict error in the CO2 value. By using a surrogate goal of Mean Monthly STDEV, the goal is to reduce the retrieved CO2 scatter rather than solving the harder problem of reducing CO2 error. This lends itself to improved interpretability and performance. This software reduces the scatter of retrieved CO2 values globally based on a minimum number of input features. It can be used as a prefilter to reduce the number of soundings requested, or as a post-filter to label data quality. The use of the MMS (Mean Monthly Standard deviation) provides a much cleaner, clearer filter than the standard ABS(CO2-truth) metrics previously employed by competitor methods. The software's main strength lies in a clearer (i.e., fewer features required) filter that more efficiently reduces scatter in retrieved CO2 rather than focusing on the more complex (and easily removed) bias issues.

Mandrake, Lukas↗
NASA NTRS2014

Using Historical Data to Automatically Identify Air-Traffic Control Behavior

This project seeks to develop statistical-based machine learning models to characterize the types of errors present when using current systems to predict future aircraft states. These models will be data-driven - based on large quantities of historical data. Once these models are developed, they will be used to infer situations in the historical data where an air-traffic controller intervened on an aircraft's route, even when there is no direct recording of this action.

trajectory generation↗
NASA NTRS2019

Surrogate Modeling and Optimization of a Combustor with an Interdigitated Flushwall Injector

Design and Analysis of Computer Experiments (DACE) methods are applied and used to perform surrogate modeling and optimization of a simplified combustor flowpath with an interdigitated flushwall injector. The objectives of the optimization are the thrust potential and combustion efficiency, which are evaluated across a range of flight Mach numbers, duct heights, spanwise spacings, and injection angles. The focus of this work is to highlight the application of a sequential learning approach, in order to learn about the responses of the objective functions over the design space and to identify local regions of interest for further analysis. This approach is contrasted to a previous effort where only a single sampling set was used to fit surrogate models and perform optimization. The optimal solutions resulting from the previous and present approaches are different, due to surrogate model-guided local refinement of the design space allowed by the sequential learning method. The values of the global error estimates between the previous and present approaches are comparable, but the sequential method proved more computationally cost-effective. Further refinement in the optimal regions might be needed to improve predictive capability of surrogate models and to obtain the optimal solutions sets.

Rajiv R Shenoy↗
NASA NTRS1986

Active controllers and the time duration to learn a task

An active controller was used to help train naive subjects involved in a compensatory tracking task. The controller is called active in this context because it moves the subject's hand in a direction to improve tracking. It is of interest here to question whether the active controller helps the subject to learn a task more rapidly than the passive controller. Six subjects, inexperienced to compensatory tracking, were run to asymptote root mean square error tracking levels with an active controller or a passive controller. The time required to learn the task was defined several different ways. The results of the different measures of learning were examined across pools of subjects and across controllers using statistical tests. The comparison between the active controller and the passive controller as to their ability to accelerate the learning process as well as reduce levels of asymptotic tracking error is reported here.

Repperger, D. W.↗