Search NASA⌕ Search

SEARCH · Search NASA

Results for “Adaptive learning”

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 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
NASA NTRS1988

Blindness in designing intelligent systems

New investigations of the foundations of artificial intelligence are challenging the hypothesis that problem solving is the cornerstone of intelligence. New distinctions among three domains of concern for humans--description, action, and commitment--have revealed that the design process for programmable machines, such as expert systems, is based on descriptions of actions and induces blindness to nonanalytic action and commitment. Design processes focusing in the domain of description are likely to yield programs like burearcracies: rigid, obtuse, impersonal, and unable to adapt to changing circumstances. Systems that learn from their past actions, and systems that organize information for interpretation by human experts, are more likely to be successful in areas where expert systems have failed.

Denning, Peter J.↗
NASA NTRS1990

The structure and formation of natural categories

Categorization and concept formation are critical activities of intelligence. These processes and the conceptual structures that support them raise important issues at the interface of cognitive psychology and artificial intelligence. The work presumes that advances in these and other areas are best facilitated by research methodologies that reward interdisciplinary interaction. In particular, a computational model is described of concept formation and categorization that exploits a rational analysis of basic level effects by Gluck and Corter. Their work provides a clean prescription of human category preferences that is adapted to the task of concept learning. Also, their analysis was extended to account for typicality and fan effects, and speculate on how the concept formation strategies might be extended to other facets of intelligence, such as problem solving.

Fisher, Douglas↗
NASA NTRS1994

Building on the foundation for an engineering career

A predictable and preventable hurdle stops a majority of young women from entering the scientific and technical fields. This cuts down the individual's career possibilities and cuts in half the pool of potential U.S. engineers later available to industry. The waste of talent does not advance our country's competitive position. The typical American adolescent girl has acquired all the basic mathematical skills needed to pursue science and math, but, from adolescence on, she does not build the foundation of science and math courses that she would need later in life to work in engineering. Several questions are addressed: Why are some young women stopped cold in their mathematical tracks during adolescence? What is the influence of psychology, including discussion of the personality traits quantifiably shared by women in technical fields? and How should the school system adapt to keep their female charges learning math and science?

White, Susan↗
NASA NTRS1995

A fuzzy logic intelligent diagnostic system for spacecraft integrated vehicle health management

Due to the complexity of future space missions and the large amount of data involved, greater autonomy in data processing is demanded for mission operations, training, and vehicle health management. In this paper, we develop a fuzzy logic intelligent diagnostic system to perform data reduction, data analysis, and fault diagnosis for spacecraft vehicle health management applications. The diagnostic system contains a data filter and an inference engine. The data filter is designed to intelligently select only the necessary data for analysis, while the inference engine is designed for failure detection, warning, and decision on corrective actions using fuzzy logic synthesis. Due to its adaptive nature and on-line learning ability, the diagnostic system is capable of dealing with environmental noise, uncertainties, conflict information, and sensor faults.

Wu, G. Gordon↗
NASA NTRS2006

Demonstration of Self-Training Autonomous Neural Networks in Space Vehicle Docking Simulations

Neural Networks have been under examination for decades in many areas of research, with varying degrees of success and acceptance. Key goals of computer learning, rapid problem solution, and automatic adaptation have been elusive at best. This paper summarizes efforts at NASA's Marshall Space Flight Center harnessing such technology to autonomous space vehicle docking for the purpose of evaluating applicability to future missions.

Patrick, M. Clinton↗
NASA NTRS2010

Hybrid Decompositional Verification for Discovering Failures in Adaptive Flight Control Systems

Adaptive flight control systems hold tremendous promise for maintaining the safety of a damaged aircraft and its passengers. However, most currently proposed adaptive control methodologies rely on online learning neural networks (OLNNs), which necessarily have the property that the controller is changing during the flight. These changes tend to be highly nonlinear, and difficult or impossible to analyze using standard techniques. In this paper, we approach the problem with a variant of compositional verification. The overall system is broken into components. Undesirable behavior is fed backwards through the system. Components which can be solved using formal methods techniques explicitly for the ranges of safe and unsafe input bounds are treated as white box components. The remaining black box components are analyzed with heuristic techniques that try to predict a range of component inputs that may lead to unsafe behavior. The composition of these component inputs throughout the system leads to overall system test vectors that may elucidate the undesirable behavior

Thompson, Sarah↗
NASA NTRS1991

Neural Architectures for Control

The cerebellar model articulated controller (CMAC) neural architectures are shown to be viable for the purposes of real-time learning and control. Software tools for the exploration of CMAC performance are developed for three hardware platforms, the MacIntosh, the IBM PC, and the SUN workstation. All algorithm development was done using the C programming language. These software tools were then used to implement an adaptive critic neuro-control design that learns in real-time how to back up a trailer truck. The truck backer-upper experiment is a standard performance measure in the neural network literature, but previously the training of the controllers was done off-line. With the CMAC neural architectures, it was possible to train the neuro-controllers on-line in real-time on a MS-DOS PC 386. CMAC neural architectures are also used in conjunction with a hierarchical planning approach to find collision-free paths over 2-D analog valued obstacle fields. The method constructs a coarse resolution version of the original problem and then finds the corresponding coarse optimal path using multipass dynamic programming. CMAC artificial neural architectures are used to estimate the analog transition costs that dynamic programming requires. The CMAC architectures are trained in real-time for each obstacle field presented. The coarse optimal path is then used as a baseline for the construction of a fine scale optimal path through the original obstacle array. These results are a very good indication of the potential power of the neural architectures in control design. In order to reach as wide an audience as possible, we have run a seminar on neuro-control that has met once per week since 20 May 1991. This seminar has thoroughly discussed the CMAC architecture, relevant portions of classical control, back propagation through time, and adaptive critic designs.

Peterson, James K.↗
NASA NTRS1994

Customizing the JPL Multimission Ground Data System: Lessons learned

The Multimission Ground Data System (MGDS) at NASA's Jet Propulsion Laboratory has brought improvements and new technologies to mission operations. It was designed as a generic data system to meet the needs of multiple missions and avoid re-inventing capabilities for each new mission and thus reduce costs. It is based on adaptable tools that can be customized to support different missions and operations scenarios. The MGDS is based on a distributed client/server architecture, with powerful Unix workstations, incorporating standards and open system architectures. The distributed architecture allows remote operations and user science data exchange, while also providing capabilities for centralized ground system monitor and control. The MGDS has proved its capabilities in supporting multiple large-class missions simultaneously, including the Voyager, Galileo, Magellan, Ulysses, and Mars Observer missions. The Operations Engineering Lab (OEL) at JPL has been leading Customer Adaptation Training (CAT) teams for adapting and customizing MGDS for the various operations and engineering teams. These CAT teams have typically consisted of only a few engineers who are familiar with operations and with the MGDS software and architecture. Our experience has provided a unique opportunity to work directly with the spacecraft and instrument operations teams and understand their requirements and how the MGDS can be adapted and customized to minimize their operations costs. As part of this work, we have developed workstation configurations, automation tools, and integrated user interfaces at minimal cost that have significantly improved productivity. We have also proved that these customized data systems are most successful if they are focused on the people and the tasks they perform and if they are based upon user confidence in the development team resulting from daily interactions. This paper will describe lessons learned in adapting JPL's MGDS to fly the Voyager, Galileo, and Mars Observer missions. We will explain how powerful, existing ground data systems can be adapted and packaged in a cost effective way for operations of small and large planetary missions. We will also describe how the MGDS was adapted to support operations within the Galileo Spacecraft Testbed. The Galileo testbed provided a unique opportunity to adapt MGDS to support command and control operations for a small autonomous operations team of a handful of engineers flying the Galileo Spacecraft flight system model.

Murphy, Susan C.↗
NASA NTRS2000

Adaptivity in Agent-Based Routing for Data Networks

Adaptivity, both of the individual agents and of the interaction structure among the agents, seems indispensable for scaling up multi-agent systems (MAS s) in noisy environments. One important consideration in designing adaptive agents is choosing their action spaces to be as amenable as possible to machine learning techniques, especially to reinforcement learning (RL) techniques. One important way to have the interaction structure connecting agents itself be adaptive is to have the intentions and/or actions of the agents be in the input spaces of the other agents, much as in Stackelberg games. We consider both kinds of adaptivity in the design of a MAS to control network packet routing. We demonstrate on the OPNET event-driven network simulator the perhaps surprising fact that simply changing the action space of the agents to be better suited to RL can result in very large improvements in their potential performance: at their best settings, our learning-amenable router agents achieve throughputs up to three and one half times better than that of the standard Bellman-Ford routing algorithm, even when the Bellman-Ford protocol traffic is maintained. We then demonstrate that much of that potential improvement can be realized by having the agents learn their settings when the agent interaction structure is itself adaptive.

Wolpert, David H.↗
NASA NTRS1993

Indirect decentralized repetitive control

Learning control refers to controllers that learn to improve their performance at executing a given task, based on experience performing this specific task. In a previous work, the authors presented a theory of indirect decentralized learning control based on use of indirect adaptive control concepts employing simultaneous identification and control. This paper extends these results to apply to the indirect repetitive control problem in which a periodic (i.e., repetitive) command is given to a control system. Decentralized indirect repetitive control algorithms are presented that have guaranteed convergence to zero tracking error under very general conditions. The original motivation of the repetitive control and learning control fields was learning in robots doing repetitive tasks such as on an assembly line. This paper starts with decentralized discrete time systems, and progresses to the robot application, modeling the robot as a time varying linear system in the neighborhood of the desired trajectory. Decentralized repetitive control is natural for this application because the feedback control for link rotations is normally implemented in a decentralized manner, treating each link as if it is independent of the other links.

Lee, Soo Cheol↗
NASA NTRS1992

Perceptual adaptation in the use of night vision goggles

The image intensification (I sup 2) systems studied for this report were the biocular AN/PVS-7(NVG) and the binocular AN/AVS-6(ANVIS). Both are quite impressive for purposes of revealing the structure of the environment in a fairly straightforward way in extremely low-light conditions. But these systems represent an unusual viewing medium. The perceptual information available through I sup 2 systems is different in a variety of ways from the typical input of everyday vision, and extensive training and practice is required for optimal use. Using this sort of system involves a kind of perceptual skill learning, but is may also involve visual adaptations that are not simply an extension of normal vision. For example, the visual noise evident in the goggles in very low-light conditions results in unusual statistical properties in visual input. Because we had recently discovered a strong and enduring aftereffect of perceived texture density which seemed to be sensitive to precisely the sorts of statistical distortions introduced by I sup 2 systems, it occurred to use that visual noise of this sort might be a very adapting stimulus for texture density and produce an aftereffect that extended into normal vision once the goggles were removed. We have not found any experimental evidence that I sup 2 systems produce texture density aftereffects. The nature of the texture density aftereffect is briefly explained, followed by an accounting of our studies of I sup 2 systems and our most recent work on the texture density aftereffect. A test for spatial frequency adaptation after exposure to NVG's is also reported, as is a study of perceived depth from motion (motion parallax) while wearing the biocular goggles. We conclude with a summary of our findings.

Durgin, Frank H.↗
NASA NTRS2006

Biologically Inspired Technology Using Electroactive Polymers (EAP)

Evolution allowed nature to introduce highly effective biological mechanisms that are incredible inspiration for innovation. Humans have always made efforts to imitate nature's inventions and we are increasingly making advances that it becomes significantly easier to imitate, copy, and adapt biological methods, processes and systems. This brought us to the ability to create technology that is far beyond the simple mimicking of nature. Having better tools to understand and to implement nature's principles we are now equipped like never before to be inspired by nature and to employ our tools in far superior ways. Effectively, by bio-inspiration we can have a better view and value of nature capability while studying its models to learn what can be extracted, copied or adapted. Using electroactive polymers (EAP) as artificial muscles is adding an important element to the development of biologically inspired technologies.

robotics↗
NASA NTRS1994

Video Game Adapts To Brain Waves

Electronic training system based on video game developed to help children afflicted with attention-deficit disorder (ADD) learn to prolong their attention spans. Uses combination of electroencephalography (EEG) and adaptive control to encourage attentiveness. Monitors trainee's brain-wave activity: if EEG signal indicates attention is waning, system increases difficulty of game, forcing trainee to devote more attention to it. Game designed to make trainees want to win and, in so doing, learn to pay attention for longer times.

Pope, Alan T.↗
DOE OSTI2024

A core lighting curriculum for university students and lighting professionals

Here, in collaboration with a group of lighting professionals, learning outcomes were defined, prioritized, organized and mapped to a three-course sequence of lighting courses within a Bachelor of Science in Architectural Engineering degree programme. Syllabi and educational exercises were developed to support the learning outcomes – including classroom activities, homework assignments and design projects. The learning exercises balance the technical foundations of applied illuminating engineering with the artistic aspects of applied lighting design and are intended to promote significant and lasting learning by providing students with education that is useful and relevant to current lighting practice. The process for identifying and prioritizing lighting content is described, a process that could be adapted by other lighting educators to other pedagogical contexts. The syllabi and learning exercises are disseminated for reuse or adaptation, or for self-study by independent learners.

42 ENGINEERING↗
NASA NTRS2022

Neurosymbolic Hybrid Approach to Driver Collision Warning

There are two main algorithmic approaches to autonomous driving systems: (1) An end-to-end system in which a single deep neural network learns to map sensory input directly into appropriate warning and driving responses. (2) A mediated hybrid recognition system in which a system is created by combining independent modules that detect each semantic feature. While some researchers believe that deep learning can solve any problem, others believe that a more engineered and symbolic approach is needed to cope with complex environments with less data. Deep learning alone has achieved state-of-the-art results in many areas, from complex gameplay to predicting protein structures. In particular, in image classification and recognition, deep learning models have achieved accuracies as high as humans. But sometimes it can be very difficult to debug if the deep learning model doesn't work. Deep learning models can be vulnerable and are very sensitive to changes in data distribution. Generalization can be problematic. It's usually hard to prove why it works or doesn't. Deep learning models can also be vulnerable to adversarial attacks. Here, we combine deep learning-based object recognition and tracking with an adaptive neurosymbolic network agent, called the Non-Axiomatic Reasoning System (NARS), that can adapt to its environment by building concepts based on perceptual sequences. We achieved an improved intersection-over-union (IOU) object recognition performance of 0.65 in the adaptive retraining model compared to IOU 0.31 in the COCO data pre-trained model. We improved the object detection limits using RADAR sensors in a simulated environment, and demonstrated the weaving car detection capability by combining deep learning-based object detection and tracking with a neurosymbolic model.

Wang, Pei↗
DOE OSTI2026

A transfer learning approach to energy-efficient control of small and medium-sized commercial buildings

Model-free reinforcement learning (RL) provides a data-driven and adaptive approach to optimize building energy use while satisfying occupant comfort. This powerful tool does not need any prior knowledge about the environment and system it is optimizing and can adapt its policy based on the changes in captures. Like any other data-driven tool, it faces high training costs due to the extensive agent-environment interactions required to capture long-term building dynamics and user comfort. Transfer learning, particularly policy distillation, offers a promising way to accelerate training by leveraging pretrained RL agents in different building and system types. Here, this study investigates online student distillation, in which the student model updates its neural network weights using outputs from teacher models. The work introduces a student distillation strategy designed for efficient knowledge transfer, along with a teacher selection method that ensures high-quality guidance. The approach is validated using a highly calibrated whole building energy model for a small/medium commercial building test facility. Results show substantial reductions in training time and data requirements while surpassing the performance of ASHRAE Guideline 36, an advanced rule-based control strategy. The distilled RL model required 45% less data and achieved 20% higher cumulative rewards than a state-of-the-art RL model, with faster convergence and lower energy consumption. These outcomes demonstrate that effective transfer learning enables a scalable and data-efficient energy management solution for commercial buildings.

ASHRAE guideline 36↗
DOE OSTI2025

SIDDA: SInkhorn Dynamic Domain Adaptation for image classification with equivariant neural networks

Modern neural networks (NNs) often do not generalize well in the presence of a ‘covariate shift’; that is, in situations where the training and test data distributions differ, but the conditional distribution of classification labels given the data remains unchanged. In such cases, NN generalization can be reduced to a problem of learning more robust, domain-invariant features. Domain adaptation (DA) methods include a broad range of techniques aimed at achieving this; however, these methods have struggled with the need for extensive hyperparameter tuning, which then incurs significant computational costs. In this work, we introduce SInkhorn Dynamic Domain Adaptation (SIDDA), an out-of-the-box DA training algorithm built upon the Sinkhorn divergence, that can achieve effective domain alignment with minimal hyperparameter tuning and computational overhead. We demonstrate the efficacy of our method on multiple simulated and real datasets of varying complexity, including simple shapes, handwritten digits, real astronomical observations, and remote sensing data. These datasets exhibit covariate shifts due to noise, blurring, differences between telescopes, and variations in imaging wavelengths. SIDDA is compatible with a variety of NN architectures, and it works particularly well in improving classification accuracy and model calibration when paired with symmetry-aware equivariant NNs (ENNs). We find that SIDDA consistently enhances the generalization capabilities of NNs, achieving up to a ${\approx}40\%$ improvement in classification accuracy on unlabeled target data, while also providing a more modest performance gain of $\lesssim 1\%$ on labeled source data. We also study the efficacy of DA on ENNs with respect to the varying group orders of the dihedral group DN, and find that the model performance improves as the degree of equivariance increases. Finally, if SIDDA achieves proper domain alignment, it also enhances model calibration on both source and target data, with the most significant gains in the unlabeled target domain—achieving over an order of magnitude improvement in the expected calibration error and Brier score. SIDDA’s versatility across various NN models and datasets, combined with its automated approach to domain alignment, has the potential to significantly advance multi-dataset studies by enabling the development of highly generalizable models.

79 ASTRONOMY AND ASTROPHYSICS↗