SEARCH · Search NASA

Results for “Reasoning AI”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3

All sources NASA NTRS DOE OSTI

NASA NTRS1986

Applications of artificial intelligence to scientific research

Artificial intelligence (AI) is a growing field which is just beginning to make an impact on disciplines other than computer science. While a number of military and commercial applications were undertaken in recent years, few attempts were made to apply AI techniques to basic scientific research. There is no inherent reason for the discrepancy. The characteristics of the problem, rather than its domain, determines whether or not it is suitable for an AI approach. Expert system, intelligent tutoring systems, and learning programs are examples of theoretical topics which can be applied to certain areas of scientific research. Further research and experimentation should eventurally make it possible for computers to act as intelligent assistants to scientists.

Prince, Mary Ellen↗

DOE OSTI2023

Benchmarking for AI for Science

AI has been instrumental for recent developments in a number of domains of the sciences. With several hundred machine learning (ML) algorithms and models, and numerous AI-specific hardware platforms, a common quest for all scientists working on AI for Science is around the selection of machine learning algorithm(s) to solve their domain-specific scientific problems. A number of different initiatives around AI Benchmarking have been set up and have been useful in understanding the benefits of different ML algorithms for different tasks.However, with the majority of these AI Benchmarking initiatives focusing on the conventional notions of benchmarking, where the focus is purely runtime performance (such as training time or inference time), their suitability for benchmarking different ML algorithms for solving scientific problems has been viewed as a performance problem even though both are hardly the same. To make reasonable, explainable, and justifiable advancements in science using AI, it is critical to focus on the merits of these algorithms in handling different domain science problems. In other words, more emphasis must be given on Benchmarking for AI for Science than AI Benchmarking. The vision of the former is not only to assess the performance of ML algorithms, but also to assess, and understand the benefits and merits of different ML algorithms in handling scientific problems. Benchmarking for AI for Science, instead of pure performance focused AI Benchmarking, has several benefits: (i) it has the potential to offer advances in the sciences, much more rapidly than through pure performance-based AI methods, (ii) it will encourage the community to focus on developing better domain-specific AI techniques, particularly given the provision for being able to benchmark different techniques, and (iii) it will encourage hardware manufacturers to focus on developing science-specific hardware subsystems.

Thiyagalingam, Jeyan↗

NASA NTRS1988

In defense of An inquiry into computer understanding

Cheeseman responds to responses to his previous essay 'An inquiry into computer understanding'. It is concluded that probability (not logic) is the best way to represent and reason about common sense in constructing an AI system.

Cheeseman, Peter↗

NASA NTRS1993

An intelligent planning and scheduling system for the HST servicing missions

A new, intelligent planning and scheduling system has been delivered to NASA-Goddard Space Flight Center (GSFC) to provide support for the up-coming Hubble Space Telescope (HST) Servicing Missions. This new system is the Servicing Mission Planning and Replanning Tool (SM/PART). SM/PART is written in C and runs on a UNlX-based workstation (IBM RS/6000) under Motif. SM/PART effectively automates the complex task of building or rebuilding integrated timelines and command plans which are required by HST Servicing Mission personnel at their consoles during the missions. SM/PART is able to quickly build or rebuild timelines based on information stored in a Knowledge Base (KB) by using an Artificial Intelligence (AI) tool called the Planning And Resource Reasoning (PARR) shell. After a timeline has been built in the batch mode, it can be displayed and edited in an interactive mode with help from the PARR shell. Finally a detailed command plan is generated. The capability to quickly build or rebuild timelines and command plans provides an additional safety factor for the HST, Shuttle and Crew.

Johnson, Jay↗

NASA NTRSDate not supplied

Artificial Neural Networks and AI in high Assurance Applications: Gaps and Techniques

In recent years, capabilties of Deep Neural Networks (DNN) and Artificial Intelligence (AI) systems have grown tremendously. They are now applied in many areas ranging from game playing, social media, science, to robotics, automotive, and aerospace applications.Based upon requirements for safety of DNN and AI in high assurance automotive and aerospace applications, I will discuss the necessity to ensure that AI technqiues for the analysis of Earth observation data and reasoning are working correctly and reliably.In this talk I will present modern techniques for the verification and validation (V&V) of DNN and other AI components as well as approaches for interpretable AI. I will discuss how these techniques can help to ensure quality of the AI results, improve confidence in their application, and facilitate human-AI interaction and collaboration.

Johann Schumann↗

NASA NTRS2026

Hydrology Copilot: A Cloud-Native Ai System for Hydrological Data Analysis

The emergence of AI-driven Earth observation systems promises to broaden access to petabyte-scale geospatial data beyond domain specialists. However, translating this vision into operational scientific infrastructure requires addressing fundamental challenges in data virtualization, code transparency, and domain-specific reasoning. We present Hydrology Copilot, a cloud-native AI framework for natural-language-driven analysis of Earth observation data. To demonstrate operational capabilities at scale, we implement the system using NASA's North American Land Data Assimilation System version 3 (NLDAS-3), which provides surface meteorological forcing and land-surface model output across North and Central America at 1-km resolution, from which drought diagnostics are derived. The system integrates five core contributions: (1) scalable data virtualization using Kerchunk-based cloud optimized access, achieving a 1.5 to 4.6 times improvement in I/O latency across benchmark queries spanning regional single-day extractions (4.6 times speedup) to continental monthly aggregations (1.5 times speedup); (2) transparent code generation through Microsoft Azure AI Foundry agents that expose executable Python workflows for scientific verification; (3) persistent conversational memory enabling multi-turn analytical discourse across sessions; (4) intelligent query validation that enforces dataset boundaries and resolves ambiguous requests before execution; and (5) a multi-agent architecture coordinating query parsing, code generation, and visualization. We evaluate the system through drought-monitoring workflows, demonstrating reliable code generation, accurate results validated against reference computations and the operational U.S. Drought Monitor, and efficient operation across increasingly complex tasks. By bridging natural-language interfaces with rigorous hydrological analysis, Hydrology Copilot advances beyond proof-of-concept demonstrations to provide a deployable framework for operational Earth science applications.

Data virtualization↗

NASA NTRS1988

Implementing embedded artificial intelligence rules within algorithmic programming languages

Most integrations of artificial intelligence (AI) capabilities with non-AI (usually FORTRAN-based) application programs require the latter to execute separately to run as a subprogram or, at best, as a coroutine, of the AI system. In many cases, this organization is unacceptable; instead, the requirement is for an AI facility that runs in embedded mode; i.e., is called as subprogram by the application program. The design and implementation of a Prolog-based AI capability that can be invoked in embedded mode are described. The significance of this system is twofold: Provision of Prolog-based symbol-manipulation and deduction facilities makes a powerful symbolic reasoning mechanism available to applications programs written in non-AI languages. The power of the deductive and non-procedural descriptive capabilities of Prolog, which allow the user to describe the problem to be solved, rather than the solution, is to a large extent vitiated by the absence of the standard control structures provided by other languages. Embedding invocations of Prolog rule bases in programs written in non-AI languages makes it possible to put Prolog calls inside DO loops and similar control constructs. The resulting merger of non-AI and AI languages thus results in a symbiotic system in which the advantages of both programming systems are retained, and their deficiencies largely remedied.

Feyock, Stefan↗

DOE OSTI2026

Hybrid Quantum–Classical Graph Transformers for Efficient Sentiment Analysis

Quantum Machine Learning (QML) offers a promising paradigm that leverages quantum computing principles to develop efficient and expressive models for learning from complex and structured data. Recent advances in natural language processing (NLP) and artificial intelligence (AI) have demonstrated capabilities in understanding, generating, and reasoning over linguistic and multimodal information. In this work, we present the Quantum Graph Transformer (QGT), a hybrid quantum–classical architecture that extends graph transformer capabilities through quantum self-attention. The QGT models variable-length sentences as token graphs, where both the embedding encoding and the self-attention mechanisms are implemented using parameterized quantum circuits (PQCs), enabling efficient contextual learning with significantly fewer trainable parameters. We train QGT using both fully connected and 𝑘 -nearest-neighbor graph structures and evaluate it on five benchmark sentiment-classification datasets. Experimental results show that QGT consistently achieves higher or comparable accuracy to existing quantum NLP models and outperforms a Classical Graph Transformer (CGT) baseline with identical architecture, achieving 29.4 × fewer parameters while requiring 3–5 × fewer samples to reach comparable performance. These findings highlight the potential of graph-based quantum models as scalable and data-efficient architectures for natural language understanding.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRS1985

Analysis of AIS Data of the Recluse Oil Field, Recluse, Wyoming

Airborne Imaging Spectrometer (AIS) data were flown over the Recluse, Wyoming oil field on September 9, 1984. Processing software was developed at Earth Satellite Corporation (EarthSat) for interactive analysis of the AIS data. EarthSat's AIS processing capabilities include destriping, solar irradiance corrections, residual calculations, geometric resampling, equal energy normalization, interactive spectral classifications and a variety of compressive algorithms to reduce the data to 8-bit format with a minimum of information loss. The in-house photolab facilities of EarthSat can routinely produce high-quality color renditions of the enhanced AIS data. A total of 80 lithologic samples were collected under the AIS flight lines. Correlation (within the atmospheric windows) between the laboratory and the AIS spectra of sample sites was generally poor. Reasonable correlation was only possible in large, freshly plowed fields. Mixed pixels and contrast between the natural and sample's surfaces were believed responsible for the poor correlation. Finally, a drift of approximately three channels was observed in the diffraction grating position within the 1.8 to 2.1 micron quadrant.

Dykstra, J. D.↗

NASA NTRS1992

The achievement of spacecraft autonomy through the thematic application of multiple cooperating intelligent agents

A description is given of UNICORN, a prototype system developed for the purpose of investigating artificial intelligence (AI) concepts supporting spacecraft autonomy. UNICORN employs thematic reasoning, of the type first described by Rodger Schank of Northwestern University, to allow the context-sensitive control of multiple intelligent agents within a blackboard based environment. In its domain of application, UNICORN demonstrates the ability to reason teleologically with focused knowledge. Also presented are some of the lessons learned as a result of this effort. These lessons apply to any effort wherein system level autonomy is the objective.

Rossomando, Philip J.↗

NASA NTRS1989

SHARP: A multi-mission AI system for spacecraft telemetry monitoring and diagnosis

The Spacecraft Health Automated Reasoning Prototype (SHARP) is a system designed to demonstrate automated health and status analysis for multi-mission spacecraft and ground data systems operations. Telecommunications link analysis of the Voyager II spacecraft is the initial focus for the SHARP system demonstration which will occur during Voyager's encounter with the planet Neptune in August, 1989, in parallel with real-time Voyager operations. The SHARP system combines conventional computer science methodologies with artificial intelligence techniques to produce an effective method for detecting and analyzing potential spacecraft and ground systems problems. The system performs real-time analysis of spacecraft and other related telemetry, and is also capable of examining data in historical context. A brief introduction is given to the spacecraft and ground systems monitoring process at the Jet Propulsion Laboratory. The current method of operation for monitoring the Voyager Telecommunications subsystem is described, and the difficulties associated with the existing technology are highlighted. The approach taken in the SHARP system to overcome the current limitations is also described, as well as both the conventional and artificial intelligence solutions developed in SHARP.

Lawson, Denise L.↗

NASA NTRS1989

SHARP - A multi-mission AI system for spacecraft telemetry monitoring and diagnosis

Lawson, Denise L.↗

DOE OSTI2026

NuGraph2 with explainability: post-hoc explanations for geometric neural network predictions

With the growing popularity of artificial intelligence (AI) used for scientific applications, the ability of attribute a result to a reasoning process from the network is in high demand for robust scientific generalizations to hold. In this work we aim to motivate the need for and demonstrate the use of post-hoc explainability methods when applied to AI methods used in scientific applications. To this end, we introduce explainability add-ons to the existing graph neural network (GNN) for neutrino tagging, NuGraph2. The explanations take the form of a suite of techniques examining the output of the network (node classifications) and the edge connections between them, and probing of the latent space using novel general-purpose tools applied to this network. We show how none of these methods are singularly sufficient to show network ‘understanding’, but together can give insights into the processes used in classification. While these methods are tested on the NuGraph2 application, they can be applied to a broad range of networks, not limited to GNNs. The code for this work is publicly available on GitHub at https://github.com/voetberg/XNuGraph.

Voetberg, Margaret [Fermilab] (ORCID:0009000527154↗

NASA NTRS1993

An integrated information retrieval and document management system

This paper describes the requirements and prototype development for an intelligent document management and information retrieval system that will be capable of handling millions of pages of text or other data. Technologies for scanning, Optical Character Recognition (OCR), magneto-optical storage, and multiplatform retrieval using a Standard Query Language (SQL) will be discussed. The semantic ambiguity inherent in the English language is somewhat compensated-for through the use of coefficients or weighting factors for partial synonyms. Such coefficients are used both for defining structured query trees for routine queries and for establishing long-term interest profiles that can be used on a regular basis to alert individual users to the presence of relevant documents that may have just arrived from an external source, such as a news wire service. Although this attempt at evidential reasoning is limited in comparison with the latest developments in AI Expert Systems technology, it has the advantage of being commercially available.

Coles, L. Stephen↗

NASA NTRS1986

Artificial intelligence and the space station software support environment

In a software system the size of the Space Station Software Support Environment (SSE), no one software development or implementation methodology is presently powerful enough to provide safe, reliable, maintainable, cost effective real time or near real time software. In an environment that must survive one of the most harsh and long life times, software must be produced that will perform as predicted, from the first time it is executed to the last. Many of the software challenges that will be faced will require strategies borrowed from Artificial Intelligence (AI). AI is the only development area mentioned as an example of a legitimate reason for a waiver from the overall requirement to use the Ada programming language for software development. The limits are defined of the applicability of the Ada language Ada Programming Support Environment (of which the SSE is a special case), and software engineering to AI solutions by describing a scenario that involves many facets of AI methodologies.

Marlowe, Gilbert↗

DOE OSTI2024

Evaluating the Effectiveness of Retrieval-Augmented Large Language Models in Scientific Document Reasoning

Despite the dramatic progress in Large Language Model (LLM) development, LLMs often provide seemingly plausible but not factual information, often referred as hallucinations. Retrieval-augmented LLMs provide a non-parametric approach to solve these issues by retrieving relevant information from external data sources and augment the training process. These models helps to trace evidence from an externally provided knowledge base allowing the model predictions to be better interpreted and verified. In this work, we critically evaluate these models in their ability to perform in scientific document reasoning tasks. To this end, we tuned multiple such model variants with science-focused instructions and evaluated them on a scientific document reasoning benchmark for the usefulness of the retrieved document passages. Our findings suggest that models justify predictions in science tasks with fabricated evidence and leveraging scientific corpus as pretraining data does not alleviate the risk of evidence fabrication.

• Artificial intelligence (AI) / machine learning ↗

NASA NTRS1985

Development of an Expert System for Representing Procedural Knowledge

A high level of automation is of paramount importance in most space operations. It is critical for unmanned missions and greatly increases the effectiveness of manned missions. However, although many functions can be automated by using advanced engineering techniques, others require complex reasoning, sensing, and manipulatory capabilities that go beyond this technology. Automation of fault diagnosis and malfunction handling is a case in point. The military have long been interested in this problem, and have developed automatic test equipment to aid in the maintenance of complex military hardware. These systems are all based on conventional software and engineering techniques. However, the effectiveness of such test equipment is severely limited. The equipment is inflexible and unresponsive to the skill level of the technicians using it. The diagnostic procedures cannot be matched to the exigencies of the current situation nor can they cope with reconfiguration or modification of the items under test. The diagnosis cannot be guided by useful advice from technicians and, when a fault cannot be isolated, no explanation is given as to the cause of failure. Because these systems perform a prescribed sequence of tests, they cannot utilize knowledge of a particular situation to focus attention on more likely trouble spots. Consequently, real-time performance is highly unsatisfactory. Furthermore, the cost of developing test software is substantial and time to maturation is excessive. Significant advances in artificial intelligence (AI) have recently led to the development of powerful and flexible reasoning systems, known as expert or knowledge-based systems. We have devised a powerful and theoretically sound scheme for representing and reasoning about procedural knowledge.

Georgeff, Michael P.↗

DOE OSTI2025

Foundation Models for Zero-Shot Segmentation of Scientific Images without AI-Ready Data

Zero-shot and prompt-based models have excelled at visual reasoning tasks by leveraging large-scale natural image corpora, but they often fail on sparse and domain-specific scientific image data. We introduce Zenesis, a no-code interactive computer vision platform designed to reduce data readiness bottlenecks in scientific imaging workflows. Zenesis integrates lightweight multimodal adaptation for zero-shot inference on raw scientific data, human-in-the-loop refinement, and heuristic-based temporal enhancement. We validate our approach on Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) datasets of catalyst-loaded membranes. Zenesis outperforms baselines, achieving an average accuracy of 0.947, Intersection over Union (IoU) of 0.858, and Dice score of 0.923 on amorphous catalyst samples; and 0.987 accuracy, 0.857 IoU, and 0.923 Dice on crystalline samples. These results represent a significant performance gain over conventional methods such as Otsu thresholding and standalone models like the Segment Anything Model (SAM). Zenesis enables effective image segmentation in domains where annotated datasets are limited, offering a scalable solution for scientific discovery.

Mukherjee, Shubhabrata↗