Search NASA⌕ Search

SEARCH · Search NASA

Results for “multitasking”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Seismicity-constrained fault detection and characterization with a multitask machine learning model

Geological fault detection and characterization are crucial for understanding subsurface dynamics across scales. While methods for fault delineation based on either seismicity location analysis or seismic image reflector discontinuity are well-established, a systematic approach that integrates both data types remains absent. We develop a novel machine learning model that unifies seismic reflector images and seismicity location information to automatically identify geological faults and characterize their geometrical properties. The model encodes a seismic image and a seismicity location image separately, and fuses the encoded features with a spatial-channel attention fusion module to improve the learning of important features in both inputs. We design an automated strategy to generate high-quality synthetic training data and labels. To improve the realism of the seismicity location image, we include random seismicity noise and missing seismicity location associated with some of the faults. We validate the model’s efficacy and accuracy using synthetic data examples and two field data examples. Moreover, we show that fine-tuning the trained model with a small, domain-specific dataset enhances its fidelity for field data applications. The results demonstrate that integrating seismicity location and seismic images into a unified framework allows the end-to-end neural network to achieve higher fidelity and accuracy in delineating subsurface faults and their geometrical properties compared with image-only fault detection methods. Our approach offers an adaptive data-driven tool for geological fault characterization and seismic hazard mitigation, bridging the gap between seismicity location and image-based fault detection methods.

58 GEOSCIENCES↗
DOE OSTI2024

Deep Multitask Learning Models for Radiation Estimation at High Energy Accelerator Facility

Controlling the dose of radiation exposure in potential radioactive facilities is critical for ensuring the safety of staff and the public. Here, in this paper, we developed machine learning models to estimate radiation exposure efficiently at the Thomas Jefferson National Accelerator Facility (JLab), aiming to enhance safety in both accelerator facilities and public areas. Multiple sensors were deployed around the three experimental halls at JLab. Data on single-beam currents, energy levels, and radiation values at the sensor locations were collected during accelerator operation. We proposed a multi-task learning model for radiation estimation, utilizing either one-dimensional convolutional neural networks (1-D CNNs) or long short-term memory networks (LSTMs) as the backbone. The proposed model was trained to simultaneously estimate radiation levels at the sensor locations. Experimental results demonstrated that the proposed model with LSTM backbone achieved the best estimation performance, with an average R 2 score of 0.7557 for estimation within the same year and 0.7157 for estimation across different years. These results significantly surpassed those of competing models.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2019

Accurate and transferable multitask prediction of chemical properties with an atoms-in-molecules neural network

Atomic and molecular properties could be evaluated from the fundamental Schrodinger’s equation and therefore represent different modalities of the same quantum phenomena. Here, we present AIMNet, a modular and chemically inspired deep neural network potential. We used AIMNet with multitarget training to learn multiple modalities of the state of the atom in a molecular system. The resulting model shows on several benchmark datasets state-of-the-art accuracy, comparable to the results of orders of magnitude more expensive DFT methods. It can simultaneously predict several atomic and molecular properties without an increase in the computational cost. With AIMNet, we show a new dimension of transferability: the ability to learn new targets using multimodal information from previous training. The model can learn implicit solvation energy (SMD method) using only a fraction of the original training data and an archive median absolute deviation error of 1.1 kcal/mol compared to experimental solvation free energies in the MNSol database.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Replication Data for: Multitask Machine Learning of Collective Variables for Enhanced Sampling of Rare Events

The data underlying this published work have been made publicly available in this repository as part of the IMASC Data Management Plan. This work was supported as part of the Integrated Mesoscale Architectures for Sustainable Catalysis (IMASC), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0012573.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS1980

Dynamic decisions and work load in multitask supervisory control

A paradigm is developed for the problem of allocating in time a single resource to multiple simultaneous task demands which appear randomly, last for various periods, and offer varying rewards for service. Based upon a dynamic optimizing algorithm plus an estimator, and including response time and future discounting constraints, a model of the human decisionmaker is compared to experimental results for human subjects performing such a task at a computer-graphics terminal. Results indicate a reasonable fit, under various model parameters and task conditions, and suggest interesting hypotheses about the nature of human 'planning ahead' and mental work load.

Tulga, M. K.↗
NASA NTRS1988

POPEYE: A production rule-based model of multitask supervisory control (POPCORN)

Recent studies of relationships between subjective ratings of mental workload, performance, and human operator and task characteristics have indicated that these relationships are quite complex. In order to study the various relationships and place subjective mental workload within a theoretical framework, we developed a production system model for the performance component of the complex supervisory task called POPCORN. The production system model is represented by a hierarchial structure of goals and subgoals, and the information flow is controlled by a set of condition-action rules. The implementation of this production system, called POPEYE, generates computer simulated data under different task difficulty conditions which are comparable to those of human operators performing the task. This model is the performance aspect of an overall dynamic psychological model which we are developing to examine and quantify relationships between performance and psychological aspects in a complex environment.

Townsend, James T.↗
NASA NTRS1987

Efficient Ada multitasking on a RISC register window architecture

This work addresses the problem of reducing context switch overhead on a processor which supports a large register file - a register file much like that which is part of the Berkeley RISC processors and several other emerging architectures (which are not necessarily reduced instruction set machines in the purest sense). Such a reduction in overhead is particularly desirable in a real-time embedded application, in which task-to-task context switch overhead may result in failure to meet crucial deadlines. A storage management technique by which a context switch may be implemented as cheaply as a procedure call is presented. The essence of this technique is the avoidance of the save/restore of registers on the context switch. This is achieved through analysis of the static source text of an Ada tasking program. Information gained during that analysis directs the optimized storage management strategy for that program at run time. A formal verification of the technique in terms of an operational control model and an evaluation of the technique's performance via simulations driven by synthetic Ada program traces are presented.

Kearns, J. P.↗
NASA NTRS1991

Efficient multitasking of Choleski matrix factorization on CRAY supercomputers

A Choleski method is described and used to solve linear systems of equations that arise in large scale structural analysis. The method uses a novel variable-band storage scheme and is structured to exploit fast local memory caches while minimizing data access delays between main memory and vector registers. Several parallel implementations of this method are described for the CRAY-2 and CRAY Y-MP computers demonstrating the use of microtasking and autotasking directives. A portable parallel language, FORCE, is used for comparison with the microtasked and autotasked implementations. Results are presented comparing the matrix factorization times for three representative structural analysis problems from runs made in both dedicated and multi-user modes on both computers. CPU and wall clock timings are given for the parallel implementations and are compared to single processor timings of the same algorithm.

Overman, Andrea L.↗
NASA NTRS1993

Multiple-User, Multitasking, Virtual-Memory Computer System

Computer system designed and programmed to serve multiple users in research laboratory. Provides for computer control and monitoring of laboratory instruments, acquisition and anlaysis of data from those instruments, and interaction with users via remote terminals. System provides fast access to shared central processing units and associated large (from megabytes to gigabytes) memories. Underlying concept of system also applicable to monitoring and control of industrial processes.

Generazio, Edward R.↗
NASA NTRS1994

A multitasking behavioral control system for the Robotic All Terrain Lunar Exploration Rover (RATLER)

An alternative methodology for designing an autonomous navigation and control system is discussed. This generalized hybrid system is based on a less sequential and less anthropomorphic approach than that used in the more traditional artificial intelligence (AI) technique. The architecture is designed to allow both synchronous and asynchronous operations between various behavior modules. This is accomplished by intertask communications channels which implement each behavior module and each interconnection node as a stand-alone task. The proposed design architecture allows for construction of hybrid systems which employ both subsumption and traditional AI techniques as well as providing for a teleoperator's interface. Implementation of the architecture is planned for the prototype Robotic All Terrain Lunar Explorer Rover (RATLER) which is described briefly.

Klarer, P.↗
NASA NTRS2011

Health Monitor for Multitasking, Safety-Critical, Real-Time Software

Health Manager can detect Bad Health prior to a failure occurring by periodically monitoring the application software by looking for code corruption errors, and sanity-checking each critical data value prior to use. A processor s memory can fail and corrupt the software, or the software can accidentally write to the wrong address and overwrite the executing software. This innovation will continuously calculate a checksum of the software load to detect corrupted code. This will allow a system to detect a failure before it happens. This innovation monitors each software task (thread) so that if any task reports "bad health," or does not report to the Health Manager, the system is declared bad. The Health Manager reports overall system health to the outside world by outputting a square wave signal. If the square wave stops, this indicates that system health is bad or hung and cannot report. Either way, "bad health" can be detected, whether caused by an error, corrupted data, or a hung processor. A separate Health Monitor Task is started and run periodically in a loop that starts and stops pending on a semaphore. Each monitored task registers with the Health Manager, which maintains a count for the task. The registering task must indicate if it will run more or less often than the Health Manager. If the task runs more often than the Health Manager, the monitored task calls a health function that increments the count and verifies it did not go over max-count. When the periodic Health Manager runs, it verifies that the count did not go over the max-count and zeroes it. If the task runs less often than the Health Manager, the periodic Health Manager will increment the count. The monitored task zeroes the count, and both the Health Manager and monitored task verify that the count did not go over the max-count.

Zoerner, Roger↗
NASA NTRS2011

Locomotor Adaptation Improves Balance Control, Multitasking Ability and Reduces the Metabolic Cost of Postural Instability

During exploration-class missions, sensorimotor disturbances may lead to disruption in the ability to ambulate and perform functional tasks during the initial introduction to a novel gravitational environment following a landing on a planetary surface. The overall goal of our current project is to develop a sensorimotor adaptability training program to facilitate rapid adaptation to these environments. We have developed a unique training system comprised of a treadmill placed on a motion-base facing a virtual visual scene. It provides an unstable walking surface combined with incongruent visual flow designed to enhance sensorimotor adaptability. Greater metabolic cost incurred during balance instability means more physical work is required during adaptation to new environments possibly affecting crewmembers? ability to perform mission critical tasks during early surface operations on planetary expeditions. The goal of this study was to characterize adaptation to a discordant sensory challenge across a number of performance modalities including locomotor stability, multi-tasking ability and metabolic cost. METHODS: Subjects (n=15) walked (4.0 km/h) on a treadmill for an 8 -minute baseline walking period followed by 20-minutes of walking (4.0 km/h) with support surface motion (0.3 Hz, sinusoidal lateral motion, peak amplitude 25.4 cm) provided by the treadmill/motion-base system. Stride frequency and auditory reaction time were collected as measures of locomotor stability and multi-tasking ability, respectively. Metabolic data (VO2) were collected via a portable metabolic gas analysis system. RESULTS: At the onset of lateral support surface motion, subj ects walking on our treadmill showed an increase in stride frequency and auditory reaction time indicating initial balance and multi-tasking disturbances. During the 20-minute adaptation period, balance control and multi-tasking performance improved. Similarly, throughout the 20-minute adaptation period, VO2 gradually decreased following an initial increase after the onset of support surface motion. DISCUSSION: Resu lts confirmed that walking in discordant conditions not only compromises locomotor stability and the ability to multi-task, but comes at a quantifiable metabolic cost. Importantly, like locomotor stability and multi-tasking ability, metabolic expenditure while walking in discordant sensory conditions improved during adaptation. This confirms that sensorimotor adaptability training can benefit multiple performance parameters central to the successful completion of critical mission tasks.

Bloomberg, J. J.↗