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.

SourceResearch connectionFrom yearThrough year

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

At least 73 records · Page 4

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Investigating the influence of latent lifestyles on productive travels: Insights into designing autonomous transit system

As a special case of multitasking, travel-based multitasking typically refers to conducting a set of in-vehicle activities while traveling. Travel-based multitasking has an indisputable influence on offering a pleasant travel experience to transit users during their rides, given that they can use their travel time to perform desirable activities and gain benefits in various form. For instance, the in-activities could help the rider free up time from his/her schedule for the day (i.e., a worthwhile use of travel time). In this study, we investigate how the worthwhileness of a travel-based multitasking could be under the influence of: (1) the transit user’s lifestyle, and (2) socio-demographics, and (3) the characteristics of the transit trip. Towards this, we conducted an intercept survey focusing on the transit trips in the Chicago metropolitan area and analyzed it using latent class modeling approach. Per the results, two classes of transit users could be identified: (1) worthwhileness seekers, productively travelers and (2) leisure seekers, occasional worthwhile travelers. The results also suggest travel time, waiting time and walking distance to the transit station, and the set of in-vehicle activities as significant predictors of worthwhile use of travel time. Finally, the findings provide insights to policymakers for improving public transit systems in the current form, as well as designing an autonomous mobility system as the future form of public transit.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

NASA NTRS1986

Comparing host and target environments for distributed Ada programs

The Ada programming language provides a means of specifying logical concurrency by using multitasking. Extending the Ada multitasking concurrency mechanism into a physically concurrent distributed environment which imposes its own requirements can lead to incompatibilities. These problems are discussed. Using distributed Ada for a target system may be appropriate, but when using the Ada language in a host environment, a multiprocessing model may be more suitable than retargeting an Ada compiler for the distributed environment. The tradeoffs between multitasking on distributed targets and multiprocessing on distributed hosts are discussed. Comparisons of the multitasking and multiprocessing models indicate different areas of application.

Paulk, Mark C.↗

DOE OSTI2023

Do people spend travel time the way they think they would? a comparative study of generic and trip-specific travel time allocation using hybrid multiple discrete continuous (MDC) framework

Unlike driving alone, public transportation allows one to engage in extraneous activities while traveling – often referred to as travel-based multitasking. Research involving travel-based multitasking often relies on either generic (i.e. not related to an actual trip) or revealed (i.e. related to an executed trip) data. The data collected for this study provided a unique opportunity to compare the generic and trip-specific preferences while traveling on public transportation. To this end, the study develops two integrated choice and latent variable models with multiple discrete-continuous (MDC) kernels to simultaneously model the activity selection and the time allocation for generic and trip-specific data, respectively. According to the results the trip-specific data could identify more nuances in the travel-based multitasking behavior than the generic data. Finally, the heterogeneity identified by the developed models will be helpful for the transit operators in providing appropriate facilities (e.g. internet access, reading lights) along various transit corridors.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2020

Fast and stable deep-learning predictions of material properties for solid solution alloys

We present a novel deep learning (DL) approach to produce highly accurate predictions of macroscopic physical properties of solid solution binary alloys and magnetic systems. The major idea is to make use of the correlations between different physical properties in alloy systems to improve the prediction accuracy of neural network (NN) models. We use multitasking NN models to simultaneously predict the total energy, charge density and magnetic moment. These physical properties mutually serve as constraints during the training of the multitasking NN, resulting in more reliable DL models because multiple physics properties are correctly learned by a single model. Two binary alloys, copper–gold (CuAu) and iron–platinum (FePt), were studied. Our results show that once the multitasking NN's are trained, they can estimate the material properties for a specific configuration hundreds of times faster than first-principles density functional theory calculations while retaining comparable accuracy. We used a simple measure based on the root-mean-squared errors to quantify the quality of the NN models, and found that the inclusion of charge density and magnetic moment as physical constraints leads to more stable models that exhibit improved accuracy and reduced uncertainty for the energy predictions.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

NASA NTRS2016

Why Do People Make Mistakes?

Multitasking is endemic in modern life and work: drivers talk on cell phones, office workers type while answering phone calls, students do homework while text messaging, nurses prepare injections while responding to doctors calls, and air traffic controllers direct aircraft in one sector while handling additional traffic in another. Whether in daily life or at work, we are constantly bombarded with multiple, concurrent interruptions and demands and we have all somehow come to believe in the myth that we can, and in fact are expected to, easily address them all - without any repercussions. However, accumulating scientific evidence is now suggesting that multitasking increases the probability of human error. This talk presents a set of NASA studies that characterize concurrent demands in one work domain, routine airline cockpit operations, in order to illustrate the ways operational task demands together with the proclivity to manage them all concurrently make human performance in this and in any domain vulnerable to potentially serious errors and to accidents.

multitasking↗

DOE OSTI2020

Optimal carbon storage reservoir management through deep reinforcement learning

Model-based optimization plays a central role in energy system design and management. The complexity and high-dimensionality of many process-level models, especially those used for geosystem energy exploration and utilization, often lead to formidable computational costs when the dimension of decision space is also large. This work adopts elements of recently advanced deep learning techniques to solve a sequential decision-making problem in applied geosystem management. Specifically, a deep reinforcement learning framework was formed for optimal multiperiod planning, in which a deep Q-learning network (DQN) agent was trained to maximize rewards by learning from high-dimensional inputs and from exploitation of its past experiences. To expedite computation, deep multitask learning was used to approximate high-dimensional, multistate transition functions. Both DQN and deep multitask learning are pattern based. As a demonstration, the framework was applied to optimal carbon sequestration reservoir planning using two different types of management strategies: monitoring only and brine extraction. Both strategies are designed to mitigate potential risks due to pressure buildup. Results show that the DQN agent can identify the optimal policies to maximize the reward for given risk and cost constraints. Finally, experiments also show that knowledge the agent gained from interacting with one environment is largely preserved when deploying the same agent in other similar environments.

15 GEOTHERMAL ENERGY↗

DOE OSTI2026

Global Explainability of A Deep Abstaining Classifier for Cancer Pathology Reports

We present a global explainability method to characterize sources of errors in a real-world multitask deep abstaining classifier (DAC), in the context of cancer histology prediction. Our multitask classifier, currently deployed for automated annotation of cancer pathology reports from NCI-SEER registries, was trained and evaluated on 1.04 million hand-annotated samples and makes simultaneous predictions of cancer site, subsite, histology, laterality, and behavior for each report. The DAC framework enables the model to abstain on ambiguous reports and confusing classes to achieve the target accuracy on the retained (non-abstained) samples, but at the cost of decreased coverage. Requiring 97% accuracy on the histology task caused our model to retain only 22% of all samples, mostly the less ambiguous and common classes. Local explainability with the GradInp technique provided a computationally efficient way of obtaining contextual reasoning for hundreds of thousands of individual predictions. Our method, involving dimensionality reduction of approximately 13000 aggregated local explanations (ALE), offers a tractable path to true global explainability. It enabled identification of sources of errors in histology classification, globally, as hierarchical complexity among classes, label noise, insufficient information, and conflicting evidence. This suggests several strategies for iterative improvement of our DAC, including well-designed exclusion criteria, focused annotation, and reduced penalties for errors involving hierarchically related classes.

59 BASIC BIOLOGICAL SCIENCES↗

NASA NTRS1981

Real time software tools and methodologies

Real time systems are characterized by high speed processing and throughput as well as asynchronous event processing requirements. These requirements give rise to particular implementations of parallel or pipeline multitasking structures, of intertask or interprocess communications mechanisms, and finally of message (buffer) routing or switching mechanisms. These mechanisms or structures, along with the data structue, describe the essential character of the system. These common structural elements and mechanisms are identified, their implementation in the form of routines, tasks or macros - in other words, tools are formalized. The tools developed support or make available the following: reentrant task creation, generalized message routing techniques, generalized task structures/task families, standardized intertask communications mechanisms, and pipeline and parallel processing architectures in a multitasking environment. Tools development raise some interesting prospects in the areas of software instrumentation and software portability. These issues are discussed following the description of the tools themselves.

Christofferson, M. J.↗

NASA NTRS1985

Experiences with Cray multi-tasking

The issues involved in modifying an existing code for multitasking is explored. They include Cray extensions to FORTRAN, an examination of the application code under study, designing workable modifications, specific code modifications to the VAX and Cray versions, performance, and efficiency results. The finished product is a faster, fully synchronous, parallel version of the original program. A production program is partitioned by hand to run on two CPUs. Loop splitting multitasks three key subroutines. Simply dividing subroutine data and control structure down the middle of a subroutine is not safe. Simple division produces results that are inconsistent with uniprocessor runs. The safest way to partition the code is to transfer one block of loops at a time and check the results of each on a test case. Other issues include debugging and performance. Task startup and maintenance (e.g., synchronization) are potentially expensive.

Miya, E. N.↗

NASA NTRS1991

Asynchronous, macrotasked relaxation strategies for the solution of viscous, hypersonic flows

A point-implicit, asynchronous macrotasked relaxation of the steady, thin-layer, Navier-Stokes equations is presented. The method employs multidirectional, single-level storage Gauss-Seidel relaxation sweeps, which effectively communicate perturbations across the entire domain in 2n sweeps, where n is the dimension of the domain. In order to enhance convergence the application of relaxation factors to specific components of the Jacobian is examined using a stability analysis of the advection and diffusion equations. Attention is also given to the complications associated with asynchronous multitasking. Solutions are generated for hypersonic flows over blunt bodies in two and three dimensions with chemical reactions, utilizing single-tasked and multitasked relaxation strategies.

Gnoffo, Peter A.↗

NASA NTRS1993

Performance analysis of a large-grain dataflow scheduling paradigm

A paradigm for scheduling computations on a network of multiprocessors using large-grain data flow scheduling at run time is described and analyzed. The computations to be scheduled must follow a static flow graph, while the schedule itself will be dynamic (i.e., determined at run time). Many applications characterized by static flow exist, and they include real-time control and digital signal processing. With the advent of computer-aided software engineering (CASE) tools for capturing software designs in dataflow-like structures, macro-dataflow scheduling becomes increasingly attractive, if not necessary. For parallel implementations, using the macro-dataflow method allows the scheduling to be insulated from the application designer and enables the maximum utilization of available resources. Further, by allowing multitasking, processor utilizations can approach 100 percent while they maintain maximum speedup. Extensive simulation studies are performed on 4-, 8-, and 16-processor architectures that reflect the effects of communication delays, scheduling delays, algorithm class, and multitasking on performance and speedup gains.

Young, Steven D.↗

NASA NTRS1991

MAX - An advanced parallel computer for space applications

MAX is a fault-tolerant multicomputer hardware and software architecture designed to meet the needs of NASA spacecraft systems. It consists of conventional computing modules (computers) connected via a dual network topology. One network is used to transfer data among the computers and between computers and I/O devices. This network's topology is arbitrary. The second network operates as a broadcast medium for operating system synchronization messages and supports the operating system's Byzantine resilience. A fully distributed operating system supports multitasking in an asynchronous event and data driven environment. A large grain dataflow paradigm is used to coordinate the multitasking and provide easy control of concurrency. It is the basis of the system's fault tolerance and allows both static and dynamical location of tasks. Redundant execution of tasks with software voting of results may be specified for critical tasks. The dataflow paradigm also supports simplified software design, test and maintenance. A unique feature is a method for reliably patching code in an executing dataflow application.

Lewis, Blair F.↗

NASA NTRS1993

Modeling strategic behavior in human-automation interaction - Why an 'aid' can (and should) go unused

Task-offload aids (e.g., an autopilot, an 'intelligent' assistant) can be selectively engaged by the human operator to dynamically delegate tasks to automation. Introducing such aids eliminates some task demands but creates new ones associated with programming, engaging, and disengaging the aiding device via an interface. The burdens associated with managing automation can sometimes outweigh the potential benefits of automation to improved system performance. Aid design parameters and features of the overall multitask context combine to determine whether or not a task-offload aid will effectively support the operator. A modeling and sensitivity analysis approach is presented that identifies effective strategies for human-automation interaction as a function of three task-context parameters and three aid design parameters. The analysis and modeling approaches provide resources for predicting how a well-adapted operator will use a given task-offload aid, and for specifying aid design features that ensure that automation will provide effective operator support in a multitask environment.

Kirlik, Alex↗

NASA NTRS1992

A new programming metaphor for image processing procedures

Most image processing systems, besides an Application Program Interface (API) which lets users write their own image processing programs, also feature a higher level of programmability. Traditionally, this is a command or macro language, which can be used to build large procedures (scripts) out of simple programs or commands. This approach, a legacy of the teletypewriter has serious drawbacks. A command language is clumsy when (and if! it attempts to utilize the capabilities of a multitasking or multiprocessor environment, it is but adequate for real-time data acquisition and processing, it has a fairly steep learning curve, and the user interface is very inefficient,. especially when compared to a graphical user interface (GUI) that systems running under Xll or Windows should otherwise be able to provide. ll these difficulties stem from one basic problem: a command language is not a natural metaphor for an image processing procedure. A more natural metaphor - an image processing factory is described in detail. A factory is a set of programs (applications) that execute separate operations on images, connected by pipes that carry data (images and parameters) between them. The programs function concurrently, processing images as they arrive along pipes, and querying the user for whatever other input they need. From the user's point of view, programming (constructing) factories is a lot like playing with LEGO blocks - much more intuitive than writing scripts. Focus is on some of the difficulties of implementing factory support, most notably the design of an appropriate API. It also shows that factories retain all the functionality of a command language (including loops and conditional branches), while suffering from none of the drawbacks outlined above. Other benefits of factory programming include self-tuning factories and the process of encapsulation, which lets a factory take the shape of a standard application both from the system and the user's point of view, and thus be used as a component of other factories. A bare-bones prototype of factory programming was implemented under the PcIPS image processing system, and a complete version (on a multitasking platform) is under development.

Smirnov, O. M.↗

NASA NTRS1999

Formal Verification of a Power Controller Using the Real-Time Model Checker UPPAAL

A real-time system for power-down control in audio/video components is modeled and verified using the real-time model checker UPPAAL. The system is supposed to reside in an audio/video component and control (read from and write to) links to neighbor audio/video components such as TV, VCR and remote-control. In particular, the system is responsible for the powering up and down of the component in between the arrival of data, and in order to do so in a safe way without loss of data, it is essential that no link interrupts are lost. Hence, a component system is a multitasking system with hard real-time requirements, and we present techniques for modeling time consumption in such a multitasked, prioritized system. The work has been carried out in a collaboration between Aalborg University and the audio/video company B&O. By modeling the system, 3 design errors were identified and corrected, and the following verification confirmed the validity of the design but also revealed the necessity for an upper limit of the interrupt frequency. The resulting design has been implemented and it is going to be incorporated as part of a new product line.

Havelund, Klaus↗

NASA NTRS2004

Situation Awareness Implications of Adaptive Automation of Air Traffic Controller Information Processing Functions

The goal of this research was to define a measure of situation awareness (SA) in an air traffic control (ATC) task and to assess the influence of adaptive automation (AA) of various information processing functions on controller perception, comprehension and projection. The measure was also to serve as a basis for defining and developing an approach to triggering dynamic control allocations, as part of AA, based on controller SA. To achieve these objectives, an enhanced version of an ATC simulation (Multitask (copyright)) was developed for use in two human factors experiments. The simulation captured the basic functions of Terminal Radar Approach Control (TRACON) and was capable of presenting to operators four different modes of control, including information acquisition, information analysis, decision making and action implementation automation, as well as a completely manual control mode. The SA measure that was developed as part of the research was based on the Situation Awareness Global Assessment Technique (SAGAT), previous goal-directed task analyses of enroute control and TRACON, and a separate cognitive task analysis on the ATC simulation. The results of the analysis on Multitask were used as a basis for formulating SA queries as part of the SAGAT-based approach to measuring controller SA, which was used in the experiments. A total of 16 subjects were recruited for both experiments. Half the subjects were used in Experiment #1, which focused on assessing the sensitivity and reliability of the SA measurement approach in the ATC simulation. Comparisons were made of manual versus automated control. The remaining subjects were used in the second experiment, which was intended to more completely describe the SA implications of AA applied to specific controller information processing functions, and to describe how the measure could ultimately serve as a trigger of dynamic function allocations in the application of AA to ATC. Comparisons were made of the sensitivity of the SA measure to automation manipulations impacting both higher-order information processing functions, such as information analysis and decision making, versus lower-order functions, including information acquisition and action implementation. All subjects were exposed to all forms of AA of the ATC task and the manual control condition. The approach to AA used in both experiments was to match operator workload, assessed using a secondary task, to dynamic control allocations in the primary task. In total, the subjects in each experiment participated in 10 trials with each lasting between 45 minutes and 1 hour. In both experiments, ATC performance was measured in terms of aircraft cleared, conflicting, and collided. Secondary task (gauge monitoring) performance was assessed in terms of a hit-to-signal ratio. As part of the SA measure, three simulation freezes were conducted during each trial to administer queries on Level 1, 2, and 3 SA.

Kaber, David B.↗

DOE OSTI2022

Model Choice Metrics to Optimize Profile-QSAR Performance

Predicting molecular activity against protein targets is difficult because of the paucity of experimental data. Approaches like multitask modeling and collaborative filtering seek to improve model accuracy by leveraging results from multiple targets, but are limited because different compounds are measured with different assays, leading to sparse data matrices. Profile-QSAR (pQSAR) 2.0 addresses this problem by fitting a series of partial least squares models for each target, using as features the predictions from single-task models on the remaining targets. Here, this method has been shown to produce better results than single task and multitask models. However, the factors determining the success of pQSAR 2.0 have as yet not been characterized. In this paper we examine the experimental conditions that lead to better pQSAR models. We limit the amount of data available to the method by retraining with decreasing amounts of data and explore the model’s ability to generalize to compounds that have never been assayed. Finally, we look at the properties of training data needed to demonstrate pQSAR improvement.

Biological and medical sciences, Computer science↗

DOE OSTI2024

GPS-supported smartphone app-based integrated travel diary and time-use data collection: challenges and lessons learned

Travel behaviour and time-use data are two vital data sources for travel demand modelling. Travel behaviour is traditionally collected through household travel surveys, enhanced by using GPS-supported smartphone apps for passive location data collection. However, recruiting individuals willing to install these apps with sustained motivation to continue participation has been a critical challenge. This paper shares insights from a travel and time-use data collection procedure in Chicago and Sydney using the Fourstep app. Social media platforms were utilised as a solution to recruit participants in Chicago, where an international market research company failed to accomplish the task. This paper also discusses the challenges we faced and suggests ways to overcome them, offering valuable guidance to researchers in recruiting participants for smartphone application-based data collection. It also offers an analysis of travel, time-use, and travel-based multitasking behaviours based on the data collected from the Chicago and Sydney samples.

GPS-supported smartphone apps↗