SEARCH · Search NASA
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.
Graphical method requires calculations and table lookup. Distribution established from only three points: mean upper and lower confidence bounds and lower confidence bound of standard deviation. Method requires only few calculations with simple equations. Graphical procedure establishes best-fit line for measured data and bounds for selected confidence level and any distribution percentile.
The Magnifying Glass Transformation (MGT) technique is proposed, as a multichannel spectral operation yielding visual imagery which is enhanced in a specified spectral vicinity, guided by the statistics of training samples. An application example is that in which the discrimination among spectral neighbors within an interactive display may be increased without altering distant object appearances or overall interpretation. A direct histogram specification technique is applied to the channels within the multispectral image so that a subset of the spectral domain occupies an increased fraction of the domain. The transformation is carried out by obtaining the training information, establishing the condition of the covariance matrix, determining the influenced solid, and initializing the lookup table. Finally, the image is transformed.
Tangents and normals to ellipse derived for situations with two correlated aerodynamic force or moment variables having normally (Gaussian) distributed uncertainty. Although transformations and vector calculations tedious when done by hand, equations in closed form and give rapid results when performed by computers or programable desk calculators. Probability extremes of ratios and magnitude obtained directly, without repeated table lookup or other complication.
An overview of magnetic bearing control and linearization approaches which have been considered for annular magnetically suspended devices is presented. These devices include the Annular Momentum Control Device and the Annular Suspension and Pointing System. Two approaches were investigated for controlling the magnetic actuator. One approach involves controlling the upper and lower electromagnets differentially about a bias flux. The bias flux can either be supplied by permanent magnets in the magnetic circuit or by bias currents. In the other approach, either the upper electromagnet or the lower electromagnet is controlled depending on the direction of force required. One advantage of the bias flux is that for small gap perturbations about a fixed operating point, the force-current characteristic is linear. Linearization approaches investigated for individual element control include an analog solution of the nonlinear electromagnet force equation and a microprocessor-based table lookup method.
A software program for the production and analysis of data from the Dynamics Explorer-A (DE-A) satellite was maintained and modified and new software initiated. A capability was developed to process DE-A plasma-wave instrument mission analysis files on the Tektronic 4027 color CRT, for which two programs were written. The algorithm for the calibration lookup table for the plasma-wave instrument data was modified and verified, and a production program to generate color FR-80 spectrograms was written.
Digital values in a moving window are compared by an operator having nine comparators connected to line buffers for receiving a succession of central pixels together with eight neighborhood pixels. A single bit of program control determines whether the neighborhood pixels are to be compared with the central pixel or a threshold value. The central pixel is always compared with the threshold. The omparator output plus 2 bits indicating odd-even pixel/line information about the central pixel addresses a lookup table to provide 14 bits of information, including 2 bits which control a selector to pass either the central pixel value, the other 12 bits of table information, or the bit-wise logical OR of all nine pixels through circuit that implements a very wide OR gate.
Research conducted on a traffic advisory and collision avoidance system (TCAS 2) mounted on a curved ground plane is described. It is found that a curved finite ground plane can be used as a good simulation model for the fuselage of an aircraft but may not be good enough to model a whole aircraft due to the shadowing of the vertical stabilizer, wings, etc. The surface curvature of this curved disc significantly affects the monopulse characteristics in the azimuth plane but not as much in the elevation plane. These variations of the monopulse characteristics verify the need of a lookup table for the 64 azimuth beam positions. The best location of a TCAS 2 array on a Boeing 737 is to move it as far from the vertical stabilizer as possible.
The method and apparatus constructed in accordance with this invention permits a plurality of parties to speak to each other on a conference line with a minimum of interference. The apparatus digitizes audio signals. Each of the parties has an audio transmitter and receiver provided for transmitting and receiving audio signals. The audio signals are converted to a PCM companded eight-bit parallel signal followed by a conversion to a serial signal for transmitting to a remote location and then reconverting each of the companded signals to a first-eight-bit parallel signal. The eight-bit parallel signal is fed to one input of a pre-programmed ROM. This eight-bit signal provides one-half of a sixteen-bit address of a lookup ROM. The other half of the sixteen-bit ROM address is supplied by another suscriber over an identical circuit.
A framework to build simulation models for aircraft dynamic systems integration is described. The objective of the framework is increased simulation model fidelity and reduced time required to develop and modify these models. The equations of motion for an elastic aircraft and their impact on the framework are discussed in broad terms. A software tool which automatically generates FORTRAN routines for tabular data lookups, the language used to develop a simulation model, and the structures for passing information into a simulation are discussed. A simulation variable nomenclature is presented. The framework has been applied to build an open-loop F/A-18 simulation model. This example model is used to illustrate model reduction issues. Current deficiencies in the framework are identified as areas for future research.
Light flashes when shaft reaches predetermined angle. Synchronization system facilitates visualization of flow in high-speed internal-combustion engines. Designed for cinematography and holographic interferometry, system synchronizes camera and light source with predetermined rotational angle of engine shaft. 10-bit resolution of absolute optical shaft encoder adapted, and 2 to tenth power combinations of 10-bit binary data computed to corresponding angle values. Pre-computed angle values programmed into EPROM's (erasable programmable read-only memories) to use as angle lookup table. Resolves shaft angle to within 0.35 degree at rotational speeds up to 73,240 revolutions per minute.
The results of an investigation into the feasibility of using the MPP for direct and large eddy simulations of the Navier-Stokes equations is presented. A major part of this study was devoted to the implementation of two of the standard numerical algorithms for CFD. These implementations were not run on the Massively Parallel Processor (MPP) since the machine delivered to NASA Goddard does not have sufficient capacity. Instead, a detailed implementation plan was designed and from these were derived estimates of the time and space requirements of the algorithms on a suitably configured MPP. In addition, other issues related to the practical implementation of these algorithms on an MPP-like architecture were considered; namely, adaptive grid generation, zonal boundary conditions, the table lookup problem, and the software interface. Performance estimates show that the architectural components of the MPP, the Staging Memory and the Array Unit, appear to be well suited to the numerical algorithms of CFD. This combined with the prospect of building a faster and larger MMP-like machine holds the promise of achieving sustained gigaflop rates that are required for the numerical simulations in CFD.
A robot system called RAMBO is being developed. It is equipped with a camera, which, given a sequence of simple tasks, can perform these tasks on a moving object. RAMBO is given a complete geometric model of the object. A low level vision module extracts and groups characteristic features in images of the object. The positions of the object are determined in a sequence of images, and a motion estimate of the object is obtained. This motion estimate is used to plan trajectories of the robot tool to relative locations nearby the object sufficient for achieving the tasks. More specifically, low level vision uses parallel algorithms for image enchancement by symmetric nearest neighbor filtering, edge detection by local gradient operators, and corner extraction by sector filtering. The object pose estimation is a Hough transform method accumulating position hypotheses obtained by matching triples of image features (corners) to triples of model features. To maximize computing speed, the estimate of the position in space of a triple of features is obtained by decomposing its perspective view into a product of rotations and a scaled orthographic projection. This allows the use of 2-D lookup tables at each stage of the decomposition. The position hypotheses for each possible match of model feature triples and image feature triples are calculated in parallel. Trajectory planning combines heuristic and dynamic programming techniques. Then trajectories are created using parametric cubic splines between initial and goal trajectories. All the parallel algorithms run on a Connection Machine CM-2 with 16K processors.
The amount of programming required to develop a PATRAN-NASTRAN translator was surprisingly small. The approach taken produced a highly flexible translator comparable with the PATNAS translator and superior to the PATCOS translator. The coding required varied from around ten lines for a shell element to around thirty for a bar element, and the time required to add a feature to the program is typically less than an hour. The use of a lookup table for element names makes the translator also applicable to other versions of NASTRAN. The saving in time as a result of using PDA's Gateway utilities was considerable. During the writing of the program it became apparent that, with a somewhat more complex structure, it would be possible to extend the element data file to contain all data required to define the translation from PATRAN to NASTRAN by mapping of data between formats. Similar data files on property, material and grid formats would produce a completely universal translator from PATRAN to any FEA program, or indeed any CAE system.
Neural networks have been applied to tasks in several areas of artificial intelligence, including vision, speech, and language. Relatively little work has been done in the area of problem solving. Two approaches to path-finding are presented, both using neural network techniques. Both techniques require a training period. Training under the back propagation (BPL) method was accomplished by presenting representations of (current position, goal position) pairs as input and appropriate actions as output. The Hebbian/interactive activation (HIA) method uses the Hebbian rule to associate points that are nearby. A path to a goal is found by activating a representation of the goal in the network and processing until the current position is activated above some threshold level. BPL, using back-propagation learning, failed to learn, except in a very trivial fashion, that is equivalent to table lookup techniques. HIA, performed much better, and required storage of fewer weights. In drawing a comparison, it is important to note that back propagation techniques depend critically upon the forms of representation used, and can be sensitive to parameters in the simulations; hence the BPL technique may yet yield strong results.
Neural networks have been applied to tasks in several areas of artificial intelligence, including vision, speech, and language. Relatively little work has been done in the area of problem solving. Two approaches to path-finding are presented, both using neural network techniques. Both techniques require a training period. Training under the back propagation (BPL) method was accomplished by presenting representations of current position, goal position pairs as input and appropriate actions as output. The Hebbian/interactive activation (HIA) method uses the Hebbian rule to associate points that are nearby. A path to a goal is found by activating a representation of the goal in the network and processing until the current position is activated above some threshold level. BPL, using back-propagation learning, failed to learn, except in a very trivial fashion, that is equivalent to table lookup techniques. HIA, performed much better, and required storage of fewer weights. In drawing a comparison, it is important to note that back propagation techniques depend critically upon the forms of representation used, and can be sensitive to parameters in the simulations; hence the BPL technique may yet yield strong results.
An experiment was conducted to determine the effects of training on the characteristic direction errors that are observed when subjects estimate exocentric directions on perspective displays. Changes in five subjects' perceptual errors were measured during a training procedure designed to eliminate the error. The training was provided by displaying to each subject both the sign and the direction of his judgment error. The feedback provided by the error display was found to decrease but not eliminate the error. A lookup table model of the source of the error was developed in which the judgement errors were attributed to overestimates of both the pitch and the yaw of the viewing direction used to produce the perspective projection. The model predicts the quantitative characteristics of the data somewhat better than previous models did. A mechanism is proposed for the observed learning, and further tests of the model are suggested.
Interaction with tumbling objects will become more common as human activities in space expand. Attempting to interact with a large complex object translating and rotating in space, a human operator using only his visual and mental capacities may not be able to estimate the object motion, plan actions or control those actions. A robot system (RAMBO) equipped with a camera, which, given a sequence of simple tasks, can perform these tasks on a tumbling object, is being developed. RAMBO is given a complete geometric model of the object. A low level vision module extracts and groups characteristic features in images of the object. The positions of the object are determined in a sequence of images, and a motion estimate of the object is obtained. This motion estimate is used to plan trajectories of the robot tool to relative locations rearby the object sufficient for achieving the tasks. More specifically, low level vision uses parallel algorithms for image enhancement by symmetric nearest neighbor filtering, edge detection by local gradient operators, and corner extraction by sector filtering. The object pose estimation is a Hough transform method accumulating position hypotheses obtained by matching triples of image features (corners) to triples of model features. To maximize computing speed, the estimate of the position in space of a triple of features is obtained by decomposing its perspective view into a product of rotations and a scaled orthographic projection. This allows use of 2-D lookup tables at each stage of the decomposition. The position hypotheses for each possible match of model feature triples and image feature triples are calculated in parallel. Trajectory planning combines heuristic and dynamic programming techniques. Then trajectories are created using dynamic interpolations between initial and goal trajectories. All the parallel algorithms run on a Connection Machine CM-2 with 16K processors.
Digital technique for generation of slowly moving video image of sinusoidal grating avoids difficulty of transferring full image data from disk storage to image memory at conventional frame rates. Depends partly on trigonometric identity by which moving sinusoidal grating decomposed into two stationary patterns spatially and temporally modulated in quadrature. Makes motion appear smooth, even at speeds much less than one-tenth picture element per frame period. Applicable to digital video system in which image memory consists of at least 2 bits per picture element, and final brightness of picture element determined by contents of "lookup-table" memory programmed anew each frame period and indexed by coordinates of each picture element.