Search NASA⌕ Search

SEARCH · Search NASA

Results for “approximate dynamic programming”

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 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
NASA NTRS2004

Automatic Generation of Algorithms for the Statistical Analysis of Planetary Nebulae Images

Analyzing data sets collected in experiments or by observations is a Core scientific activity. Typically, experimentd and observational data are &aught with uncertainty, and the analysis is based on a statistical model of the conjectured underlying processes, The large data volumes collected by modern instruments make computer support indispensible for this. Consequently, scientists spend significant amounts of their time with the development and refinement of the data analysis programs. AutoBayes [GF+02, FS03] is a fully automatic synthesis system for generating statistical data analysis programs. Externally, it looks like a compiler: it takes an abstract problem specification and translates it into executable code. Its input is a concise description of a data analysis problem in the form of a statistical model as shown in Figure 1; its output is optimized and fully documented C/C++ code which can be linked dynamically into the Matlab and Octave environments. Internally, however, it is quite different: AutoBayes derives a customized algorithm implementing the given model using a schema-based process, and then further refines and optimizes the algorithm into code. A schema is a parameterized code template with associated semantic constraints which define and restrict the template s applicability. The schema parameters are instantiated in a problem-specific way during synthesis as AutoBayes checks the constraints against the original model or, recursively, against emerging sub-problems. AutoBayes schema library contains problem decomposition operators (which are justified by theorems in a formal logic in the domain of Bayesian networks) as well as machine learning algorithms (e.g., EM, k-Means) and nu- meric optimization methods (e.g., Nelder-Mead simplex, conjugate gradient). AutoBayes augments this schema-based approach by symbolic computation to derive closed-form solutions whenever possible. This is a major advantage over other statistical data analysis systems which use numerical approximations even in cases where closed-form solutions exist. AutoBayes is implemented in Prolog and comprises approximately 75.000 lines of code. In this paper, we take one typical scientific data analysis problem-analyzing planetary nebulae images taken by the Hubble Space Telescope-and show how AutoBayes can be used to automate the implementation of the necessary anal- ysis programs. We initially follow the analysis described by Knuth and Hajian [KHO2] and use AutoBayes to derive code for the published models. We show the details of the code derivation process, including the symbolic computations and automatic integration of library procedures, and compare the results of the automatically generated and manually implemented code. We then go beyond the original analysis and use AutoBayes to derive code for a simple image segmentation procedure based on a mixture model which can be used to automate a manual preproceesing step. Finally, we combine the original approach with the simple segmentation which yields a more detailed analysis. This also demonstrates that AutoBayes makes it easy to combine different aspects of data analysis.

Fischer, Bernd↗
NASA NTRSDate not supplied

Aerothermal Analysis of the Dragonfly Titan Entry

The Dragonfly mission will send a rotorcraft lander to the surface of Saturn’s moon Titan as part of the New-Frontiers program. This will be the first spacecraft to land on Titan since the Huygens probe’s descent in January of 2005 and only the second spacecraft to enter Titan’s atmosphere. The Dragonfly entry capsule is significantly larger than the Huygens probe and will experience higher aerothermal environments. This poster will provide an overview of the aerothermodynamic models used in the Dragonfly aeroshell design process. Titan provides a unique entry environment that has several fundamental differences from the environments at the more common entry destinations of Earth and Mars. The lack of atmospheric oxygen significantly reduces heatshield recession (expected to be negligible), which simplifies material response analysis but also reduces the efficiency of the thermal protection system. The atmos-phere is composed of mostly nitrogen and trace amounts of methane. During hypersonic entry this methane dissociates and leads to the formation of the molecule CN, which is known to radiate strongly in the shock layer. Current Dragonfly heating predictions estimate that radiation contributes about 50% of the total heat flux along the forebody and up to 90% on the aftbody, making radiation a key element of aerothermal analysis. To properly address the importance of radiation, a unique radiative-heating correlation was developed that incorporates normal-shock equilibrium chemistry predictions based on work from the Mars 2020 program. These new heating correlations agree well with Computational Fluid Dynamics (CFD) predictions and enable large-scale trajectory analyses by providing rapid heating predictions along candidate trajectories. Dragonfly aerothermal environments are generated with the Data Parallel Line Relaxation (DPLR) finite-volume Navier-Stokes solver alongside the Nonequilibrium Radiative Transport and Spectra (NEQAIR) radiation transport code. DPLR produces shock-aligned flowfields that incorporate the effects of both chemical and thermal nonequilibrium while NEQAIR solves the radiation transport equations along lines-of-sight through the flowfield solutions to predict the radiative heating on the aeroshell. Titan’s atmosphere produces complex thermal and chemical nonequilibria and so classical radiation approximations that extrapolate from a single line-of-sight, such as the Tangent-Slab method, were found to overpredict peak radiative heat fluxes by 10-20%. Therefore, all radiative heating is directly calculated using 3-D transport for both the forebody and aftbody to reduce unnecessary conservatism. A 21-species finite-rate chemistry model, including electrons and ionized species, is used in the CFD to ensure accurate vehicle heating predictions (the level of ionization is particularly for accurate radiative flux predictions). This poster will cover: 1) The models and assumptions used to simulate entry into Titan’s atmosphere, 2) Dragonfly aerothermal engineering correlations alongside statistical heating values, and 3) Driving aerothermal considerations for a Titan entry such as CN formation and radiation-flowfield coupling.

EDL↗
DOE OSTI2026

Near-Optimal Performance of Stochastic Model Predictive Control

Here, this article presents a regret analysis for stochastic model predictive control (SMPC) in linear systems with quadratic performance index and additive and multiplicative uncertainties. Under a finite support assumption, the problem can be cast as a finite-dimensional quadratic program, but the problem becomes quickly intractable as the problem size grows exponentially in the horizon length. SMPC aims to compute approximate solutions by solving a sequence of problems with truncated prediction horizons and committing the solution in a receding-horizon fashion. Although this approach is widely used in practice, its performance relative to the optimal solution is not well understood. This article reports for the first time a rigorous near-optimal performance guarantee of SMPC: under stabilizability and detectability conditions, the regret of SMPC is exponentially small in the prediction horizon length, allowing SMPC to achieve near-optimal performance at a substantially reduced computational expense.

93E20, 93B45↗
NASA NTRS2014

Wind Tunnel Testing of a 120th Scale Large Civil Tilt-Rotor Model in Airplane and Helicopter Modes

In April 2012 and October 2013, NASA and the U.S. Army jointly conducted a wind tunnel test program examining two notional large tilt rotor designs: NASA's Large Civil Tilt Rotor and the Army's High Efficiency Tilt Rotor. The approximately 6%-scale airframe models (unpowered) were tested without rotors in the U.S. Army 7- by 10-foot wind tunnel at NASA Ames Research Center. Measurements of all six forces and moments acting on the airframe were taken using the wind tunnel scale system. In addition to force and moment measurements, flow visualization using tufts, infrared thermography and oil flow were used to identify flow trajectories, boundary layer transition and areas of flow separation. The purpose of this test was to collect data for the validation of computational fluid dynamics tools, for the development of flight dynamics simulation models, and to validate performance predictions made during conceptual design. This paper focuses on the results for the Large Civil Tilt Rotor model in an airplane mode configuration up to 200 knots of wind tunnel speed. Results are presented with the full airframe model with various wing tip and nacelle configurations, and for a wing-only case also with various wing tip and nacelle configurations. Key results show that the addition of a wing extension outboard of the nacelles produces a significant increase in the lift-to-drag ratio, and interestingly decreases the drag compared to the case where the wing extension is not present. The drag decrease is likely due to complex aerodynamic interactions between the nacelle and wing extension that results in a significant drag benefit.

Wind Tunnel Testing of a 1/20th Scale↗
NASA NTRS1976

Preliminary analysis of long-range aircraft designs for future heavy airlift missions

A computerized design study of very large cargo aircraft for the future heavy airlift mission was conducted using the Aircraft Synthesis program (ACSYNT). The study was requested by the Air Force under an agreement whereby Ames provides computerized design support to the Air Force Flight Dynamics Laboratory. This effort is part of an overall Air Force program to study advanced technology large aircraft systems. Included in the Air Force large aircraft program are investigations of missions such as heavy airlift, airborne missile launch, battle platform, command and control, and aerial tanker. The Ames studies concentrated on large cargo aircraft of conventional design with payloads from 250,000 to 350,000 lb. Range missions up to 6500 n.mi. and radius missions up to 3600 n.mi. have been considered. Takeoff and landing distances between 7,000 and 10,000 ft are important constraints on the configuration concepts. The results indicate that a configuration employing conventional technology in all disciplinary areas weighs approximately 2 million pounds to accomplish either a 6500-n.mi. range mission or a 3600-n.mi. radius mission with a 350,000-lb payload.

Nelms, W. P., Jr.↗
NASA NTRS1981

Manufacturing in space: Fluid dynamics numerical analysis

Natural convection in a spherical container with cooling at the center was numerically simulated using the Lockheed-developed General Interpolants Method (GIM) numerical fluid dynamic computer program. The numerical analysis was simplified by assuming axisymmetric flow in the spherical container, with the symmetry axis being a sphere diagonal parallel to the gravity vector. This axisymmetric spherical geometry was intended as an idealization of the proposed Lal/Kroes growing experiments to be performed on board Spacelab. Results were obtained for a range of Rayleigh numbers from 25 to 10,000. For a temperature difference of 10 C from the cooling sting at the center to the container surface, and a gravitional loading of 0.000001 g a computed maximum fluid velocity of about 2.4 x 0.00001 cm/sec was reached after about 250 sec. The computed velocities were found to be approximately proportional to the Rayleigh number over the range of Rayleigh numbers investigated.

Robertson, S. J.↗
NASA NTRS1991

Techniques for developing approximate optimal advanced launch system guidance

An extension to the authors' previous technique used to develop a real-time guidance scheme for the Advanced Launch System is presented. The approach is to construct an optimal guidance law based upon an asymptotic expansion associated with small physical parameters, epsilon. The trajectory of a rocket modeled as a point mass is considered with the flight restricted to an equatorial plane while reaching an orbital altitude at orbital injection speeds. The dynamics of this problem can be separated into primary effects due to thrust and gravitational forces, and perturbation effects which include the aerodynamic forces and the remaining inertial forces. An analytic solution to the reduced-order problem represented by the primary dynamics is possible. The Hamilton-Jacobi-Bellman or dynamic programming equation is expanded in an asymptotic series where the zeroth-order term (epsilon = 0) can be obtained in closed form.

Feeley, Timothy S.↗
DOE OSTI2010

Feedback-directed page placement for ccNUMA via hardware-generated memory traces

Non-uniform memory architectures with cache coherence (ccNUMA) are becoming increasingly common, not just for large-scale high performance platforms but also in the context of multi-core architectures. Under ccNUMA, data placement may influence overall application performance significantly as references resolved locally to a processor/core impose lower latencies than remote ones. This work develops a novel hardware-assisted page placement paradigm based on automated tracing of the memory references made by application threads. Two placement schemes, modeling both singlelevel and multi-level latencies, allocate pages near processors that most frequently access that memory page. These schemes leverage performance monitoring capabilities of contemporary microprocessors to efficiently extract an approximate trace of memory accesses. This information is used to decide page affinity, i.e., the node to which the page is bound. The method operates entirely in user space, is widely automated, and handles not only static but also dynamic memory allocation. Experiments show that this method, although based on lossy tracing, can efficiently and effectively improve page placement, leading to an average wall-clock execution time saving of over 20% for the tested benchmarks on the SGI Altix with a 2x remote access penalty and 12% on AMD Opterons with a 1.3–2.0x access penalty. This is accompanied by a one-time tracing overhead of 2.7% over the overall original program wallclock time.

Computer Science↗
DOE OSTI2014

Tensor to scalar ratio and large scale power suppression from pre-slow roll initial conditions

We study the corrections to the power spectra of curvature and tensor perturbations and the tensor-to-scalar ratio r in single field slow roll inflation with standard kinetic term due to initial conditions imprinted by a ''fast-roll'' stage prior to slow roll. For a wide range of initial inflaton kinetic energy, this stage lasts only a few e-folds and merges smoothly with slow-roll thereby leading to non-Bunch-Davies initial conditions for modes that exit the Hubble radius during slow roll. We describe a program that yields the dynamics in the fast-roll stage while matching to the slow roll stage in a manner that is independent of the inflationary potentials. Corrections to the power spectra are encoded in a ''transfer function'' for initial conditions T{sub α}(k), P{sub α}(k) = P{sup BD}{sub α}(k)T{sub α}(k), implying a modification of the ''consistency condition'' for the tensor to scalar ratio at a pivot scale k{sub 0}: r(k{sub 0}) = −8n{sub T}(k{sub 0}) [T{sub T}(k{sub 0})/T{sub R}(k{sub 0})]. We obtain T{sub α}(k) to leading order in a Born approximation valid for modes of observational relevance today. A fit yields T{sub α}(k) = 1+A{sub α}k{sup −p}cos [2πωk/H{sub sr}+φ{sub α}], with 1.5∼<p∼<2, ω ≅ 1 and H{sub sr} the Hubble scale during slow roll inflation, where curvature and tensor perturbations feature the same p,ω for a wide range of initial conditions. These corrections lead to both a suppression of the quadrupole and oscillatory features in both P{sub R}(k) and r(k{sub 0}) with a period of the order of the Hubble scale during slow roll inflation. The results are quite general and independent of the specific inflationary potentials, depending solely on the ratio of kinetic to potential energy κ and the slow roll parameters ε{sub V}, η{sub V} to leading order in slow roll. For a wide range of κ and the values of ε{sub V}; η{sub V} corresponding to the upper bounds from Planck, we find that the low quadrupole is consistent with the results from Planck, and the oscillations in r(k{sub 0}) as a function of k{sub 0} could be observable if the modes corresponding to the quadrupole and the pivot scale crossed the Hubble radius very few (2–3) e-folds after the onset of slow roll. We comment on possible impact on the recent BICEP2 results.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2017

CXSFIT Code Application to Process Charge-Exchange Recombination Spectroscopy Data at the T-10 Tokamak

The applicability of the CXSFIT code to process experimental data from Charge-eXchange Recombination Spectroscopy (CXRS) diagnostics at the T-10 tokamak is studied with a view to its further use for processing experimental data at the ITER facility. The design and operating principle of the CXRS diagnostics are described. The main methods for processing the CXRS spectra of the 5291-Å line of C{sup 5+} ions at the T-10 tokamak (with and without subtraction of parasitic emission from the edge plasma) are analyzed. The method of averaging the CXRS spectra over several shots, which is used at the T-10 tokamak to increase the signal-to-noise ratio, is described. The approximation of the spectrum by a set of Gaussian components is used to identify the active CXRS line in the measured spectrum. Using the CXSFIT code, the ion temperature in ohmic discharges and discharges with auxiliary electron cyclotron resonance heating (ECRH) at the T-10 tokamak is calculated from the CXRS spectra of the 5291-Å line. The time behavior of the ion temperature profile in different ohmic heating modes is studied. The temperature profile dependence on the ECRH power is measured, and the dynamics of ECR removal of carbon nuclei from the T-10 plasma is described. Experimental data from the CXRS diagnostics at T-10 substantially contribute to the implementation of physical programs of studies on heat and particle transport in tokamak plasmas and investigation of geodesic acoustic mode properties.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1984

Inversion Algorithms for Water Vapor Radiometers Operating at 20.7 and 31.4 Ghz

Eight water vapor radiometers (WVRs) were constructed as research and development tools to support the Advanced System Programs in the Deep Space Network and the Crustal Dynamics Project. These instruments are intended to operate at the stations of the Deep Space Network (DSN), various radio observatories, and obile facilities that participate in very long baseline interferometric (VLBI) experiments. It is expected that the WVRs will operate in a wide range of meteorological conditions. Several algorithms are discussed that are used to estimate the line-of-sight path delay due to water vapor and columnar liquid water rom the observed microwave brightness temperatures provided by the WVRs. In particular, systematic effects due to site and seasonal variations are examined. The accuracy of the estimation as indicated by a simulation calculation is approximately 0.3 cm for a noiseless WVR in clear and moderately cloudy weather. With a realistic noise model of WVR behavior, the inversion accuracy is approximately 0.6 cm.

Resch, G. M.↗
NASA NTRS1992

Navier-Stokes calculation of transonic flow past the NTF 65-deg delta wing

Viscous flow past a wind tunnel model of a 65-degree swept angle Delta wing at transonic speeds is being studied. The model was tested in the 8-foot cryogenic transonic wind tunnel at the National Transonic Facility. Aerodynamic forces and wing surface pressure data were obtained at various angles of attack, Mach numbers, and Reynold's numbers for four different leading edges of the wing. The objectives of the present investigation are: (1) to perform numerical modeling of the flow around the wing; (2) to validate the experimental data with a Navier-Stokes computational fluid dynamics code and vice versa; (3) to investigate the effects of the sting mount of the wing; (4) to evaluate the effects of leading edge radius on the flow; and (5) to explain the Reynold's number effect as indicated by the test data. Several computer programs were developed to define the surfaces of the wing, the four leading edges, and the sting mount. Based on these geometric databases, the surface grids of a single-block computational domain was generated interactively on the IRIS workstation using the GRIDGEN2D module of GRIDGEN. To refine the grids and to avoid excessive loss of grid points due to collapsed edges, a 9-block computational domain containing approximately 750,000 grid points was developed with the GRIDBLOCK module to replace the single-block grid.

Wu, Chivey↗
NASA NTRS2014

Magnetospheric MultiScale (MMS) System Manager

The Magnetospheric MultiScale (MMS) mission is an ambitious NASA space science mission in which 4 spacecraft are flown in tight formation about a highly elliptical orbit. Each spacecraft has multiple instruments that measure particle and field compositions in the Earths magnetosphere. By controlling the members relative motion, MMS can distinguish temporal and spatial fluctuations in a way that a single spacecraft cannot.To achieve this control, 2 sets of four maneuvers, distributed evenly across the spacecraft must be performed approximately every 14 days. Performing a single maneuver on an individual spacecraft is usually labor intensive and the complexity becomes clearly increases with four. As a result, the MMS flight dynamics team turned to the System Manager to put the routine or error-prone under machine control freeing the analysts for activities that require human judgment.The System Manager is an expert system that is capable of handling operations activities associated with performing MMS maneuvers. As an expert system, it can work off a known schedule, launching jobs based on a one-time occurrence or on a set reoccurring schedule. It is also able to detect situational changes and use event-driven programming to change schedules, adapt activities, or call for help.

Ground System Expert Automation↗
NASA NTRS2014

New Worlds / New Horizons Science with an X-ray Astrophysics Probe

In 2013 NASA commenced a design study for an X-ray Astrophysics Probe to address the X-ray science goals and program prioritizations of the Decadal Survey New World New Horizons (NWNH) with a cost cap of approximately $1B. Both the NWNH report and 2011 NASA X-ray mission concept study found that high-resolution X-ray spectroscopy performed with an X-ray microcalorimeter would enable the most highly rated NWNH X-ray science. Here we highlight some potential science topics, namely: 1) a direct, strong-field test of General Relativity via the study of accretion onto black holes through relativistic broadened Fe lines and their reverberation in response to changing hard X-ray continuum, 2) understanding the evolution of galaxies and clusters by mapping temperatures, abundances and dynamics in hot gas, 3) revealing the physics of accretion onto stellar-mass black holes from companion stars and the equation of state of neutron stars through timing studies and time-resolved spectroscopy of X-ray binaries and 4) feedback from AGN and star formation shown in galaxy-scale winds and jets. In addition to these high-priority goals, an X-ray astrophysics probe would be a general-purpose observatory that will result in invaluable data for other NWNH topics such as stellar astrophysics, protostars and their impact on protoplanetary systems, X-ray spectroscopy of transient phenomena such as high-z gamma-ray bursts and tidal capture of stars by massive black holes, and searches for dark matter decay.

Smith, Randall K.↗
NASA NTRS1987

Structure and dynamics of amorphous water ice

Further insight into the structure and dynamics of amorphous water ice, at low temperatures, was obtained by trapping in it Ar, Ne, H2, and D2. Ballistic water-vapor deposition results in the growth of smooth, approximately 1 x 0.2 micrometer2, ice needles. The amorphous ice seems to exist in at least two separate forms, at T < 85 K and at 85 < T < 136.8 K, and transform irreversibly from one form to the other through a series of temperature-dependent metastable states. The channels formed by the water hexagons in the ice are wide enough to allow the free penetration of H2 and D2 into the ice matrix even in the relatively compact cubic ice, resulting in H2-(D2-) to-ice ratios (by number) as high as 0.63. The larger Ar atoms can penetrate only into the wider channels of amorphous ice, and Ne is an intermediate case. Dynamic percolation behavior explains the emergence of Ar and Ne (but not H2 and D2) for the ice, upon warming, in small and big gas jets. The big jets, each containing approximately 5 x 10(10) atoms, break and propel the ice needles. Dynamic percolation also explains the collapse of the ice matrix under bombardment by Ar , at a pressure exceeding 2.6 dyn cm-2, and the burial of huge amounts of gas inside the collapsed matrix, up to an Ar-to-ice of 3.3 (by number). The experimental results could be relevant to comets, icy satellites, and icy grain mantles in dense interstellar clouds.

NASA Program Exobiology↗
DOE OSTI2020

Tractable minor-free generalization of planar zero-field Ising models

In this work, we present a new family of zero-field Ising models over N binary variables/spins obtained by consecutive 'gluing' of planar and O(1)-sized components and subsets of at most three vertices into a tree. The polynomial time algorithm of the dynamic programming type for solving exact inference (computing partition function) and exact sampling (generating i.i.d. samples) consists of sequential application of an efficient (for planar) or brute-force (for O(1)-sized) inference and sampling to the components as a black box. To illustrate the utility of the new family of tractable graphical models, we first build a polynomial algorithm for inference and sampling of zero-field Ising models over K 33 -minor-free topologies and over K 5 -minor-free topologies—both of which are extensions of the planar zero-field Ising models—which are neither genus- nor treewidth-bounded. Second, we empirically demonstrate an improvement in the approximation quality of the NP-hard problem of inference over the square-grid Ising model in a node-dependent nonzero 'magnetic' field.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1981

Test techniques for determining laser ranging system performance

Procedures and results of an on going test program intended to evaluate laser ranging system performance levels in the field as well as in the laboratory are summarized. Tests show that laser ranging system design requires consideration of time biases and RMS jitters of individual system components. All simple Q switched lasers tested were found to be inadequate for 10 centimeter ranging systems. Timing discriminators operating over a typical 100:1 dynamic signal range may introduce as much as 7 to 9 centimeters of range bias. Time interval units commercially available today are capable of half centimeter performance and are adequate for all field systems currently deployed. Photomultipliers tested show typical tube time biases of one centimeter with single photoelectron transit time jitter of approximately 10 centimeters. Test results demonstrate that NASA's Mobile Laser Ranging System (MOBLAS) receiver configuration is limiting system performance below the 100 photoelectron level.

Zagwodzki, T. W.↗
NASA NTRS1987

Down-to-earth supercomputing at NASA Ames

The requirements for computers for aerospace vehicle design problems, in particular fluid dynamics, are examined. The levels of approximation to Navier-Stokes equations are: (1) linearized inviscid form, (2) nonlinear inviscid equation, (3) Reynolds-averaged Navier-Stokes equation, (4) large eddy simulation with models for subgrid scale turbulence, and (5) full Navier-Stokes equations to calculate laminar/turbulent transition and turbulence dissipation. The Cray 2 with 256 megaword memory and 4.1 nsec clock and the Cray X-MP/48 with a 128 megaword memory and a 9.5 nsec clock of the Ames facility are described, and an example revealing their applicability is presented. The use of zonal methods to simulate aerodynamic flow fields numerically is discussed. Consideration is given to the development of a program for simulation of aerodynamic flow fields, the compilers, symbolic debuggers, the scheduling of multiple CPUs, and the analysis and storage of the data.

Stevens, Kenneth G., Jr.↗