Search NASASearch

SEARCH · Search NASA

Results for “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 145 records · Page 8

A forward method for optimal stochastic nonlinear and adaptive control

A computational approach is taken to solve the optimal nonlinear stochastic control problem. The approach is to systematically solve the stochastic dynamic programming equations forward in time, using a nested stochastic approximation technique. Although computationally intensive, this provides a straightforward numerical solution for this class of problems and provides an alternative to the usual dimensionality problem associated with solving the dynamic programming equations backward in time. It is shown that the cost degrades monotonically as the complexity of the algorithm is reduced. This provides a strategy for suboptimal control with clear performance/computation tradeoffs. A numerical study focusing on a generic optimal stochastic adaptive control example is included to demonstrate the feasibility of the method.

Bayard, David S.

Neural Architectures for Control

The cerebellar model articulated controller (CMAC) neural architectures are shown to be viable for the purposes of real-time learning and control. Software tools for the exploration of CMAC performance are developed for three hardware platforms, the MacIntosh, the IBM PC, and the SUN workstation. All algorithm development was done using the C programming language. These software tools were then used to implement an adaptive critic neuro-control design that learns in real-time how to back up a trailer truck. The truck backer-upper experiment is a standard performance measure in the neural network literature, but previously the training of the controllers was done off-line. With the CMAC neural architectures, it was possible to train the neuro-controllers on-line in real-time on a MS-DOS PC 386. CMAC neural architectures are also used in conjunction with a hierarchical planning approach to find collision-free paths over 2-D analog valued obstacle fields. The method constructs a coarse resolution version of the original problem and then finds the corresponding coarse optimal path using multipass dynamic programming. CMAC artificial neural architectures are used to estimate the analog transition costs that dynamic programming requires. The CMAC architectures are trained in real-time for each obstacle field presented. The coarse optimal path is then used as a baseline for the construction of a fine scale optimal path through the original obstacle array. These results are a very good indication of the potential power of the neural architectures in control design. In order to reach as wide an audience as possible, we have run a seminar on neuro-control that has met once per week since 20 May 1991. This seminar has thoroughly discussed the CMAC architecture, relevant portions of classical control, back propagation through time, and adaptive critic designs.

Peterson, James K.

An assessment of separable fluid connector system parameters to perform a connector system design optimization study

The key to optimization of design where there are a large number of variables, all of which may not be known precisely, lies in the mathematical tool of dynamic programming developed by Bellman. This methodology can lead to optimized solutions to the design of critical systems in a minimum amount of time, even when there are a great number of acceptable configurations to be considered. To demonstrate the usefulness of dynamic programming, an analytical method is developed for evaluating the relationship among existing numerous connector designs to find the optimum configuration. The data utilized in the study were generated from 900 flanges designed for six subsystems of the S-1B stage of the Saturn 1B space carrier vehicle.

Prasthofer, W. P.

Satellite-tracking and Earth dynamics research programs

The Arequipa station obtained a total of 31,989 quick-look range observations on 719 passes in the six months. Data were acquired from Metsahovi, San Fernando, Kootwijk, Wettzell, Grasse, Simosato, Graz, Dodaira and Herstmonceux. Work progressed on the setup of SAO 1. Discussions were also initiated with the Israelis on the relocation of SAO-3 to a site in southern Israel in FY-1984. Arequipa and the cooperating stations continued to track LAGEOS at highest priority for polar motion and Earth rotation studies, and for other geophysical investigations, including crustal dynamics, earth and ocean tides, and the general development of precision orbit determination. SAO completed the revisions to its field software as a part of its recent upgrading program. With cesium standards Omega receivers, and other timekeeping aids, the station was able to maintain a timing accuracy of better than plus or minus 6 to 8 microseconds.

Source record

Free vibration characteristics of multiple load path blades by the transfer matrix method

The determination of free vibrational characteristics is basic to any dynamic design, and these characteristics can form the basis for aeroelastic stability analyses. Conventional helicopter blades are typically idealized as single-load-path blades, and the transfer matrix method is well suited to analyze such blades. Several current helicopter dynamic programs employ transfer matrices to analyze the rotor blades. In this paper, however, the transfer matrix method is extended to treat multiple-load-path blades, without resorting to an equivalent single-load-path approximation. With such an extension, these current rotor dynamic programs which employ the transfer matrix method can be modified with relative ease to account for the multiple load paths. Unlike the conventional blades, the multiple-load-path blades require the introduction of the axial degree-of-freedom into the solution process to account for the differential axial displacements of the different load paths. The transfer matrix formulation is validated through comparison with the finite-element solutions.

Murthy, V. R.

TSS subsatellite attitude dynamics and control laws verification programs

A dynamic model of the Tethered Satellite System (TSS) and of the relevant simulation program, developed in order to provide the dynamic analysis support for the design verification of the subsatellite attitude control, is presented. Special care was spent in the satellite attitude dynamic analysis and the model was specifically conceived to this aim. The way in which the simulation results can be utilized for the verification and testing of the attitude control is also presented.

Venditti, Floriano

Modeling human decision making behavior in supervisory control

An optimal decision control model was developed, which is based primarily on a dynamic programming algorithm which looks at all the available task possibilities, charts an optimal trajectory, and commits itself to do the first step (i.e., follow the optimal trajectory during the next time period), and then iterates the calculation. A Bayesian estimator was included which estimates the tasks which might occur in the immediate future and provides this information to the dynamic programming routine. Preliminary trials comparing the human subject's performance to that of the optimal model show a great similarity, but indicate that the human skips certain movements which require quick change in strategy.

Tulga, M. K.

Satellite tracking and earth dynamics research programs

The SAO laser site in Arequipa continued routine operations throughout the reporting period except for the months of March and April when upgrading was underway. The laser in Orroral Valley was operational through March. Together with the cooperating stations in Wettzell, Grasse, Kootwikj, San Fernando, Helwan, and Metsahove the laser stations obtained a total of 37,099 quick-look observations on 978 passes of BE-C, Starlette, and LAGEOS. The Network continued to track LAGEOS at highest priority for polar motion and Earth rotation studies, and for other geophysical investigations, including crustal dynamics, Earth and ocean tides, and the general development of precision orbit determination. The Network performed regular tracking of BE-C and Starlette for refined determinations of station coordinate and the Earth's gravity field and for studies of solid earth dynamics. Monthly statistics of the passes and points are given by station and by satellite.

Source record

Linear and nonlinear dynamic analysis of redundant load path bearingless rotor systems

The bearingless rotorcraft offers reduced weight, less complexity and superior flying qualities. Almost all the current industrial structural dynamic programs of conventional rotors which consist of single load path rotor blades employ the transfer matrix method to determine natural vibration characteristics because this method is ideally suited for one dimensional chain like structures. This method is extended to multiple load path rotor blades without resorting to an equivalent single load path approximation. Unlike the conventional blades, it isk necessary to introduce the axial-degree-of-freedom into the solution process to account for the differential axial displacements in the different load paths. With the present extension, the current rotor dynamic programs can be modified with relative ease to account for the multiple load paths without resorting to the equivalent single load path modeling. The results obtained by the transfer matrix method are validated by comparing with the finite element solutions. A differential stiffness matrix due to blade rotation is derived to facilitate the finite element solutions.

Murthy, V. R.

Andean tectonics: Implications for Satellite Geodesy

Current knowledge and theories of large scale Andean tectonics as they relate to site planning for the NASA Crustal Dynamics Program's proposed high precision geodetic measurements of relative motions between the Nazca and South American plates are summarized. The Nazca Plate and its eastern margin, the Peru-Chile Trench, is considered a prototype plate marked by rapid motion, strong seismicity and well defined boundaries. Tectonic activity across the Andes results from the Nazca Plate subducting under the South American plate in a series of discrete platelets with different widths and dip angles. This in turn, is reflected in the tectonic complexity of the Andes which are a multitutde of orogenic belts superimposed on each other since the Precambrian. Sites for Crustal Dynamics Program measurements are being located to investigate both interplate and extraplate motions. Observing operations have already been initiated at Arequipa, Peru and Easter Island, Santiago and Cerro Tololo, Chile. Sites under consideration include Iquique, Chile; Oruro and Santa Cruz, Bolivia; Cuzco, Lima, Huancayo and Bayovar, Peru; and Quito and the Galapagos Islands, Ecuador. Based on scientific considerations, Santa Cruz, Huancayo (or Lima), Quito and the Galapagos Islands should be replaced by Isla San Felix, Chile; Brazilia or Petrolina, Brazil; and Guayaquil, Ecuador. If resources permit, additional important sites would be Buenaventura and Villavicencio or Puerto La Concordia, Colombia; and Mendoza and Cordoba, Argentina.

Allenby, R. J.

Exploring Applications of Machine Learning for Wildfire Monitoring and Detection using Unmanned Aerial Vehicles

Wildfires are increasing in frequency and severity around the world, including the United States. The losses caused by wildfires could be mitigated if high-risk areas, hotspots, and flare-ups could be monitored continuously, such as through the use of Unmanned Aerial Vehicles (UAVs). This paper documents exploratory efforts using machine learning to determine efficient flight paths for UAVs and to detect wildfires using image classification. On path planning, three machine learning techniques—Genetic Algorithm, Simulated Annealing, and Dynamic Programming—were explored. Genetic Algorithm was found to be an effective approach for path planning for wildfire monitoring and surveillance by UAVs. For a scenario of 25 locations in a circular arrangement, the algorithm was able to return the optimal path. The accuracy and execution time was found to be sensitive to the algorithm hyperparameters selected, which was especially evident in scenarios with hundreds or thousands of locations. Simulated Annealing was also found to be an effective approach for UAV path planning, with a major benefit of avoiding getting trapped in local minima and being straightforward to implement. Like Genetic Algorithm, the performance of Simulated Annealing was also found to be sensitive to the algorithm hyperparameters selected. By comparison, Dynamic Programming guarantees optimality for any number of locations, but it was found to be less practical in terms of execution time for scenarios with more than about a couple dozen locations. On wildfire detection, image classification using deep learning with a convolutional neural network was explored. Transfer learning was found to be a useful technique to efficiently train deep learning models. Also, it was determined that GPU processing can increase training speed by an order of magnitude, which enables significantly faster development. For a validation test set of 500 images, there were only two false negatives and zero false positives. These results demonstrate that detecting wildfires in static cameras using machine learning is feasible and establish a baseline for using images captured by UAVs in flight for wildfire detection.

Wildfire management

Land landing couch dynamics computer program

Computer programs perform landing stability studies of mechanical impact system designs for advanced spacecraft. The programs consider variation in spacecraft vertical and horizontal velocity, attitude and orientation, shock strut load-stroke characteristics, and ground coefficient of friction.

Herting, D. N.

Satellite-tracking and earth-dynamics research programs

Observations and research progress of the Smithsonian Astrophysical Observatory are reported. Satellite tracking networks (ground stations) are discussed and equipment (Baker-Nunn cameras) used to observe the satellites is described. The improvement of the accuracy of a laser ranging system of the ground stations is discussed. Also, research efforts in satellite geodesy (tides, gravity anomalies, plate tectonics) is discussed. The use of data processing for geophysical data is examined, and a data base for the Earth and Ocean Physics Applications Program is proposed. Analytical models of the earth's motion (computerized simulation) are described and the computation (numerical integration and algorithms) of satellite orbits affected by the earth's albedo, using computer techniques, is also considered. Research efforts in the study of the atmosphere are examined (the effect of drag on satellite motion), and models of the atmosphere based on satellite data are described.

Source record

Satellite-tracking and earth-dynamics research programs

The activities carried out by the Smithsonian Astrophysical Observatory (SAO) for the National Aeronautics and Space Administration (NASA) during the period 1 July to 31 December 1978 are described. All four SAO laser sites were in routine operation during the reporting period; they obtained a 6-month total of 182,529 range observations on 3610 passes.

Source record

Satellite tracking and Earth dynamics research programs

Following an upgrading program, ranging performance capabilities of a satellite-tracking pulsed laser system were assessed in terms of range accuracy, range noise, data yield, and reliability. With a shorter laser pulse duration (2.5 to 3.0 NSEC) and a new analog pulse processing system, the systematic range errors were reduced to 3 to 5 cm and range noise was reduced to 5 to 16 cm and range noise was reduced to 5 to 15 cm on Starlette and BE-C, and 10 to 18 cm on LAGEOS. Maximum pulse repetition rate was increased to 30 pulses per minute and significant improvement was made in signal to noise ratio by installing a 3 A interference filter and by reducing the range gate window to 200 to 400 nsec. The solution to a problem involving leakage of a fraction of the laser oscillator pulse through the pulse chopper was outlined.

Pearlman, M. R.

Calculation of rotor impedance for use in design analysis of helicopter airframe vibrations

Excessive vibration is one of the most prevalent technical obstacles encountered in the development of new rotorcraft. The inability to predict these vibrations is primarily due to deficiencies in analysis and simulation tools. The Langley Rotorcraft Structural Dynamics Program was instituted in 1984 to meet long term industry needs in the area of rotorcraft vibration prediction. As a part of the Langley program, this research endeavors to develop an efficient means of coupling the rotor to the airframe for preliminary design analysis of helicopter airframe vibrations. The main effort was to modify the existing computer program for modeling the dynamic and aerodynamic behavior of rotorcraft called DYSCO (DYnamic System COupler) to calculate the rotor impedance. DYSCO was recently developed for the U.S. Army and has proven to be adaptable for the inclusion of new solution methods. The solution procedure developed to use DYSCO for the calculation of rotor impedance is presented. Verification of the procedure by comparison with a known solution for a simple wind turbine model is about 75 percent completed, and initial results are encouraging. After the wind turbine impedance is confirmed, the verification effort will continue by comparison to solutions of a more sophisticated rotorcraft model. Future work includes determination of the sensitivity of the rotorcraft airframe vibrations to helicopter flight conditions and rotor modeling assumptions. When completed, this research will ascertain the feasibility and efficiency of the impedance matching method of rotor-airframe coupling for use in the analysis of airframe vibrations during the preliminary rotorcraft design process.

Nygren, Kip P.

Program Structure Combines Segmentation and Dynamic Storage

Programing techniques incorporate advantages of overlaying into segmented loads while retaining all dynamic load advantages of segmentation, employing those capabilities that best suit mode of operation, whether batch or interactive. User is allowed to load a program automatically in a variable manner, based solely on a single data input to the program, to maintain minimal field lengths for interactive use.

Tiffany, S. H.