2005 Student Abstracts
Ames Research Center
Nicholas Mattei
University of Kentucky
The K10 rover is an autonomous robot designed and built by the Intelligent Robotics Group (IRG) at the NASA Ames Research Center. The robot is designed to provide an experimental test bed for human-robot interaction on future lunar missions. The robot has four wheels and is capable of speeds up to one meter per second. The only way to stop the robot in an emergency situation is to run up to the robot and press a large red emergency stop button. Due to the robot’s speed IRG requested the ability to stop the robot through wireless means. The Electronic Systems and Controls Branch provides quality, professional engineering solutions in support of the various projects at NASA Ames Research Center. The Electronic Systems and Controls Branch has a standing contract with IRG to provide support for the robot development whenever needed. With this in mind the development and implementation of a wireless emergency stop fell to the Electronic Systems and Controls Branch.
K10 has four wheels which each have their own associated motor. In order to stop the robot all four wheels, and their independent signals must stop together. To achieve this goal the control logic for the motor driver board was interrupted immediately in front of the final output signal. Thus the final output signal, which is logic low for drive and logic high for stop, is ORed with the input signal from the wireless controller. The wireless controller allows a user to set the other input of the OR gate to a logic high at any time thus stopping the rover. The add on circuitry for K10 consists of a daughterboard that attaches to the underside of each of the four existing motor driver boards and a wireless controller board that is affixed to the top of the rover and handles the wireless input and output. After an initial configuration of the emergency stop had been assembled and tested the requirements changed to include that the robot should not operate if the emergency stop was not in an operational state. This required the addition of some additional circuitry to the existing daughterboard design so that the output signal would be held high even if the wireless controller were no longer connected. After the modifications, the daughterboards were constructed using a Quick Circuit milling machine. Once the boards were created and populated they were affixed to the existing motor controller boards inside K10. The wireless controller card was placed in a box on the roof of K10 and attached to the daughterboards with a custom build connector. Testing proved successful and up to standard for IRG and the wireless emergency stop is now a permanent feature of K10 and future iterations of the hardware.
Glenn Research Center
Adrian Accurso
Dartmouth College
Diels-Alder trapping of a photochemically generated o-quinodimethane intermediate was employed to produce a novel perylene bisimide dye with a z-shaped structure. This compound was characterized by absorption and emission spectroscopy to determine radiative lifetime, quantum yield, and the extinction coefficient of the material. The synthesis of a methoxy derivative of the compound was also explored in an attempt to determine the modulating effect of varying electron donating and withdrawing groups on excited state properties of the dye. Future derivatives of interest and synthetic routes will also be discussed.
April L. Bohannan
University of Central Oklahoma
In this experiment, an interferometer technique that allows the simultaneous sorting and detecting of Laguerre-Gaussian modes was demonstrated. It utilizes two Mach-Zehnder interferometers in conjunction with a computer-generated hologram know as a kinoform. The orbital angular momentum associated with the Laguerre-Gaussian modes allow for the sorting and detecting which form the foundation of this design. Secondly, a three mode sorting system was demonstrated using this interferometer technique. This technique is an area of interest because the orbital angular momentum of a Laguerre-Gaussian beam has no theoretical limit on the number of quantum information bits that can be encoded on a single photon.
David Jackson
Georgia Institute of Technology
This paper discusses a new adaptive on-board modeling technique for aircraft engine performance diagnostics. First, the concept of model-based performance diagnostics is introduced and the conventional approach of using a Kalman filter estimator to adapt (tune) the on-board model is described. A shortcoming of the conventional on-board modeling technique is that the problem is typically underdetermined - there are more model unknowns than sensor measurements. To address this deficiency designers typically permit only a subset of model parameters to adjust. While this approach provides suitable model-to-engine output matching, the accuracy of non-measurable engine parameters is often poor. Previously it was shown that singular value decomposition (SVD) can be applied to define an optimal set of model tuning parameters which yield improved model estimation accuracy. Whereas previous work optimized the tuning parameters for a single engine operating point, this paper will describe an extension to the SVD technique, which will define a global set of engine tuning parameters permitting the technique to function over multiple operating points. Establishing a global consistent set of model tuning parameters is desirable for fault diagnostic purposes because this will enable the model to be updated to account for slow varying engine degradation, without absorbing abrupt faults into the model update. Estimation results are compared for the conventional Kalman filter approach, single point SVD-based estimation, and global SVD-based estimation. The global SVD-based tuner approach is shown to provide significantly improved state estimation over the conventional tuning approach, with only slightly degraded state estimation accuracy compared to single-point SVD-based estimation. The conventional approach was found to provide slightly better output estimation accuracy than the two SVD-based tuning approaches. Estimation accuracy of updating the model once per flight is also presented. The global SVD-based tuning technique greatly outperformed the conventional tuning technique when used for this purpose. While the global SVD-based tuning technique performed quite well at certain operating points it provided relatively large maximum estimation errors at other points making its practicality questionable. Potential enhancements to address this issue are being investigated.
Jet Propulsion Laboratory
Erinna Chen
Princeton University
A significant problem in the use of electric thrusters in spacecraft is the formation of low-energy ions in the thruster plume. Low-energy ions are formed in the plume via random collisions between high-velocity ions ejected from the thruster and slow-moving neutral atoms of propellant effusing from the engine. The sputtering of spacecraft materials due to interactions with low-energy ions may result in erosion or contamination of the spacecraft. The trajectory of these ions is determined primarily by the plasma potential of the plume. Thus, accurate characterization of the plasma potential is essential to predicting low-energy ion contamination. Emissive probes can be utilized to determine the plasma potential. When the ion and electron currents to the probe are balanced, the potential of such probes float to the plasma potential. Two emissive probes were fabricated; one utilizing a DC power supply, another utilizing a rectified AC power source. Labview programs were written to coordinate and automate probe motion in the thruster plume.
Employing handshaking interaction, these motion programs were synchronized to various data acquisition programs to ensure precision and accuracy of the measurements.
Comparing these experimental values to values from theoretical models will allow for a more accurate prediction of low-energy ion interaction.
Johnson Space Center
Aurora A. Robeson
Oregon Institute of Technology
Inhibited Propylene Glycol was tested at low temperatures to observe the freezing characteristics of the fluid. Thermal Control System (TCS) fluids can reach low temperature ranges during lunar missions and can affect the performance of equipment as well as fluid flow. Propylene Glycol may be used in spacecraft TCS for future missions to the moon and mars to transport heat from the spacecraft equipment and other sources, to the radiators where it is rejected to space. Dowfrost HD and water mixtures of 35, 50, and 60% were tested for density, viscosity and freezing/thaw properties and then compared to the manufacture’s data. All of three tests provided important information to determine if inhibited propylene glycol will be able to be used in a future TCS. The density test determined how much expansion or contraction would occur as a function of temperature. The density data showed that the fluids became denser at lower temperatures and the densities were in good agreement with published values. The viscosity test was used to determine at what temperatures the fluid could not be pumped. Viscosity proved to increase dramatically as temperature decreased, in some cases the texture of the fluid thickened to a non-pumpable state as temperature decreased. The freeze/thaw test has determined the most desirable glycol concentrations for burst protection of tubing or pipe used. This test was to quantify the forces generated as the fluid freezes and thaws. 50 and 60% solutions did not generate as much force as water or the 35% solution as they were subjected to 5 freeze/thaw cycles. The results obtained in these tests indicate that both the 50 and 60% solutions displayed viscosities, freezing temperatures, and low freeze/thaw forces which would make them desirable for use as a TCS fluid in future vehicles.
Kennedy Space Center
Curtis E. Groves
West Virginia University
The scope of this project was to complete a simulation (kinematic) of the, space shuttle, Orbiter to External Tank (ET) Mate. This simulation represents a horizontal and vertical transfer of the Orbiter through the Vehicle Assembly Building (VAB) and movement of major and extensible working Platforms. These simulations were created, as mechanisms, using the Design Mechanism module of Pro/ENGINEER. These mechanisms were completed by converting static constrains to dynamic connections and placing servo motors to control degrees of freedom. A number of parts and assemblies were completed to make the simulation more realistic. An assembly model of the Orbiter Transporter System (OTS) was completed. Assemblies of the OTS Orbiter forward and aft attach fixtures, Orbiter sling, and High Bay 3 Platforms were built. The following Orbiter sling parts were modeled: slide link, forward adapter, slide shoe, jack arm, cross beams, and hand wheels. Steel structure was added to the VAB. Emphasis was placed upon completing parts and assemblies that would enhance the virtual representation of the VAB environment. Computer file size was managed by using “simplified reps and shrinkwrap.” The Orbiter to ET Mate mechanism assembly was then used to compare and analyze the Orbiter to Platform critical clearance distances using different Orbiter hang angles. It was found that a 15 minute arc angle change in Orbiter hang angle affected distance at the platform critical spots as much as .550 inches. A Moving Picture Experts Group (MPEG) film clip of the Orbiter to ET Mate was developed, as a training aid, using Windows Movie Maker. The film clip may be viewed using Windows Media Player. This simulation may be used by Ground Support Equipment (GSE) personnel to emulate scenes and scenarios commonplace to the Orbiter to ET Mate. These parts, assemblies, and mechanisms may be viewed using Product View Express (a part, assembly, and drawing viewer), Pro/ENGINEER (a hybrid parametric solid modeling software), or Pro/INTRALINK (an engineering part and assembly management system). These parts, assemblies, mechanisms and movie clip may be used by engineers as “decision-making” tools.
Langley Research Center
Genny A. Pang
University of California at Los Angeles
The motivation for this study is to facilitate the generation of computational grids for complex geometries on the MSL by using unstructured grids and to learn how to apply the Fun3d code to an MSL RCS jet flow. A procedure was developed to generate an unstructured grid over the jet and obtain a flow solution. The unstructured-grid generation process took 95% less time than the structured-grid generation process. The solutions generated exhibited the expected behavior of a reaction control jet problem. However, a grid convergence study shows that further grid refinement is necessary. The methods developed in this study will be used to simulate the full MSL vehicle with Fun3d and the results will be compared with forthcoming experimental wind-tunnel data.
Los Alamos National Laboratory
Charles Beer
University of Nebraska-Lincoln
Magnetized target fusion is a concept for achieving fusion energy in a regime that is between the mainstream approaches (low-pressure magnetic fusion energy, MFE, and high pressure inertial confinement fusion, ICF). One issue with this intermediate method is the creation of suitable target plasma and the movement of that plasma into an area in which it can be physically compressed. A target with trapped magnetic field and closed magnetic flux surfaces is used in MTF because it confines the charged particles. Thus, the closed field lines in an MTF target greatly reduce particle and therefore energy losses. One very promising solution is known as a field-reversed configuration (FRC), a high plasma pressure equilibrium. An FRC is formed when a plasma that contains a magnetic field is suddenly exposed to a much stronger magnetic field shock in the opposite direction, causing the internal field lines to reconnect and form closed flux surfaces. This particular configuration is very advantageous because it remains stable with fairly high plasma pressure even when translated by an external magnetic field. With a computer code, I have simulated the ability of the plasma to translate, and examined the effects of different source geometries used to accelerate the plasma on formation, stability, and translation speed.
Michelle McMillan
Northern Arizona University
The effect of grazing on carbon sequestration in semiarid grasslands is still unknown. In this study we collected plant, soil, and air samples to measure their δ13C and δ15N values. The samples were collected from three exclosures in the Valles Caldera National Preserve. The exclosures represent heavy, light, and un-grazed areas. Within each exclosure we sampled at three locations, close to the river, at about 10 ft away from the river and at about 20 ft away from the river (or upland sites). The ratio of 13C to 12C (δ13C) is an indicator of stress in vegetation so we expect that with increased grazing impact and soil dryness, a signal should be apparent between exclosures as well as between the river and upland. The nitrogen isotopes give information on nitrogen source to the plants so any fertilizer left from the grazers should be observed in the sampled plants. We also took CO2 fluxes in each exclosure to compare the rate of carbon escaping into the atmosphere. The results from this study should help determine grazing impact to local ecologic systems and possibly provide information on carbon movement in grasslands due to grazing.
Marshall Space Flight Center
Geoffrey A. Hollinger
Swarthmore College
In this paper, a genetic algorithm (GA) is used to design fault-tolerant analog controllers for a piezoelectric micro-robot. A second-order function is developed to model the robot’s piezoelectric actuators, and the GA is used to evolve closed-loop controllers for this model. The GA is first used to assist in traditional PID design and is later used to synthesize variable topology analog controllers. Through the use of a compact circuit representation, runtimes are minimized and controllers are synthesized with minimum population sizes and components. Fault-tolerance is built into the fitness function to facilitate the design of controllers robust to both actuator failure and component failure. The GA is successfully used to design synthetic controllers and to optimize a traditional PID design. This research shows the advantages of GA assisted design when applied to robot-control problems.
Shayla Swain
Tuskegee University
In order for mankind to have long duration visits to the Moon and then venture to other planets such as Mars, habitats need to be developed for the crews to live and work in. One of the goals of the Thin Films and Inflatable (TFI) team of the In Situ Fabrication Resource (ISFR) Habitat Structures group at Marshall Space Fight Center is the development of polymer liners for habitats. This paper presents the results of an initial study that examines the potential use of EVOH for this application. EVOH is a copolymer of ethylene and vinyl alcohol that is a leading candidate because of its excellent oxygen barrier properties. Thin films were fabricated by rolling and compression molding and sealed by hot-plate welding and impulse sealing, to form a model habitats. Films that were sealed using the hot-plate welding method had excellent strength. When a sample was pulled apart, failure occurred in other portions of the film rather than the weld. In addition, a Hualian FS-100 Impulse Sealer was used to seal commercially available EVOH-nylon thin films; however, tests have to be performed to determine the strength of the seal.
Stennis Space Center
Ratessiea Lett
Jefferson Davis Community College
The E-Complex is one of Stennis Space Center’s most complex rocket propulsion testing facilities. Proper training and safety awareness courses are mandatory before any physical work can be done at the test stand, not only to remain safe but to better understand the hazards associated with particular operations. The current Integrated Powerhead Demonstration Engine in place at the E-1 stand uses Triethylaluminum/Triethylborane as an ignition source along with liquid oxygen and liquid hydrogen as propellants. The duties of a Junior Test Operations Engineer are nearly as demanding as those of other Test Operations Engineers in the field. Some duties include writing work authorization documents such as Test Preparation Sheets and Detailed Operating Procedures. Other endeavors completed during the research of a Test Operations Engineer consisted of producing an engine hot fire test database which includes valve timings and updating the IPD Engine camera map for E-1. With sufficient background knowledge of the cryogenics used at the stand and writing documents relating to maintenance of the stand, one can make a beneficial contribution to the test operations team.
Arden Moore
Louisiana Tech University
The purpose of this work is to evaluate the performance of the pressure-based, finite-rate viscous flow solver known as Loci-STREAM. Use of parallel computing resources was measured and determined to be most efficient when the number of computational cells per processor was set at 2,500. The solver’s performance on a problem with realistic conditions was shown using a test case of a 4-inch split-body valve similar to those used in rocket propulsion testing at Stennis Space Center. Two-dimensional axisymmetric simulations were completed for valve openings ranging from 5% to 100%. A three-dimensional half-plane simulation was also obtained for the valve at 100% open. The results predicted by Loci-STREAM for the valve flow coefficient, Cv, were then compared to those from another CFD code, FDNS (Finite-Difference Navier Stokes), and experimental data. For the 2D axisymmetric simulations, Loci-STREAM’s results were in better agreement with experimental values than FDNS for 7 of the 11 cases tested, but still failed to accurately model the valve at operating percentages over 40%. For the 3D simulation, Loci-STREAM’s results were within 15.1% of the experimental value of Cv and within 8% of the results returned by two other CFD codes, FDNS and CRUNCH.
Wallops Flight Facility
Hamsa Jaganathan
Purdue University
Phaeocystis antarctica is a phytoplankton that lives in the Antarctic Ocean. This species is constantly tested at NASA’s Hydrospheric and Biospheric Sciences Laboratory in Wallops Flight Facility. The f/2 culturing medium is used to grow the Phaeocystis antarctica at NASA. However, other institutes have claimed that the L1 culturing medium is better. This experiment tests the growth rates of Phaeocystis antarctica in the L1 medium and the f/2 medium. By measuring the fluorescence emitted by Phaeocystis antarctica in the two different media each day, the growth rate was calculated. The rate of the L1 medium was compared to the rate of the f/2 medium. Also, using statistical analysis, the significance of the increase of fluorescence each day was found. After ten days of experiments, results showed that the Phaeocystis antarctica in f/2 medium had a higher rate of growth. In the 95% confidence level of the statistical test, the results showed no significant difference of fluorescence emitted from Phaeocystis antarctica between the two media.
White Sands Test Facility
Suzanne Hayward
University of Washington
This paper addresses the goal of investigating various means of improving the signal-to-noise ratio of the 17-caliber two stage light gas gun’s velocity measuring system. With enough improvement this system could also be used to trigger the high-speed cameras recording the projectile impact. The different methods investigated include DC blocking, polarization, reflection, and optical correlation using Fraunhofer patterns. So far the best candidate techniques are Fraunhofer pattern optical correlation and circular polarization, though they require more study. Additionally, shrouding the detectors may be helpful in blocking out background noise and improving the clarity of signal dips.
