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.
Two schemes for reception of weak radio signals conveying digital data via phase modulation provide for mutual coupling of multiple receivers, and coherent combination of outputs of receivers. In both schemes, optimal mutual-coupling weights computed according to Kalman-filter theory, but differ in manner of transmission and combination of outputs of receivers.
A design method for a servo compensator is developed in the frequency domain using singular values. The method is applied to a reusable rocket engine. An intelligent control system for reusable rocket engines was proposed which includes a diagnostic system, a control system, and an intelligent coordinator which determines engine control strategies based on the identified failure modes. The method provides a means of generating various linear multivariable controllers capable of meeting performance and robustness specifications and accommodating failure modes identified by the diagnostic system. Command following with set point control is necessary for engine operation. A Kalman filter reconstructs the state while loop transfer recovery recovers the required degree of robustness while maintaining satisfactory rejection of sensor noise from the command error. The approach is applied to the design of a controller for a rocket engine satisfying performance constraints in the frequency domain. Simulation results demonstrate the performance of the linear design on a nonlinear engine model over all power levels during mainstage operation.
A detailed discussion of Markov parameters in system identification is given. Different forms of input-output representation of linear discrete-time systems are reviewed and discussed. Interpretation of sampled response data as Markov parameters is presented. Relations between the state-space model and particular linear difference models via the Markov parameters are formulated. A generalization of Markov parameters to observer and Kalman filter Markov parameters for system identification is explained. These extended Markov parameters play an important role in providing not only a state-space realization, but also an observer/Kalman filter for the system of interest.
During the past few years considerable insight was gained into the QUEST algorithm both as a maximum likelihood estimator and as a Kalman filter/smoother for systems devoid of dynamical noise. The new algorithms and software are described and analytical comparisons are made with the more conventional attitude Kalman filter. It is also described how they may be accommodated to noisy dynamical systems.
The uncertainty in the geopotential model is currently one of the major error sources in the orbit determination of low-altitude Earth-orbiting spacecraft. The results of an investigation of different geopotential error models and modeling approaches currently used for operational orbit error analysis support at the Goddard Space Flight Center (GSFC) are presented, with emphasis placed on sequential orbit error analysis using a Kalman filtering algorithm. Several geopotential models, known as the Goddard Earth Models (GEMs), were developed and used at GSFC for orbit determination. The errors in the geopotential models arise from the truncation errors that result from the omission of higher order terms (omission errors) and the errors in the spherical harmonic coefficients themselves (commission errors). At GSFC, two error modeling approaches were operationally used to analyze the effects of geopotential uncertainties on the accuracy of spacecraft orbit determination - the lumped error modeling and uncorrelated error modeling. The lumped error modeling approach computes the orbit determination errors on the basis of either the calibrated standard deviations of a geopotential model's coefficients or the weighted difference between two independently derived geopotential models. The uncorrelated error modeling approach treats the errors in the individual spherical harmonic components as uncorrelated error sources and computes the aggregate effect using a combination of individual coefficient effects. This study assesses the reasonableness of the two error modeling approaches in terms of global error distribution characteristics and orbit error analysis results. Specifically, this study presents the global distribution of geopotential acceleration errors for several gravity error models and assesses the orbit determination errors resulting from these error models for three types of spacecraft - the Gamma Ray Observatory, the Ocean Topography Experiment, and the Cosmic Background Explorer.
This work introduces, examines, and compares several quaternion normalization algorithms, which are shown to be an effective stage in the application of the additive extended Kalman filter (EKF) to spacecraft attitude determination, which is based on vector measurements. Two new normalization schemes are introduced. They are compared with one another and with the known brute force normalization scheme, and their efficiency is examined. Simulated satellite data are used to demonstrate the performance of all three schemes. A fourth scheme is suggested for future research. Although the schemes were tested for spacecraft attitude determination, the conclusions are general and hold for attitude determination of any three dimensional body when based on vector measurements, and use an additive EKF for estimation, and the quaternion for specifying the attitude.
Advanced aircraft will require flight critical computer systems for stability augmentation as well as guidance and control that must perform reliably in adverse, as well as nominal, operating environments. Digital system upset is a functional error mode that can occur in electromagnetically harsh environments, involves no component damage, can occur simultaneously in all channels of a redundant control computer, and is software dependent. A strategy is presented for dynamic upset detection to be used in the evaluation of critical digital controllers during the design and/or validation phases of development. Critical controllers must be able to be used in adverse environments that result from disturbances caused by an electromagnetic source such as lightning, high intensity radiated field (HIRF), and nuclear electromagnetic pulses (NEMP). The upset detection strategy presented provides dynamic monitoring of a given control computer for degraded functional integrity that can result from redundancy management errors and control command calculation error that could occur in an electromagnetically harsh operating environment. The use is discussed of Kalman filtering, data fusion, and decision theory in monitoring a given digital controller for control calculation errors, redundancy management errors, and control effectiveness.
Nine cooperating rule-based systems, collectively called AUTOCREW, were designed to automate functions and decisions associated with a combat aircraft's subsystem. The organization of tasks within each system is described; performance metrics were developed to evaluate the workload of each rule base, and to assess the cooperation between the rule-bases. Each AUTOCREW subsystem is composed of several expert systems that perform specific tasks. AUTOCREW's NAVIGATOR was analyzed in detail to understand the difficulties involved in designing the system and to identify tools and methodologies that ease development. The NAVIGATOR determines optimal navigation strategies from a set of available sensors. A Navigation Sensor Management (NSM) expert system was systematically designed from Kalman filter covariance data; four ground-based, a satellite-based, and two on-board INS-aiding sensors were modeled and simulated to aid an INS. The NSM Expert was developed using the Analysis of Variance (ANOVA) and the ID3 algorithm. Navigation strategy selection is based on an RSS position error decision metric, which is computed from the covariance data. Results show that the NSM Expert predicts position error correctly between 45 and 100 percent of the time for a specified navaid configuration and aircraft trajectory. The NSM Expert adapts to new situations, and provides reasonable estimates of hybrid performance. The systematic nature of the ANOVA/ID3 method makes it broadly applicable to expert system design when experimental or simulation data is available.
Pilot aiding during low-altitude flight depends on the ability to detect and locate obstacles near the helicopter's intended flightpath. Computer-vision-based methods provide one general approach for obstacle detection and range estimation. Several algorithms have been developed for this purpose, but have not been tested with actual flight data. This paper presents results obtained using helicopter flight data with a feature-based range estimation algorithm. A method for recursively estimating range using a Kalman filter with a monocular sequence of images and knowledge of the camera's motion is described. The helicopter flight experiment and four resulting datasets are discussed. Finally the performance of the range estimation algorithm is explored in detail based on comparison of the range estimates with true range measurements collected during the flight experiment.
The 40-50-d oscillation in length-of-day (LOD) and AAM is investigated using the 12-yr overlap between two records: AAM data, compiled from the NMC, and LOD variation from the JPL Kalman-filtered earth rotation series. The NMC records are analyzed by latitude belts, in the light of the UCLA GCM results, in order to identify possibly distinct sources of the AAM oscillation in the midlatitudes and the tropics. Results suggest that two intraseasonal oscillations exist in the earth-atmosphere system: a tropical 50-d oscillation associated with the convectively driven waves of the type first described by Madden and Julian (1971) and a midlatitude 40-d oscillation associated with the interaction of nonzonal flow with topography.
Free, equatorially trapped sinusoidal wave solutions to a linear model on an equatorial beta plane are used to fit the Geosat altimetric sea level observations in the tropical Pacific Ocean. The Kalman filter technique is used to estimate the wave amplitude and phase from the data. The estimation is performed at each time step by combining the model forecast with the observation in an optimal fashion utilizing the respective error covariances. The model error covariance is determined such that the performance of the model forecast is optimized. It is found that the dominant observed features can be described qualitatively by basin-scale Kelvin waves and the first meridional-mode Rossby waves. Quantitatively, however, only 23 percent of the signal variance can be accounted for by this simple model.
Equatorial atmospheric angular momentum (AAM) excitation functions and polar motion excitation functions (derived by Kalman filtering Very Long Baseline Interferometry polar motion estimates) are compared with the times of 1984-mid-1988 large earthquakes (magnitude greater than or equal to 7.5). There is a moderate correlation between times of large earthquakes and peaks in polar motion excitation. A strong correlation exists between the times of large earthquakes and large peaks in equatorial AAM amplitude; such a correlation is evident for six out of the eight large earthquakes occurring over the studied time interval. The AAM results indicate potential for the temporal prediction of large/great earthquakes.
A system identification model that combines generalized-spline function approximation with a nonlinear control system is described. The complete control system contains three main elements: a nonlinear-inverse-dynamic control law that depends on a comprehensive model of the plant, a state estimator whose outputs drive the control law, and a function approximation scheme that models the system dynamics. The system-identification task, which combines an extended Kalman filter with a function approximator modeled as an artificial neural network, is considered. The results of an application of the identification techniques to a nonlinear transport aircraft model are presented.
Data from a worldwide Global Positioning System (GPS) tracking experiment have been used to determine variations in earth rotation (UT1-UTC) over a time period of three weeks. Kalman filtering and smoothing enabled changes in UT1-UTC over intervals of 2 to 24 hrs to be detected with the GPS data. Internal consistency checks and comparisons with other solutions from very long baseline interferometry (VLBI) and satellite laser ranging (SLR) indicate that the GPS UT1-UTC estimates are accurate to about 2 cm. Comparison of GPS-estimated variations in UT1-UTC with 2-hr time resolution over 4 days with predicted variations computed from diurnal and semidiurnal oceanic tidal contributions strongly suggests that the observed periodic sub-daily variations of about 0.1 msec are largely of tidal origin.
The estimation of dynamic states for feedback control of structural systems using second-order differential equations and acceleration measurements is described. The formulation of the observer model, and the design of the observer gains is discussed in detail. It is shown the second-order observer is highly stable because the stability constraints on the observer gains are model independent. The limitation of the proposed observer is the need for 'nearly' collocated actuators and accelerometers. Experimental results using a control-structure interaction testbed are presented that show the second-order observer provided more stability than a Kalman filter estimator without decreasing closed-loop performance.
This paper describes the development of an autonomous integrated spacecraft navigation system which provides multiple modes of navigation, including relative and absolute navigation. The system provides attitude information from GPS or INS, or by tightly integrating the two systems. Interferometric GPS techniques are used when multiple antennas and integrated Doppler measurements are available. An important aspect of this research is the autonomously reconfigurable Kalman filter, controlled by an embedded knowledge base, designed to respond to component degradation and changes in mission goals.
The research airdata system of an instrumented F-104 aircraft has been calibrated to measure winds aloft in support of the Space Shuttle wind measurement investigation at the National Aeronautics and Space Administration Ames Research Center Dryden Flight Research Facility. For this investigation, wind measurement accuracies comparable to those obtained from Jimsphere balloons were desired. This required an airdata calibration more accurate than needed for most aircraft research programs. The F-104 aircraft was equipped with a research pilot-static noseboom with integral angle-of-attack and flank angle-of-attack vanes and a ring-laser-gyro inertial reference unit. Tower fly-bys and radar acceleration-decelerations were used to calibrate Mach number and total temperature. Angle of attack and angle of side slip were calibrated with a trajectory reconstruction technique using a multiple-state linear Kalman filter. The F-104 aircrat and instrumentation configuration, flight test maneuvers, data corrections, calibration techniques, and resulting calibrations and data repeatability are presented. Recommendations for future airdata systems on aircraft used to measure winds aloft are also given.
Methods are presented to normalize the attitude quaternion in two extended Kalman filters (EKF), namely, the multiplicative EKF (MEKF) and the additive EKF (AEKF). It is concluded that all the normalization methods work well and yield comparable results. In the AEKF, normalization is not essential, since the data chosen for the test do not have a rapidly varying attitude. In the MEKF, normalization is necessary to avoid divergence of the attitude estimate. All of the methods of the methods behave similarly when the spacecraft experiences low angular rates.