APPLICATION OF SELF-ADAPTIVE CONTROL TECHNIQUES TO THE FLEXIBLE SUPERSONIC TRANSPORT
Limit-cycle and model-reference, self-adaptive techniques for designing the flight control systems of flexible supersonic transport aircraft
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.
Limit-cycle and model-reference, self-adaptive techniques for designing the flight control systems of flexible supersonic transport aircraft
Significance of solar geomagnetic processes in determining distribution of ionospheric components and implications for reference models of ionosphere
Description of design reference model for space shuttle propulsion system onboard checkout and monitoring equipment to establish design criteria - Vol. 2
An algorithm for design via parameter optimization has been developed for linear-time-invariant control systems based on the model reference adaptive control concept. A cost functional is defined to evaluate the system response relative to nominal, which involves in general the error between the system and nominal response, its derivatives and the control signals. A program for the practical implementation of this algorithm has been developed, with the computational scheme for the evaluation of the performance index based on Lyapunov's theorem for stability of linear invariant systems.
During intervals in 1967 to 1970, the OGO-4 and 6 spacecraft made over 2000 traversals over the equatorial electrojet in the altitude range 400-800 km when local times were between 9 and 15 hours. These spacecraft carried total field magnetometers making measurements to an accuracy of 2 gamma with a sample rate greater than once a second. Delta F values, the deviations from these observations, were formed from an internal reference model. The results were plotted for a 30 deg band about the equator, and the characteristics of the electrojet effect in the data were investigated. This effect was characterized by a sharp negative V-signature of some 16-19 deg in width and a variable amplitude. The position of this minimum was found to lie within 0.5 deg of the dip equator. A slight northward shift was noted at the longitude of Huancayo. The jet amplitudes were normalized to 400 km amplitudes and observed to be highly variable in time. Amplitudes over the longitude range 50 to 90 deg W averaged 60% higher than elsewhere, as expected, due to the weaker main field. However, though the scatter of amplitudes is high, the expected minima in east Asia was not evident. It was speculated that this could be due to a less conducting upper mantle in this area.
In order to minimize the effects of parameter variations in the dynamics of an optical tracking telescope, a model referenced parameter adaptive control system is described that - in conjunction with more traditional forms of compensation - achieves a reduction of rms pointing error by more than a factor of six. The adaptive compensation system utilizes open loop compensation, closed loop compensation, and model reference compensation to provide the precise input to force telescope axis velocity to follow the ideal velocity.
Wind tunnel tests to determine the longitudinal and lateral-directional effects of control surface modifications on the space shuttle orbiter aerodynamic characteristics are discussed. A total of 103 data runs were made which consisted of pitch runs through a range of zero to 28 degrees at a zero yaw angle and yaw runs from minus 6 to plus 6 degrees at various fixed pitch angles. At each data point, data from an internal strain gage balance was sampled with the digital data system. Also recorded were the model angles of pitch and yaw and the test section static pressure. Results are presented in the form of tabulated aerodynamic coefficient data about the model reference center.
The design specifications for the programs and modules within the NASA Aerospace Safety Information System (NASIS) are presented. The purpose of the design specifications is to standardize the preparation of the specifications and to guide the program design. Each major functional module within the system is a separate entity for documentation purposes. The design specifications contain a description of, and specifications for, all detail processing which occurs in the module. Sub-models, reference tables, and data sets which are common to several modules are documented separately.
The adaptive control and identification problems for single input, single output, time invariant systems with unknown parameters are considered here. A model reference is used and Liapunov's direct method provides the mechanisms for deriving the adaptive and identification algorithms. The Kalman Lemma (derived in connection with the Lure' problem) is shown to be extremely helpful in the design of these systems.
During intervals in 1967-1970, the OGO-4 and 6 spacecraft made over 2000 traversals over the equatorial electrojet in the altitude range 400-800 km when local times were between 9 and 15 hr. These spacecraft carried total field magnetometers making measurements to an accuracy of 2 gammas with a sample rate greater than once a second. Delta-F values, the deviations of these observations from an internal reference model, were plotted for a 30 deg band about the Equator, and the characteristics of the electrojet effect in the data were investigated. This effect was characterized by a sharp negative V-signature of some 16-19 deg in width and a variable amplitude.
The past contributions of NASA to adaptive control technology are reviewed. The review places emphasis on aircraft applications although spacecraft and launch vehicle control applications are included. Particular emphasis is given to the adaptive control system used in the X-15 research aircraft. Problem areas that limited the realizable performance of this adaptive system are discussed. Current technological capabilities are used to extrapolate the present-day potential for adaptive flight control. Specifically, the potential created by use of the modern high-speed digital computer in flight control is discussed. Present plans for research in digital adaptive control systems for the NASA F8-C digital fly-by-wire program are presented. These plans are currently envisioned to include research in at least two types of adaptive controls, the system identification/on-line design type, and the model reference type.
Transonic wind-tunnel studies were conducted with modified cross, hemisflo, and disk-gap-band parachute models in the wake of a cone-cylinder shape forebody. The basic cross design was modified with the addition of a circumferential constraining band at the lower edge of the canopy panels. The tests covered a Mach number range of 0.3 to 1.2 and a dynamic pressure range from 479 Newtons per square meter to 5746 Newtons per square meter. The parachute models were flexible textile-type structures and were tethered to a rigid forebody with a single flexible riser. Different size models of the modified cross and disk-gap-band canopies were tested to evaluate scale effects. Model reference diameters were 0.30, 0.61, and 1.07 meters (1.0, 2.0, and 3.5 ft) for the modified cross; and nominal diameters of 0.25 and 0.52 meter (0.83 and 1.7 ft) for the disk-gap-band; and 0.55 meter (1.8 ft) for the hemisflo. Reefing information is presented for the 0.61-meter-diameter cross and the 0.52-meter-diameter disk-gap-band. Results are presented in the form of the variation of steady-state average drag coefficient with Mach number. General stability characteristics of each parachute are discussed. Included are comments on canopy coning, spinning, and fluttering motions.
Conventional and adaptive attitude control of spacecraft which use control moment gyros (CMG's) as torque sources are discussed. Control laws predicated on the assumption of a linear system are used since the spacecraft equations of motion are formulated in an 'eigenaxis system' so that they are essentially linear during 'slow' maneuvers even if large angles are involved. The overall control schemes are 'optimal' in several senses. Eigenaxis rotations and a weighted pseudo-inverse CMG steering law are used and, in the adaptive case, a Model Reference Adaptive System (MRAS) controller based on Liapunov's Second Method is adopted. To substantiate the theory, digital simulation results obtained using physical parameters of a Large Space Telescope type spacecraft are presented. These results indicate that an adaptive control law is often desirable.
Use is made of linearized error characteristic equation for model-reference adaptive systems to determine a parameter adjustment rule for obtaining time-invariant error dynamics. Theoretical justification of error stability is given and an illustrative example included to demonstrate the utility of the proposed technique.
A class of dynamical systems in a conflict situation is formulated and discussed, and the formulation is applied to the study of an important class of systems in the presence of uncertainty. The uncertainty is deterministic and the only assumption is that its value belongs to a known compact set. Asymptotic stability is fully discussed with application to variable structure and model reference control systems.
Model reference adaptive controllers for multivariable plants are designed using only input and output signals rather than the complete state vector. The design is based on Liapunov's direct method, and the Meyer-Kalman-Yacubovich lemma. State variable filters are employed to avoid differentiating output signals. Augmented error signals are used in deriving stable adaptive control laws which assure that the normally used response error signals approach zero asymptotically.
A new nonlinear stability criterion is developed by use of a class of Lyapunov functionals for model-reference adaptive systems (MRAS). Results are compared with traditional results, and a comparative design technique is used to illustrate its function in improving the transient response of an MRAS controller. For a particular system structure and class of input signals, the new stability criterion is shown to include traditional sufficiency stability conditions as a special case. An example is cited to illustrate the use of the nonlinear criterion and its definite advantages in helping improve the adaptive error transient response of a system. Analysis of results is effected by use of a linearization technique on the resulting adaptive equations.
The 1977 altimetry data bank is analyzed for the geometrical shape of the sea surface expressed as surface spherical harmonics after referral to the higher reference model defined by GEM 9. The resulting determination is expressed as quasi-stationary dynamic SST. Solutions are obtained from different sets of long arcs in the GEOS-3 altimeter data bank as well as from sub-sets related to the September 1975 and March 1976 equinoxes assembled with a view to minimizing seasonal effects. The results are compared with equivalent parameters obtained from the hydrostatic analysis of sporadic temperature, pressure and salinity measurements of the oceans and the known major steady state current systems with comparable wavelengths. The most clearly defined parameter (the zonal harmonic of degree 2) is obtained with an uncertainty of + or - 6 cm. The preferred numerical value is smaller than the oceanographic value due to the effect of the correction for the permanent earth tide. Similar precision is achieved for the zonal harmonic of degree 3. The precision obtained for the fourth degree zonal harmonic reflects more closely the accuracy expected from the level of noise in the orbital solutions.