Results for “uncertain systems” — Page 7

NASA NTRS1991

A robust momentum management and attitude control system for the space station

A game theoretic controller is synthesized for momentum management and attitude control of the Space Station in the presence of uncertainties in the moments of inertia. Full state information is assumed since attitude rates are assumed to be very assurately measured. By an input-output decomposition of the uncertainty in the system matrices, the parameter uncertainties in the dynamic system are represented as an unknown gain associated with an internal feedback loop (IFL). The input and output matrices associated with the IFL form directions through which the uncertain parameters affect system response. If the quadratic form of the IFL output augments the cost criterion, then enhanced parameter robustness is anticipated. By considering the input and the input disturbance from the IFL as two noncooperative players, a linear-quadratic differential game is constructed. The solution in the form of a linear controller is used for synthesis. Inclusion of the external disturbance torques results in a dynamic feedback controller which consists of conventional PID (proportional integral derivative) control and cyclic disturbance rejection filters. It is shown that the game theoretic design allows large variations in the inertias in directions of importance.

Speyer, J. L.↗

NASA NTRS1992

Robust momentum management and attitude control system for the Space Station

A game theoretic controller is synthesized for momentum management and attitude control of the Space Station in the presence of uncertainties in the moments of inertia. Full state information is assumed since attitude rates are assumed to be very accurately measured. By an input-output decomposition of the uncertainty in the system matrices, the parameter uncertainties in the dynamic system are represented as an unknown gain associated with an internal feedback loop (IFL). The input and output matrices associated with the IFL form directions through which the uncertain parameters affect system response. If the quadratic form of the IFL output augments the cost criterion, then enhanced parameter robustness is anticipated. By considering the input and the input disturbance from the IFL as two noncooperative players, a linear-quadratic differential game is constructed. The solution in the form of a linear controller is used for synthesis. Inclusion of the external disturbance torques results in a dynamic feedback controller which consists of conventional PID (proportional integral derivative) control and cyclic disturbance rejection filters. It is shown that the game theoretic design allows large variations in the inertias in directions of importance.

Rhee, Ihnseok↗

NASA NTRS1991

On-board fault management for autonomous spacecraft

The dynamic nature of the Cargo Transfer Vehicle's (CTV) mission and the high level of autonomy required mandate a complete fault management system capable of operating under uncertain conditions. Such a fault management system must take into account the current mission phase and the environment (including the target vehicle), as well as the CTV's state of health. This level of capability is beyond the scope of current on-board fault management systems. This presentation will discuss work in progress at TRW to apply artificial intelligence to the problem of on-board fault management. The goal of this work is to develop fault management systems. This presentation will discuss work in progress at TRW to apply artificial intelligence to the problem of on-board fault management. The goal of this work is to develop fault management systems that can meet the needs of spacecraft that have long-range autonomy requirements. We have implemented a model-based approach to fault detection and isolation that does not require explicit characterization of failures prior to launch. It is thus able to detect failures that were not considered in the failure and effects analysis. We have applied this technique to several different subsystems and tested our approach against both simulations and an electrical power system hardware testbed. We present findings from simulation and hardware tests which demonstrate the ability of our model-based system to detect and isolate failures, and describe our work in porting the Ada version of this system to a flight-qualified processor. We also discuss current research aimed at expanding our system to monitor the entire spacecraft.

Fesq, Lorraine M.↗

NASA NTRSDate not supplied

APIS: Honeybee Foraging Task Assignment for Use in Uncertain and Unreliable Environments

Multiagent Cyber-Physical-Human (CPH) systems in realistic environments operate under uncertain conditions. Communication among agents, aimed at reducing the uncertainty, is itself subject to uncertainty. We propose to manage uncertainties in autonomous, long-duration operations of multiagent systems via a modified Honeybee Foraging (HBF) behavioral scheme. The resulting system, Autonomous Persistent Intelligent Swarm (APIS),incorporates two new behaviors to ameliorate informational uncertainty. When “scouting”, agents are tasked based on informational quality and reliability rather than solely on priorities. When “dancing”, agents are tasked to rendezvous with other dancing agents to exchange information at close range, where successful communication is guaranteed. When coupled with uncertainty-aware modeling across agents, these behaviors improve situational awareness and resilience of the system, enabling it to function under more uncertain conditions arising during long-duration missions.

Autonomous systems↗

NASA NTRSDate not supplied

APIS: Honeybee Foraging Task Assignment for Use in Uncertain and Unreliable Environments

Multiagent Cyber-Physical-Human (CPH) systems in realistic environments operate under uncertain conditions. Communication among agents, aimed at reducing the uncertainty, is itself subject to uncertainty. We propose to manage uncertainties in autonomous, long-duration operations of multiagent systems via a modified Honeybee Foraging (HBF) behavioral scheme. The resulting system, Autonomous Persistent Intelligent Swarm (APIS),incorporates two new behaviors to ameliorate informational uncertainty. When “scouting”, agents are tasked based on informational quality and reliability rather than solely on priorities. When “dancing”, agents are tasked to rendezvous with other dancing agents to exchange information at close range, where successful communication is guaranteed. When coupled with uncertainty-aware modeling across agents, these behaviors improve situational awareness and resilience of the system, enabling it to function under more uncertain conditions arising during long-duration missions.

Autonomous systems↗

NASA NTRS1992

Autonomous Frequency-Domain Indentification

Test and data-processing system determines plant models and uncertainties. Integrated system of methods, digital signal-processing, and algorithms identifies parametric model of large, flexible structures. Experiments in development of system conducted on laboratory model intended to represent large space antenna or flexible spacecraft. Also applicable to many terrestrial systems, robust control of dynamic plants and processes, robust control of systems about which knowledge uncertain or incomplete, decentralized control, and knowledge-based control systems.

Mettler, Edward↗

NASA NTRS1996

Iterative LQG Controller Design Through Closed-Loop Identification

This paper presents an iterative Linear Quadratic Gaussian (LQG) controller design approach for a linear stochastic system with an uncertain open-loop model and unknown noise statistics. This approach consists of closed-loop identification and controller redesign cycles. In each cycle, the closed-loop identification method is used to identify an open-loop model and a steady-state Kalman filter gain from closed-loop input/output test data obtained by using a feedback LQG controller designed from the previous cycle. Then the identified open-loop model is used to redesign the state feedback. The state feedback and the identified Kalman filter gain are used to form an updated LQC controller for the next cycle. This iterative process continues until the updated controller converges. The proposed controller design is demonstrated by numerical simulations and experiments on a highly unstable large-gap magnetic suspension system.

Hsiao, Min-Hung↗

NASA NTRS2003

Integration of Online Parameter Identification and Neural Network for In-Flight Adaptive Control

An indirect adaptive system has been constructed for robust control of an aircraft with uncertain aerodynamic characteristics. This system consists of a multilayer perceptron pre-trained neural network, online stability and control derivative identification, a dynamic cell structure online learning neural network, and a model following control system based on the stochastic optimal feedforward and feedback technique. The pre-trained neural network and model following control system have been flight-tested, but the online parameter identification and online learning neural network are new additions used for in-flight adaptation of the control system model. A description of the modification and integration of these two stand-alone software packages into the complete system in preparation for initial flight tests is presented. Open-loop results using both simulation and flight data, as well as closed-loop performance of the complete system in a nonlinear, six-degree-of-freedom, flight validated simulation, are analyzed. Results show that this online learning system, in contrast to the nonlearning system, has the ability to adapt to changes in aerodynamic characteristics in a real-time, closed-loop, piloted simulation, resulting in improved flying qualities.

Hageman, Jacob↗

NASA NTRS2003

Integration of Online Parameter Identification and Neural Network for In-Flight Adaptive Control

An indirect adaptive system has been constructed for robust control of an aircraft with uncertain aerodynamic characteristics. This system consists of a multilayer perceptron pre-trained neural network, online stability and control derivative identification, a dynamic cell structure online learning neural network, and a model following control system based on the stochastic optimal feedforward and feedback technique. The pre-trained neural network and model following control system have been flight-tested, but the online parameter identification and online learning neural network are new additions used for in-flight adaptation of the control system model. A description of the modification and integration of these two stand-alone software packages into the complete system in preparation for initial flight tests is presented. Open-loop results using both simulation and flight data, as well as closed-loop performance of the complete system in a nonlinear, six-degree-of-freedom, flight validated simulation, are analyzed. Results show that this online learning system, in contrast to the nonlearning system, has the ability to adapt to changes in aerodynamic characteristics in a real-time, closed-loop, piloted simulation, resulting in improved flying qualities.

Hageman, Jacob J.↗

NASA NTRS1978

Insensitive control technology development

THe investigation of two insensitive controller synthesis techniques was reported. The finite dimensional inverse approach produces a time varying insensitive controller and/or parameter identifier by constructing inverse functions derived from a finite number of input output pair relationships. The MD/IM concept relies on the information matrix theory that was developed in the estimation and identification field. The MD/IM synthesis technique is based on the hypothesis that minimizing the information matrix will reduce system identifiability and consequently system sensitivity to uncertain parameters. The controllers designed with both techniques were evaluated on a realistic C-5A aircraft flight control problem. Results indicate that the FDI controller is more suited to trajectory type problems because of its time varying nature. The MD/IM controller performed as well as the top-rated controllers of the initial effort and has direct application to aircraft flight control problems.

Harvey, C. A.↗

NASA NTRS1985

Payload isolation and precision pointing for the 1990's

The design of a pointing system that is applicable for a variety of payloads is examined. The system employs a very soft interface coupled with inertial control torques which use reaction wheels or control moment gyros. The fundamental stability and disturbance rejection characteristics of simple gimbal pointing systems and the soft mounted inertially reacting concept are evaluated and compared. It is observed that in simple and two-stage gimbal systems there is dynamic interaction with the basebody and these types of systems are not applicable for a Space Station/Space Platform environment in which system dynamics are uncertain; however, the soft mounted inertially reacting concept minimizes the dynamic interaction with the basebody and retains stability. It is concluded that the soft mounted inertially reacting concept has the pointing accuracy and disturbance isolation of a free flying spacecraft while still obtaining power, communication, orbit maintenance, and servicing from a basebody.

Sirlin, S. W.↗

NASA NTRS1988

Robustness margin calculation with dynamic and real parametric uncertainty

The problem of robust stability in feedback control systems with real uncertain parameters and unmodeled dynamics, where the latter takes the form of one complex block dynamic uncertainty, is considered. A robustness margin Tm is defined with respect to such model uncertainty structure and robust stability is characterized in terms of it. An algorithm to calculate Tm and the closely related multivariable stability margin km and structured singular value mu measures of robust stability is developed. It is shown that other control problems, such as robust performance for real parametric uncertainty, can also be handled in this framework.

Sideris, Athanasios↗

NASA NTRS1989

A survey on the structured singular value

The structured singular value, U, is an important linear algebra tool to study a class of matrix perturbation problems. It is useful for analyzing the robustness of stability and performance of uncertain, (nominally) linear systems. Computation of (M) is difficult, and usually, upper and lower bounds are all that can be reliably computed. Upper bounds give conservative estimates of the sizes of allowable perturbations. The maximum singular value of a matrix M is an upper bound for (M). As an upper bound, it can be improved by finding a transformations to the data (i.e. M) which do not change the structured singular value, but do reduce the maximum singular value. Typically, upper bound algorithms involve searches over sets of transformations to yield the tightest bound. Lower bound algorithms are intelligent searches for minimum-norm solutions to multivariable polynomial equations, and are based on various optimality conditions that hold at the global (and, unfortunately, some local) minima. The current methods to compute both of these types of bounds are reviewed. Theoretical justification and extensive numerical experience with the various algorithms are covered.

Packard, Andy↗

NASA NTRS1999

The Dependence on Grid Resolution of Numerically Simulated Convective Cloud Systems Using Ice Microphysics

Mesoscale research and forecast models are increasingly being used at horizontal resolutions of 1-8 km to simulate a variety of precipitating systems. When the model is used to simulate convective systems, it is uncertain to what extent the dynamics and microphysics of convective updrafts can be resolved with grids larger than 1 km. In this study, two- and three-dimensional versions of the Goddard Cumulus Ensemble model are used to determine the impact of horizontal grid resolution on the behavior of the simulated storms and on the characteristics of the cloud microphysical fields. It will be shown that as resolution decreases from about 1 km to greater than 3 km, there is a fairly rapid degradation of the storm structure in the form of reduced convective mass fluxes, updraft tilts, and cloud microphysics. A high-resolution simulation of hurricane outer rainbands using the MM5 mesoscale model shows also that there can be a substantial modification of the key microphysical processes that contribute to rainfall as a result of reducing the horizontal resolution.

Braun, Scott A.↗

NASA NTRS2008

Approximation of Failure Probability Using Conditional Sampling

In analyzing systems which depend on uncertain parameters, one technique is to partition the uncertain parameter domain into a failure set and its complement, and judge the quality of the system by estimating the probability of failure. If this is done by a sampling technique such as Monte Carlo and the probability of failure is small, accurate approximation can require so many sample points that the computational expense is prohibitive. Previous work of the authors has shown how to bound the failure event by sets of such simple geometry that their probabilities can be calculated analytically. In this paper, it is shown how to make use of these failure bounding sets and conditional sampling within them to substantially reduce the computational burden of approximating failure probability. It is also shown how the use of these sampling techniques improves the confidence intervals for the failure probability estimate for a given number of sample points and how they reduce the number of sample point analyses needed to achieve a given level of confidence.

Giesy. Daniel P.↗

NASA NTRSDate not supplied

Design Under Uncertainty with Design-Dependent Uncertain Variables

Uncertainty quantification (UQ) can provide a more robust understanding of a system, leading to better informed decisions earlier in the design process. The additional information that UQ provides can be leveraged during a design optimization process known as design under uncertainty that, when incorporated with multidisciplinary design and optimization, can become computationally infeasible due to the large number of responses required for meaningful results. Previous work addressed reducing the computational expense in design under uncertainty by incorporating analytic derivatives throughout polynomial chaos expansion. Although this allows design under uncertainty to be feasible for more systems, some multidisciplinary systems have design-dependent uncertain variables. This paper details an implementation of design dependent uncertain variables in a manner than preserves derivatives required for efficient gradient-based optimization throughout the process. Two analytic examples of design-dependent uncertain variables are given: the first transforms a uniform uncertain variable with one design variable and the second transforms a normal uncertain variable with two design variables. The polynomial chaos expansion (PCE) results are comparable to both the Monte Carlo (MC) results and the analytic results for the two examples. A case study that maximizes the lift-to-drag ratio with a design-dependence between the wing leading edge sweep angle and uncertain parameter percentage of laminar flow is compared to a MC and alternative optimization formulations. This paper demonstrates that design-dependent uncertain variables are valid and hold throughout PCE.

Joanna N Schmidt↗

NASA NTRSDate not supplied

Design Under Uncertainty with Design-Dependent Uncertain Variables

Uncertainty quantification (UQ) can provide a more robust understanding of a system, leading to better informed decisions earlier in the design process. The additional information that UQ provides can be leveraged during a design optimization process known as design under uncertainty that, when incorporated with multidisciplinary design and optimization, can become computationally infeasible due to the large number of responses required for meaningful results. Previous work addressed reducing the computational expense in design under uncertainty by incorporating analytic derivatives throughout polynomial chaos expansion. Although this allows design under uncertainty to be feasible for more systems, some multidisciplinary systems have design-dependent uncertain variables. This paper details an implementation of design dependent uncertain variables in a manner than preserves derivatives required for efficient gradient-based optimization throughout the process. Two analytic examples of design-dependent uncertain variables are given: the first transforms a uniform uncertain variable with one design variable and the second transforms a normal uncertain variable with two design variables. The polynomial chaos expansion (PCE) results are comparable to both the Monte Carlo (MC) results and the analytic results for the two examples. A case study that maximizes the lift-to-drag ratio with a design-dependence between the wing leading edge sweep angle and uncertain parameter percentage of laminar flow is compared to a MC and alternative optimization formulations. This paper demonstrates that design-dependent uncertain variables are valid and hold throughout PCE.

robust design↗

NASA NTRS1983

Control and modeling of a CELSS (Controlled Ecological Life Support System)

Research topics that arise from the conceptualization of control for closed life support systems which are life support systems in which all or most of the mass is recycled are discussed. Modeling and control of uncertain and poorly defined systems, resource allocation in closed life support systems, and control structures or systems with delay and closure are emphasized.

Auslander, D. M.↗