SEARCH · Engineering Papers
Results for “Robust control”
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.
Hypersonic vehicle model and control law development using H(infinity) and micron synthesis
The control system design for a Single Stage To Orbit (SSTO) air breathing vehicle will be central to a successful mission because a precise ascent trajectory will preserve narrow payload margins. The air breathing propulsion system requires the vehicle to fly roughly halfway around the Earth through atmospheric turbulence. The turbulence, the high sensitivity of the propulsion system to inlet flow conditions, the relatively large uncertainty of the parameters characterizing the vehicle, and continuous acceleration make the problem especially challenging. Adequate stability margins must be provided without sacrificing payload mass since payload margins are critical. Therefore, a multivariable control theory capable of explicitly including both uncertainty and performance is needed. The H(infinity) controller in general provides good robustness but can result in conservative solutions for practical problems involving structured uncertainty. Structured singular value mu framework for analysis and synthesis is potentially much less conservative and hence more appropriate for problems with tight margins. An SSTO control system requires: highly accurate tracking of velocity and altitude commands while limiting angle-of-attack oscillations, minimized control power usage, and a stabilized vehicle when atmospheric turbulence and system uncertainty are present. The controller designs using H(infinity) and mu-synthesis procedures were compared. An integrated flight/propulsion dynamic mathematical model of a conical accelerator vehicle was linearized as the vehicle accelerated through Mach 8. Vehicle acceleration through the selected flight condition gives rise to parametric variation that was modeled as a structured uncertainty. The mu-analysis approach was used in the frequency domain to conduct controller analysis and was confirmed by time history plots. Results demonstrate the inherent advantages of the mu framework for this class of problems.
Direct Adaptive Aircraft Control Using Dynamic Cell Structure Neural Networks
A Dynamic Cell Structure (DCS) Neural Network was developed which learns topology representing networks (TRNS) of F-15 aircraft aerodynamic stability and control derivatives. The network is integrated into a direct adaptive tracking controller. The combination produces a robust adaptive architecture capable of handling multiple accident and off- nominal flight scenarios. This paper describes the DCS network and modifications to the parameter estimation procedure. The work represents one step towards an integrated real-time reconfiguration control architecture for rapid prototyping of new aircraft designs. Performance was evaluated using three off-line benchmarks and on-line nonlinear Virtual Reality simulation. Flight control was evaluated under scenarios including differential stabilator lock, soft sensor failure, control and stability derivative variations, and air turbulence.
Extensions of output variance constrained controllers to hard constraints
Covariance Controllers assign specified matrix values to the state covariance. A number of robustness results are directly related to the covariance matrix. The conservatism in known upperbounds on the H infinity, L infinity, and L (sub 2) norms for stability and disturbance robustness of linear uncertain systems using covariance controllers is illustrated with examples. These results are illustrated for continuous and discrete time systems. **** ONLY 2 BLOCK MARKERS FOUND -- RETRY *****
Using fuzzy logic to integrate neural networks and knowledge-based systems
Outlined here is a novel hybrid architecture that uses fuzzy logic to integrate neural networks and knowledge-based systems. The author's approach offers important synergistic benefits to neural nets, approximate reasoning, and symbolic processing. Fuzzy inference rules extend symbolic systems with approximate reasoning capabilities, which are used for integrating and interpreting the outputs of neural networks. The symbolic system captures meta-level information about neural networks and defines its interaction with neural networks through a set of control tasks. Fuzzy action rules provide a robust mechanism for recognizing the situations in which neural networks require certain control actions. The neural nets, on the other hand, offer flexible classification and adaptive learning capabilities, which are crucial for dynamic and noisy environments. By combining neural nets and symbolic systems at their system levels through the use of fuzzy logic, the author's approach alleviates current difficulties in reconciling differences between low-level data processing mechanisms of neural nets and artificial intelligence systems.
An Analysis of the Optimal Control Modification Method Applied to Flutter Suppression
Unlike basic Model Reference Adaptive Control (MRAC)l, Optimal Control Modification (OCM) has been shown to be a promising MRAC modification with robustness and analytical properties not present in other adaptive control methods. This paper presents an analysis of the OCM method, and how the asymptotic property of OCM is useful for analyzing and tuning the controller. We begin with a Lyapunov stability proof of an OCM controller having two adaptive gain terms, then the less conservative and easily analyzed OCM asymptotic property is presented. Two numerical examples are used to show how this property can accurately predict steady state stability and quantitative robustness in the presence of time delay, and relative to linear plant perturbations, and nominal Loop Transfer Recovery (LTR) tuning. The asymptotic property of the OCM controller is then used as an aid in tuning the controller applied to a large scale aeroservoelastic longitudinal aircraft model for flutter suppression. Control with OCM adaptive augmentation is shown to improve performance over that of the nominal non-adaptive controller when significant disparities exist between the controller/observer model and the true plant model.
Robust eigenstructure assignment by a projection method: Application using multiple optimization criteria
New ideas which lead to feedback control laws for large flexible structures which are insensitive to model uncertainty are presented. A pole placement method is presented which leads to near-unitary closed loop eigenvectors, and a new method is introduced to design the control while simultaneously considering three competing measures of optimality. Robustness versus Integral algorithms are applicable to at least moderately high-dimensioned systems. In the present discussion, controls for two coupled flexible bodies are considered. A 6x24 gain matrix is designed to control a 12 modes system using 6 actuators. Researchers also developed control laws for the R2P2 simulator at Martin Marietta; in this case 3 actuators are used to control a 12th order system. Simulation studies indicate that researchers indeed achieved robust designs without significant difficulties associated with spillover into the uncontrolled modes. Here, several key ideas and numerical results are given. In the references, details of the formulation, discussions of salient features, and connection to the availiable literature are given.
Multicriteria Gain Tuning for Rotorcraft Flight Controls (also entitled The Development of the Conduit Advanced Control System Design and Evaluation Interface with a Case Study Application Fly by Wire Helicopter Design)
Handling qualities analysis and control law design would seem to be naturally complimenting components of aircraft flight control system design, however these two closely coupled disciplines are often not well integrated in practice. Handling qualities engineers and control system engineers may work in separate groups within an aircraft company. Flight control system engineers and handling quality specialists may come from different backgrounds and schooling and are often not aware of the other group's research. Thus while the handling qualities specifications represent desired aircraft response characteristics, these are rarely incorporated directly in the control system design process. Instead modem control system design techniques are based on servo-loop robustness specifications, and simple representations of the desired control response. Comprehensive handling qualities analysis is often left until the end of the design cycle and performed as a check of the completed design for satisfactory performance. This can lead to costly redesign or less than satisfactory aircraft handling qualities when the flight testing phase is reached. The desire to integrate the fields of handling qualities and flight,control systems led to the development of the CONDUIT system. This tool facilitates control system designs that achieve desired handling quality requirements and servo-loop specifications in a single design process. With CONDUIT, the control system engineer is now able to directly design and control systems to meet the complete handling specifications. CONDUIT allows the designer to retain a preferred control law structure, but then tunes the system parameters to meet the handling quality requirements.
Towards Autonomous Lunar Resource Excavation via Deep Reinforcement Learning
To support sustainable infrastructure on the Moon, NASA needs to leverage lunar resources for in-situ processing and construction. NASA’s Regolith Advanced Surface Systems Operations Robot (RASSOR) is principally designed to mine and deliver regolith for these tasks. To reliably perform these operations on the lunar surface, RASSOR's sensors and control systems need to be robust and maximize information extracted from a reduced sensor payload. Herein, we present our findings from the Intelligent Capabilities Enhanced RASSOR project. We created reduced-order simulation environments in which we applied reinforcement learning algorithms to learn autonomous trenching controllers and produced state estimation architectures. We developed two simulations: a 2D excavation simulation used to facilitate parameter selection, and a 3D simulation developed using a game physics engine to simulate simplified soil interactions and incorporate robotic agents parameterized by dynamic models. Within these simulations, we learned autonomous excavation routines that exceed excavation efficiency measures as compared against RASSOR's existing control and teleoperation-based methods.
Towards Autonomous Lunar Resource Excavation via Deep Reinforcement Learning
To support sustainable infrastructure on the Moon, NASA needs to leverage lunar resources for in-situ processing and construction. NASA’s Regolith Advanced Surface Systems Operations Robot (RASSOR) is principally designed to mine and deliver regolith for these tasks. To reliably perform these operations on the lunar surface, RASSOR's sensors and control systems need to be robust and maximize information extracted from a reduced sensor payload. Herein, we present our findings from the Intelligent Capabilities Enhanced RASSOR project. We created reduced-order simulation environments in which we applied reinforcement learning algorithms to learn autonomous trenching controllers and produced state estimation architectures. We developed two simulations: a 2D excavation simulation used to facilitate parameter selection, and a 3D simulation developed using a game physics engine to simulate simplified soil interactions and incorporate robotic agents parameterized by dynamic models. Within these simulations, we learned autonomous excavation routines that exceed excavation efficiency measures as compared against RASSOR's existing control and teleoperation-based methods.
Application of Artificial Intelligence Techniques in Unmanned Aerial Vehicle Flight
This paper describes the development of an application of Artificial Intelligence for Unmanned Aerial Vehicle (UAV) control. The project was done as part of the requirements for a class in Artificial Intelligence (AI) at Nova southeastern University and as an adjunct to a project at NASA Goddard Space Flight Center's Wallops Flight Facility for a resilient, robust, and intelligent UAV flight control system. A method is outlined which allows a base level application for applying an AI method, Fuzzy Logic, to aspects of Control Logic for UAV flight. One element of UAV flight, automated altitude hold, has been implemented and preliminary results displayed. A low cost approach was taken using freeware, gnu, software, and demo programs. The focus of this research has been to outline some of the AI techniques used for UAV flight control and discuss some of the tools used to apply AI techniques. The intent is to succeed with the implementation of applying AI techniques to actually control different aspects of the flight of an UAV.
Magnet Architectures and Active Radiation Shielding Study (MAARSS)
This report describes the analysis of several magnetic radiation shield architectures and the spacecraft systems associated with them, such as quench detection, thermal management, and compensation coils. This work was performed as part of the NASA Institute of Advanced Concepts (NIAC) Grant Phase II effort. Because radiation exposure during long duration space missions poses a significant risk for crewed space-flight, Advanced Magnetic Lab, Inc. (AML) and NASA are studying magnetic radiation shielding, which may generate low-mass protection. Recent technological developments in high-temperature superconducting technology suggest the possibility of such shielding. This report analyzes some of the technical difficulties involved in such a system and identifies technology areas where further investment would be warranted. Most analysis was performed on a baseline configuration (8-meter diameter coils, with 1 tesla field). However, a baseline design with an expandable coil configuration (16 meter diameter coils, 1.5 tesla field) has a significant potential to increase shielding efficiency. The magnetic forces on the magnetic shield's components are large enough to require detailed structural design. Analysis of the thermal structural response of a yttrium-barium-copper-oxide (YBCO) high-temperature, superconducting (HTS) tape in a large-scale solenoid magnet showed that structural and thermal performance improved by replacing Hastelloy® with graphene in the tape. Integrating graphene into the high-strength fiber (HSF) support structure brings even greater benefits. The energy stored in a shield system's magnetic field must be safely dissipated if a quench-type failure occurs. Analysis of quench detection by fiber-optic thermal sensing and improvement of quench characteristics using graphene showed quench is manageable. Low-temperature superconductors can quench when subjected to movement, but lab tests of coil expansion testing showed that HTS materials do not quench when subject to the movement associated with coil expansion. Compensation coils are necessary to reduce the magnetic field in the crew habitat. Analysis showed that subdividing the compensation coil into individually controlled sections allows for a robust, adjustable system. AML analyzed the forces on an approaching capsule that were a result of eddy currents in its structure induced by its motion through the magnetic field from the coils. AML also analyzed coil-to-coil forces. Thermal control to maintain the HTS material at a suitable temperature is challenging. The sunshield technology review revealed a greater possibility of thermal insulation of the coils and use of passive cooling. The availability of suitable cryocoolers was also reviewed. A parametric graph estimating mass and shielding effectiveness for strength and thickness was developed, and allows spacecraft designers to assess the tradeoffs of using a magnetic radiation shield.
Multi-Timescale Optimal Operation Framework for Integrated Economic and Reliability Analysis of Hybrid Power Plants
This paper introduces a hierarchical modeling framework for hybrid power plants (HPP) to facilitate the operation of HPP in power systems similar to conventional generators (Congens) in the integrated multi-timescale optimal operation framework. To consider the uncertainties of HPP renewable power in the day-ahead scheduling, distributionally robust optimization (DRO) is used. To ensure that the state-of-charge (SOC) of energy storage systems in HPPs aligns closely with the planned value for long-term reliability, real-time SOC management is incorporated. In addition, an adjustable real-time control is designed for the robust delivery of HPP real-time services. Case studies performed on a revised IEEE 39-bus system demonstrate the effectiveness of the proposed framework for HPP operation. Simulation results highlight that the proposed framework not only can help operators schedule HPP similar to Congens in varying weather conditions but can also maintain the frequency reliability of the system.
Active flutter suppression using Linear Quadratic Gaussian theory
This paper describes the application of Linear Quadratic Gaussian (LQG) methodology to the design of active control systems for suppression of aerodynamic flutter. A full-size wind tunnel model of a supercritical wing with associated sensors and actuators comprises the system to be controlled. Results of a synthesis methodology that provide small values of rms response, insensitivity to flight condition, and robust stability are presented. Results of control surface and sensor position optimization are also presented. Both frequency response matching and residualization are used to obtain practical flutter controllers.
Switching State-Feedback LPV Control with Uncertain Scheduling Parameters
This paper presents a new method to design Robust Switching State-Feedback Gain-Scheduling (RSSFGS) controllers for Linear Parameter Varying (LPV) systems with uncertain scheduling parameters. The domain of scheduling parameters are divided into several overlapped subregions to undergo hysteresis switching among a family of simultaneously designed LPV controllers over the corresponding subregion with the guaranteed H-infinity performance. The synthesis conditions are given in terms of Parameterized Linear Matrix Inequalities that guarantee both stability and performance at each subregion and associated switching surfaces. The switching stability is ensured by descent parameter-dependent Lyapunov function on switching surfaces. By solving the optimization problem, RSSFGS controller can be obtained for each subregion. A numerical example is given to illustrate the effectiveness of the proposed approach over the non-switching controllers.
Robustness analysis based on passivity
A technique of robust stability analysis for multivariable control systems is introduced. Existing tools in the literature are mostly based on small gain type of conditions such as the H(infinity)-norm or mu measure. In contrast, the approach presented is motivated by passivity theory. The natural energy Lyapunov function associated with a passive system allows robustness to be analyzed with Lyapunov stability theory. Consequently, nonlinear and time-varying perturbations can be considered within the same framework. For unstructured uncertainties, the robustness margin is characterized in terms of an index related to the Hermitian part of a transfer function. If the uncertainties contain additional diagonal structure, stability margins are obtained for variations of each uncertain block in both positive and negative directions. Applications of this robustness analysis technique are illustrated.
A middle man approach to knowledge acquisition in expert systems
The Weed Control Advisor (WCA) is a robust expert system that has been successfully implemented on an IBM AT class microcomputer in CLIPS. The goal of the WCA was to demonstrate the feasibility of providing an economical, efficient, user friendly system through which Texas rice producers could obtain expert level knowledge regarding herbicide application for weed control. During the development phase of the WCA, an improved knowledge acquisition method which we call the Middle Man Approach (MMA) was applied to facilitate the communication process between the domain experts and the knowledge engineer. The MMA served to circumvent the problems associated with the more traditional forms of knowledge acquisition by placing the Middle Man, a semi-expert in the problem domain with some computer expertise, at the site of system development. The middle man was able to contribute to system development in two major ways. First, the Middle Man had experience working in rice production and could assume many of the responsibilities normally performed by the domain experts such as explaining the background of the problem domain and determining the important relations. Second, the Middle Man was familiar with computers and worked closely with the system developers to update the rules after the domain experts reviewed the prototype, contribute to the help menus and explanation portions of the expert system, conduct the testing that is required to insure that the expert system gives the expected results answer questions in a timely way, help the knowledge engineer structure the domain knowledge into a useable form, and provide insight into the end user's profile which helped in the development of the simple user friendly interface. The final results were not only that both time expended and costs were greatly reduced by using the MMA, but the quality of the system was improved. This papa will introduce the WCA system and then discuss traditional knowledge acquisition along with some of the problems often associated with it, the MMA methodology, and its application to the WCA development.
A formulation of stability for spacecraft formations
A formulation of stability for a set of spacecraft in formation flight is presented. First, a formation is defined in a precise mathematical form in terms of control interactions. Then, stability is formulated based on input-to-output stability with respect to a partitioning of the formation dynamics. This formulation of stability is shown to be useful in characterizing disturbance propagation in the formation as a function of the partition interconnection topology, and also in analyzing the robustness of sensing, communication, and control topologies. Stability results are presented for hierarchical, cyclic, and disturbance attenuating formations.