Keith K. Denoyer

Advanced smart structures flight experiments for precision spacecraft

Abstract This paper presents an overview as well as data from four smart structures flight experiments directed by the U.S. Air Force Research Laboratorys Space Vehicles Directorate in Albuquerque, New Mexico. The Middeck Active Control Experiment

Passive vibroacoustic isolation for payload containers

Flight II (MACE II) is a space shuttle flight experiment designed to investigate modeling and control issues for achieving high precision pointing and vibration control of future spacecraft. The Advanced Controls Technology Experiment (ACTEX-I) is an experiment that has demonstrated active vibration suppression using smart composite structures with embedded piezoelectric sensors and actuators. The Satellite Ultraquiet Isolation Technology Experiment (SUITE) is an isolation platform that uses active piezoelectric actuators as well as damped mechanical flexures to achieve hybrid passive/active isolation. The Vibration Isolation, Suppression, and Steering Experiment (VISS) is another isolation platform that uses viscous dampers in conjunction with electromagnetic voice coil actuators to achieve isolation as well as a steering capability for an infra-red telescope.

Hybrid structural/acoustic control of a subscale payload fairing

This paper presents analysis and experimental results which examine key issues related to passive vibroacoustic isolation for container type structures. The key noise reduction principle examined is the passive application of a characteristic impedance mismatch in conjunction with a vibration isolation suspension system to limit structural transmission. The characteristic impedance mismatch is created by imposing a near vacuum condition between partitions of a container structure. Unlike active boundary control techniques, this approach is insensitive to the grazing angle of the source acoustics, is simple, and avoids technology challenges such as the need to develop efficient large stroke actuators. The analysis for this problem is performed by extending Fahys double-leaf partition model. Using the extended model, numerical simulations are conducted to study the effect of various design parameters on acoustic transmission. Guidelines developed from this study are then used to construct an experiment to show the viability of the concept. The experiment demonstrates that at least a 19 dB reduction in sound pressure level (SPL) may be achieved, with a modest level of vacuum.

Dynamics and Control of Slewing Active Beam

During launch, spacecraft experience severe acoustic and vibration loads. Acoustic loads are primarily transmitted through the shroud or payload fairing of the launch vehicle. In recent years, there has been a trend towards using lighter weight and extremely stiff structures such as sandwich construction and grid-stiffened composites in the manufacturing of payload fairings. While substantial weight savings can be achieved using these materials, the problem of acoustic transmission is exacerbated. For this reason, the Air Force Research Laboratory has been actively engaged in vibroacoustic research aimed at reducing the acoustic and vibration levels seen by payloads during launch. This paper presents experimental results for the simultaneous structural and acoustic cavity mode control of a sub-scale composite isogrid payload fairing structure. In this experiment, actuation is performed through the use of both an internal speaker as well as piezoceramic strain actuators located on the outer skin of the composite structure. Sensing is accomplished using a microphone as well as a piezoelectric strain sensor. The control approach presented in this paper is a decentralized frequency domain approach which makes use of a series of independent control loops. One loop uses the microphone and speaker, while additional loops use the piezoelectric sensors and actuators. The control algorithm consists of independent second-order Positive Position Feedback (PPF) controllers tuned to reduce the magnitude of each cavity mode. A PPF filter in conjunction with an extremely sharp bandpass filter is used on the structural mode of limit spillover. This approach leads to a substantial reduction in the acoustic transmission in the range of 0 - 800 Hz. Transmission coincident with the primary cavity modes of the system are reduced in magnitude by 26 and 9 dB respectively while the structural model that is responsible for the majority of transmission is reduced by approximately 7 dB.

Experiences with smart structures for on-orbit Vibration Isolation

that for all cases of the D-model simulation the matrix enclosed by a broken line in Eq. (2) is almost singular, suggesting that the D model is the correct structure. On the other hand, Fig. 4b shows that the singular value ratio takes significant values in the frequency range of interest if the inner-loop gain Kpe is large enough. A close look at Fig. 4b and an analysis can point out the following: a peak exists around the short period mode natural frequency, the peak magnitude increases as the magnitude of Yph increases, and the frequency band where the singular value ratio takes significant values shrinks as the pilot remnant intensity increases.

On-orbit experiments and applications of shape memory alloy mechanisms

The Air Force Research Laboratory is currently conducting a number of space flight experiments with the goal of demonstrating and transitioning smart structures technology to the operational user. Three of these experiments have focused on approaches for providing high-performance on- orbit isolation to precision spacecraft payloads. This paper will describe the design and performance of two systems that are slated for a 2000 launch; the vibration isolation, suppression, and steering experiments and the satellite ultra-quiet isolation technology experiment. Additionally, this paper will provide an overview of a third program initiated in 1999, the miniature vibration isolation system.

Experimental sensor and actuator location procedure for control of dynamically complex smart structures

Spacecraft require a variety of mechanisms to accomplish mission-related functions such as deployment, articulation, and positioning. Current off-the-shelf devices such as pyrotechnic separation nuts, paraffin actuators, and other electro-mechanical devices may not be able to meet future satellite requirements, such as low shock and vibration, and zero contamination. The Air Force Research Laboratory (AFRL), with corporate and government partners, has developed Shape Memory Alloy (SMA) spacecraft release mechanisms and hinges as alternatives. In order to meet future goals, the SMA devices have been designed to reduce shock and vibration, reduce parts, and eliminate pyrotechnics. This paper will focus on descriptions and results of on-orbit SMA mechanism experiments and applications. AFRL has flown SMA release devices as part of the Shape Memory Alloy Release Device (SMARD) experiment on MightSat I. The SMARD experiment, that compared the shock and release times of two SMA devices with those of current off-the-shelf devices, was conducted in May 1999 with extremely successful results. In addition, four AFRL funded SMA release mechanisms successfully deployed the Air Force Academy FalconSat spacecraft from the Orbital Sub-Orbital Program Space Launch Vehicle in January 00. AFRL has also conducted an on-orbit experiment with SMA hinges. The hinges were flown as part of the Lightweight Flexible Solar Array program, that was a joint AFRL/DARPA/NASA/Lockheed Martin program to develop innovative solar array technologies. Six SMA hinges were launched as part of the LFSA experiment on the Space Shuttle Columbia in July 1999 with successful results.

MACE II: A SPACE SHUTTLE EXPERIMENT FOR INVESTIGATING ADAPTIVE CONTROL OF FLEXIBLE SPACECRAFT

In choosing positions for sensors and actuators for structural control, the first step is usually to develop a model that describes the motion of the structure in response to an excitation. The next step depends on the type of sensors and actuators used. If displacement or acceleration sensors and shakers are used, the model serves as a guide to find locations on the structure where displacement is large for a given disturbance. If in-plane strain-based smart sensors and actuators are used, the model is used to identify locations with large in-plane strain. If the structure is relatively complex, there is a good chance that the initial model will not predict motion that agrees completely with the measured motion of the structure. This initial model is then typically adjusted so that the behavior it predicts agrees with a measured modal analysis of the structure. This process can be extremely time consuming, and while the reconciled modes often agree well with a modal analysis, there can be large errors with respect to in-plane strain. Prediction of in-plane is necessary for accurate location of smart sensors and actuators like piezoceramics. In this paper an experimental method is introduced which uses in-plane sensors to find good smart sensor and actuator locations to control acoustic excitation of a complex structure. Experimental results are also presented which demonstrate the proposed technique.

Model inversion tracking control for UltraLITE using neural networks

This paper presents an overview of the Middeck Active Control Experiment - Flight II (MACE II). MACE is a space shuttle flight experiment designed to investigate modeling and control issues for achieving high precision pointing and vibration control of future spacecraft. MACE was developed by NASA Langley Research Center, the Massachusetts Institute of Technology, and Payload Systems, Inc. The experiment was successfully flown on STS-67 in March 1995. The Air Force Research Laboratory (AFRL) has initiated a program to refly the MACE hardware to investigate the use of adaptive control algorithms for precision structural control. MACE II will answer key questions about the ability of adaptive algorithms to perform with respect to the constraints and uncertainties associated with space flight. It will also provide a basis for comparing these adaptive techniques with the fixed-gain linear control approach employed by MACE I.

AN INFORMATION FILTER APPROACH TO RAPID SYSTEM IDENTIFICATION: CONVERGENCE SPEED AND NOISE SENSITIVITY

This paper presents the analytical methodology and initial numerical simulation results for autonomous neural control of the Ultra-Lightweight Imaging Technology Experiment (UltraLITE) Phase I test article. The UltraLITE Phase I test article is a precision deployable structure currently under development at the United States Air Force Research Laboratory (AFRL). Its purpose is to examine control and hardware integration issues related to large deployable sparse optical array spacecraft systems. In this paper, a multi-stage control architecture is examined which incorporates artificial neural networks for model inversion tracking control. The emphasis in the control design approach is to exploit the known nonlinear dynamics of the system in the synthesis of a model inversion tracking controller and to augment the nonlinear controller with an adaptive neuro-controller to accommodate for changing dynamics, failures, and model uncertainties.