Search

Human-based object detection in synthetic aperture RADAR (SAR) imagery is complex and technical, laboriously slow but time critical—the perfect application for machine learning (ML). Training an ML network for object detection requires very large image datasets with imbedded objects that are accurately and precisely labeled. Unfortunately, no such SAR datasets exist. Therefore, this paper proposes a method to synthesize wide field of view (FOV) SAR images by combining two existing datasets: SAMPLE, which is composed of both real and synthetic single-object chips, and MSTAR Clutter, which is composed of real wide-FOV SAR images. Synthetic objects are extracted from SAMPLE using threshold-based segmentation before being alpha-blended onto patches from MSTAR Clutter. To validate the novel synthesis method, individual object chips are created and classified using a simple convolutional neural network (CNN); testing is performed against the measured SAMPLE subset. A novel technique is also developed to investigate training activity in deep layers. The proposed data augmentation technique produces a 17% increase in the accuracy of measured SAR image classification. This improvement shows that any residual artifacts from segmentation and blending do not negatively affect ML, which is promising for future use in wide-area SAR synthesis.

Human-based object detection in synthetic aperture RADAR (SAR) imagery is complex and technical, laboriously slow but time critical—the perfect application for machine learning (ML). Training an ML network for object detection requires very large image datasets with imbedded objects that are accurately and precisely labeled. Unfortunately, no such SAR datasets exist. Therefore, this paper proposes a method to synthesize wide field of view (FOV) SAR images by combining two existing datasets: SAMPLE, which is composed of both real and synthetic single-object chips, and MSTAR Clutter, which is composed of real wide-FOV SAR images. Synthetic objects are extracted from SAMPLE using threshold-based segmentation before being alpha-blended onto patches from MSTAR Clutter. To validate the novel synthesis method, individual object chips are created and classified using a simple convolutional neural network (CNN); testing is performed against the measured SAMPLE subset. A novel technique is also developed to investigate training activity in deep layers. The proposed data augmentation technique produces a 17% increase in the accuracy of measured SAR image classification. This improvement shows that any residual artifacts from segmentation and blending do not negatively affect ML, which is promising for future use in wide-area SAR synthesis., Outstanding Thesis, Major, United States Air Force, Approved for public release. Distribution is unlimited.

We present a complementarity that complements relationships among the elements in the neutrino mixing matrix and address its physical implications. First we show how a complementarity with a phase being introduced as an extra parameter can be held in the nine independent schemes of parameterizing the matrix introducing a discrete parameter symmetry and a combination of sine functions, a part of Jarlskog invariant, within a certain size of uncertainty. Then, for the first time, we show that we can use the uncertainty associated with the complementarity as an empirical constraint complementing that among the diagonal elements in the neutrino mixing matrix. We discuss its physical implication in relation to the size of the uncertainty among the elements in the end.

We propose that the sound wave coming to the inner side of Miniboone detector could be one of the sources for our having the neutrino anomaly in the experiment. We start with presenting a rough estimate for the size of the energy associated with the sound wave coming into the detector, the size of the loss of energy associate with the save wave as it travels down to the target medium due to their gravitationally interacting with the sound wave in a classical sense. After that, we describe that the neutrino anomaly could be due to the sound wave interacting with the detector material under the pressure due to the presence of the mineral oil and their producing phonon-induced electrons via a process such as the thermionic emission, which may lead more events to be identified as electron-like events in the experiment. We also address that the sound wave may scatter with the electrons produced from the electron-photon shower.

Target pose estimation and aimpoint selection is crucial in direct energy weapon systems, as it allows the system to point to a specific and strategic area of the target. However, it is a challenging task because a dedicated attitude sensor is required. Motivated by new emerging deep learning capabilities, the present work proposes a deep learning model to estimate a target spacecraft attitude in terms of Euler angles. Data for the deep learning model were experimentally generated from 3D UAV models, incorporating effects such as atmospheric backgrounds and turbulence. The targets pose was derived from the training, validation, and prediction of 2D keypoints. With a keypoint detection model it is possible to detect interest points in an image, which allows us to estimate pose, angles, and dimensions of the target in question. Utilizing a weak-perspective direct linear transformation algorithm, the pose of a 3D object with respect to a camera from 3D to 2D correspondences could be determined. Additionally, from this correspondence, an aimpoint, mimicking laser tracking could be determined on the target. This work evaluates these methods and their accuracy against experimentally generated data with simulated real-world environments.

High energy laser (HEL) systems are susceptible to atmospheric turbulence when focusing on targets down range. Current HEL systems use wavefront sensors and complex adaptive optics systems to compensate for these aberrations. The primary objective of this thesis is to investigate target image aberration compensation techniques using machine learning algorithms, eliminating the need for complex wavefront sensing hardware. Target imagery will be obtained from the High Energy Laser Beam Control Research Testbed (HBCRT) and imagery aberrations will be simulated to provide necessary datasets for training and validation of the image aberration compensation methods. The performance of these techniques will be evaluated for military imaging applications.

Target pose estimation and aimpoint selection is crucial in direct energy weapon systems, as it allows the system to point to a specific and strategic area of the target. However, it is a challenging task because a dedicated attitude sensor is required. Motivated by new emerging deep learning capabilities, the present work proposes a deep learning model to estimate a target spacecraft attitude in terms of Euler angles. Data for the deep learning model were experimentally generated from 3D UAV models, incorporating effects such as atmospheric backgrounds and turbulence. The targets pose was derived from the training, validation, and prediction of 2D keypoints. With a keypoint detection model it is possible to detect interest points in an image, which allows us to estimate pose, angles, and dimensions of the target in question. Utilizing a weak-perspective direct linear transformation algorithm, the pose of a 3D object with respect to a camera from 3D to 2D correspondences could be determined. Additionally, from this correspondence, an aimpoint, mimicking laser tracking could be determined on the target. This work evaluates these methods and their accuracy against experimentally generated data with simulated real-world environments., Outstanding Thesis, Ensign, United States Navy, Approved for public release. Distribution is unlimited.

High energy laser (HEL) systems are susceptible to atmospheric turbulence when focusing on targets down range. Current HEL systems use wavefront sensors and complex adaptive optics systems to compensate for these aberrations. The primary objective of this thesis is to investigate target image aberration compensation techniques using machine learning algorithms, eliminating the need for complex wavefront sensing hardware. Target imagery will be obtained from the High Energy Laser Beam Control Research Testbed (HBCRT) and imagery aberrations will be simulated to provide necessary datasets for training and validation of the image aberration compensation methods. The performance of these techniques will be evaluated for military imaging applications., Ensign, United States Navy, Approved for public release. Distribution is unlimited.

Satellite imaging performance can degrade due to optical aberrations. To maximize a satellite’s imaging output over its useful lifespan, deep learning presents a cost-effective alternative to traditional adaptive optics for deblurring satellite images. This is because deep learning is essentially a post-processing technique that relies on algorithms and a large dataset. This research focuses on applying deep learning algorithms based on the UNET Convolutional Neural Network, which is widely used in the bio-medical imaging field, to deblur optically aberrated satellite imagery. The XVIEW dataset, which is composed of images taken by the Worldview-3 satellite, is used. The XVIEW images are then simulated with optical aberrations (defocus and spherical) using Zernike polynomials. The blurred images are subsequently deblurred with UNET and UNET variants (UNET++ and UNET3+) before final performance evaluation with various image quality metrics. The results showed that (1) UNET algorithms can effectively deblur optically aberrated satellite images, and (2) UNET3+ modified with additional convolutional layers (deep-UNET3+) provided the best deblurring performance. Based on the positive results, this thesis recommends that the UNET algorithm be applied on actual field cases of optically aberrated satellite imagery and be further developed to perform better even on super-resolution applications., Military Expert 5, Singapore Army, Approved for public release. Distribution is unlimited.

Atmospheric turbulence significantly degrades the performance of High Energy Laser (HEL) beams. The three key undesirable effects are: (1) degraded target images used for target tracking; (2) inaccurate HEL pointing; and (3) reduction in HEL power during propagation to the target. The current approach for compensating for these turbulence effects uses adaptive optics to measure atmospheric turbulence and compensate the aberration in the optical beam. However, an adaptive optics system has limited performance in strong turbulence and an optical system makes the HEL system more complex. With improvements in Deep Learning algorithms and further development in Artificial Intelligence, we used Deep Learning and Convolutional Neural Networks to predict the atmospheric turbulence and compensate for its negative effects on laser beams. The predicted turbulence can be used for image correction and HEL beam correction using a deformable mirror to reduce turbulence effects during propagation., Military Expert 5, Republic of Singapore Navy, Approved for public release. Distribution is unlimited.

Rendezvous and proximity operations are an essential component of both military and commercial space missions and are rising in complexity. This dissertation presents an applied reachability analysis and develops a computationally feasible autonomous guidance algorithm for the purpose of spacecraft rendezvous and proximity maneuvers around a tumbling object. Recent advancements enable the use of more sophisticated, computation-based algorithms, instead of traditional control methods. These algorithms are desirable for autonomous applications due to their ability to optimize performance and explicitly handle constraints (e.g., safety, control limits). In an autonomous setting, however, some important questions must be answered before an algorithm implementation can be realized. First, the feasibility of a maneuver is addressed by analyzing the fundamental spacecraft relative dynamics. Particularly, a set of initial relative states is computed and visualized from which the desired rendezvous state can be reached (i.e., backward reachability analysis). Second, with the knowledge that a maneuver is feasible, the Model Predictive Control (MPC) framework is utilized to design a stabilizing feedback control law that optimizes performance and incorporates constraints such as control saturation limits and collision avoidance. The MPC algorithm offers a computationally efficient guidance strategy that could potentially be implemented in real-time on-board a spacecraft., http://archive.org/details/autonomousspacec1094560364, Major, United States Air Force, Approved for public release; distribution is unlimited.

Rendezvous and proximity operations are an essential component of both military and commercial space missions and are rising in complexity. This dissertation presents an applied reachability analysis and develops a computationally feasible autonomous guidance algorithm for the purpose of spacecraft rendezvous and proximity maneuvers around a tumbling object. Recent advancements enable the use of more sophisticated, computation-based algorithms, instead of traditional control methods. These algorithms are desirable for autonomous applications due to their ability to optimize performance and explicitly handle constraints (e.g., safety, control limits). In an autonomous setting, however, some important questions must be answered before an algorithm implementation can be realized. First, the feasibility of a maneuver is addressed by analyzing the fundamental spacecraft relative dynamics. Particularly, a set of initial relative states is computed and visualized from which the desired rendezvous state can be reached (i.e., backward reachability analysis). Second, with the knowledge that a maneuver is feasible, the Model Predictive Control (MPC) framework is utilized to design a stabilizing feedback control law that optimizes performance and incorporates constraints such as control saturation limits and collision avoidance. The MPC algorithm offers a computationally efficient guidance strategy that could potentially be implemented in real-time on-board a spacecraft.

For certain imaging satellite missions, a large aperture with wide field-of-view is needed. In order to achieve diffraction limited performance, the mirror surface Root Mean Square (RMS) error has to be less than 0.05 waves. In the case of visible light, it has to be less than 30 nm. This requirement is difficult to meet as the large aperture will need to be segmented in order to fit inside a launch vehicle shroud. To reduce this requirement and to compensate for the residual wavefront error, Micro-Electro-Mechanical System (MEMS) deformable mirrors can be considered in the aft optics of the optical system. MEMS deformable mirrors are affordable and consume low power, but are small in size. Due to the major reduction in pupil size for the deformable mirror, the effective field angle is magnified by the diameter ratio of the primary and deformable mirror. For wide field of view imaging, the required deformable mirror correction is field angle dependant, impacting the required parameters of a deformable mirror such as size, number of actuators, and actuator stroke. In this paper, a representative telescope and deformable mirror system model is developed and the deformable mirror correction is simulated to study the impact of the large field angles in correcting a wavefront error using a deformable mirror in the aft optics.

Imaging satellites have several challenges in dynamics and control to meet high performance requirements. As an example, Hubble Space Telescope pointing requirements are of 0.012 arc-sec and jitter requirements not to exceed 0.007 arc-sec. These satellites require fast slew maneuver with minimum slew time. Flexibility/control interaction becomes critical for these satellites. These satellites require fine mirror surfaces for diffraction limited performance. For visible images, the surface mirror accuracy requirements are a minimum of 30 nm. In order to meet these performance requirements, advance dynamics and control techniques and actuators and sensors are required. Meeting these requirements becomes even more challenging for future imaging satellites as the diameter of primary mirrors increases, resulting in deployable mirrors. This paper provides an overview of advanced dynamics and control techniques for jitter control, flexibility/control interactions, slew maneuvers, and reflector active surface control.

Several future imaging missions will require large space mirrors, 10-20 m in diameter. This is a challenging problem on how to build cost-effective large aperture mirrors with surface error accuracy in the rage of 30 nanometer RMS. In order to achieve these performance, use of several smart structures technologies are proposed. One concept is to use multiple satellites, called collectors with small apertures and coherently combines the images by a combiner spacecraft, so the system acts like a large aperture satellite. However, the technologies for coherently combing the images are challenging. First, the collector satellites have to operate in orbit formation flying with nanometer level position control. Attitude control for the satellites will be needed in micro-radian level. Next, accurate metrology system is needed for measuring the distance and angular motion between collectors and combiner. Technologies have been demonstrated at component level. However, they have not been demonstrated at system level. At NPS, a test bed is under development to demonstrate sparse aperture concept at system level. This paper will discuss the design of sparse aperture testbed. In the sparse aperture test bed, white light point source and scene images are used to measure the performance of the sparse aperture system. Initially, multiple satellites are simulated with two separate spacecraft simulation platforms with 6 axis translational and rotation actuation. 6-Axis NanoMaxTM NanoPositioning Flexure Stages by ThorLabs could provide a maximum travel range of 4 mm in translation and 6 degrees in rotation with a minimum resolution of 1.0 nm in translation and 0.018 micro- rad in rotation. Since the combiner spacecraft can be placed far away from each aperture, spherical mirrors with a long focal length (f ¿ 2 meters) will be used as a sparse aperture on each simulated spacecraft platform. This will reduce the complexity of the phasing and alignment compared to the parabolic or hyper

The terahertz (THz) spectral range remains a relatively untapped portion of the electromagnetic spectrum. THz radiation's unique ability to penetrate non-metallic materials presents an exciting opportunity for many imaging applications. The purpose of this research is to investigate a unique imaging method using a THz radiation source and metamaterial absorber. By using a metamaterial absorber, the THz detection frequency of interest can be tuned by controlling geometrical parameters with nearly 100% absorption. THz sensing can be achieved by integrating a metamaterial absorber with bi-material legs to form a sensor. Moveable mirror-like surfaces on the backside of the metamaterial under THz absorption can cause a deflection of visible light and from it, the original image can be reconstructed using an optical readout system. In this thesis, the construction of the optical readout system for characterization of sensor pixels as well as THz imaging is described.

Adaptive optics offer the potential to reduce the high cost and long lead time associated with manufacturing mirrors for spaced-based telescopes by allowing lighter materials to be substituted. These lighter materials lack the optical performance of traditional space-based mirrors. Deformable mirrors could be used to correct for surface aberrations in order to improve the optical quality by altering their surface to adjust the wavefront. Research focused on placing a deformable mirror at the exit pupil of a simulated telescope. Experimental work first studied a severely degraded one-meter carbon fiber reinforced polymer mirror to establish a baseline. Simulations were conducted to see how a notional deformable mirror would be able to negate the optical effects due to a distorted mirror in combination with field angle effects. Results from the investigation showed that a deformable mirror yielded the greatest benefit when applied to a distorted mirror surface. Increasing the actuator count on the deformable mirror boosted the root mean square performance across all field angles. Increasing the actuator stroke yielded minimal benefits after a certain reduction in wavefront had already been achieved. Further research is recommended to focus on using a continuous deformable mirror to account for field angle effects.

In recent years, carbon fiber reinforced polymer (CFRP) mirrors been proposed for use in future imaging satellites. Compared to traditional glass-based mirrors, CFRP mirrors offer reduced manufacturing times, lower coefficients of thermal expansion, lower areal density, and higher strength-to-weight ratios. Shorter manufacturing times promise to reduce program schedule requirements and cost. These advantages come at the expense of surface quality, which results in wavefront errors that are outside of the diffraction limit for optical imaging. To compensate for the reduced surface quality of CFRP mirrors, a deformable mirror (DM) is required in the optical path. During this research, the surface quality of a CFRP mirror was evaluated to establish a root-mean-square (RMS) error threshold for the DM corrections. An integral DM control law that employed a constrained least-squares solution was utilized to reduce the overall system wavefront error to below the specified CFRP error threshold. The application of this control law yielded a 38% reduction in RMS wavefront error (as compared to the CFRP error threshold), thus reducing the CFRP RMS surface performance requirements by the same amount. Reducing the surface performance requirements of CFRP mirrors is a critical step toward employing these mirrors in future imaging satellites.

The purpose of this thesis is to provide a system level performance analysis for an imaging spacecraft. In an imaging spacecraft, an attitude control subsystem’s function is to orient the spacecraft’s body to acquire a target through the use of an actuator. In practice, reaction wheels commonly perform this function by producing a reactive torque on the spacecraft. Consequently, due to the static and dynamic imbalances in individual reaction wheels, an undesired vibration, called jitter, is generated during operation and causes variations in the spacecraft’s attitude. Focusing on missions and payloads operating in the infrared band, optical jitter effects on target detection and tracking performance need to be investigated. Using a quaternion error feedback design, jitter produced by the reaction wheels was recorded while performing a standard spacecraft maneuver. Simulating a low earth orbiting satellite, the vibrations generated a significant optical jitter blur due to a line-of-sight motion. After implementing the optical jitter blur in a baseline high resolution image, the simulation considerably reduced the frame’s spatial resolution and intensity. The simulation demonstrated the jitter blur’s effects on spatial resolution and intensity, which significantly decreased the system’s ability to detect and track objects-of-interest.

The objective of this research is to investigate and evaluate detection and tracking algorithms suitable for Overhead Persistent InfraRed (OPIR) coverage of moving ground targets. One of the largest hurdles is operating with a low signal-to-noise ratio (SNR) in a cluttered environment. The local contrast method (LCM) and principal component analysis (PCA) detection algorithms will be explored and tested while centroid and correlation tracking algorithms will be discussed.Kalman and alpha-beta filters will be compared and contrasted as viable track prediction techniques. This work will also provide a solid knowledge base for future research on the High Energy Laser (HEL) Beam Control Research Testbed that the Naval Postgraduate School is developing in partnership with Boeing Directed Energy Systems. While they are different applications, both HELs and OPIR share common detection and tracking strategies. Simulation results show that the LCM is superior to PCA. However, the best results are obtained by combining the two.Kalman and alpha-beta filters handle single targets with a constant velocity or acceleration with ease, but advanced tracking methods like the velocity matched filter to provide constraints would provide a more robust solution when performing multiple target tracking.

Current laser weapon systems are limited to close range encounters because the laser beam attenuates quickly within the atmosphere. A phenomenon known as deep turbulence is characterized by strong scintillation and branch points in the wave-front phase. Many wave-front sensors perform poorly in the presence of deep turbulence, and are unable to accurately reconstruct the wave-front. This paper examines a wave-front sensor, the self-referencing interferometer (SRI) that is theoretically immune to the effects of deep turbulence. The SRI is both simulated mathematically and constructed in the lab for comparison between analytical and experimental results. Performance of the SRI is analyzed in the presence of realistic deep turbulence effects generated by a spatial light modulator, and realistic noise effects introduced by the digital imaging system. Simulated results show a significant loss of signal level as turbulence is increased, but a resilience of the wave-front sensor above a signal-to-noise ratio of two. Analogously, in the experimental results the signal drops off rapidly with increasing levels of turbulence, and reaches unacceptably low levels above a Rytov number of 0.4. A qualitative analysis of the wave-front reconstruction shows remarkable similarity between simulated and experimental results, though the experimental results contain far more error induced branch points than in the simulation. Methods are being explored to boost the signal and reduce the noise at the camera to allow the system to handle higher levels of turbulence.

The military, and specifically the Navy, is making a move toward the use of directed energy weapons as a tactical advantage over typical kinetic weapons. Directed energy weapons require different controls and delivery methods to deal with differences in interaction of the atmosphere and the laser, as compared to traditional kinetic weapons. This thesis looks at a Noise Variance Weighted Complex Exponential Reconstruction method, and compares with typical zonal and modal least squares methods, to determine the actual wavefront in the presence of disturbances characteristic of the deep turbulence experienced in the maritime environment. The ability of each reconstructor to handle the effects of intensity dropout, branch points, and branch cuts is analyzed along with the effects of signal strength of the sensor, sensor grid size, and level of the intensity experienced.

A deformable mirror is proposed as a solution to correct aberrations on a satellite primary mirror surface. The deformable mirror is used in a double-pass configuration with an interferometer and a mirror segment on the Naval Postgraduate School Segmented Mirror Telescope to test the ability of a deformable mirror to correct inherent surface figure aberrations of an imaging satellite mirror surface. Manual, iterative feedback, and constrained optimization control methods are used to control the deformable mirror to correct surface figure error on the Segmented Mirror Telescope mirror segments to achieve 55% root mean square improvement for the primary mirror segment surface.

The concept of controlling optical laser beams on spacecraft for acquisition, tracking and pointing purposes is quickly becoming a reality. As a result, fine pointing of laser beams is turn out to be an increasingly important research topic. A unique testbed was constructed in order to study and develop new methods for laser beam control. This testbed, the Moving Target-Source Test Fixture (MTSTF), hosts a laser source, the Extremely Agile Relaying Laser Source (EARLS), which has the capability of automatically acquiring and directing a laser beam onto a satellite simulator while in motion. The purpose of this thesis is to make the EARLS platform operational by developing a tracking control system. The ultimate goal is to point the laser beam at the satellite simulatorâ s receiving telescope and maintain the laser within the telescopeâ s limits in the presence of structural disturbance induced by the EARLS motion., http://archive.org/details/beamcontrolofext109452630, Approved for public release; distribution is unlimited.

Two objectives dominate consideration of control moment gyroscopes for spacecraft maneuvers: high torque (equivalently momentum) and singularity-free operations. This paper adds to the significant body of research toward these two goals using a minimal three-control-moment-gyroscope array to provide significant singularity-free momentum performance increase spherically (in all directions) by modification of control-moment-gyroscope skew angles, compared with the ubiquitous pyramid geometry skewed at 54.73 deg. Spherical 1H (one control moment gyroscope’s worth momentum) singularity-free momentum is established with bidirectional 1H and 2H in the third direction in a baseline configuration. Next, momentum space reshaping is shown via mixed skew angles permitting orientation of maximum singularity-free angular momentum into the desired direction of maneuver (yaw in this study). Finally, a decoupled gimbal angle calculation technique is shown to avoid loss of attitude control associated with singular matrix inversion. This technique permits 3H (maximal) yaw maneuvers without loss of attitude control despite passing through singularity. These claims are demonstrated analytically, then heuristically, and finally validated experimentally.

The objective of this research was to develop beam jitter control methods for a High Energy Laser (HEL) testbed. The first step was to characterize the new HEL testbed at NPS. This included determination of natural frequencies and component models which were used to create a Matlab/Simulink model of the testbed. Adaptive filters using Filtered-X Least Mean Squares (FX-LMS) and Filtered-X Recursive Least Square (FX-RLS) were then implemented. Disturbance sources included narrowband mechanical vibration of the optical bench as well as broadband atmospheric turbulence (simulation only). A feedback controller with adaptive filter (or feedbacktype adaptive filter) was applied to a multi-rate video tracking loop, which required precise plant identification to prevent instabilities. A strapdown type Inertial Reference Unit (IRU) system was investigated using a reference laser, Position Sensing Device (PSD), Fiber Optic Gyro (FOG), and Fast Steering Mirror (FSM). The controller of the strapdown type IRU system consisted of a feedforward gyro signal in parallel with a feedback PSD signal. A Video tracking control system was simulated and tested on the HEL testbed. The IRU strapdown controller was simulated and preliminary tests conducted. Simulation and experimental results demonstrated superior performance over classical control methods., http://archive.org/details/improvedbeamjitt109454439, Outstanding Thesis, US Navy (USN) author., Approved for public release; distribution is unlimited.

Phase Diversity Wavefront Sensing (PD WFS) is a wavefront reconstruction technique used in adaptive optics, which takes advantage of the curvature conjugating analog physical properties of a deformable mirror (MMDM or Bi-morph) such that the computational intensity required for correcting an aberrated wavefront, becomes simplified over traditional Shack-Hartmann WFS techniques. By looking at an image reflected off a deformable mirror by two cameras placed on either side of focus of a lens, intensity differences, indicating wavefront aberration in the beam, can be detected by the cameras acting together as a WFS and analyzed by a computer providing control to the actuators of a deformable mirror such that any detected difference in intensities between the two cameras can be minimized. This process of mirror surface conjugation serves to correct for the aberrated or curved wavefront by reflecting a new wavefront, compensated for curvature, such that its reflection is approximately planar. The theory of PD WFS is well documented however, there is very little quantifiable information regarding the specific challenges in designing a functioning PD WFS. In this research a PD WFS was designed and the concept proven such that a wavefront could be corrected through a computer controlled closed loop conjugation of a deformable mirror. The results were analyzed using a traditional Shack-Hartmann WFS and off-the-shelf "Front Surfer" wavefront analysis software to verify the validity of the experimental data. PD WFS has become critical in the development of segmented mirror adaptive optical systems where traditional wavefront reconstruction using Shack-Hartmann wavefront sensing tends to break down at the mirror segment edges. The Naval Postgraduate School, Spacecraft Research and Design Center (SRDC) intends to explore the use of a segmented mirror adaptive optical systems for space based applications., http://archive.org/details/phasediversitywa109453140, US Navy (USN) author., Approved for public release; distribution is unlimited.

Future long dwell high resolution imagery satellites and space telescopes will require very large flexible primary mirrors. These large mirrors face many challenges including optical surface imperfections, structural vibrations, and jitter. A flexible mirror can overcome some of these challenges by applying adaptive optics techniques to correct mirror deformations and aberrations to produce image quality data. This paper examines and develops control techniques to control a deformable mirror subjected to a disturbance. The experimental portion of the work uses discrete time proportional integral control with second order filters to control disturbances in a deformable mirror and correct aberrations in an adaptive optics system using laser light. Using an adaptive optics testbed containing two deformable mirrors, two fast steering mirrors, two wave front sensors, a position sensor, and a combination of lenses the system corrects a simulated dynamic disturbance induced in the deformable mirror. Experiments using the described testbed successfully demonstrate wavefront control methods, including a combined iterative feedback and gradient control technique. This control technique results in a three fold improvement in RMS wavefront error over the individual controllers correcting from a biased mirror position. Second order discrete time notch filters are also used to remove induced low frequency actuator and sensor noise at 0.8 Hz, 2 Hz and 5 Hz. Additionally a 2 Hz structural disturbance is simulated on a Micromachined Membrane Deformable Mirror and removed using discrete time notch filters combined with a modal iterative closed loop feedback controller, showing a 36 fold improvement in RMS wavefront error over the iterative closed loop feedback alone., http://archive.org/details/wavefrontcontrol109453085, Outstanding Thesis, US Air Force (USAF) author., Approved for public release; distribution is unlimited.

The Segmented Mirror Telescope (SMT) housed at the Naval Postgraduate School is a unique, state-of-the-art optical instrument built to explore new technologies needed for future spacebased telescopes. A discrete Fourier transform wavefront reconstruction technique developed at Lawrence Livermore National Laboratory is discussed in this thesis as applied to a hexagonal aperture. A Fourier domain implementation of a spatial-frequency modal controller for a simple spring-mass model of a deformable mirror surface is provided by this thesis. This technique avoids more difficult time-domain solutions, is computationally efficient and scalable to much larger multi-input, multi-output systems than the SMT., http://archive.org/details/segmentedmirrort109455614, Approved for public release; distribution is unlimited.

The Segmented Mirror Telescope (SMT) housed at the Naval Postgraduate School is a unique, state-of-the-art optical instrument built to explore new technologies needed for future spacebased telescopes. A discrete Fourier transform wavefront reconstruction technique developed at Lawrence Livermore National Laboratory is discussed in this thesis as applied to a hexagonal aperture. A Fourier domain implementation of a spatial-frequency modal controller for a simple spring-mass model of a deformable mirror surface is provided by this thesis. This technique avoids more difficult time-domain solutions, is computationally efficient and scalable to much larger multi-input, multi-output systems than the SMT.

Air-bearing-based spacecraft simulators are widely used to develop and verify spacecraft control techniques required by modern spacecraft applications. To create a spacelike environment with ground spacecraft simulators, the effects of gravity should be minimized. For a spherical air-bearing system with 3 rotational degrees of freedom, the center of rotation of the spacecraft simulator should be exactly aligned with the center of gravity. This paper presents the automatic mass balancing method, which compensates for the center of gravity offset from the center of rotation by actuating three proof masses on linear motion stages. Adaptive control of the automatic mass balancing system is used while the balancing masses are actuated in real time. The proposed techniques are implemented on the ground-based three-axis spacecraft simulator for the bifocal relay mirror spacecraft., Presented as Paper 6595 at the AIAA Guidance, Navigation, and Control Conference and Exhibit, Keystone, Colorado, 21-24 August 2006. Published in Journal of Guidance, Control, and Dynamics, v32 n3 p1005-1017, May-Jun 2009.

The objective of this research was to develop beam jitter control methods for a High Energy Laser (HEL) testbed. The first step was to characterize the new HEL testbed at NPS. This included determination of natural frequencies and component models which were used to create a Matlab/Simulink model of the testbed. Adaptive filters using Filtered-X Least Mean Squares (FX-LMS) and Filtered-X Recursive Least Square (FX-RLS) were then implemented. Disturbance sources included narrowband mechanical vibration of the optical bench as well as broadband atmospheric turbulence (simulation only). A feedback controller with adaptive filter (or feedbacktype adaptive filter) was applied to a multi-rate video tracking loop, which required precise plant identification to prevent instabilities. A strapdown type Inertial Reference Unit (IRU) system was investigated using a reference laser, Position Sensing Device (PSD), Fiber Optic Gyro (FOG), and Fast Steering Mirror (FSM). The controller of the strapdown type IRU system consisted of a feedforward gyro signal in parallel with a feedback PSD signal. A Video tracking control system was simulated and tested on the HEL testbed. The IRU strapdown controller was simulated and preliminary tests conducted. Simulation and experimental results demonstrated superior performance over classical control methods.

Two objectives dominate consideration of control moment gyroscopes (CMGs) for spacecraft maneuvers: High torque (or equivalently momentum) and singularity-free operations. Utilizing a 3/4 CMG skewed-pyramid in the optimal singularity-free configuration, this paper develops a decoupled control strategy to reduce the remaining singular conditions. Analysis is provided to justify the argument with promising simulations followed by experimental verification using a free-floating satellite simulator.

High energy laser (HEL) weapons are some of most challenging military applications in the future battle fields since the speed of light delivery enables the war fighter to engage very distant targets immediately. The issues of the technology on the HEL system include various types of high energy laser devices, beam control systems, atmospheric propagation, and target lethality. Among them, precision pointing of laser beam and high-bandwidth rejection of jitters produced by platform vibrations are one of the key technologies for the emerging fields of laser communications and HEL systems. HEL testbed has been developed to support the research environments on the precision beam control technology including acquisition, tracking, and pointing. The testbed incorporates optical table, two axis gimbal, high speed computers, and a variety of servo components, sensors, optical components, and software. In this report, system configuration and operation modes of the testbed are briefly introduced. The results of the experiments and integrated modeling from component to system level are described and discussed. Based on these results, new control algorithms are designed and it is shown that the algorithm can improve the pointing performance of the system.

High energy laser (HEL) weapons are some of most challenging military applications in the future battle fields since the speed of light delivery enables the war fighter to engage very distant targets immediately. The issues of the technology on the HEL system include various types of high energy laser devices, beam control systems, atmospheric propagation, and target lethality. Among them, precision pointing of laser beam and high-bandwidth rejection of jitters produced by platform vibrations are one of the key technologies for the emerging fields of laser communications and HEL systems. HEL testbed has been developed to support the research environments on the precision beam control technology including acquisition, tracking, and pointing. The testbed incorporates optical table, two axis gimbal, high speed computers, and a variety of servo components, sensors, optical components, and software. In this report, system configuration and operation modes of the testbed are briefly introduced. The results of the experiments and integrated modeling from component to system level are described and discussed. Based on these results, new control algorithms are designed and it is shown that the algorithm can improve the pointing performance of the system., Approved for public release; distribution is unlimited.

High energy laser (HEL) weapons are some of most challenging military applications in the future battle fields since the speed of light delivery enables the war fighter to engage very distant targets immediately. The issues of the technology on the HEL system include various types of high energy laser devices, beam control systems, atmospheric propagation, and target lethality. Among them, precision pointing of laser beam and high-bandwidth rejection of jitters produced by platform vibrations are one of the key technologies for the emerging fields of laser communications and HEL systems. HEL testbed has been developed to support the research environments on the precision beam control technology including acquisition, tracking, and pointing. The testbed incorporates optical table, two axis gimbal, high speed computers, and a variety of servo components, sensors, optical components, and software. In this report, system configuration and operation modes of the testbed are briefly introduced. The results of the experiments and integrated modeling from component to system level are described and discussed. Based on these results, new control algorithms are designed and it is shown that the algorithm can improve the pointing performance of the system.

Phase Diversity Wavefront Sensing (PD WFS) is a wavefront reconstruction technique used in adaptive optics, which takes advantage of the curvature conjugating analog physical properties of a deformable mirror (MMDM or Bi-morph) such that the computational intensity required for correcting an aberrated wavefront, becomes simplified over traditional Shack-Hartmann WFS techniques. By looking at an image reflected off a deformable mirror by two cameras placed on either side of focus of a lens, intensity differences, indicating wavefront aberration in the beam, can be detected by the cameras acting together as a WFS and analyzed by a computer providing control to the actuators of a deformable mirror such that any detected difference in intensities between the two cameras can be minimized. This process of mirror surface conjugation serves to correct for the aberrated or curved wavefront by reflecting a new wavefront, compensated for curvature, such that its reflection is approximately planar. The theory of PD WFS is well documented however, there is very little quantifiable information regarding the specific challenges in designing a functioning PD WFS. In this research a PD WFS was designed and the concept proven such that a wavefront could be corrected through a computer controlled closed loop conjugation of a deformable mirror. The results were analyzed using a traditional Shack-Hartmann WFS and off-the-shelf "Front Surfer" wavefront analysis software to verify the validity of the experimental data. PD WFS has become critical in the development of segmented mirror adaptive optical systems where traditional wavefront reconstruction using Shack-Hartmann wavefront sensing tends to break down at the mirror segment edges. The Naval Postgraduate School, Spacecraft Research and Design Center (SRDC) intends to explore the use of a segmented mirror adaptive optical systems for space based applications.

Future long dwell high resolution imagery satellites and space telescopes will require very large flexible primary mirrors. These large mirrors face many challenges including optical surface imperfections, structural vibrations, and jitter. A flexible mirror can overcome some of these challenges by applying adaptive optics techniques to correct mirror deformations and aberrations to produce image quality data. This paper examines and develops control techniques to control a deformable mirror subjected to a disturbance. The experimental portion of the work uses discrete time proportional integral control with second order filters to control disturbances in a deformable mirror and correct aberrations in an adaptive optics system using laser light. Using an adaptive optics testbed containing two deformable mirrors, two fast steering mirrors, two wave front sensors, a position sensor, and a combination of lenses the system corrects a simulated dynamic disturbance induced in the deformable mirror. Experiments using the described testbed successfully demonstrate wavefront control methods, including a combined iterative feedback and gradient control technique. This control technique results in a three fold improvement in RMS wavefront error over the individual controllers correcting from a biased mirror position. Second order discrete time notch filters are also used to remove induced low frequency actuator and sensor noise at 0.8 Hz, 2 Hz and 5 Hz. Additionally a 2 Hz structural disturbance is simulated on a Micromachined Membrane Deformable Mirror and removed using discrete time notch filters combined with a modal iterative closed loop feedback controller, showing a 36 fold improvement in RMS wavefront error over the iterative closed loop feedback alone.

In this paper, experimental verifications of various jerk-limited control designs are presented. The focus is on a large slew maneuver of a flexible spacecraft where the fast and accurate pointing is required at the end of the maneuver. Various control profiles are investigated including input shaping and jerk-limited optimal control profiles, especially to account the effect of jerk in the control profile. The control design is verified by the experiment using the Flexible Spacecraft Simulator (FSS)., Presented at the AIAA Guidance, Navigation, and Control Conference and Exhibit held in Keystone, Colorado on 21-24 August 2006.

The ground-based spacecraft simulator is a useful tool to develop and verify new control laws required by modern spacecraft applications. In order to simulate the space environment with the ground-based spacecraft simulator, the effects of gravity should be minimized. In this paper, a three-axis rotational rigid spacecraft simulator on a spherical air-bearing system is considered. The method of estimating the inertial properties of a spacecraft as well as the location of the center of mass is presented with the batch least- square estimation. The identified center of mass location is used to actuate the automatic mass balancing system to compensate for the center of mass offset. Adaptive control of the automatic mass balancing system is also presented when the balancing masses are actuated in real-time to eliminate the center of mass offset from the center of rotation. The proposed technique is implemented on the Three Axis Simulator 2 (TAS2) which is a ground-based experimental testbed for the Bifocal Relay Mirror Spacecraft (BRMS)., Presented at the AIAA Guidance, Navigation, and Control Conference and Exhibit held in Keystone, Colorado on 21-24 August 2006.

Current laser weapon systems are limited to close range encounters because the laser beam attenuates quickly within the atmosphere. A phenomenon known as deep turbulence is characterized by strong scintillation and branch points in the wave-front phase. Many wave-front sensors perform poorly in the presence of deep turbulence, and are unable to accurately reconstruct the wave-front. This paper examines a wave-front sensor, the self-referencing interferometer (SRI) that is theoretically immune to the effects of deep turbulence. The SRI is both simulated mathematically and constructed in the lab for comparison between analytical and experimental results. Performance of the SRI is analyzed in the presence of realistic deep turbulence effects generated by a spatial light modulator, and realistic noise effects introduced by the digital imaging system. Simulated results show a significant loss of signal level as turbulence is increased, but a resilience of the wave-front sensor above a signal-to-noise ratio of two. Analogously, in the experimental results the signal drops off rapidly with increasing levels of turbulence, and reaches unacceptably low levels above a Rytov number of 0.4. A qualitative analysis of the wave-front reconstruction shows remarkable similarity between simulated and experimental results, though the experimental results contain far more error induced branch points than in the simulation. Methods are being explored to boost the signal and reduce the noise at the camera to allow the system to handle higher levels of turbulence., http://archive.org/details/performanceofsel1094538959, Lieutenant, United States Navy, Approved for public release; distribution is unlimited.

The terahertz (THz) spectral range remains a relatively untapped portion of the electromagnetic spectrum. THz radiation's unique ability to penetrate non-metallic materials presents an exciting opportunity for many imaging applications. The purpose of this research is to investigate a unique imaging method using a THz radiation source and metamaterial absorber. By using a metamaterial absorber, the THz detection frequency of interest can be tuned by controlling geometrical parameters with nearly 100% absorption. THz sensing can be achieved by integrating a metamaterial absorber with bi-material legs to form a sensor. Moveable mirror-like surfaces on the backside of the metamaterial under THz absorption can cause a deflection of visible light and from it, the original image can be reconstructed using an optical readout system. In this thesis, the construction of the optical readout system for characterization of sensor pixels as well as THz imaging is described., http://archive.org/details/memsterahertzfoc1094549469, Captain, United States Marine Corps, Approved for public release; distribution is unlimited.

The objective of this research is to investigate and evaluate detection and tracking algorithms suitable for Overhead Persistent InfraRed (OPIR) coverage of moving ground targets. One of the largest hurdles is operating with a low signal-to-noise ratio (SNR) in a cluttered environment. The local contrast method (LCM) and principal component analysis (PCA) detection algorithms will be explored and tested while centroid and correlation tracking algorithms will be discussed.Kalman and alpha-beta filters will be compared and contrasted as viable track prediction techniques. This work will also provide a solid knowledge base for future research on the High Energy Laser (HEL) Beam Control Research Testbed that the Naval Postgraduate School is developing in partnership with Boeing Directed Energy Systems. While they are different applications, both HELs and OPIR share common detection and tracking strategies. Simulation results show that the LCM is superior to PCA. However, the best results are obtained by combining the two.Kalman and alpha-beta filters handle single targets with a constant velocity or acceleration with ease, but advanced tracking methods like the velocity matched filter to provide constraints would provide a more robust solution when performing multiple target tracking., http://archive.org/details/acomparisonofdet1094544600, Lieutenant Commander, United States Navy, Approved for public release; distribution is unlimited.

In recent years, carbon fiber reinforced polymer (CFRP) mirrors been proposed for use in future imaging satellites. Compared to traditional glass-based mirrors, CFRP mirrors offer reduced manufacturing times, lower coefficients of thermal expansion, lower areal density, and higher strength-to-weight ratios. Shorter manufacturing times promise to reduce program schedule requirements and cost. These advantages come at the expense of surface quality, which results in wavefront errors that are outside of the diffraction limit for optical imaging. To compensate for the reduced surface quality of CFRP mirrors, a deformable mirror (DM) is required in the optical path. During this research, the surface quality of a CFRP mirror was evaluated to establish a root-mean-square (RMS) error threshold for the DM corrections. An integral DM control law that employed a constrained least-squares solution was utilized to reduce the overall system wavefront error to below the specified CFRP error threshold. The application of this control law yielded a 38% reduction in RMS wavefront error (as compared to the CFRP error threshold), thus reducing the CFRP RMS surface performance requirements by the same amount. Reducing the surface performance requirements of CFRP mirrors is a critical step toward employing these mirrors in future imaging satellites., http://archive.org/details/reducingsurfacep1094547894, Lieutenant Commander, United States Navy, Approved for public release; distribution is unlimited.

Adaptive optics offer the potential to reduce the high cost and long lead time associated with manufacturing mirrors for spaced-based telescopes by allowing lighter materials to be substituted. These lighter materials lack the optical performance of traditional space-based mirrors. Deformable mirrors could be used to correct for surface aberrations in order to improve the optical quality by altering their surface to adjust the wavefront. Research focused on placing a deformable mirror at the exit pupil of a simulated telescope. Experimental work first studied a severely degraded one-meter carbon fiber reinforced polymer mirror to establish a baseline. Simulations were conducted to see how a notional deformable mirror would be able to negate the optical effects due to a distorted mirror in combination with field angle effects. Results from the investigation showed that a deformable mirror yielded the greatest benefit when applied to a distorted mirror surface. Increasing the actuator count on the deformable mirror boosted the root mean square performance across all field angles. Increasing the actuator stroke yielded minimal benefits after a certain reduction in wavefront had already been achieved. Further research is recommended to focus on using a continuous deformable mirror to account for field angle effects., http://archive.org/details/opticalcorrectio1094551588, Lieutenant, United States Navy, Approved for public release; distribution is unlimited.

A deformable mirror is proposed as a solution to correct aberrations on a satellite primary mirror surface. The deformable mirror is used in a double-pass configuration with an interferometer and a mirror segment on the Naval Postgraduate School Segmented Mirror Telescope to test the ability of a deformable mirror to correct inherent surface figure aberrations of an imaging satellite mirror surface. Manual, iterative feedback, and constrained optimization control methods are used to control the deformable mirror to correct surface figure error on the Segmented Mirror Telescope mirror segments to achieve 55% root mean square improvement for the primary mirror segment surface., http://archive.org/details/correctingsurfac1094539033, Outstanding Thesis, Commander, United States Navy, Approved for public release; distribution is unlimited.

The military, and specifically the Navy, is making a move toward the use of directed energy weapons as a tactical advantage over typical kinetic weapons. Directed energy weapons require different controls and delivery methods to deal with differences in interaction of the atmosphere and the laser, as compared to traditional kinetic weapons. This thesis looks at a Noise Variance Weighted Complex Exponential Reconstruction method, and compares with typical zonal and modal least squares methods, to determine the actual wavefront in the presence of disturbances characteristic of the deep turbulence experienced in the maritime environment. The ability of each reconstructor to handle the effects of intensity dropout, branch points, and branch cuts is analyzed along with the effects of signal strength of the sensor, sensor grid size, and level of the intensity experienced., http://archive.org/details/comparisonofnois1094534707, Ensign, United States Navy, Approved for public release; distribution is unlimited.

The purpose of this thesis is to provide a system level performance analysis for an imaging spacecraft. In an imaging spacecraft, an attitude control subsystem’s function is to orient the spacecraft’s body to acquire a target through the use of an actuator. In practice, reaction wheels commonly perform this function by producing a reactive torque on the spacecraft. Consequently, due to the static and dynamic imbalances in individual reaction wheels, an undesired vibration, called jitter, is generated during operation and causes variations in the spacecraft’s attitude. Focusing on missions and payloads operating in the infrared band, optical jitter effects on target detection and tracking performance need to be investigated. Using a quaternion error feedback design, jitter produced by the reaction wheels was recorded while performing a standard spacecraft maneuver. Simulating a low earth orbiting satellite, the vibrations generated a significant optical jitter blur due to a line-of-sight motion. After implementing the optical jitter blur in a baseline high resolution image, the simulation considerably reduced the frame’s spatial resolution and intensity. The simulation demonstrated the jitter blur’s effects on spatial resolution and intensity, which significantly decreased the system’s ability to detect and track objects-of-interest., http://archive.org/details/opticaljittereff1094547946, Lieutenant, United States Navy, Approved for public release; distribution is unlimited.