U.S. Air Force Research Lab Summer Faculty Fellowship Program

U.S. Air Force Research Lab Summer Faculty Fellowship Program

U.S. Air Force Research Lab Summer Faculty Fellowship Program

AFIT (AFIT Wright-Patterson Air Force Base, Ohio )

SF.50.24.B10219: Development of Efficient Hydrogen Storage and Utilization Technologies for Possible Decarbonization of Aircraft Propulsion

Rahman, Muhammad - 937-656-5254

Hydrogen is being considered as a potential fuel source for military aircraft propulsion. It has the potential to greatly reduce the environmental impact of commercial as well as military aviation. Aviation sector is a major contributor of global climate change with commercial aircraft
contributing to approximately 3% and military operations contributing up to 15% of the total greenhouse gas emission. This is one of the most difficult sectors to decarbonize due to challenges associated with storage of hydrogen and lack of infrastructure to support any large-scale operation. Therefore, the developments of new technologies for sustainable aviation will play a key role in achieving Net Zero by 2050. Hydrogen may be burned directly in modified gas turbine engines or can be used in fuel cells to generate electricity which then powers electric motors to drive propellers. Hydrogen has much lower density compared to current aviation fuels, meaning much larger tanks are needed to store enough fuel for long-range flights, adding significant volume and weight to the aircraft. This research study is intended to explore pathways to improve cryogenic storage of liquid hydrogen fuel needed for long-haul flights as well as exploring more efficient chemical storage of hydrogen for possible applications in fuel cells to fly all electric sub-regional and regional aircrafts. The investigation may span over a comprehensive review of the state-of-the art, technological feasibility study, cryogenic storage tank design and optimization, hydrogen combustor design, integrated thermal management, fuel cells and electrical architecture systems, and special safety issues related to using hydrogen as a fuel.

SF.50.24.B10218: Development of Low-Cost and Low-Energy Intensive Atmospheric Water Extraction System for Military Applications

Rahman, Muhammad - 937-656-5254

The idea of this project is to explore possible pathways to develop an atmospheric water extraction system that can be used to supply potable water for a range of atmospheric conditions (temperature and humidity) when warfighters are cut off from the supply chain and no external liquid water source such as groundwater, seawater, river, or lake is available. The target is to build a device that can work with low energy consumption (of the order of 0.1 kWh electrical or thermal) for each liter of water extracted from the atmosphere. This game-changing device is expected to help
Department of Defense as it moves towards more mobile, flexible, and self-sufficient operations, including the Expeditionary Advanced Base Operations (EABO, U.S. Marine Corps), the Multi-Domain Operations (MDO, Army), and the Agile Combat Employment (ACE, U.S. Air Force)
concepts. Broadly, the development of technology for generating fresh water from atmospheric air will be a quite effective solution in overcoming the global fresh and clean water crisis. This research and development effort will review the latest developments in the field, critically evaluate
the competing technological concepts, and put together the design and development of a device that can produce clean water with high energy efficiency and large capacity.

SF.50.24.B10170: Sustainable Construction Methods and Materials

Abdelraheem, Mohamed - 937-255-3636

This research program is dedicated to the development and evaluation of sustainable construction methods and materials. Our projects encompass a wide range of sustainability-related topics, including resilient infrastructure, green building practices, waste recycling, life cycle cost analysis, and the use of supplementary cementitious materials. Our primary focus is on sustainable and resilient pavements and the cement heat of hydration.

AFIT is seeking a highly qualified scientist or engineering professor to contribute to exploratory experimentation and the development of future research proposals. The ideal candidate will have, extensive experience in sustainable construction and materials research, a strong record of publications in peer-reviewed journals, and Proven success in securing research grants.

SF.50.24.B10169: Kernel Methods for the Solution of Partial Differential Equations (PDEs)

Reeger, Jonah - 937-255-3636

This research program will develop novel theoretical and computational results related to the use of so-called kernel methods. Kernel methods employ a basis that includes shifts of a chosen conditionally-positive definite function--the kernel--and supplemental polynomial terms in the process of approximating a function. The ideas behind kernel methods allow for efficient function approximation with high orders of accuracy even in the presence of variable node spacing. Such methods are also easily generalized to high dimensions and adaptable to a wide variety of domains. The use of kernel methods will be directed toward the solution of PDEs, with the interplay between lasers and the medium that they are propagating through an application of interest.

SF.50.24.B10157: Artificial Intelligence in National Defense: Analysis of Opportunities and Challenges

Li, Kunpeng - 937-255-6309

Digital technologies have become increasingly important for national defense. Artificial intelligence (AI) significantly enhances national defense by proactively addressing security challenges, improving threat detection and assessment, and strengthening response capabilities. The applications of AI in national defense are multifaceted, ranging from detecting and responding to cyber threats, developing autonomous weapons systems, supporting critical national infrastructures and logistics systems, to facilitating combat simulation and training. This research aims to analyze the critical role of AI in modern national defense strategies, exploring the inherent opportunities and challenges.

SF.50.24.B10156: Multidisciplinary Analysis & Design for Hypersonic Vehicles

Camberos, Jose - 937-656-4826

The basic objective of multidisciplinary design is to integrate the various disciplines that constitute the environment of aerospace vehicles. The goal of modern design is to optimize the total system rather than the individual components, permitting the conflicting requirements of the subsystems to be handled much more effectively in getting optimal solutions. High-Speed Aerospace Vehicle design is a large optimization problem consisting of libraries of variables, constraints, and performance functions. By expanding and contracting these libraries, we can explore the inherent coupling between subsystems and the disciplines and their impact on system level performance. Topics of interest include (1) simultaneous design with multiple constraints; (2) structural requirements derived from strength, stiffness, and frequency considerations; (3) static and dynamic aero-thermo-elastic requirements; (4) requirements from acoustic and thermal environments; (5) linear and nonlinear aerodynamic interactions with the propulsion, structure, and control sub-systems; (6) tailoring of composites and other new high-temperature materials; (7) shape and topology optimization; (8) development and testing of efficient optimization methods; (8) sensitivity analyses; (9) Uncertainty Quantification for Design; (10) System Modeling and Discretization for Design, (11) Multi-Fidelity Analysis for Design; and (12) Optimization of transient systems.

key words

Hypersonic vehicle conceptual design; Systems integration; Aeroelasticity; Aerothermoelasticity; Aeroservoelasticity; Composite materials; Computer-aided design; Materials science; Design optimization; Structural mechanics; Uncertainty quantification for design; High-speed systems;

SF.50.23.B10147: Single Photon Based Quantum Technologies and Applications

Wyman, Keith - 937 656-0661

As a graduate school focused on Department of Defense relevant research and education, the Air Force Institute of Technology (AFIT) is at the forefront of optics and quantum programs. Specifically, we are studying free space and fiber-based quantum networks which can enable global-scale quantum communication via satellite-based nodes, quantum ground transceivers, and regional fiber networks. To enable building of robust global quantum networks, it is critical to learn how the state of the qubit is transformed while propagating through the atmosphere or through fiber. To that end, AFIT is seeking a scientist/engineering professor for its summer faculty fellowship who would perform experimental or computational research in:
1. Quantum technology development
a. single photon sources
b. detectors
c. quantum memory
2. Quantum optics, sensors and informatics
3. Optical Beam manipulation for propagation through atmosphere
4. Experiments involving foundations of quantum mechanics

SF.50.23.B10146: Artificially Intelligent Decision-making for Resilient Smart City Microgrids (AID-GRID)

Langhals, Brent - 937 656-4579

Climate change places an immense stress on existing vulnerabilities within our current energy infrastructure via two primary challenges. First, operational management and control of smart city microgrids is becoming more and more challenging due to the high levels of uncertainties and intermittencies originating from new renewable energy penetration. Second, given the increased frequency of extreme climate events and uncertainties around the complex nature of energy systems, ensuring resilience against system anomalies and climate catastrophes is becoming progressively more difficult. To address these challenges, this project will develop artificial intelligence (AI) to dynamically update dispatch decisions in smart city microgrids in real-time based on operational conditions.

SF.50.23.B10143: Computational Exploration of Metallic Alloys and Their Oxides in Extreme Environments

Samin, Adib - 937-656-5721

This opportunity focuses on the computational exploration of metallic alloys and their oxides in extreme environments. The endeavor seeks to investigate the behavior of structural materials under high stresses, elevated temperatures, intense radiation, and corrosive environments. In addition, the coupling between these various phenomena and their effects on structure-property relationships is of the utmost importance. The aim is to assess the role that different mechanisms can play in controlling thermodynamic and kinetic aspects of material behavior and response.

SF.50.23.B10142: Automated Fusion of Titanium Alloy Imaging Data for Microtexture Region (MTR) Characterization Using Deep Learning Techniques

Gaw, Nathan - 937-255-6418

This research project aims to address a critical challenge in the characterization of microtexture regions (MTRs) within titanium alloys, which are integral to the early onset creep fatigue failure of rotating turbomachinery components. The fusion of various inspection methods, Scanning Electron Microscopy (SEM), Polarized Light Microscopy (PLM), Scanning Acoustic Microscopy (SAM), and Eddy Current Testing (ECT), has been proposed by researchers at the Air Force Research Laboratory (AFRL) to characterize MTRs efficiently. However, successful fusion of these imaging data requires accurate image registration, which remains a significant challenge due to the absence of baseline approaches in the literature. The visiting professor will lead a research effort to develop advanced neural network architectures for the automated registration/fusion of SEM, PLM, SAM, and/or ECT images for MTR characterization. By contributing to the ongoing challenges in titanium MTR characterization, this research project will have a direct impact on improving the materials state awareness of critical aircraft components, thereby enhancing the reliability and safety of aerospace systems. This research will provide opportunities for future collaboration with various institutions, including Air Force Institute of Technology (AFIT) and the Air Force Research Laboratory (AFRL).

SF.50.23.B10141: Heat Exchangers for Aircraft Systems

Gorla , Rama - 937-656-6103

Develop an efficient heat exchanger for aircraft systems. Future Autonomous Collaboration (ACP) will need non-conventional aircraft thermal management systems to remove challenging heat loads. Historically, unmanned aircraft thermal management systems have relied on combination of ram air and single phase loops for cooling aircraft systems. The heat exchangers used in those legacy systems tend to be large, built with a bulky shape and fabricated using conventional manufacturing methods. These bulky rectangular shaped heat exchangers are not adaptable to the unique volume spaces available on ACP. To address these needs, heat exchangers with enhanced power to weight and power to volume are needed. In addition, ACP heat exchangers may need to be easily integrated into existing structures with unusual shapes and or in unused spaces in the fuselage or weapon pods. Triply Periodic Minimal Surfaces (TPMS) have a high compression strength to weight ratio and can provide a lot of surface area. With careful design of a TPMS structure, flow paths can be separated, making these surfaces potentially useful as a heat transfer and structural medium. This work aims to research solutions that will enable the use of additive manufactured conformal heat exchangers on ACP in order to maintain optimal operating conditions. Models will demonstrate the ability to accurately predict the thermal performance of heat exchangers for future design configurations. Heat exchanger topology analysis will identify different configurations and designs of heat exchangers to determine benefits and drawbacks of specific topologies. Experimental data will support the modeling tasking by validating the design dimensional parameters and determining the feasibility of topology configurations.

Research Classification/Restrictions: Unclassified

SF.50.23.B10140: Uncertainty Quantification for Heat Transfer/Thermal Management

Gorla , Rama - 937-656-6103

Develop a scalable and holistic UQ framework that enables the simultaneous inclusion of multi-physics and multi-fidelity models as well as experimental data at varying levels of trust is necessary to address UQ in Thermal Management for aircraft. Such a framework must leverage existing methods and develop new approaches to address increased computational accuracy requirements of UQ within a design environment. Historically, Uncertainty Quantification (UQ) is performed late in the design cycle, when mitigation of deficiencies is costly or may result in a penalty to performance or capability. These late defects and faults may be critical due to unanticipated interdisciplinary couplings or due to the uncertain nature of anticipated interdisciplinary quantities of interest. Types of uncertainty may include, but are not limited to: parameter uncertainties, such as model or design parameters, geometric or material variables, and parameters associated with environment and process control; model uncertainties, such as from physics-based models from simple to complex, empirical models based on experiments, couplings/interfaces between disciplines, and model boundary conditions; data uncertainties, including noise, measurement errors, and missing data; requirements or usage uncertainty, including uncertainty in constraints; and, uncertainties arising from simulation, including discretization errors, round-off errors, and algorithmic errors.

Research Classification/Restrictions: Unclassified

SF.50.21.B0004: Fundamental Laser-Matter Interactions, Sensors and Quantum Technologies

Patnaik, Anil - 937-656-5729

As a graduate school focused on Air Force relevant research and education, the Air Force Institute of Technology (AFIT) is at the forefront of optics and sensing programs. The advent of novel high power, high pulse-repetition rate and short-pulse duration (up to a few femtoseconds) lasers are pushing the boundaries of diagnostics and sensing capabilities. Also, recent developments in miniaturization of sensors are driving the LMI boundaries into quantum regime, opening up a whole new world of science that allows to us to imagine technologies beyond the fundamental limits currently thought to be imposed by nature. This pursuit is an opportunity for any ambitious researcher to pursue a research career in AFIT`s vibrant environment on any of the following broad theoretical and experimental topics (but not limited to):
1) Quantum Technologies:
a. Quantum optics, sensors and informatics
b. Manipulating non-linear dispersion, slow and fast light and applications
c. Foundations of quantum mechanics
2) Laser-matter Interaction Applications:
a. Ultrafast laser-matter interaction, nonlinear optics, laser-induced plasma and intense-laser matter interactions
b. Hypersonic material and electro-magnetics

SF.50.21.B0003: High energy-density physics research

Dexter, Michael - 937-656-6196

AFIT is seeking a scientist/engineering professor for its summer faculty fellowship from a US university who would perform experimental and computational research in relativistic intensity laser plasma interaction, high energy density physics, high repetition rate targets for such experiments, detection and characterization of high energy particles and radiation that are generated in such interactions. The Summer Fellow is expected to work in the group of Extreme Light Laboratory (ELL), which works on fundamental research and applications involving high intensity laser plasma interaction (LPI), which has broad applicability in problems that are of interest to DoD in particular and the scientific and industrial community at large. Of Particular interest is probing Electric and Magnetic field structures generated during intense LPI, studying feasibility of developing a proton radiography source using ELL high repetition rate laser targetry platform, and development of a machine learning (ML) platform for statistical analysis of high volume of LPI data collected at the lab to predict the influence of statistical and systematic fluctuation in laser and target parameters on LPI based x-ray, electron, proton and neutron generation.

SF.50.21.B0002: Hypersonic Vehicle Technologies

Grandhi, Ramana - 937-656-6105


With the advent of new design methodologies, computing power, propulsion systems, materials, and manufacturing processes, there is now a tremendous potential to develop revolutionary hypersonic vehicles. Traditional design and analysis assumptions face challenges due to tight integration of the propulsion system into a hypersonic airframe, presence of high temperature loads in hypersonic flight, and nonlinear responses of aerodynamic and vibroacoustic loadings. Research efforts are required to yield higher fidelity designs to identify the vehicle configurations that will enable new tactical capabilities for the warfighter.
AFIT proposes advancing design methodologies by incorporating higher-fidelity physics-based models representing nonlinear behavior of the engineering disciplines in aerodynamics, propulsion, structures, materials, and heat transfer. The impact of this research will be in the creation and testing of hypersonic vehicle sub-systems and configurations for representing realistic operational conditions validated through physical testing.
U.S. Citizens only

SF.50.21.B0001: Multidisciplinary Optimization of Aerospace Vehicles

Grandhi, Ramana - 937-656-6105

Computational optimization methodologies are growing in popularity for use in aircraft design as cross-disciplinary physical interactions are better understood and system requirements continue to increase in complexity. Future air dominance demands increased capabilities in speed, range, survivability, mission versatility, and reliability. To satisfy these demands, one must achieve synergy between aircraft constituent sub-systems including, among others, propulsion, structures, flight controls, and materials. These methodologies offer the potential of increased system performance through enabling numerous design options to be explored systematically. While optimization methods have traditionally been predominant in the latter stages of a design process (preliminary and detailed), there is a growing interest and need for the utilization of higher-fidelity physics in the earlier stages of design (conceptual). Similarly topology optimization and 3D printing are having increased presence for realization of complex shapes and physical testing, respectively. However, simulation models based on these higher-fidelity physics tend to have higher computational cost in comparison to their lower-fidelity counterparts. Advances in machine learning, artificial intelligence and quantum computing have to be incorporated in reducing the higher computational cost of physics-based models. In addition, designers need to account for uncertainties present in modeling, mission operation, experiments, acquisition, etc. for producing resilient systems. U.S. Citizens only

SF.50.20.B0005: Impact of Flow Field of High-Speed Air Vehicles on Wave Propagation and Sensing

Dexter, Michael - 937-656-6196

High-speed vehicles create special types of flow structures and trails in the atmosphere. It is of interest to understand the interaction of different types of sensor waves with such flow fields. The scope of this problem is quite large because atmosphere constitutes an enormous spatial domain with widely varying conditions. In particular, the phenomenology will rapidly vary with altitude, and the vehicle velocity can vary over a wide range. This will create vastly varying flow field structures, sometimes partially ionized and sometimes neutral. Even in the case of neutral flow, there can be different conditions because the flow constituents can be in a different states of excitation. Furthermore, the different types of sensor waves, and the wide range of frequencies that can be involved adds additional complexity and scope. There is thus a variety of physical processes that one will encounter. The study will involve theoretical models, numerical simulations, and experimental data collection using scale models. We are interested in fundamental research in all aspects of wave propagation and scattering associated with this problem.

SF.50.20.B0004: Development of Advanced Radiation Detection Algorithms using Machine Learning and Artificial Intelligence

Holland, Darren - 937-656-5952

The primary objective of this project is to enable enhanced detection of radiation and radiation sources using advanced algorithms, artificial intelligence, and machine learning to improve national security and fundamental nuclear measurements. Our research group is actively pursuing improvements to imaging algorithms, directional detection, spectral deconvolution and neutron spectroscopy. Interested SFFP applicants would work with AFIT faculty and students to advance one or more of these areas during the fellowship period. Some details of each area follow.
Inference of pre-detonation properties of a nuclear explosive can be achieved by analysis of the post-detonation radioactive debris (i.e. fallout) using in-field gamma spectroscopic measurements. The composition of fallout from a nuclear explosion depends significantly on the pre-detonation composition of the nuclear explosive’s fissile material (e.g. uranium-235 vs. plutonium-239), the spectrum of neutrons that burned the fissile material (e.g. a fission spectrum vs. a fusion spectrum), weapon design and composition, environmental composition near the detonation size, and factors affecting fractionation such as weather and height of burst. As a result, the gamma-ray spectra associated with fallout are complex and significant convolution can occur for key isotopes. This area of research will pursue advanced deconvolution approaches using a sample that was analyzed both with “gold standard” radiochemistry and gamma spectroscopy.
An integrated gamma-ray and neutron imaging system using the rotating scatter mask concept has been developed at AFIT. The ultimate goal of this system is to enable medium to high fidelity neutron and gamma-ray imaging and spectroscopy in a single, portable system. Current research is developing new and improved machine learning based imaging and post-processing algorithms to improve radiological search performance of the system for discrete volumetric sources. This area of research could be expanded to mapping continuous or semi-continuous radiation fields generated from fallout from a radiation dispersal device, nuclear weapon, or nuclear accident such as Fukushima.
For many defense applications, neutron spectroscopy is difficult to impossible to do the stochastic response of neutron energy deposition in most radiation detectors. However, neutron spectroscopy can be used to reduce false negatives for point-of-entry screening, improve source identification for counter-proliferation operations, enable positive identification for counter-force applications, and ensure confidence in treaty verification applications. In all of these cases, neutron spectroscopy is often achieved with neutron spectrum unfolding using organic recoil scintillators or activation foils. However, most of these algorithms fail to account for systematic uncertainties, require an a priori guess at the observed spectrum, or are very limited in spectral resolution that can be achieved. This area of research would seek to address some of these limitations for foil activation unfolding analyses.

SF.50.20.B0003: Isotopic Determination via Spectroscopy of Laser Produced Plasmas

Patniak, Anil - 937-656-5729

This research program seeks to investigate techniques to improve the determine isotopic content of samples via spectroscopic analysis of Laser Produced Plasmas (LPPs) This work includes standard versions of Laser Induced Breakdown Spectroscopy (LIBS) to determine the abundance of solid state and gaseous samples of both stable and radioactive isotopes of interest in a range of environmental conditions. The work will then expand on those baseline capabilities using more advanced techniques for signal enhancement such as double-pulse laser-induced breakdown spectroscopy (DP-LIBS), LIBS coupled with Laser Induced Fluorescence (LIBS-LIF) or Laser Absorption Spectroscopy (LIBS-LAF). The research will also seek to find optimal post processing routines including chemometric analysis techniques, such as principal components regression, partial least squares, and artificial neural networks, to analyze spectra. Results will be compared to spectra simulated from first principles.

SF.50.20.B0002 : Autonomy and Navigation Technology

Taylor, Clark - 937-255-6221

The Autonomy and Navigation Technology (ANT) is an inter-department, multi-disciplinary research center supported by over 25 AFIT faculty and 15 full-time staff members. The ANT Center conducts research in three thrust areas: Autonomous and Cooperative Systems, GPS-denied Navigation, and GNSS Navigation & Navigation Warfare.
Autonomous and Cooperative Systems: We are working to increase autonomy for military systems, in particular small unmanned ground and aerial systems, using classical control methods, artificial intelligence, machine learning, and human-machine teaming. We are interested in wide variety of projects and topics in this area, including algorithm development (e.g. path planning or task assignment), test and evaluation and cooperative control and navigation.
GPS-denied Navigation: The ANT Center has investigated over 16 different areas for navigation without GPS, including common ones like vision, sound, lidar, radar, and magnetic anomaly fields and uncommon ones like odor and lightning. We are interested in work related to new methods of navigation as well as improvements to existing methods.
GNSS Navigation & Navigation Warfare: We research methods for obtaining trusted GNSS for military systems. These solutions exists in the intersection of signal authentication, increased signal availability, and signal integrity. We are interested in a wide range of research topics within these domains.

SF.50.20.B0001: Exploitation of Material Signatures of Nuclear Fuel Cycle Processes to Support Nonproliferation Efforts

Bickley, Abigail - 937-656-5704

Ultra-trace actinide bearing particulate is a by-product of many processes in the nuclear fuel cycle. The composition of these particles is expected to be associated with both geographical location and fuel cycle step. The identification of elemental patterns available as indicators of proliferant activity requires statistical analysis of the composition of large quantities of particulate, a process which is time and labor intensive for human analysts. This research project consists of two components: laboratory analysis of microscopic actinide bearing particulate and the development of computational tools to exploit the laboratory measurements using machine learning and artificial intelligence.
X-Ray Fluorescence spectroscopy (XRF) and scanning electron microscopy (SEM-EDS) can be used to determine the elemental composition of microscopic particulate. Raman spectroscopy can be used to determine chemical compounds present in the particles. These tools will be used to identify the composition of unknown samples of interest. A variety of sample sets are available for analysis ranging from laboratory grown samples to soil samples collected from accident and test sites. Using the data collected, a statistical analysis will be performed to establish linkages between material origin, formation conditions and chemical processes.
It is proposed that data mining techniques be used to identify ultra-trace elemental signatures that can be associated with geographical location and nuclear fuel cycle step. This project will be largely computational in nature and will require research and implementation of modern data fusion and high performance computing tools for identifying process signatures within the data.
US citizenship is required to participate in this research.

SF.50.19.B0007: Self-tuning, Robust Estimators with Accurate Uncertainty Outputs

Taylor, Clark - 937-255-6221

State estimation lies at the heart of most dynamic or reasoning systems. As these systems have become more complex, state estimators have advanced from the early Kalman filter implementations to estimators that handle larger and more complex state vectors (including mixed discrete and continuous states), accept significantly non-linear dynamic and measurement models, and handle more complicated probability distributions.
In addition to these improvements in estimators, we are particularly interested in new estimation algorithms and approaches that address the following problems:
1. Estimators that produce accurate uncertainty estimates. Much work on making better estimators has focused on computing the most accurate state estimate possible given the observed data. In many applications though, the uncertainty associated with that state estimate is of equal import to the state estimate itself. Therefore, we are interested in improvements to estimation algorithms that yield more accurate uncertainty estimates of the estimated state. One area of particular interest are computer vision algorithms where the inputs have large number of outliers. While algorithms have long been used to screen out these outliers, these algorithms’ effects on the uncertainty outputs of the estimators is not well understood or characterized.
2. Self-tuning estimators that can adjust to different uncertainties on the input signals. For example, a navigation estimator that takes in both GPS (global positioning system) and inertial measurement unit (IMU) inputs is typically tuned to the performance of a specific IMU. However, even when an IMU of the same series is put into the system, its measurements may have different uncertainty characteristics than the unit that the GPS/IMU estimator was tuned for. This can have significant effects on the performance and uncertainty estimates of the estimation system. Therefore, we are interested in estimation approaches that can self-tune to the uncertainty characteristics of the inputs, providing more accurate state and uncertainty estimates.
3. Robust estimators that can detect and appropriately handle conflicting or erroneous data. While uncertainty in the inputs is generally expected with estimators, minimum squared error estimators are significantly affected by outliers. Estimators that recognize erroneous inputs and discard them in principled ways are of significant interest.
The goal of this research is to enable theoretical or applied improvements to estimation systems, with an emphasis on one or more of the areas described above.

SF.50.19.B0004: Novel Orbits for Military Space Mission Design in a Multi-Body Environment

Little, Bryan - 937-656-6129

The faculty, graduate students, and staff of AFIT`s Center for Space Research and Assurance (CSRA) conduct research to deliver space capabilities needed by the Department of Defense and the Intelligence Community.
High-altitude parking orbits provide resiliency to the military space infrastructure by providing redundancy in key assets, allowing for rapid reconstitution of under performing satellites, both individual and formations. Novel orbits and their associated dynamics can be exploited to provide unique trajectories and designs unobservable in lower-order models. Mission design and CONOPS development in a multi-body dynamical environment may be essential to maintaining space superiority and responsiveness and securing the ultimate high ground.
A Summer Faculty Fellow will work with the CSRA to advance the state-of-the-art in the following research areas:
- Investigate high-altitude orbits as a means for formation reconstitution
- Construct open-loop optimal guidance policy in a dynamical environment using heuristic and pseudospectral methods
- Provide alternative tactics in orbital engagement scenarios requiring finite and/or impulsive maneuvers
- Investigate relative satellite motion in a multi-body dynamical environment
- Evaluate the merits of high-altitude orbits for the purpose of space surveillance

SF.50.19.B0003: Numerical Simulation of Nonlinear Waves

Akers, Benjamin - 937-255-6297

This research program develops new asymptotic and numerical methods for the study of nonlinear wave phenomena. Numerical methods which are both flexible to problem type and highly (or spectrally) accurate are of interest. Both dynamic and steady state problems are to be considered.
These numerical methods will be developed for general problems as well as tailored to application area. Two application areas of interest are interfacial fluid dynamics and high energy laser propagation. In the former, bifurcation and stability will be research themes. In the latter, there is opportunity to coordinate with ongoing modeling and experiment.

SF.50.18.B0002: Applications of Algebraic Number Theory to Combinatorial Designs

Bulutoglu, Dursun - 937-656-6265

Let X be an N by N matrix of +-1s, where N is not divisible by 4. Hadamard`s maximum determinant problem seeks to find X that maximizes the value of Det(X^TX). Such an X is called a D-optimal design. The current state of knowledge on D-optimal designs is tabulated on the webpage:
http://www.indiana.edu/~maxdet/.
Finding achievable upper bounds for Det(X^TX) is essential in finding D-optimal designs. The best known upper bounds for Det(X^TX) are the Barba bound for N=1 (mod 4), Ehlich/Wojtas bound for N=2 (mod 4), and Ehlich bound for N=3 (mod 4).
Det(X^TX) is the product of eigenvalues of X^TX, where each eigenvalue is an algebraic integer in R. The first part of this research will use the approach of Cheng [1978. Optimality of certain asymmetrical experimental designs. Ann. Stat. 6, 1239-1261] and the properties of algebraic integers to improve the aforementioned best known upper bounds for Det(X^TX).
Let l be an odd integer. Binary Legendre pairs of length l can be used can be used to construct a Hadamard matrix of size 2l+2 Fletcher, Gysin, and Seberry [2001. Application of the discrete Fourier transform to the search for generalized Legendre pairs and Hadamard matrices. Australas. J. Combin. 23, 75-86]. Let
z_1k=u_0+u_1w^k+u_2w^2k+...+u_(l-1)w^(l-1)k, and
z_2k=v_0+v_1w^k+v_2w^2k+...+v_(l-1)w^(l-1)k, where
w is a primitive l`th root of unity and {u_i} and {v_i} are binary sequences of length l. Then binary Legendre pairs exists if and only if the system
|z_1k|^2+|z_2k|^2=(l+1)/2 for k=1,2,...,l
has a solution. The number of solutions to this system of equations appears to grow exponentially Fletcher, Gysin, and Seberry [2001. Application of the discrete Fourier transform to the search for generalized Legendre pairs and Hadamard matrices. Australas. J. Combin. 23, 75-86]. Both z_1k and z_2k are sums of roots of unity in a cyclotomic field extension of Q. The second part of this research will focus on exploring ways of exploiting properties of sums of roots of unity to find binary Legendre pairs for l>=49. In particular, theory of cyclotomic field extensions may be useful in studying this problem.

SF.50.18.B0001: Networks / Security / Critical Infrastructure Protection / Applied Machine Learning / Remote Sensing

Hopkinson, Kenneth - 937-656-5533

Our lab conducts research efforts involving Network Optimization, Network Security, SCADA / Critical Infrastructure Protection, Cognitive Radios and Cognitive Radio Networks, Applied Machine Learning, and Remote Sensing. The main goal is to use situational awareness, acquired via distributed sensor information, to enhance operations and security. Our applied machine learning work looks to advance beyond existing algorithms via machine learning in cases where we have or can acquire enough data to train effectively.

SF.50.17.B0001: Environmental Sensing and Modeling of Methane Emissions via Unmanned Aerial Vehicles (UAVs)

Slagley, Jeremy - 937-255-8305

Methane emissions have a devastating effect on the atmosphere. Methane has been shown a far more effective greenhouse gas than carbon dioxide. Methane emissions sensing and modeling is very important to understand source apportionment and aid in developing policies and technologies to control emissions. There are two main methods in methane inventorying and apportionment: atmospheric sampling and modeling (top down), and source sampling and modeling (bottom up). The source sampling and modeling lends itself better to apportionment, but there are disparities in taking relatively few samples and relying on ranges of assumptions in the models. More sampling data from sources refine the models, but the spatial and temporal distribution of source emissions precludes exhaustive study. It is time-consuming and expensive to have environmental scientists in the field collecting data. Unmanned Aerial Vehicles (UAVs) offer an opportunity for source emissions sampling at remote sites which extends the capacity of the field environmental scientist. However, there are several research questions to resolve to enable using the sampling data in source emissions modeling. 1. Effects of “prop wash” on sampling measurements and techniques to employ UAVs to minimize adverse effects 2. Payload tradeoff to achieve sufficient measurement resolution/limit of detection 3. Georeferencing and flight velocity/sensor response time error resolution.

SF.50.16.B0003: Creep Deformation and Durability of Ultra High Temperature Ceramics in Extreme Environments

Ruggles-Wrenn, M.B. - (937)255-3636 x4641

The ultra-high temperature ceramics (UHTCs) are candidates for such aerospace applications as sharp leading edges and thermal protection systems for reusable atmospheric re-entry vehicles and hypersonic flight vehicles. Before UHTCs can be used in applications, their structural integrity and environmental durability must be assured. To provide that assurance, the mechanical behavior of UHTCs at relevant service temperatures and environments must be thoroughly understood and characterized. Recent research at AFIT developed, constructed and validated a unique facility for mechanical testing of UHTCs in air or argon at 1500-1700°C. We developed and validated a method to perform compression creep tests of small UHTC samples in air or argon at 1500-1700°C.
Ongoing research is focused on investigating mechanical behavior of the UHTCs at temperatures ranging from 1300 to 1700 °C in laboratory air or inert gas. We aim to provide fundamental analysis of high-temperature deformation of UHTCs and to identify the controlling creep mechanisms. It is envisioned that experimental results obtained in compression creep tests of UHTCs at 1300-1700°C in air and in argon will provide a basis for evaluating creep rates, creep activation energies, and identifying operating creep mechanisms. Emphasis is on assessing the interaction between oxidation and compression creep processes. Unknown deformation and failure mechanisms may be discovered. Results of this research will provide experimental foundation to extend the models for the oxidation of the UHTCs to include the effects of mechanical load on oxidation.
We are interested in experimental investigation as well as in modeling of the material response subjected to mechanical loading in extreme environmental conditions. Unique experimental facilities are available in the Mechanics of Advanced Aerospace Materials Laboratory (MAAML) at AFIT.

SF.50.16.B0002: Human Machine Systems

Miller, Michael - 937-656-4641

As humans, we use machines to augment our physical and cognitive capabilities. In general, my research focuses on augmenting human capabilities and safety through the improved design of supportive systems. The preponderance of this work focuses on augmenting human cognition by enhancing the design of the interface between functional software agents and the human operator. This can include the improved design of software or hardware interfaces, as well as the design of software agents, which enhance system performance. Ideally this research yields design tools which facilitate the development of computing systems which are capable of performing as collaborative partners with human operators to support robust and efficient system operation.

SF.50.14.B1125: Biological Process Research for Environmental Applications

Harper, Willie - 937-656-4545

My research explores biological processes that are important in a range of environmental applications, with a primary focus on water quality. Currently-sponsored projects are focused on the removal of organic chemicals, biosensing, and resource recovery. Research activity combines traditional research approaches, such as mathematical modeling and laboratory-scale experimentation, with the modern tools from chemistry and microbiology, and research based on this combination uncovers knowledge and provides exciting opportunities for interdisciplinary collaboration. Although individual projects might emphasize experimentation, modeling, or microbiological aspects, all research involves quantification, the key to making the research results relevant to engineers.

The objectives of our ongoing projects are: 1) to understand and predict the fate of chemical warfare agents and industrial chemicals in engineered water treatment systems, 2) investigate novel biosensors and hyperspectral imaging technology to detect hazardous substances, and 3) evaluate resource recovery paradigms using systems thinking.

SF.50.02.B7123: Fatigue and Fracture of Advanced Materials

Ruggles-Wrenn, Marina - (937)255-3636 X4641

Active research is in progress to characterize the deformation mechanisms, fracture and fatigue behavior for structural materials including conventional polymeric composites, high temperature composites, and hybrid materials. We are interested in the experimental investigation as well as in modeling of mechanical response of and damage mechanisms in materials under myriad of loading conditions, such as high cycle fatigue, low-cycle fatigue, fretting, foreign object damage, creep, thermo-mechanical fatigue, etc. Unique experimental facilities for testing are available. Research focuses on developing the scientific base and fundamental understanding.

SF.50.01.B7843: Radio Tomographic Imaging

Martin, Richard - 937-656-5545

Device free localization is the process of tracking users who are not emitting a radio signal. An emerging method of doing this is radio tomographic imaging (RTI). RTI involves setting up a dense network of radio sensors. When a user physically enters the network, it will obstruct a subset of the network links. By measuring the change in signal strength on all network links, it is possible to compute a 3D image indicating which voxels are obstructed. This can in turn be used for target tracking and identification. Of particular military interest is the fact that RTI can be used for imaging through walls and foliage; for example, work at AFIT has demonstrated imaging capabilities through foot-thick concrete walls.

Current RTI research at AFIT includes (i) improving the physical model relating the presence of a user to the change in radio signal strength, while accounting for multipath, (ii) improving the performance of the imaging algorithm, (iii) improving the system implementation by reducing computations or designing an application-specific communication protocol for the sensors, and (iv) developing target tracking and identification tools.

SF.50.01.B6134: Combustion Dynamics for Novel Combustor Systems

Polanka, Marc - 937 656-6140

As future requirements lead toward compact, efficient engine designs, conventional gas turbine component design methodology will become more integrated to provide higher performance systems. Several concepts are being
explored to obtain lighter weight, more efficient, lower fuel consumption combustors. One example of this integration of components is the Ultra Compact Combustor (UCC). In this configuration, fuel is deliberately added circumferentially above the vane geometry to accomplish combustion simultaneously while the flow is turned by the vane. Research areas have focused on the combustion mechanisms at high g-loading and radial migration of the hot combustion gases into the integrated vane along with investigations into Rayleigh losses associated with higher Mach number combustion. With optical diagnostics such as PIV, PLIF, and TLAS in place in the laboratory, the capability to completely understand these complex burning configurations exist. Future efforts will continue to understand the integration issues with the compressor and turbine. New efforts specifically geared at understanding how to cool the turbine appropriately in this high equivalence ratio environment will also be developed.
Another research area focused on the high temperature effects of film cooling of turbine vanes. These investigations have focused on attempting to understand the impact of temperature on the properties of the coolant and how cooling effectiveness results scale from low temperature investigations to high temperatures. A single vane facility exists that can change the freestream temperature from ambient to 1600K. Investigations into both internal and external cooling configurations are possible over a range of Reynolds numbers and blowing ratios.
Keywords: Combustion, Diagnostics, Novel Combustors, Film Cooling, Turbines

SF.50.01.B5171: Mission Assurance: Impact Assessment and Situational Awareness

Grimaila, Michael - 937-255-1074

Virtually all modern organizations have embedded information systems and networking technologies into their core processes as a means to increase operational efficiency, improve decision making quality, reduce delays, and/or maximize profit. Unfortunately, this dependence can place the organization's mission at risk when an information incident (e.g., the loss or degradation of the confidentiality, integrity, availability, non-repudiation, or authenticity of a critical information resource or flow) occurs. This research focuses on developing solutions to provide decision makers with timely notification and relevant impact assessment, in terms of mission objectives, following an information incident.

SF.50.01.B4576: Data Analytics for Additive Manufacturing

Li, Kunpeng - (937) 255-3636

Successful development and deployment of additive manufacturing (AM) products require efficient and effective data analytics to transfer product characteristics to manufacturing software to drive the operations of a 3D Printer. This research involves data modeling of component shapes, characteristics, and/or profiles. The end goal of this research is to use systems engineering models and techniques to design additive manufacturing products that can meet rigorous assessment metrics for intricate product design, evaluation, justification, and integration. This research will support the activities of the Additive Manufacturing Laboratory at AFIT. The transition from traditional manufacturing to additive manufacturing calls for innovative data analytics techniques. Researchers participating in this topic must have strong analytical skills, math modeling background, software capabilities, and an interest in systems simulation. It is expected that the outputs from the research will contribute to the AFIT goal of driving and advancing innovation in new product development for defense applications.

SF.50.00.B5167: Molecular Reaction Dynamics

Weeks, D.E - 937-656-5731

The detailed analysis of a wide variety of chemical reactions plays a central role in a number of Air Force and DOD applications ranging from the chemical oxygen iodine laser, to upper atmospheric chemistry, to the development of new high energy density materials. To support these efforts, we are developing new computational methods to characterize chemical reactions. Our approach employs time dependent wave package dynamics to calculate scattering matrix elements and associated reaction rates and cross sections. Initial efforts have focused on developing this new time dependent technique through the analysis of inelastic collinear reactions of type A + BC -> C, incorporating the translational and vibrational degrees of freedom. More recent efforts have successfully incorporated the rotational degree of freedom and we are currently focusing on the non-adiabatic reaction B + H2. For these calculations, we are including the rotational and vibrational degrees of freedom of the hydrogen molecule together with the electronic degrees of freedom of the Boron atom. Future efforts include the extension of the technique to four atom reactions, and the continued refinement of time dependent techniques for computing scattering matrix elements. Researchers with experience in computational physics, molecular dynamics, wave packet propagation, or related areas are encouraged to apply.

SF.50.00.B5157: Chemical, Nuclear, and Biochemical Measurements and Computations Applied to CBRN Objectives

Burggraf, L.W - (937) 656-5863

Experimental and theoretical methods of chemical physics are applied to CBRN proliferation problems. Three projects illustrate the wide range of research interests: (1) characterization and inactivation of Ba and Bt bacterial spores, (2) surface chemistry of uranium oxides and contaminant metals (3) gamma imaging using Compton backscatter and gamma absorption.
We have demonstrated that topological images and phase images and chemical force measurements using atomic force microscopy (AFM) can distinguish surface properties of living and inactivated spores of bacillus anthracis from closely related bacterial spores. We are developing dynamic models of these nano-mechanical AFM measurements. We apply AFM techniques and other techniques to compare differences in properties of viable and inactivated bacillus spores. Inactivation of spores by ionizing radiation, UV radiation and thermal treatments are compared.
Uranium dioxide from nuclear fuel processes or depleted uranium munitions may be dispersed into environments. Particles of uranium dioxide react further in the atmosphere by oxidation and formation of complexes (hydrates, hydroxides, and carbonates), increasing the mobility and bioavailability of uranium, contaminant metals and fission isotopes. Spectroscopy and kinetics surface species on UO2 single crystals are measured, using spectroscopy tools including: positron spectrometry, photoluminescence (LIBS), Raman spectroscopy, Fourier transform infrared (FTIR), secondary ion mass spectrometry (SIMS), x-ray photoelectron spectroscopy (XPS) and x-ray diffraction (XRD). Spectroscopy signatures of various oxidation states and crystalline forms of uranium oxides, hydroxides, and carbonates are measured using these spectroscopy tools. Quantum methods are being developed to model spectroscopy of UxOy ions and point defects in solid state systems.
We are developing methods to employ planar high-purity germanium (HPGe) strip detectors to imaging applications including positron annihilation measurements for ACAR/DBAR and Compton/absorption gamma imaging. We are constructing a gamma spectrometer to simultaneously measure DBAR (Doppler broadened annihilation radiation) and ACAR (angular correlation annihilation radiation) spectra. We are developing a low-cost, low-bandwidth gamma imaging technique for mobile platforms using rotating scatter mask techniques. This approach is of interest for nuclear weapons inspection and field-detection of special nuclear materials using portable detectors.

SF.50.00.B0814: Optimal Packings of Subspaces

Fickus, Matthew - 937-255-6277

In various applications including coding theory, quantum information theory, and compressed sensing, the following problem arises: how should we arrange a given number of subspaces (of a given dimension) of a Hilbert space (of some other given dimension) so that the minimum distance between any two of these subspaces is as large as possible? That is, what are the optimal packings in the corresponding Grassmannian manifold? In the special case where the subspaces are lines (i.e., are one-dimensional) it suffices for them to form an equiangular tight frame (ETF). More generally, when the dimension of the subspaces is greater than one, equi-chordal tight fusion frames (ECTFFs) give packings that are optimal with respect to chordal distance, while equi-isoclinic tight fusion frames (EITFFs) are optimal with respect to spectral distance.
This project focuses on these ideas and other closely related concepts. Research priorities include: (1) explicit constructions of new ETFs, ECTFFs and EITFFs; (2) the discovery of new necessary conditions on the existence of such objects; (3) explicit constructions of collections of subspaces that are optimal packings in situations when no ETFs/ECTFFs/EITFFs exists (e.g., subspaces that meet the orthoplex bound); (4) applications of these ideas to other mathematical fields (e.g., combinatorial design, where certain types of ETFs are closely related to strongly regular graphs, difference sets, balanced incomplete block designs, generalized quadranges, distance regular covers of complete graphs, and association schemes); (5) design of ETFs/ECTFFs/EITFFs that meet other, real-world-application motivated constraints.

AFIT

Dr. Graham, Scott
Dean for Research
AFIT/CZ
Wright Patterson Air Force Base, Ohio 45433-7765
Telephone: 937-255-3633
Email: Scott.Graham@afit.edu