Project Description:

This project aims to develop novel integration and hybridization micro-fabrication methods for large arrays of electro-optic devices based on different material systems and platforms, such as Si CMOS and III-V light emitters. This type of devices have been widely used as high definition UV/Visible/IR micro-displays and scene projectors. The development of advanced high-resolution Micro-LED scene projector requires a very-large-scale integration of the Si CMOS electronic driver array with the GaN-based LED emitter pixel array to generate radiometrically-correct images and videos. Traditional aligned hybridization techniques are challenge to achieve precise bonding of millions of sub-pixels. Advanced hybridization methods such as laser liftoff and transfer technology will be investigated. The students will be involved in bonding metal deposition, patterning and bonding, laser liftoff and transfer, GaN LED structure fabrication etc.

Citizenship Required:

US-Persons Only

In-Office Time Required: 100%

Competencies:

Electrical engineering, Mechanical Engineering, Materials Science and Engineering

Project Description:

Spacecraft systems and subsystems are generally vulnerable to the harsh conditions of space, including energetic particle radiation, atomic oxygen (AO) exposure, solar radiation, high vacuum, thermal cycling, and micrometeoroid impacts. Environmental energy deposition causes chemical changes in materials, resulting in alterations to various physical properties such as optical behavior, mechanical strength, electrical conductivity, and chemical reactivity. These changes degrade the functionality of both the materials and the spacecraft systems they form, ultimately shortening the spacecraft's operational lifespan. Therefore, understanding how material properties evolve over the course of a mission is crucial for the design of long-term space missions. This project aims to gain insights into the characteristics of novel materials intended for future USSF missions, evaluating their durability under electron irradiation of the simulated Geostationary Orbit (GEO) and cislunar environments, as well as AO exposure of low Earth orbit (LEO).

This project will be executed in close collaboration with the Kirtland Air Force Research Laboratory (AFRL), underscoring the significance of a cooperative approach in harnessing expertise and resources. The synergy between our research team and AFRL is crucial for ensuring the success and relevance of the study, fostering a dynamic exchange of knowledge, and facilitating the seamless integration of experimental findings into practical applications for future USSF missions.

During this project, GRIP students will engage in a comprehensive exploration of the morphological, chemical, and optical characteristics of various spacecraft materials exposed to simulated space environments, using a variety of experimental techniques optical microscopy accompanied by a laser-based elemental analyzer, bi-directional reflectance spectroscopy (BRDF) and Fourier Transform Infrared (FTIR) spectroscopy.

Learning objectives of the proposed project include: 1.) Training of interns on the usage of the respective equipment; 2.) Material characterization measurement of the respective samples; 3.) Compiling a comprehensive final report that delineates the experimental results and actively participating in effective communication with AFRL researchers. The findings of research performed by Summer interns are expected to be included in a follow-up conference or journal publication prepared by a GRIP mentor.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

Material science, physics, optics, chemistry

Project Description:

Simulation of imagery is critical to many scientific fields. For example, it is far easier to build a synthetic database to train machine learning models due to the ease of labeling ray-traced imagery. Unfortunately, simulators such as Unreal Engine utilize reflectance models that are not physically accurate but rather are meant to provide artistic representations. While these simulators can produce convincing imagery, they are therefore not well suited for scientific applications. As a further complication, databases of materials of interest to scientific communities are often sparse. It is rare that a material of interest has been optically characterized and even rarer that the data is readily available.

To address this gap, it is critical that Georgia Tech Research Institute (GTRI) build up an in-house robotic system to study the directional reflectance of materials in a rapid and automated manner. This will open up the potential to develop more accurate reflectance models for emerging materials across a wide array of scientific communities. For example, GTRI is working to understand how spacecraft reflectance signatures can be used to assess the health of spacecraft; such spacecraft materials have exotic reflectance properties that cannot be described by existing models in the computer rendering community.

In this internship, students will develop the initial prototype for GTRI's automated robotic spectral goniometer. This goniometer system will consist of a solar-simulator to illuminate samples, a spectroradiometer to measure multi-spectral reflectance, and rotation stages to orient samples along the full hemisphere of reflectance. Students will perform initial design of numerous hardware and software components, including: a system to house all components, a motor controller system to angularly rotate the sample relative to the sensor, automation software to coherently rotate all components during a measurement sequence, and an automated system to record reflectance measurements.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

Computer Engineering, Computer Science, Optics, Mechanical Engineering, Robotics

Project Description:

Photonic interconnects bypass the bandwidth bottlenecks of copper interconnects. In the world of chiplets and 2D+ integration, DoD and commercial applications need to leverage the large bandwidths of photonics to pass Tb/s data rates between chips and packages. However, photonic chiplets are still in infancy, and no high yield, manufacturable optical direct chip-chip coupling methods currently exist. Current methods require expensive and slow fiber-attach methods to interconnect chips. Small packages of heterogeneous or multi-chip photonic systems can bypass fiber altogether by designing chip-to-chip coupling structures on the die. This project will 1) survey literature on optical coupling methods, 2) design coupling structures and methods suitable for foundry fabrication, 3) optimize designs for efficiency and against chiplet placement tolerance, and 4) document the research, potentially for external publication.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Remote Option: No

Competencies:

Electrical Engineering, Physics, Optics, Electromagnetics

Project Description:

We aim to develop novel colloidal robots based on advanced energetic nanomaterials for precision medicine applications. These robots will serve as targeted therapy agents for cancer treatments such as cytolysis, photothermal therapy, and the removal of stones inside the body. Additionally, these advanced energetic materials can be utilized in various other fields, including:

Project Description:

Project Objective:
Overview: Modern free space optical communication (FSOC) systems are a critical technology for rapidly establishing field-ready and scalable communication links that transmit information at the speed of light.

Challenge: However, modern FSOC is fundamentally exploitable as it relies on established lines of sights between transmitters and receivers, operates at known spectral bands, unshielded light signals, and uses simple encoding schemes such as shift-keying to communicate information. This leads to discovery risk motivating the need for security measures. While encryption strategies like quantum keying exist, the current methods notably rely on costly, specialized and difficult to manufacture custom optics that inhibit perfect encryption, and experience degradation due to environmental factors such as turbulence and decorrelation of the optical field over long (km-scale) distances.

Innovation: To address these limitations, this GRIP proposes to innovate on recent efforts in optical computing to open a new design potential for designing low cost, low SWaP, and secured platforms at GTRI, with a focus on optical comms. Optical computing is a form of physics-informed optimization that relies designing chains of simple, low-cost and well-understood custom optical, such as diffractive optics, as learnable phase encoders as well as off-the-shelf solutions to create optical systems that offer unprecedented control over optical fields to achieve specific tasks. Already, these strategies have been shown to unlock completely new potential to use all-optical systems to perform high-level computations such as image encryption or statistical inference with low cost and significantly reduced size, weight and power requirements. In addition, these strategies can learn over representative physical simulators to discover novel systems that are natively robust to expected perturbations such as misalignment, optical turbulence and incoherence of light. For these reasons, optical computing is well-poised to revolutionize optical design and FSOC potential at GTRI. To setup GTRI as a leader in this intersection, this GRIP proposes using a team of three students to accomplish three tasks:

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

Electrical Engineering, Optical Engineering, Physics, Computer Science

Project Description:

This project is for prototyping or simulating an innovative active remote sensing system by integrating an event-based camera into a lidar (Light Detection and Ranging) platform. Depending on your team's skills and interests, we will focus on developing either a tracking lidar system capable of following moving objects in the sky or a mapping lidar system designed to create detailed three-dimensional representations of environments.

Recent GTRI / GRIP research has introduced alternative methods that eliminate the reliance on Time-of-Flight (ToF) measurements. The technology involves geometrically mapping range bins onto a sensor array. In this approach, the optical system is designed so that light from different distances (range bins) is focused onto specific areas of the sensor. This spatial encoding of depth information allows for instantaneous range measurements without the need for time-resolved detection.

Event-based cameras, also known as dynamic vision sensors or neuromorphic cameras, represent a significant shift from traditional imaging technologies. Unlike conventional cameras that capture full images at fixed frame rates, event-based cameras asynchronously record changes in brightness at individual pixels. Each pixel independently detects and reports an event whenever it senses a significant change in intensity, providing data with microsecond temporal resolution.

Intentionally, this project combines the work of two previous successful GRIPs that a) created a Python package for simulating and processing event data and b) validated the new approach to lidar that does not rely on timing laser pulses.

The project offers significant innovation potential by advancing computational imaging techniques and providing an alternative to conventional lidar technology. Students will develop skills in optical engineering, computational imaging, signal processing, software development, and hardware prototyping through hands-on experience. Expected deliverables include a demonstration of tracking or mapping capabilities, technical documentation, developed software algorithms, and a final poster presentation.

Support provided includes mentorship from the project advisor, access to laboratory facilities with optical and computational resources, technical literature and existing tools, and constructive feedback from GTRI experts throughout the project.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

optical engineering, electrical engineering, software development, signal processing, mechanical engineering

Project Description:

The planned human presence on the Moon in upcoming years presents risk factors exacerbated by austere cislunar conditions. In particular, early lunar missions will depend on Earth-Moon communications with components on Earth, in space, and on the lunar surface. GRIP students will have the opportunity to contribute to vulnerability analyses of a model Earth-Moon communication network. The students will have the choice to investigate the effects of interference or jamming on the ground segment or space segment into the lunar segment. GRIP students will model transmitters, receivers, and interference sources in a program such as MATLAB or written in Python. The students will create post-processing analysis scripts in MATLAB or Python. The students will also be exposed to industry standard analysis tools for possible future work. This project will emphasize satellite uplink survivability in an emergent civil application. Key concepts of the project include space electromagnetic spectrum operations, electronic warfare, and satellite network vulnerability.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

Mathematics, physics

Project Description:

The project will provide undergraduate students with practical experience in applying MORA's standardized data formats for real-time communications and pave the way for AI/ML advancements in RF systems.

Goal: By the end of the internship, students will deliver a fully functional MORA testbed with capabilities that demonstrate the benefits of standardized in-phase and quadrature (I/Q) data, a foundational requirement for interoperability and enabling AI/ML analytics in defense applications.

Project Scope Students will work within a 2-channel MORA-enabled test setup, focusing on:

Relevance to DoD Avionics and Ground RF Systems The DoD's push toward modular, open, and adaptable architectures for RF and avionics systems is driving the need for interoperability across diverse systems, from airborne platforms to ground stations. MORA's standardized data formats, particularly for I/Q data, enable seamless integration between these systems and allow AI/ML models to process signals in real-time. The project will highlight:

Citizenship Required:

US Citizens Only

In-Office Time Required: 100%

Competencies:

electrical engineering, digital signal processing, RF engineering

Project Description:

Robotic manipulators are flexible platforms ubiquitous in today's automation industry. However, today they are largely employed with fixed repeatable tasks. This underutilizes the flexibility of general robot systems which are capable of a broad range of behaviors. Our lab specializes in researching the deployment of these robots in semi-structured and unstructured environments, such as on factory floors and in agriculture fields. This project aims to create a high-fidelity digital twin of our robotic cell to accelerate development and testing of novel control architectures and learned AI behaviors. Students will model the geometric and physical properties of the system, evaluate baseline control algorithms, and create an AI-ready training platform to develop cutting edge robot control schemes. Time permitting, we will evaluate the sim to real gap on a range of tasks to evaluate the digital twin’s ability to mirror the real system.

Students will:

Students should have experience with programming in Python and/or C++, and preferably have exposure to ROS or ROS2.

Citizenship Required:

No Restriction

In-Office Time Required: 100%

Competencies:

Computer Science, Machine Learning, Robotics

Project Description:

For robot manipulation in non-fixtured environments, low-level control performance can only be as effective as the quality of the scene perception. In any complex domain, accurate and reliable perception and scene understanding is critical to the performance of any robot. This project will involve first benchmarking existing state-of-the-art object and feature localization methods and then developing novel approaches to create an accurate digital representation of the kinematic state of a robot's environment. Key objectives include:

The outcome of this project will provide insights into the efficacy of state-of-the-art models and will contribute towards the development of more precise robot behaviors. Innovation in this core capability will improve the use of robots in industry, defense, and service applications.

Students should have experience with Python and/or C++ programming, and preferably should have experience with OpenCV, PyTorch and/or the Robot Operating System (ROS).

Citizenship Required:

No Requirement

In-Office Time Required: 100%

Competencies:

Computer Science, Electrical Engineering

Project Description:

A number of the highest challenges facilitating Earth Independent Operations (EIO) for the space industry stem the recent prioritization of Beyond Earth Orbit (BEO) mission prioritization (cislunar, Mars, Lagrangian, etc). EIO is the emerging term for Human-On-the-Loop mission operations and captures an array of operational and technical challenges. Active EIO challenges range include an array of lunar operations support areas both in support of lunar orbit-based assets and facilitating robotic/manned lunar surface missions with a number of operational gaps revolving around communications, sensing, onboard (ie., satellite) processing, power management, orbital additive manufacturing, and logistical support.

This GRIP program will focus on an in-depth system level analysis to identify top mission/technical shortfalls and resultant proposed technical solutions. The researchers will begin with conducting a risk analysis and literature review to identify primary BEO operational gaps. The team will then conduct a corresponding system assessment of current Commercial-of-the-Shelf (COTS) and modified COTS technical solutions to a subset of the critical identified gaps. Lastly, for areas where there are currently technological gaps in currently available COTS technology, the students will be connected with corresponding Subject Matter Experts (SMEs) internally and externally to discuss where the state of the art Test Readiness Level (TRL) technology maturation stands. The SME TRL assessments will be combined with the previous results and briefed to GTRI civil space researchers and stakeholders for potential further discussion with NASA.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

Engineering (mechanical, aerospace, computer, electrical), physics

Project Description:

Generative AI is capable of creating brand new content. This includes text, images, videos, and other forms of data, often in response to specific prompts. This GRIP project will explore applying generative AI modeling to the domain of 3D data in the form of high-resolution point clouds of peach trees obtained both before and after dormant pruning. The purpose is to learn the underlying rules that are being applied by human experts performing this highly skilled manual task. Once quantified, these rules may be used to generate optimal pruning decisions for never-before-seen trees to guide future automation (there is currently a shortage in skilled pruners). The student interns for this project will start the summer by learning about and researching existing generative AI methods. The use of different 3D data representations, coupled with environmental metadata and constrained by limited training sets, will be explored. As will the translation of predicted pruning decisions to desired robot actions. The project will involve collaboration with horticulturalists at UGA Griffin and field work at a nearby research peach orchard.

Citizenship Required:

No Requirements

In-Office Time Required: 100%

Competencies:

CS, ME, ECE

Project Description:

Students will design and create technology to enhance trust in artificial intelligence (AI) on systems such as IoT, smart grids, or cyber logs, focusing on data complexities and structural changes as it passes through a given model. AI reliability is often a measure of the magnitude of input data perturbation a model can withstand before changing its output, or in terms of formalized verification specification. In practice, defining specifications is not always straight-forward, especially in the case of Large Language Models (LLMs) where output is often natural language without a well defined metric. However, as data moves through a model, it is represented by vectors (or points) in a high-dimensional space, making it accessible for analysis through geometric and topological methods. In the context of trustworthy AI, geometric and topological metrics allows for AI reliability to be measured without the need for predefined mission parameters on model output.

The project aims to tackle the challenges presented by AI models when erratic behavior occurs on prediction and classification tasks, developing advanced metrics and indicators to identify (and calibrate) trust in AI models designed for seamless integration of various elements of smart infrastructure. The application of geometric and topological analysis will enable a deeper understanding of the structural evolution of data inside the AI, leading to more efficient and robust systems, and allowing for AI models to be trained and deployed on the network.

This technology has the potential to significantly impact the current cyber environment through the development of more efficient, robust, and intelligent digital infrastructure. This can lead to significant improvements in areas like energy management, resource allocation, and system resilience against faults or attacks.

Citizenship Required:

No Requirements

In-Office Time Required: 0%

Competencies:

Applied Mathematics, Computer Engineering, Computer Science, Discrete Mathematics

Project Description:

Frequency-selective limiters (FSLs) are passive magnetic devices that automatically and selectively attenuate signals above a preset power threshold level, effectively performing as a self-tuned notch filter. FSLs were invented decades ago, and generally were large bulky units with substantial bias magnets. Typical applications were onboard naval vessels to prevent radar systems from overwhelming radio receivers. Recent breakthroughs in materials engineering and device physics have led to FSLs in flat surface-mount packages as small as 4 mm x 4 mm x 1 mm, enabling their use for interference suppression in communications, as well as in other commercial and government systems. While stand-alone FSLs perform best with signals that are well separated in frequency, in dense radio-frequency environments, they cause spectral distortions and harmonics, in effect, creating interference while they suppress interference. This project investigates digital signal processing techniques to mitigate the distortion generated by FSLs to expand their utility in spectrally crowded environments.

In a previous year, GRIP interns developed simulation models in MATLAB Simulink and ported their models into a field-programmable gate array (FPGA) to implement a working prototype of real-time distortion mitigation for a frequency-selective limiter. In 2025, GRIP interns will build on this work and

The end-of-term deliverable will be a bench-top demonstration of an FSL generating severe distortion while suppressing large interference signals, then showing that engaging our custom algorithm completely eliminates the distortion in real time. This demo would include scenarios where the FSL performs flawlessly on its own, as well as signal configurations where digital signal processing is required in order to achieve a spectrally clean (useful) result that enables high-quality communications in the presence of strong interference.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

Electrical engineering, digital signal processing, FPGA

Project Description:

Both military and civil sectors are confronted with an increasingly congested, complex, and contested radio frequency electromagnetic operating environment (EMOE), in which receivers are challenged to separate signals of interest from interfering signals in real time. The computational complexity of this problem scales with the number of distinct signals, requiring an increase in the number of independent observations and receivers. Unfortunately, current signal separation algorithms do not scale well (polynomial) with the number of receivers. This project will develop a versatile approach to generate new scalable algorithms to serve multiple functions and/or missions. Simulations and measurements will be conducted to generate and evaluate scaling algorithms and compare them with conventional algorithms requiring matrix inversion. This work will allow GTRI and our sponsors to process larger problems and to facilitate larger and more capable sensor arrays.

Citizenship Required:

No Requirements

In-Office Time Required: 100%

Competencies:

computer engineering, electrical engineering, digital signal processing, software, networking, FPGA

Project Description:

This project will improve the computation time associated with simulating radar returns from varying elevations and land types of a particular region on the Earth. A toolbox will also be designed to assist with target geolocation in areas of varying elevation and radar line of sight. Students will learn about radar physics and geolocation algorithms. Students must be proficient in Matlab or Python.

Citizenship Required:

US Citizens Only

In-Office Time Required: 0%

Remote Availability: Yes

Competencies:

computer science and engineering, electrical engineering, applied mathematics

Project Description:

This project will translate the outputs of internal GTRI radar simulations into processed products in a graphical form. One example is to create a toolkit for overlaying synthetic aperture radar imagery and performance metrics onto Google Earth. Students will learn radar physics and processing as well as utilizing various coordinate systems and transforms. Students must be proficient in Matlab or Python.

Citizenship Required:

US Citizens Only

In-Office Time Required: 100%

Competencies:

computer science and engineering, electrical engineering, applied mathematics

Project Description:

The purpose of this project is to convert the output data from multiple internal, GTRI modeling and simulation suites into Government referenced data formats. Format validation and comparison’s will be made by curating a commercial dataset during the project. The modeled radar data will be converted into sensor independent products and ingested into community tools for signal and image processing. Students will learn about radar physics, data formats, synthetic aperture radar processing, and industry standards. Students must be proficient in Matlab or Python.

Citizenship Required:

US Citizens Only

In-Office Time Required: 100%

Competencies:

computer science and engineering, electrical engineering, applied mathematics

Project Description:

This project will use Umbra's open data program and other commercial radar data sources for the purpose of algorithm development. Students will learn about radar automatic target recognition (ATR), synthetic aperture radar (SAR), and ground moving target indication (GMTI). When exact datasets of interest are unavailable, students will be introduced to data emulation software to modify antenna patterns, waveforms, and other features of collected radar data according to physical principles. Students must be proficient in Matlab or Python. [Graphic license: https://creativecommons.org/licenses/by-sa/4.0/]

Citizenship Required:

US Citizenship Only

In-Office Time Required: 100%

Competencies:

electrical engineering, applied mathematics

Project Description:

Using an open standard, integrate a radar capability into a surrogate radar system. This system can be one of several different form factors - desktop, rack mount, or small-form-factor UAV. After integration, perform the function of the radar capability to the extent capable in the host system to demonste radar functionality.

Citizenship Required:

US Citizens Only

In-Office Time Required: 100%

Competencies:

Aerospace Engineering, Electrical Engineering, Computer Engineering

Project Description:

The emergence of autonomous targeting and strike execution in the war in Ukraine provides a glimpse into the future of warfare where human control is increasingly removed from the launch, targeting, and strike determination of weaponized drones. Our research seeks to explore active defensive techniques to disrupt and degrade the performance of autonomous drone systems and establish state-of-the-art techniques for rapidly adapting to changing drone targeting algorithms and techniques. We propose the use of low-cost LED screens to carry out physical adversarial attacks on drone-based computer vision systems, such as YOLOv8, Detectron2, and EfficientDet. By displaying adversarial patterns on the screens, we will explore how these disruptions can prevent the detection or misclassify vehicles in aerial images. Using the DOTA dataset for object detection in aerial views, the models will be tested in simulated real-time environments where drones send live video feeds to the object detection system. We will manipulate variables like the position, angle, and distance of the LED screens to assess the vulnerability of these models under different real-world conditions. The outcomes of this study could have direct implications for military tactics and provide a strategic advantage in the next generation of increasingly autonomous warfare. This research builds on previous studies of adversarial attacks in the physical world, and aims to identify potential vulnerabilities in modern object detection models.

Citizenship Required:

US Citizens Only

In-Office Time Required: 100%

Competencies:

Computer Vision, Computer Science, Software Engineering, Electrical Engineering, Data Science, Artificial Intelligence

Project Description:

The capacity of artificial intelligence (AI) to boost human creativity through collaborative or facilitative interaction is as of yet, not fully understood, explored, or realized. Though the pursuit of AI technologies should be engaged with caution and ethical forethought [1], as AI technology continues to evolve, promising possibilities exist for employing productive, creative human-AI teams (HAT) in pursuits that benefit the greater societal good [2]. Previous research supported by GRIP 2024 [3] designed a conceptual human digital twin (HDT) model to inform human-centered AI research (HAI) [4] for the command, control, and communications (C3) operational context, to explore ethical considerations, optimize the human decision process, enhance crew performance, and lower operational risk. In this project, we will adapt these HDT concepts and inform our ongoing HAI research [5], by exploring different methods of training an HDT model to team with human experts to collaboratively complete specific tasks that emulate emergency risk management and response planning, including the collaborative envisioning of hypothetical events from malign actors [6] (i.e., alternative futures). During the course of this project, we will leverage methods and findings of previous empirical research exploring ways of training AI/ML-based HDT models such as through reinforcement learning (RL), within a custom Minecraft sandbox environment for applications such as emergency management training, search and rescue (SaR), and others [e.g., 7,8,9,10,11,12,13,14,15]. Minecraft (https://www.minecraft.net/en-us) is a public-facing gaming platform that is widely known and relatively easy to learn through publicly available training courses. Customized Minecraft games can be created and then provided in an open forum to serve as a viable avenue of collecting crowdsourced data, data which can then train an HDT model using methods that require large amounts of example data that otherwise might be difficult to collect. In this project, we will create our own customized Minecraft sandbox and tasks designed to emulate the tasks and cognitive demands inherent to emergency risk management and response planning, using our previous research to inform the cognitive processes built into the design. We will conduct, 1) training of an HDT model using competing expert-in-the-loop techniques supported in the scholarly literature to assist a human expert in risk management and response planning, and 2) experiments to assess performance differences in the model variants and HAT concepts. In addition to providing experimental findings to inform future research endeavors, this project will produce a valuable Minecraft experimentation sandbox to use as an alternative means of developing and evaluating HDT models and HAT concepts in research for C3, intelligence analysis (IA), emergency response, search and rescue (SAR), and similar applications.

Citizenship Required:

US Citizens Only

In-Office Time Required: 100%

Competencies:

artificial intelligence, machine learning, data science, computer science, cognitive science, psychology, human-computer interaction

Project Description:

One of the primary research focuses of the Sensor Systems Engineering Division at Cobb County Research Facility is on advanced RF and Electronic Warfare (EW) system design and development. In this project, students will assist in developing logic modules that process digitized RF data within a selected FPGA device, or transfer data between various platforms through standardized board-to-board or backplane links. Students will design, implement, and deploy these logic modules in an AMD Gen-3 RFSoC evaluation platform or an AMD Versal ACAP evaluation platform. Students will learn and perform a portion or even the entire development cycle from VHDL/Verilog design entry to functional verification, integration, synthesis, place-and-route, and timing analysis. Students will also assist in system debugging and testing in a lab environment.

The outcome of this project will be a complete understanding of the logic design and development cycle, along with associated software applications for development in FPGA-based evaluation platforms.

Citizenship Required:

US Persons Only

In-Office Time Required: 100%

Competencies:

computer engineering, computer science, electrical engineering

Project Description:

This project will validate a GTRI modeling and simulation tool for synthetic aperture radar against commercial data by selecting data geometries and targets in real data to simulate in the tool. A test matrix will be designed to exercise all portions of the simulation tool and quantitative comparison metrics will be established. Students will learn radar physics and synthetic aperture radar processing. Students must be proficient in Matlab or Python.

Citizenship Required:

US Citizens Only

In-Office Time Required: 100%

Competencies:

computer science and engineering, electrical engineering, applied mathematics