Computer Engineering Commons^™

Griddle: A Novel Hardware Based Matrix Multiplier Architecture, Seth Kiefer

Master's Theses

Matrix multiplication is a computational cornerstone in modern artificial intelligence and scientific computing, yet general-purpose processors struggle to perform these operations efficiently at scale. This thesis presents Griddle, a novel hardware architecture for matrix multiplication implemented on a Xilinx Artix-7 FPGA. Griddle focuses on flexibility and scalability by adopting a purely iterative approach that supports arbitrarily shaped input matrices without requiring padding or strict dimensional constraints. The architecture uses computational pipelines to execute a multiplication operation. Each pipe consists of a multiplication core and accumulation buffer that compute matrix products in parallel. The multiplication core contains a set of multiplier …

Simulated Live Studio Audience, Theodore David Shellenberger

Computer Engineering

The Simulated Live Studio Audience is a Python based application that utilizes Vosk, Roboflow, and Llama 3.2 to provide a user with auditory feedback based upon both visual and audible input from their device's microphone and camera. This system functions with a custom trained computer vision model to detect a specified object and when individuals walk in and out of the camera frame, outputting sitcom style simulated crowd reaction sounds accordingly. The simulated studio audience program also takes in vocal input from users, converts it to text, and, using a large language model, analyzes it for content that can be …

Framework For Multi-Agent Coordination And Distributed Localization In Micro-Uavs, Minwoo Park, Nikolas Tambornini

Electrical Engineering

Micro-UAVs (unmanned aerial vehicles) due to their inexpensive nature and compact form factor have shown an increase in prevalence throughout a multitude of applications including, but not limited to: search and rescue, military reconnaissance, and agriculture monitoring. However, for a majority of these high impact applications, a swarm of micro-UAVs are required and furthermore mandate that they are able to cooperatively and autonomously coordinate with each other. For long, controlling and communicating between a user and a singular micro-UAV has been a well known and solved problem, however the same can't be said for swarms of micro-UAVs. This project seeks …

Autonomous Mapping Rover, Garrett Jones, Timothy Kyle Chu, Eugenio Caruso Pasos

Electrical Engineering

This report details the design, implementation, and evaluation of "Rovero," an autonomous mapping rover developed as a senior project. Rovero integrates state-of-the-art technologies, including ROS2 for communication, SLAM algorithms for mapping, and sensor fusion for accurate navigation. The project aimed to achieve robust autonomous operation in indoor environments, leveraging a combination of LiDAR, IMU, and encoder data for real-time decision-making. Key challenges addressed include path planning, obstacle avoidance, and integration of multiple sensor inputs. Results demonstrate successful mapping capabilities and efficient navigation performance.

Design Principles For Robotic-Controlled 3d Printing Of Soft Tissue-Mimicking Materials, Karina Mealey

Master's Theses

Geometrically and physically accurate 3D models are emerging as essential tools for surgeons in pre-operative planning and training, but current methods of producing these models are not sufficient. Existing solutions have downsides such as high costs, long and tedious production times, high complexity resulting in a lack of scalability, or an inability to meet material requirements for accurately simulating human tissue.

The objective of this work is to develop principles for a low-cost, easy-to-use 3D printing system capable of prototyping with soft tissue-mimicking materials. To pursue this goal, a MyCobot280 robot arm was used as the mechanism to explore the …

Farmnav-Uav: A Field-Ready Autonomous Uav For Tracking Static And Dynamic Trajectories In Agricultural Applications, Veera Venkata Ram Murali Krishna Rao Muvva

Department of Agricultural and Biological Systems Engineering: Dissertations, Theses, and Student Research

This study presents the design and development of a custom-built uncrewed aerial vehicle (UAV) tailored for precision agriculture. Unlike commercial UAVs, which are often constrained by proprietary systems and limited hardware customization, the proposed platform emphasizes modularity, upgradeability, and cost-effectiveness. The UAV is equipped with a Cube Blue flight controller for reliable low-level actuation and a Raspberry Pi 4 companion computer that executes a Model Predictive Control (MPC) algorithm for high-level trajectory optimization and stability enhancement.

Most conventional UAV autopilot systems rely on non-optimal control strategies, such as proportional–integral–derivative (PID) controllers, which can be inadequate for dynamic or resource-constrained environments. …

Sla Evaluation And Composition In Reconfigurable Cloud-Based Services, Michael Iannelli

Dissertations, Theses, and Capstone Projects

Given the business model of offering data and computing services in a cloud setting, a major question arises: How do the services of one cloud provider compare to those of others? With the ubiquitous use of smartphones and tablets, the ability of a cloud provider to support QoS and client mobility becomes paramount. This research proposes a methodology for evaluating service-level agreements (SLAs) between cloud providers and their consumers, with a particular focus on dynamic SLA composition to adapt to changes in the application requirements and the external environment—such as traffic surges, security threats, or evolving business models.

In one …

A Turtlebot3 Hardware Testbed For Distributed Kalman Filter Localization, Dmitri Dobrynin, Indigo T. Garcia

Electrical Engineering

This report presents the preliminary design for a distributed localization framework for a multi-robot system. Many robotics research papers provide simulations of proposed algorithms in regards to formation control and task allocation. However, it is often that these proposals are without hardware experiments, being limited only to simulation. The objective of this framework is to provide a hardware implementation of a distributed Kalman filtering algorithm for multi-agent localization, as well as provide grounds for future multi-agent experiments. The framework is implemented on a swarm of three Turtlebot3 mobile robots. The robots can accurately localize themselves with respect to other agents …

Conversational Social Robot, Julianna M. Christopoulos, Mackenzie Goldman, Cece E. Hujanen, Jared Hunter

Mechanical Engineering

Background: Social robots are used in various settings to reduce burden on human workers and expand opportunities for people in need of assistance.

Challenge: Design a humanoid robot head and torso capable of holding conversations and interacting with a user using a LLM and motion. Conversations are limited to discussing Cal Poly resources and opportunities with visitors to the Bently Research Center on campus.

Implementation Of Area-Based Aggregation On Multivariate Continuous Uncertain Data, Rahul Nair

Master's Theses

Uncertain data is incredibly widespread - from sensor data to AI-based learned information, there exists a need to associate information with a certain probability of its veracity. Traditional relational databases lack a built-in functionality to support uncertain data, instead assigning them boilerplate values. Probabilistic databases tackle this problem by assigning non-deterministic data with an associated, often discrete, probability. Variants of probabilistic databases, namely continuous uncertain databases, are used to better model data represented through ranges and distributions. This is especially applicable with sensor-based geographic data as most commercial equipment contains some inherent margin of error.

While uncertain and probabilistic databases …

Fault Tolerant Dynamic Task Allocation For Heterogeneous Multi-Robot Systems, Jack R. Cline

Master's Theses

This research presents a novel approach to dynamic task allocation in heterogeneous multi-robot systems with integrated fault detection capabilities. As multiple industries are becoming more reliant on multi-robot systems for tasks, maintaining operational efficiency despite robot failures becomes critical. We propose a framework that combines optimization-based task allocation with a Kalman filter that estimates task progress for anomaly detection to identify unreliable agents and dynamically redistribute tasks. Observing values such as the normalized innovation squared (NIS), covariance, and progress rate, the algorithm can designate a robot as faulty. Embedding information about which robots are faulty in the task algorithm allows …

Ecen191/Robo150 Robotics Tools: Evaluating The Impact Of Project-Based Learning On Teaching Effectiveness, Nikhil Satyala

UNL Faculty Course Portfolios

The course selected for this study aims to introduce the skills and knowledge needed to utilize software and hardware tools used to design and operate robots. In this introductory course, undergraduate students in the Robotics Engineering major are provided with the opportunity to broaden their skill set with focus on specializing in one of three engineering disciplines (Software, Electrical and Mechanical). The primary goal of this course is to teach how develop the proficiency in basic concepts in robotics and enable the students to build and program a small-scale fully functional robotic arm. In this portfolio study, I investigated the …

Analyzing Player Difficulty Perception In Platformers Through Procedural Level Generation, Sasank Madineni

Master's Theses

Games utilizing Procedural Level Generation (PLG), such as Roguelikes, are becoming increasingly popular in today's gaming sphere. In games employing PLG, levels are generated randomly or pseudo-randomly, and aim to retain player attention through variance in levels between playthroughs. However, when generating levels with variance in structure and design, player enjoyment is often a mixed bag. With low enjoyment, player retention for these games can dwindle. This study explores the efficacy of real-time difficulty adjustment in procedurally generated platformers, as a method for maintaining stable player enjoyment without causing frustration. This thesis focuses on creating a short user experience, MIMEVA, …

Deeppanorf: Deep Prior For Neural 3d Reconstruction From Sparse Panoramas, Edward Du

Master's Theses

Advances in neural field representations have led to a significant improvement in view synthesis quality. However, many current novel view synthesis methods rely on a dense set of input views, which can be impractical and inefficient in real-world applications. We propose DeepPanoRF, a novel method for 360◦ scene reconstruction from a sparse set of input equirectangular panoramas. Built upon K-Planes, a radiance field representation that encodes explicit features on orthogonal feature planes, our method does not directly learn feature grids. Instead, we parameterize the feature grids to enable sparse view reconstruction without pretraining or additional regularization. We implement a custom …

Adversarial Deep Reinforcement Learning For Tank Duel Simulation Using Lidar-Based Observations, Braedan Kennedy

Master's Theses

Previous research has demonstrated that reinforcement learning agents can learn to steer differential-drive robots around obstacles using 2D lidar scans as observations. However, these studies typically treat all range returns as undifferentiated obstacles—objects to avoid—without distinguishing between different object types. This thesis builds upon previous research by introducing an adversarial task in which an agent must interpret raw range readings to both avoid static obstacles and identify, pursue, and engage a hostile target.

To investigate this problem, this thesis introduces TankGame, a novel, lightweight 2D tank duel simulator. Each agent receives a 360° lidar scan, controls its motion via tread …

Distributed Formation Control Of Nonholonomic Mobile Robots: Safety-Critical Leader-Follower Approach With Obstacle Avoidance And Dynamic Reconfiguration, Kelvin C. Villago

Master's Theses

Networked control systems for multi-agent robotics have emerged as a critical paradigm for executing complex coordinated tasks in diverse environments. While formation control serves as the backbone of such systems, real-world deployment introduces significant challenges including communication constraints, environmental obstacles, and the need for adaptive reconfiguration. This research addresses these challenges by developing a novel unified framework that seamlessly integrates obstacle avoidance algorithms with dynamic formation reconfiguration capabilities, specifically designed for communication-limited networked control architectures. The proposed framework represents a significant advancement over existing approaches by simultaneously handling both static and dynamic obstacles while maintaining system cohesion under communication constraints. …

Sentry V3: Extending Context Switches On A Trusted Secure Coprocessor, Mark Kong

Master's Theses

Software correctness and integrity is only ensured through trust in the un- derlying hardware. However, modern computer systems are complex to design and secure. Thus, given the choice between performance and security, companies will often prioritize performance, resulting in vulnerable systems. This creates exploitable systems that must be patched retroactively because business value performance over security. One approach to this issue is to separate the root of security from the rest of the system to create a minimal trusted computing base. Trustguard is one instance of this. Trustguard implements a Containment Architecture with Ver- ified Output (CAVO) model which shows …

Optimizing Web Servers With Io_Uring, Mihika Nigam

Master's Theses

Modern web servers face unprecedented demands for high throughput and low latency [10] [11]. Yet, even the state-of-the-art optimizations often fail under heavy workloads on existing infrastructure. Despite advancements in hardware, communication between applications and the kernel remains a critical bottleneck.

Industry surveys reveal that over 50% of production servers still rely on traditional epoll-based architectures [5]. This research aims to investigate and characterize Linux’s new io_uring subsystem, which has the potential to overcome these challenges. Through controlled load testing of various existing architectures like event-driven, multi-process, and multi-threaded architectures (including other commercial servers), we demonstrate how io_uring can help …

Assessment Of The Imx7ulp Heterogeneous Soc For Use In A Next Generation Student Cubesat Obc, Lorenzo A. Pedroza

Master's Theses

CubeSats represent a rapidly evolving platform for space research, industry, and education, demanding increasingly sophisticated onboard computing solutions that balance performance, fault tolerance, size, weight, and power constraints. Although the design of the Cal Poly's CubeSat Lab's (PolySat) current On-Board Computer (OBC) keeps up with many of these factors, the demands of modern workloads like fine attitude determination will start to outpace available compute. While the selection of a more capable processor to address this would have traditionally resulted in increased energy usage, modern system-on-chip (SoC) architectures offer novel ways to trade available compute for power savings on-the-fly. With this …

A Fault-Tolerant, Multi-Heap Dynamic Memory Allocator For Freertos, Garrett E. O'Neill

Master's Theses

The use of dynamic memory allocation presents a significant challenge for embedded systems, particularly in applications that require high reliability. The software controlling these systems needs to perform critical operations within strict timing constraints, and memory management plays a critical role in a system’s ability to meet these constraints. Dynamic memory allocation is inherently non-deterministic: if a task requests memory, it is impossible to predict how long it will take for the memory to be allocated. If a critical task were to rely on dynamically allocated memory, its execution could become stalled leading to a missed deadline and system failure. …

Open Source Asic Design Curriculum, Francisco Wilken

Master's Theses

The ever-growing importance of Application-Specific Integrated Circuits (ASICs) in a high-compute world necessitates that college graduates entering the workforce are well prepared to design them. This thesis details the design of novel ASIC curriculum, using open source tools to teach at the undergraduate level. By moving to a higher level of abstraction than classical transistor-focused coursework, chip design material can be made accessible earlier in an undergraduate degree. Additionally, open source tools provide a powerful, free, and portable platform for students to create their own designs, solving assignments focused on industry readiness. Finally, this thesis studies the results and challenges …

Meta-Learning Hyperparameters For Foundation Model Adaptation In Remote-Sensing Imagery, Zichen Tian, Yaoyao Liu, Qianru Sun

Research Collection School Of Computing and Information Systems

Training large foundation models of remote-sensing (RS) images is almost impossible due to the limited and long-tailed data problems. Fine-tuning natural image pre-trained models on RS images is a straightforward solution. To reduce computational costs and improve performance on tail classes, existing methods apply parameter-efficient fine-tuning (PEFT) techniques, such as LoRA and AdaptFormer. However, we observe that fixed hyperparameters -- such as intra-layer positions, layer depth, and scaling factors, can considerably hinder PEFT performance, as fine-tuning on RS images proves highly sensitive to these settings. To address this, we propose MetaPEFT, a method incorporating adaptive scalers that dynamically adjust module …

Adversarial Robustness In Advanced Machine Learning Models Integrating Graph Neural Networks And Large Language Models, Mahmoud Nazzal

Dissertations

Artificial intelligence (AI) has achieved remarkable performances across various domains. In most real-world applications, data often takes relational forms, such as graphs and networks, or sequential forms, such as text and time series. As AI evolves, specialized models have emerged to handle these structures; Graph Neural Networks (GNNs) for relational mining and Large Language Models (LLMs) for sequential understanding. Despite their success, these models face challenges in security, robustness, and interpretability. GNNs excel in relational reasoning but are vulnerable to adversarial manipulation and lack interpretability, while LLMs are strong in linguistic reasoning and generalization yet struggle with relational data and …

Gamified Gait Rehabilitation Via Real-Time Biofeedback And Adaptive Hip-Exoskeleton Control, Mariya Huzaifa Tohfafarosh

Theses

Gait impairments arise from systemic diseases, age-related degeneration, musculoskeletal dysfunctions, or neurological conditions. While traditional rehabilitation can be effective, they often face challenges such as high costs, inaccessibility, and low patient engagement. To address these challenges, my work introduces a virtual reality-based rehabilitation (VRBR) system, integrating real-time motion and electromyographic (EMG) muscle activation feedback with a gamified virtual environment for enhanced adaptability and engagement. The system includes a custom-designed hip-exoskeleton that provides adaptive spring-like assistance or resistance, supporting both mobility-impaired users and strength training. Assistance levels can be tuned to match the user's progress. Additionally, a custom pressure insole was …

Codelympics: An Educational Game, Shreyas Raghunath, Riley Guioguio

Computer Science and Engineering Senior Theses

This thesis addresses the growth of Computer Science and the increasing demand for resources that educate people about programming. The central problem is that existing resources either fall short in engaging the user, or are ineffective at communicating the basics of programming. Through a year of development and testing, we developed an application to serve as a bridge between games and educational content, making programming skills fun and accessible for younger students with no prior experience.

This thesis details the development of Codelympics, an educational game designed to teach the basics of computer programming, including control flow, functions, variables, and …

Are Cycles Of Neural Activity The Algorithm Of The Brain?, Edwin Omondi Onyango

Computer Science Senior Theses

We propose that precisely timed neural activity cycles can serve as structural primitives for memory and computation in a system that exhibits associative learning like the brain. Inspired by biologically grounded mechanisms such as calcium-dependent plasticity, spike-timing-dependent learning, and phase-sensitive excitability, we construct a spiking neural network model in which repeated temporal coincidences drive the formation of self-sustaining activity loops. These cycles, once formed, persist as dynamic memory traces: not stored as static weights, but as reverberating patterns that replay in time when these loops are restarted. We show that noise alone fails to induce stable structure, but even sparse, …

True-Bsg: A True Random Bit-Stream Generator For Fast And Efficient Stochastic Computing, Mehran Shoushtari Moghadam, M. Hassan Najafi

Faculty Scholarship

Stochastic computing (SC) leverages random bitstreams to perform arithmetic operations, offering ultra-lowcost, fault-tolerant, and highly parallelizable computations. The quality of these bit-streams is crucial for the accuracy and reliability of SC. This paper introduces TRUE-BSG, a novel true random bit-stream generator designed for fast and energyefficient SC. Unlike state-of-the-art (SoTA) pseudo-random and quasi-random bit-stream generators, TRUE-BSG utilizes a highquality true random number generator (TRNG), capable of producing random bits at a rate of 1 Gigabit per second. Our TRNG ensures high entropy and minimal correlation. TRUE-BSG shows comparable accuracy to software-based generators and better energy efficiency than SoTA bit-stream generators, …

Ams-Hd: Acute Mountain Sickness Detection With Hyperdimensional Computing, M. Hassan Najafi, Mehran Shoushtari Moghadam

Faculty Scholarship

Acute mountain sickness (AMS) is a potentially life-threatening condition that affects many individuals traveling to high altitudes. Early diagnosis is crucial, especially for travelers who may not have immediate access to medical resources. While traditional machine learning (ML) methods have been used to detect AMS using biomedical data (e.g., heart rate, blood oxygen saturation, respiration rate, blood pressure, and body temperature), hyperdimensional computing (HDC) has yet to be explored for this purpose using the few of biomedical data. Previous classification methods fall short of balancing accuracy with low hardware complexity, but HDC offers a promising solution. HDC provides a hardware-efficient …

Endo Almost 3–Absorbing Sub-Modules (Modules) And Related Concepts, Wafaa H. Hanoon, Mahmood S. Fiadh, Marwah W. Allami

Iraqi Journal for Computer Science and Mathematics

The concept of Endo Almost 3-Absorbing sub-modules (modules) is presented in this study, along with observations and the connections between Endo 2-Absorbing sub-modules (modules), Endo Approximately 2-Absorbing sub-modules (modules), Endo quasi-prime sub-modules (modules), and Endo prime sub-modules (modules). The study provides a range of attributes, illustrations, and justifications for these concepts. We aim to utilize the ramifications of this research to develop new ideas based on Endo Almost 3-Absorbing sub-modules (modules). Along with observations and an exploration of the relationships between Endo 2-Absorbing sub-modules (modules), Endo Approximately 2-Absorbing sub-modules (modules), Endo quasi-prime sub-modules (modules), and Endo prime sub-modules (modules), the …

Retracted: A Group Decision-Making For Selecting Multi-Deep Face Recognition Models, Abdulbasit Alazzawi, Qahtan M. Yas, Burhan Albayati

Iraqi Journal for Computer Science and Mathematics

Deep face recognition is a significant area of biometric authentication that addresses challenges such as low resolution, varying facial expressions, and inconsistent lighting. This paper presents a robust deep-learning approach to tackle these challenges. The study aims to employ multi-criteria decision-making techniques and verify the influence of individual and group expert opinions in decision-making. However, balancing criteria such as accuracy, sensitivity, specificity, precision, and recall remain challenging across different models. To fill this gap, the study utilized a decision-support framework that included the fuzzy analytical hierarchical process to set criteria weights based on expert input and the Technique for Order …