Computer Engineering Commons^™

Predicting Vegetation Override Force For Off-Road Autonomy, Marc Nicholas Moore

Theses and Dissertations

Vegetation override is an important aspect of off-road ground vehicle mobility. An autonomous ground vehicle’s (AGV) perception system must distinguish between vegetation that can be easily driven through from vegetation that cannot. Predicting the resistance of vegetation could allow path- planning systems to make this distinction. However, despite its importance, direct measurement of vegetation resistance is rare, as most studies use indirect proprioceptive data, such as inertial measurements, as proxies for override force. Notably, there is a lack of empirical data on the override resistance of small stems (< 2.5 cm) and clusters of vegetation on medium-sized (approx. 1000kg) vehicles. To address this gap, a comprehensive dataset of override measurements was collected for clumps of small vegetation relevant to intermediate-sized AGVs navigating off-road terrain. This dataset includes over 70 recordings using the Robot Operating System (ROS) during controlled driving experiments through small trees, grasses, and bushes. The collected data includes light detection and ranging (LiDAR) scans, imagery, force measurements from integrated load cells, and simultaneous localization and mapping (SLAM) information. A key contribution of this research is the development and calibration of a custom push bar system equipped with load cells to directly measure override forces. These measurements are compared to empirical models previously developed by the U.S. Army Corps of Engineers for larger single-stem vegetation. A preprocessing pipeline was developed to automatically extract and label LiDAR and camera data according to these force measurements. This self-labeled dataset was then used to train machine learning models that predict override resistance of vegetation from LiDAR and camera scans alone. This research characterizes the relationship between override forces and the observable features of vegetation as measured by LiDAR and camera sensors. Deep learning models were developed and trained to predict override forces based on different input modalities and features derived from point clouds and images. The performance of these models was compared across various input features to investigate how deep learning can create a generalizable and accurate force prediction system.

Enhancement Of Ambient Air Quality Index Forecasting Using Optimized Ensemble Model, Vanitha M

Theses and Dissertations

Forecasting ambient air quality is essential for environmental sustainability and public health, especially in heavily populated regions such as China, India, and the United States where air pollution remains a serious concern. Traditional forecasting models often struggle to accurately represent air quality data because of its complex patterns and nonlinear interactions. To address these challenges and improve forecast performance, this research proposes a comprehensive strategy that integrates parallel heterogeneous ensemble modeling with Bayesian optimization.

The study begins with a seasonal machine learning–based imputation technique (SeasonalMLImpute) designed to handle missing data in meteorological and air quality parameters. This method is evaluated …

Generative Ai-Based Optimized Recommender System For Debt Collection Using Large Language Models, Keerthana S

Theses and Dissertations

Reducing the percentage of defaulters who often skip payments throughout the debt collection process might help minimize losses in the banking industry. The debt collection process should be optimized to reduce the rate of defaulters and improve collection rates. Traditional Machine Learning algorithms focused on credit risk analysis, defaulter prediction, and forecasting the recovery rate of debt collection. Researchers are not currently prioritizing the analysis of debt collectors’ performance. The debt collector’s primary responsibility is to retrieve outstanding debts from consumers on behalf of the debt collection firm.

Examining debt collectors’ performance is essential to enhance collection efficiency in the …

Patterns Of Interactions In Human-Machine Teams, Kazuhiko Momose

Theses and Dissertations

Increasingly capable machines, including Artificial Intelligence (AI) agents are playing a more important role in a wide range of applications, including human daily activities and safety-critical systems. They can benefit even more when humans and such machines agents work together as a team by leveraging each other's strengths and complementing each other to enhance overall performance. To design high-performing teams, it is critical to analyze the team dynamics and understand how humans and machines interact with each other. Collaboration, Coordination, and Cooperation (3Cs) are terms typically used to describe the behavior of teams. However, these terms tend to be used …

Breaking The Procrastination Barrier, Bianca Ebanks

Theses and Dissertations

Procrastination is a common barrier to productivity, impacting individuals' ability to achieve goals, especially in academic and professional settings. This study investigates a mixed-methods online intervention combining Behavioral Analysis (BA) principles with mindfulness exercises to reduce procrastination. The aim was to develop a human-centered, personalized intervention that utilizes behavioral reminders and mindfulness techniques to address procrastination in 43 participants. Participants' procrastination levels were assessed using the Irrational Procrastination Scale (IPS), and interventions were tailored based on individual procrastination tendencies. Reminders were sent via email or text, with timing adjusted to participants’ specific needs. The intervention also included mindfulness exercises designed …

Heterogeneous Collaborative Robotics: Multi-Robot Navigation In Dynamic Environments, Tyler Nicholas Raettig

Theses and Dissertations

Abstract: The challenges of multi-robot navigation in dynamic environments, focusing on uncertainties in obstacle complexities, partial observation, and the transition of policies from simulations to the real world. The proposed approach utilizes a deep reinforcement learning (DRL) framework enabling a Light Detection and Ranging (LiDAR)-equipped robot to communicate with a camera-equipped robot to achieve optimal paths despite their different sensors. The key contributions include the development of a cooperative architecture for information exchange between robots, a DRL-based framework for learning navigation policies, and a training mechanism based on dynamic randomization for enhanced real-world adaptability. Experimental validation using Gazebo simulations demonstrates …

End-To-End Learning For A Low-Cost Robotics Arm, Abhishek Chothani

Theses and Dissertations

Robotic manipulation is a cornerstone of automation, with the ultimate goal of developing versatile systems capable of executing a wide range of real-world tasks autonomously. Traditional robotics approaches, while reliable and widely adopted in industrial settings, often struggle with adaptability, perception, and dynamic task execution. This thesis explores the evolution from classical robotics techniques to modern learning-based approaches, leveraging advancements in artificial intelligence to overcome these limitations.

Initially, the thesis presents a pick-and-place pipeline built using the Drake robotics framework and the KUKA iiwa robotic arm. This system employs a pseudoinverse controller for inverse kinematics to perform structured tasks like …

Improving Robustness Of Learning-Based Approaches In Autonomous Systems And Engineering Education, Godwyll Aikins

Theses and Dissertations

This dissertation advances the development of robust learning-based approaches across two complementary domains: engineering education and autonomous systems. Through four studies, this research addresses critical challenges in preparing data-proficient engineers and developing reliable autonomous systems that can operate under uncertainty and incomplete information. The engineering education study examines how mechanical and aerospace engineering undergraduates conceptualize and develop data proficiency skills essential for modern engineering practice. Through interviews with 27 students, the research employs the How People Learn framework to analyze student perspectives on information literacy, data interpretation, and computational thinking. The findings inform pedagogical strategies for developing data proficiency in …

Tac-It An Affective Computing User Interface Design, Andrew Biron

Theses and Dissertations

TAC-IT affective computing user-interface design is an independent computer peripheral that is a tool to be utilized to obtain a user’s self-reported emotional state in real-time. Doctor Rosalind Picard first coined and used the term affective computing in her paper Affective Computing [Picard, R. (1995)]. Affective Computing is defined as the study and development of systems and devices that can recognize, interpret, process, and simulate human affects. It is an interdisciplinary field spanning computer science, psychology, and cognitive science. Since that time, areas of research have expanded exponentially, and areas of interest include how to trigger emotions in a test …

Identifying Redundant Audio Content Over Cloud Environment Using Deduplication Techniques, Venkatesh K

Theses and Dissertations

Cloud computing has become an integral part of modern internet-based services, with users relying heavily on cloud environments as primary storage solutions. However, the exponential growth in data volume presents a challenge (i.e) the proliferation of duplicated content within cloud repositories. Deduplication techniques provide a promising approach to mitigate this issue. This research focuses on detecting redundant audio content within a cloud environment, specifically targeting the sharing of extensive audio files, such as those in Waveform Audio File Format (WAV). The study proposes the Refined Super Subset Identification Algorithm (RSSIA) to efficiently identify redundant content and segments within existing audio …

Investigating Spatiotemporal Trends Using Precursory Signatures: Implications To Develop Short-Term Earthquake Forecasting Techniques In Sumatra-Andaman Region, Ramya Jeyaraman J

Theses and Dissertations

Earthquake forecasting is a challenging field due to Earth's heterogeneous nature. This research aims to develop a short-term earthquake forecasting model by analyzing spatiotemporal trends and precursory signatures in the Sumatra-Andaman region, known for its high seismic activity and tsunami risk. The study adopts an interdisciplinary approach, integrating solid earth tides (SET), micro shocks, and outgoing longwave radiation (OLR) to gain deeper insights into seismic nucleation processes. The research begins by using Singular Spectral Analysis (SSA) to identify potential seismically vulnerable areas through the analysis of irregularities in SET.

A spatiotemporal analysis of micro shocks is conducted to assess the …

Nature Inspired Optimization For Spectrum Sensing And Allocation In Cognitive Radio Networks, Saravanan R

Theses and Dissertations

Cognitive radio (CR) refers to intelligent radio technology that scans its environment to optimize spectrum use and adjusts its parameters accordingly. It employs a communication system that is aware of its surroundings, including spectrum usage and availability. A key aspect of CR is identifying idle channels by analyzing traffic patterns using effective learning strategies.

However, CRNs face challenges such as cross-layer design issues, spectrum sensing errors, hidden node problems, and complex spectrum management. Spectrum sensing is critical for accessing unused radio spectrum while minimizing interference. Efficient sensing techniques must be cost-effective, fast, and capable of detecting weak primary signals. Although …

Impaired Speech Recognition Of Neurological Disorder Persons Using Machine Learning And Deep Learning Techniques, Vishnika Veni S

Theses and Dissertations

Speech Assistive Tools have emerged in recent years to support individuals with cognitive and neurological disorders in the field of assistive technology. People affected by neurological disorders such as autism, stroke, cerebral palsy, dysarthria, Parkinson’s disease, and brain injury often find it difficult to articulate desired sounds, resulting in impaired speech. As the population of impaired speakers continues to increase every year, there is a strong need to develop intelligent speech recognition systems for affected individuals. The primary objective of this research is to develop an Impaired Speech Recognition (ISR) system for the Tamil language. Word Recognition Accuracy (WRA) is …

Development Of Brain Tumor Detection And Feature Extraction Through Deep Learning Approach, Sivapathi A

Theses and Dissertations

As the body's central control system, the human brain is susceptible to a wide variety of disorders, including tumors characterized by abnormal cell growth. It is imperative to detect these tumors as early as possible to plan effective treatment and improve patient outcomes. By using contemporary medical imaging methods, this research seeks to improve the accuracy and efficiency of brain tumor detection through the careful preprocessing and analysis of images, particularly Magnetic Resonance Imaging (MRI) [1]. To provide context for the subsequent research efforts, the challenges inherent in brain tumor detection are discussed comprehensively, including segmentation accuracy, small lesion detection, …

Real-Time Robot Pose Estimation For Industry 4.0: Enhancing Motion Validation And Monitoring With A Vision-Based Framework, Jad Samaha

Theses and Dissertations

Amidst the era of Industry 4.0, robots became integral to automated production lines by performing complex tasks with high precision and adapting to changing production needs in real-time. Therefore, ensuring their precise and reliable operation is paramount for maintaining high-quality standards and operational efficiency. This invoked the need for an automated motion validation tool that guarantees accurate program task execution, by detecting deviations or anomalies that may indicate mechanical faults or software errors. Given the exponential advancements in AI, particularly in Computer Vision, a vision-based solution is ideal for ensuring the correct positioning of a robot in real-time. Operating independently …

Investigating Hardware-Based Aes Countermeasures In The Sam4l Microcontroller For Side-Channel Attack Mitigation, Turki Mesbel Alamri

Theses and Dissertations

Side-channel attacks (SCAs) represent a sophisticated method by which attackers exploit indirect pathways, such as power leakage and electromagnetic emissions to glean sensitive information from microprocessors. These attacks analyze variation in power consumption during operations like data encryption to infer protected data, potentially compromising the security of the device. For example, observing the power draw differences when a device encrypts known data can also serve defensive purposes. Attacks often overlook emissions and power patterns, while focusing on avoiding network and host-based detection systems. This oversight presents an opportunity for security professionals to use SCAs to enhance system defenses by monitoring …

Enhancing Security And Privacy For Smarter Environment Through A Robust Cyber-Physical System Framework, Ramya S

Theses and Dissertations

As digital computing paradigm and practices have emerged in disciplines, devices with processors and sensors were rudimentary, performing independent tasks with limited power. The first computer processors were slow, bulky and consuming high energy, as sensors in thermometers and pressure gauges provide original, independent measurements without effective communication, 1999. It often required powerful, energy-efficient processors and advanced sensors to enable seamless communication and sophisticated data processing.

These devices, since smart home systems to industrial automation tools, which continuously collect, analyse and share data via the internet, facilitating if real-time management, predictive maintenance, and improved seamless experience are used, transforming everyday …

Deep Learning Technique For The Classification Of Stress Among The Students Using Physiological Biomarkers With A Hybrid Feature Approach, Rajendran Vg

Theses and Dissertations

Adolescence is a crucial part in life, and the presence of stress, anxiety, depression, and health issues during this stage is a great concern. This research aims to analyze and predict the cognitive stress in students during the examination period using EEG biomarkers. In this study, raw EEG data is acquired under two different experimental conditions, before and after examination, from 14 subjects with an eight-channel Enobio device. After preprocessing of the EEG signal, the brain rhythms such as theta, alpha, and beta sub-band energies and EEG band ratios such as neural activity, heart rate, arousal index, vigilance index and …

Ai In Healthcare: Early Diagnosis Of Skin Cancer Using Medical Image Processing And Deep Neural Networks, Nirmala V

Theses and Dissertations

Several cancer types are commonly prevalent, and skin cancer is one among them, becoming even more widespread worldwide in the last few decades. To diagnose skin cancer at an early stage and obtain appropriate therapy to treat it, there is a demand to know more about the disease’s characteristics or severity. Skin cancer is caused mainly by various reasons, including damage of the sun or tanning beds by ultraviolet light exposure.

Failing to treat skin cancer might substantially impair an individual’s quality of life as the victim. They likely to experience physical issues linked with the deformities caused by psychological …

Motion Based Analysis Of Ultrasound Imaging For The Study Of Musculoskeletal Tissue Bio Mechanics, Ananth Hari R

Theses and Dissertations

Ultrasound image analysis plays an important role in diagnosing musculoskeletal injuries and monitoring rehabilitation exercises. The first and foremost step in this analysis involves segmentation of region of interest from the ultrasound images. The segmentation of the musculoskeletal tissues from the ultrasound images is challenging due to the inherent drawback present in ultrasound like : (1) Poor image quality due to image corruption by speckle noise, shadows, and attenuation. (2) Dis-continuous boundaries due to orientation dependence during the acquisition of image. (3) Low contrast between nearby anatomical structures. Hence in order to overcome these drawbacks, there is a need for …

Abnormal Event Detection Using Hypergraph Based Multiple Objects Tracking Techniques In Surveillance Videos, Palanivel S

Theses and Dissertations

Abnormal event detection aims to identify the events that deviate from expected normal patterns. This work primarily focuses on detection of rare events in public places. The existing research challenges in a video-based surveillance systems for the vehicle have been analysed and presence of abnormal objects in traffic-oriented videos have been detected. A novel approach for event summarization and rare event detection has been proposed in this work. The key ingredient in this work is the incorporation of Hypergraph (HG) matching.

Despite the reasonable amount of success achieved by a large number of researchers over the globe, distinguishing important videos …

An Adaptive Hybrid Deep Learning Architecture For Providing Guaranteed Qos In 5g Cellular Networks, Rajilal Mv Ms

Theses and Dissertations

Wireless network systems must have effective resource allocation, particularly in the context of 5G networks when flexibility is needed to meet a range of network requirements. Resource allocation is essential in cellular network contexts to guarantee equitable access to customers, partners, and cellular service users. Since resource distribution determines network performance, it offers significant advantages when executed well. One of the biggest issues with 5G technology is resource allocation, particularly when it comes to the Quality of Service (QoS) for various applications. Resources in wireless networks include items like channels, power, and spectrum; these must all be apportioned according to …

Approximate Computing And In-Memory Computing: The Best Of The Two Worlds!, Mohammed Essa Fawzy Essa

Theses and Dissertations

Machine learning (ML) has become ubiquitous, integrating into numerous real-life applications. However, meeting the computational demands of ML systems is challenging, as existing computing platforms are constrained by memory bandwidth, and technology scaling no longer yields substantial improvements in system performance. This work introduces novel hardware architectures to accelerate ML workloads, addressing both compute and memory challenges. In the compute domain, we explore various approximate computing techniques to assess their efficacy in accelerating ML computations. Subsequently, we propose the Approximate Tensor Processing Unit (APTPU), a hardware accelerator that utilizes approximate processing elements to replace direct quantization of inputs and weights …

Uav-Based Tracking And Following Of Railroad Lines, Keith Michael Lewandowski

Theses and Dissertations

Given the pivotal role of the railroad industry in modern transportation and the potential risks associated with track malfunctions, the inspection and maintenance of railroad tracks emerges as a critical concern. While existing solutions excel in performing accurate measurements and detection, they often rely on large, expensive, and time-consuming platforms for inspections. This project, however, seeks to solve the same problem with the use of an unmanned aerial vehicle (UAV), significantly reducing time and cost while maintaining detection capabilities. In particular, this solution is ideal for large-scale, high-level inspections following major events such as floods [6], hurricanes [7] or earthquakes …

Development Of The Structure And Control System Of A Stewart Platform Robot For Human Balance Recovery Interventions, Rhobenn R. Alvarez Zambrano

Theses and Dissertations

In this thesis, the process to design a Stewart platform parallel robot for balance recovery with given assembly constraints and mobility requirements is described. A Model Based Design (MBD) approach in MATLAB was used as a tool to model and optimize the design of the platform through quick and repeatable workspace and movements simulations. An algorithm based on Inverse Kinematics was used to find the most adequate Stewart platform configuration which yields a workspace that fulfills the design goals best. Solidworks was used as a 3D CAD Modeling tool to elaborate machining blueprints while ensuring that each piece fits accurately …

Development Of An Efficient Multi-Objective Approach For Secure Live Virtual Machine Migration, Venkata Subramanian N

Theses and Dissertations

Cloud computing offers organizations flexibility and cost-efficiency through pay-asyou- go services, allowing them to scale resources according to their needs and reduce expenditures. Cloud as a Service (CaaS) offloads IT management complexities, while Cloud Data Center (CDC) provides infrastructure for on-demand, scalable, and flexible services over the Internet. Virtualization improves operational efficiency by providing simultaneous access to multiple virtual machines, while Live Virtual Machine Migration enhances agility, resilience, resource allocation, and fault tolerance.

However, achieving effective VMM requires forecasting cloud resource utilization, selecting the right target host, and ensuring security. Live VM migration is inevitable for optimizing CDC resource utilization. …

Improving Tcp Performance Using P4-Programmable Data Planes, Jose Antonio Gomez Gaona

Theses and Dissertations

The Transmission Control Protocol (TCP) forms the foundation of reliable end-to-end communication and is widely used by many applications for efficient data transfers across the Internet. TCP relies on various congestion control algorithms (CCAs) to manage its response to network congestion. These algorithms ensure fairness among competing flows and optimize the use of available bandwidth. However, since TCP is deployed within end hosts, its capacity to attain visibility of network events is constrained. This limitation arises from the closed nature of non-programmable network devices, which prevents researchers from exploring customized actions in response to network events. Recently, the emergence of …

Matrix Processing With Photonic Analog Computing, James Michael Garofolo

Theses and Dissertations

In the digital age, a wide variety of engineering problems have been solved, to a great deal of success, by digital computing techniques. The flexibility of software and relatively low cost of digital computing hardware make it an ideal starting point for solving a majority of tasks, and the numerical stability of software solutions make it highly appealing as the major workhorse for computational tasks. Despite this, many problems are actually suboptimally solved by digital methods, leading to systems with high latency, low throughput, power hungry parallel processing units and an excess of memory for discretizing sensor inputs. Computational photonic …

Hardware Acceleration Of Numerical Methods For Solving Ordinary Differential Equations, Soham Bhattacharya

Theses and Dissertations

Along with the advancement in technology, the role of hardware accelerators is increasing consistently, delivering advancements in scientific simulations and data analysis in scientific computing, signal processing tasks in communication systems, matrix operations, and neural network computations in artificial intelligence and machine learning models. On the other hand, several high-speed computer applications in this era of high-performance computing often depend on ordinary differential equations (ODEs); however, their nonlinear nature can present a challenge to obtaining analytic solutions. Consequently, numerical approaches prove effective in delivering only approximate solutions to these equations. This research discusses the implementation of a customized hardware accelerator …

Navigating The Rules: Integrating Td3 And Sensor Fusion For Traffic-Aware Autonomous Vehicle Path Planning, Mahmoud Ayman Mohamed Elsayed

Theses and Dissertations

This work presents a novel algorithm for local path planning for autonomous vehicles (AVs) which prioritizes both safety and adherence to traffic regulations, addressing critical functions for AV navigation, such as navigating complex environments, avoiding obstacles, and ensuring passenger and road users safety. The algorithm integrates the Twin Delayed Deep Deterministic Policy Gradient (TD3) with sensor fusion based on Nvidia Convolutional Neural Network (NCNN). The study utilizes the CARLA simulator, and real-world datasets, including KITTI and WAYMO, to train and evaluate the proposed algorithm. The proposed algorithm leverages the complementary strengths of Imitation Learning (IL) and Deep Reinforcement Learning (DRL) …