41530 research outputs found
Sort by
Design And Analysis Of A New Buck-Boost Converter With A Low-Side Switch
Get PDFThis work explores the design, simulation, construction, and analysis of a novel Non-isolated DC-DC Buck-Boost converter which has the advantage of incorporating a low-side switch compared to the traditional buck-boost which requires a high-side switch. This allows the use of a low-side driver which further simplifies the design and operation of the converter. The proposed Buck-Boost converter was constructed to provide -24 V output from an input range of 12V-18V with 15V nominal input at 10W maximum output power utilizing 500kHz switching frequency. Findings from simulations and hardware tests verify that the converter effectively provides the desired -24 V output at varying loads with less than 3% ripple. At the nominal input voltage, the efficiency of the converter reaches 82.37% at full load and peak efficiency of 88.5% at 20% load. Moreover, the input voltage ripple of the proposed non-isolated converter reached 8.4% at full load, due to the pulsating nature of the input current. Overall, results verify the feasibility of the proposed non-isolated Buck-Boost converter as an alternative solution for the conventional buck-boost with the advantage of a low side switch while maintaining a low component count
Deep Learning Using Vision And LiDAR For Global Robot Localization
Get PDFAs the field of mobile robotics rapidly expands, precise understanding of a robot’s position and orientation becomes critical for autonomous navigation and efficient task performance. In this thesis, we present a snapshot-based global localization machine learning model for a mobile robot, the e-puck, in a simulated environment. Our model uses multimodal data to predict both position and orientation using the robot’s on-board cameras and LiDAR sensor. In an effort to minimize localization error, we explore different sensor configurations by varying the number of cameras and LiDAR layers used. Additionally, we investigate the performance benefits of different multimodal fusion strategies while leveraging the EfficientNet CNN architecture as our model’s foundation. Data collection and testing is conducted using Webots simulation software, and our results show that, when tested in a 12m x 12m simulated apartment environment, our model is able to achieve positional accuracy within 0.2m for each of the x and y coordinates and orientation accuracy within 2°, all without the need for sequential data history. Our results demonstrate the potential for accurate global localization of mobile robots in simulated environments without the need for existing maps or temporal data
Multi-Hop and Automated Testing of Wide-Area Network Radio Systems
Get PDFRequests for low-power and versatile wide-area Internet-of-Things (IoT) networks are becoming more prevalent with upcoming research in low-data rate ad-hoc mesh networks. Connections to terrestrial satellites have proven promising methods for increasing range capabilities for wide area networks (WAN). As research in communication with IoT networks and satellites progresses, reliable multi-hop mesh networks for Earth-based communication are still necessary to develop the base for satellite communication. These multi-hop communication mesh networks are being developed to allow communication where Line-of-Sight (LOS) from A to B is incapable due to environmental barriers. Hence, communication is possible without LOS by providing an intermediate C station that can view A and B. This research attempts to generate tests to confirm the functionality and efficiency of multi-hop communication over LoRa radios. These tests are aimed to determine the basic radio functionality in a short one-minute test, along with a longer test capable of calculating a more detailed set of telemetry points. With these tests, we should be able to measure the performance of the radios within the multi-hop network, allowing confidence in data from radio to radi
An Investigation Into Teaching Sports Analytics
Get PDFSports analytics arrived in the mainstream media through the novel and film Moneyball. However, its origins date back to operations researchers following World War II. Often considered a subdiscipline of statistics, sports analytics draws from statistics but also includes concepts from data science, communication, and marketing. As a passionate fan of sports, I have pursued statistics in my undergraduate and graduate education with the dream of working in sports for my career. However, educational opportunities in sports analytics are limited nationwide, and more specifically, there is no educational opportunity at my university, California Polytechnic State University in San Luis Obispo. This thesis investigates the sports analytics discipline, aiming to explain what sports analytics is, how it differs from statistics, how sports analytics is used in various organizations, what sports analysts do, and how sports analytics should be taught at the undergraduate level here at Cal Poly. To accomplish this, I have taken three online sports analytics courses, conducted interviews with professors of sports analytics and sports analysts of professional and college teams, done extensive online research and literature review, and gauged interest campus-wide in a potential sports analytics course. Ultimately, this thesis led me to conclude that sports analytics differs from statistics, and there should be a course in sports analytics at Cal Poly offered by the Statistics Department. Skills including SQL and Tableau, communication to various sports constituents, data collection and data management, machine learning methods such as classification trees and clustering, advanced statistical methods such as General Additive Models and spatial analysis, and visualization techniques are all prominent in sports analytics. Statistics students at Cal Poly do not gain a firm foundation in all of these ideas and could benefit from a course which teaches these skills. The significance of this work is that I have created a course proposal for a sports analytics course. If this course were to be adopted by the Statistics Department, students would learn essential skills to prepare them for a career in sports or any data related career. This work can advance sports analytics education and lead to the creation of other courses in the discipline down the line
Cal Poly Gen 2 BattleBot Electrical System
Get PDFBattleBots is a popular robot combat sport where engineers from all over the world design and construct a robot with the aim to disable or impair the opposing robot. The Cal Poly Gen 2 BattleBot aims to complete a robot from scratch with hopes to compete in the official competition. This report focuses on the electronic design behind the robot, specifically the printed circuit board (PCB) and component selection. The design process of this project involved choosing specific motors and microcontrollers based on cost, efficiency, and benefits with the end goal of a complete printed circuit board (PCB). Brushed motors were chosen for the drivetrain due to their ease of interface with the RoboClaw motor driver. Brushless motors were chosen for the weapon system with their superior speed and torque characteristics. The ESP32 microcontroller was chosen due to cost, Bluetooth capability, and communication interfaces. The PCB design evolved iteratively, incorporating features such as voltage regulation, safety control circuits, and motor control mechanisms. The testing plan validated the PCB’s functionality and safety and worked to ensure the functionality of motor control. The final electrical system was successfully integrated with mechanical and software systems, contributing to a fully operational BattleBot. This project demonstrates the successful applications of electrical engineering principles to the design and implementation of a complex robotic system
Identifying Ion-Scale Waves in Parker Solar Probe Sub-Alfvénic Intervals
Get PDFIdentification of coherent quasi-parallel ion-scale (1-32 Hz) wave activity in sub-Alfvénic regions of the solar wind observed by NASA\u27s Parker Solar Probe during perihelion encounters 8 to 16. Wave activity is filtered computationally by coherency, circular polarization and propagation angle with respect to the mean magnetic field. Initial statistical results are presented with suggestions for future improvements and studies. A general overview of the heliosphere and the context of ion-cyclotron waves (an ion-scale wave) in the coronal heating problem. Along the way to identify ion-scale wave activity, tables of sub-Alfvénic intervals and current sheet crossings for encounters 8 to 16 are also presented
Combining Cloud Architecting with Education
Get PDFI pursued the AWS Solutions Architect Professional Certification while applying my knowledge to build and revise technical solutions for an educational company known as EDFX
Sizing Wind Tunnel Heater For High Enthalpy Conditions
Get PDFThis paper determines the feasibility of adding a heater to an existing blowdown supersonic wind tunnel to unlock new high-enthalpy test applications, considering cost and power requirements at a variety of different states. This process includes both modeling the current range of test section properties in Cal Poly\u27s blowdown wind tunnel and determining the new range of properties that a heat exchanger could induce. These results are verified with a computational fluid dynamics study. Additionally, sublimation and ablation properties of materials are explored to create appropriate models to study atmospheric re-entry once the heat exchanger is implemented.
It is found that adding a heater to the supersonic wind tunnel would significantly increase the test section temperature. Additionally, enough heat could be added without damaging the facility to surpass the vapor pressure of camphor and naphthalene at test section conditions, allowing for the tunnel to be used for sublimation and ablation applications. Using the tunnel with the variable Mach nozzle currently installed would induce minimum heater power requirements of 75kW for a Mach 4 configuration and 200kW for the testing Mach 3.13 condition to reach this vapor pressure. However, this power requirement can be significantly reduced by installing a new nozzle that would induce flow at a Mach number of 6-8. Liquefaction is found to be avoided at every test and Mach condition, even without any heat added, while condensation cannot be avoided at any configuration, regardless of nozzle used or heat added. Therefore, we recommend that a dryer be installed to help remedy these issues
Performance Interference Detection For Cloud-Native Applications Using Unsupervised Machine Learning Models
Get PDFContemporary cloud-native applications frequently adopt the microservice architecture, where applications are deployed within multiple containers that run on cloud virtual machines (VMs). These applications are typically hosted on public cloud platforms, where VMs from multiple cloud subscribers compete for the same physical resources on a cloud server. When a cloud subscriber application running on a VM competes for shared physical resources from other applications running on the same VM or from other VMs co-located on the same cloud server, performance interference may occur when the performance of an application degrades due to shared resource contention. Detecting such interference is crucial for maintaining the Quality-of-Service of cloud-native Web applications. However, cloud subscribers lack access to underlying host-level hardware metrics traditionally used for interference detection without needing to instrument high overhead-inducing per-request response time values. Machine learning (ML) techniques have proven effective in detecting performance interference using metrics available at the subscriber level, though these techniques have predominantly focused on supervised models with pre-existing labeled data sets that can distinguish between normal and interference conditions. In contrast, this work proposes an unsupervised clustering ML approach to identify performance interference in cloud-native applications. The proposed approach implements a lightweight method for collecting container metrics in normal and interference scenarios and applies a dimensionality reduction technique to mitigate redundancy and noise in the collected dataset. We then apply a density-based clustering approach to this unlabeled data set to classify interference in two applications running on the AWS EC2 cloud: a microbenchmark Web application called Acme Air and a large-scale production-realistic Web benchmark called DeathStarBench. Results indicate that our density-based clustering approach effectively distinguishes between normal and interference conditions and achieves an average Density-Based Clustering Validation (DBCV) index of 0.781 and a cluster homogeneity of 0.875 across both applications
Gateway Decathlon: Carbon Calculations of an Offsite Residential Construction
Get PDFThis senior project delves into a comprehensive calculation of carbon emissions for both efficiency and sustainability within the realm of modular construction. A material takeoff of the building was performed, allowing for an accurate measurement of materials and their respective EPD (Environmental Product Declaration) values. After material selection, a final carbon calculation of the building was calculated. Carbon emission calculations were performed on the current Revit Model allowing for a full-scale analysis, from Cradle to Gate, of the offsite residential unit. The results show that the total initial embodied carbon emissions of the unit is 9241 (kg-CO2e), representing the total carbon emissions for the entire residential unit. The utilization of cork was efficient as this offset the carbon emissions by 6162 (kg-CO2e). These values represent the framing and finishes, and they provide a broader understanding of the environmental impact of offsite residential construction