Burton Andrews, "Design and Implementation of Cooperative Controllers for Uninhabited Autonomous Vehicles," advisor: Prof. Kevin Passino.
Cooperative Control of Uninhabited Autonomous Vehicles (UAVs) is an area that will play a significant role in many applications in the near future, some of the most important being in the military field. Much research has been conducted on the development of theoretical approaches to UAV searching techniques and task assignment; however, actual physical implementation of a network of UAVs remains a necessary goal of UAV Cooperative Control. This research project will aid in the achievement of this goal by designing a small network of UAVs that complete a set of basic coordinated procedures. The robots in the Electrical Engineering control lab will be used as UAVs and optical photodetectors will represent targets or threats. The UAVs will search for and verify potential targets, engage or kill positively identified targets, and assess whether or not an engaged target has been successfully destroyed. The project will have to address some of the design issues associated with Cooperative Control of UAVs such as communication and network topology as well as introduce new challenges that have not yet been considered. Emphasis will be placed on successful implementation of the previously mentioned search and engagement procedures in order to demonstrate the benefits of physical implementation versus computer simulation.
Vivake Asnani, "Active Control of Modulated Sounds Using Analytical and Experimental Methods," advisors: Prof. Rajendra Singh and Prof. Steve Yurkovich
In this study, active control techniques will be explored in effort to diminish modulated sounds in a duct environment. Many machines often emit quasi-steady state modulated noise signals. Examples include geared transmissions, cooling fans, electric motors with imperfect windings, and the intake system of an automotive engine. Performing experiments in a duct environment reduces the problem of active control to a single dimension. By simplifying the experiment, the conceptual problems involved in the control of modulated sounds can more easily be examined.
Yan Chan, "Navigational Control using PIC Microcontroller," advisor: Prof. Kenneth Breeding
As the designs of cars are becoming more advanced, much effort is being made to ensure a safer and better driving experience. While features such as anti-lock brakes and airbags serve to protect the driver and passenger, GPS system in some luxury cars helps the driver navigate on the road. It is possible that in the near future, a car will be capable to drive itself with just a few instructions from the user. My research project is a simplified approach to the problem mentioned above. The project involves the interfacing of a microcontroller, infrared sensors and microswitches to build a controller that navigates a small car in a maze. Similar to the real life problem, the car must be aware of the direction it travels, its current location, and how to deal with obstacles.
Cheng-Lin Chien, "Machine Vision and Image Processing," advisor: Prof Umit Ozguner
My research is on the study of the machine vision. This research sets up the machine vision by using two cameras as human eyes and a microprocessor as a human brain to recognize the object. Moreover, with the aid of artificial intelligence, the machine acts like a regular person and identifies objects. This machine vision project will be implemented on the RoboCup robot. In other words, this research creates "eyes" for the RoboCup robot. RoboCup is an international robot soccer competition. This competition promotes the study of robotics and artificial intelligence (AI) researches. Many teams from different research institutions have involved in this competition and my research will prepare for entering this competition in the future. My partner, Mr. Ting-Hsiang Chuang and I will build a robot for the Ohio State University. As for starters to this sophisticated robot building, we both will work on a robot that performs only a single penalty-kicking task. The RoboCup competition has many different divisions of competition by the complexity of the robot. Our goal is to construct a robot in the Midsize robot division of the competition. The difference between Midsize robot and other division is that Midsize robot can not be aided by a central vision from outside of the field. In other words, the Midsize robot must have its own vision and self-control during the game.
Ting-Hsiang Chang, "Robotic Movement Control System," advisor: Prof Umit Ozguner
The Robotic Movement Control is a part of Robot Soccer project. This project includes the design of kicking mechanism, interface between the Image Recognition and Movement Control and combines them with the Image Recognition to form a single penalty-kicking robot. The first part of the proposal is an introduction of the goal of this project and what can be done in this project. The second part of this proposal is how I will approach and accomplish what I propose to do. The last part of this proposal is a brief personal statement about me relative with this project.
Petru Cociorva, "Robocup Project," advisor: Prof Umit Ozguner
For my undergraduate honors research project, I propose to design and implement the player-ball interaction in an artificial intelligence soccer game. The long-term goal of the team in Professor Ozgunerıs laboratory is to build a soccer-playing robot, and compete in the Robot World Cup Soccer Games and Conferences. Currently, the prototype robot is able to move in any direction on the field, to turn left or right, to increase or decrease speed, and to notice solid objects in its path (such as other players or the ball). The goal of my project is to improve the robotıs capabilities, by developing algorithms that enable the robot to control the ball and move it in the desired direction. In order to do that, I propose to install a forward paddle (a flipper), which can rotate, and whose angle of attack is directed by the robotıs micro-controller. The micro-controller is then programmed to determine, based on Physics laws, how to rotate the flipper in order to drive the ball in the desired direction.
Greg Eckenrode, "Design and Implementation of Self-Calibrating Collaborative Signal Processing Acoustic Arrays," advisor: Prof. Randy Moses
Ju-Ling Eng, "Numerical Study of Ultrawideband Wireless Pulse Propagation," advisor: Prof. Fernando Teixeira
Ultrawideband (UWB) is an emerging technology for wireless communications that will allow for an entirely new class of electronic applications benefiting public safety, businesses and consumers. Particular areas in which this new technology will have a strong impact include telemedicine and health care. For those applications, it is very important to study the interaction of short UWB electromagnetic wireless pulses with the various human body tissues. In this project, we will study UWB electromagnetic pulse propagation using powerful computational techniques based on the finite-difference time-domain (FDTD) algorithm. These techniques serve to predict the behavior of those pulses under reflection (scattering) for various biological tissues and can be instrumental in the design of UWB antennas.
Steve Horst, "Digital IF Receiver for Radar Return Signals," advisor: Prof. Denny Burnside
This project proposes a redesign of the receiver used in the compact radar ranges at both Ohio State and MIT's Lincoln Labs. It will be funded by a grant from Lincoln Labs specifically to modernize all of the instrumentation in their range. The current system, which is only about ten years old, occupies much of a small room and uses many rigid analog techniques. Advances in digital signal processing technology will allow for a much more compact and flexible design. The proposed receiver design will process two channels simultaneously, one being the return signal and the other used as a reference. In addition, it can also look at several frequencies simultaneously, which is useful for recovering more signal power by capturing images inherent to the radar pulse. This project uses a novel approach to radar receivers, and provides an interesting challenge for a thesis.
Ahmed Jamil, "Linearization of Power Amplifier Radio Frequency Integrated Circuit," advisor: Prof. Patrick Roblin
Ling-Fung Kho, "A Numerical Study of the Electromagnetic Heating from Radio Frequency Ablation of Breast Tumors," advisor: Prof. Robert Lee
Stephen Leifer, "Examining ZnO for Room Temperature Exciton Lasers," advisor: Prof. Leonard Brillson
There are many technological offshoots for room temperature exciton lasers. They can be made using zinc oxide, but first the properties of ZnO need to be further studied. Using Surface Photovoltage Spectroscopy (SPS) the properties and energy spectrum of ZnO can be determined. The samples will be placed in an ultrahigh vacuum (UHV) chamber where they can be treated with oxygen, nitrogen, and hydrogen plasmas to remove defects and impurities or dope the material. The information gathered in this project will address the problem of why ZnO resists becoming p-type.
Andy Warnock, "Spatial mode structure of longitudinal modes in semiconductor lasers," advisor: Prof. Betty Lise Anderson
Specifications for lasers are lacking in one fundamental area, laser modes. Modes come in two types, longitudinal (along propagation path) and spatial (orthogonal to propagation path). Spatial modes influence beam shape and power distribution across the width of the beam, as well as divergence of the beams as they propagate. Since a laser may operate in one of many possible longitudinal modes or may operate in multiple longitudinal modes simultaneously, determining the behavior of spatial modes across all longitudinal modes is beneficial. This project proposes to determine for the first time whether the shapes and weights of spatial modes are the same across all longitudinal modes in a semiconductor laser.
Brent Woods, "4-D Texture Analysis of Medical Images using Distributed Computing," advisors: Prof. Bradley Clymer and Prof. Tahsin Kurc
The Haralick method of texture analysis has been shown to be an effective tool used in computer-aided diagnosis. To apply such a texture analysis technique to a four-dimensional set of medical images provides an incredible computing challenge that a few years ago would not have been possible. In this project, the author will extend current 3-D texture analysis toolkits to include 4-D analysis capability and show that texture analysis on large datasets can be accomplished through parallel computing. The author will also examine methods of partitioning data and computation in order to demonstrate the texture analysis processing can be optimized in a distributed computing environment. Last update: 30-Mar-03
Bradley Clymer / clymer.1@osu.edu