Thesis/Project Final Exam Schedule

 

Final Examination Schedule

PLEASE JOIN US AS THE FOLLOWING CANDIDATES PRESENT THEIR CULMINATING WORK.

Summer 2019
 

Thursday, July 11

Allison Gibson

Chair: Dr. Geethapriya Thamilarasu
Candidate: Master of Science in Cybersecurity Engineering
12:30 P.M.; DISC 464
Protect Your Pacemaker: Blockchain for Authentication and Consented Authorization in Implanted Medical Devices with MedIC

Implanted medical devices often lack methods for authentication and authorization for their programmers, resulting in unauthorized access. In this project, we introduce Medical Implant Consent (MedIC), a blockchain-based solution that grants authorization according to informed patient consent. We describe protocols for doctors to obtain digital medical licenses and for patients to obtain digital informed patient consent. Then we demonstrate consented authorization. MedIC used 46 MB of RAM, and had a footprint of 4436 MB. It took 200 ms to authenticate according to our patient's consent.

Tuesday, July 30

Nana Liu

Chair: Dr. Min Chen
Candidate: Master of Science in Computer Science & Software Engineering

1:15 P.M.; DISC 464
Cloud-based Versioning Control System to PELDA

We aim to design and develop web services on cloud platform to process endangered language recordings and extract audio features to facilitate effective language documentation and analysis for groups as well as individuals, that is to say, support potentially distributed users to work on same projects and keep track of any work done. To achieve this goal, in the project, I designed and implemented a version control system which manages project accessibility and annotation file versioning. We deployed it as web services into the Azure cloud platform, the RESTful APIs we designed and developed then can be integrated into any application that supports HTTP communication. We integrated google login to the system for the user management. This versioning system will not only facilitate cooperation between remote group members but also record individual accomplishments across several projects.

Friday, August 2

Julio Perez

Chair: Dr. Geethapriya Thamilarasu
Candidate: Master of Science in Cybersecurity Engineering

1:15 P.M.; DISC 464
IoT Targeted Blockchain

Internet of Things (IoT) device adoption has been growing consistently over the past decade. Unfortunately, many devices on the market are designed without consideration for security. This has manifested insecure systems that do not account for data security. To address this deficiency, the first stage of Slimload, a custom IoT focused blockchain, is proposed. This system can be added to IoT devices creating a seamless network layer that is leveraged by secondary applications to save data on a blockchain ensuring immutability. Compliance with the confidentiality, integrity, and availability triad is possible employing this blockchain. The introduced framework removes the proof-of-work constraints and introduces a distributed miner selection process that ensures all participating nodes assume the role, given enough participation. 

Monday, August 12

Dong Hun (Ryan) Kang

Chair: Dr. Min Chen
Candidate: Master of Science in Computer Science & Software Engineering

10:00 A.M.; DISC 464
TEXT-TO-IMAGE SYNTHESIS: Retrieving Visual Representation Of Natural Language Sentences Using A Mobile Application

In a world where doing business globally and traveling has become a part of everyday life, knowing a second language has become a great asset and in some cases a necessity. However, a study was done proving that students at the undergraduate level have more difficulties in the use of preposition of place and direction than the preposition of time. Thus, this led to the development of the "illustration" technique. This technique has proven to be effective when teaching the prepositions of place since the visuals play an enormous part in processing new information as it can be used as a reference point or stimulus. However, the accessibility to the illustrations is the main problem with this teaching method. The goal of this project is to build an educational software that generates a visual representation of a sentence with the preposition of place. Our developed system uses ImageNet to access a real-world image database that fetches multiple images and WordNet frameworks to tokenize the preposition of place and nouns of a sentence. The system also used the GrabCut algorithm to segment the main object from the images to create a visual representation of the text. Additionally, the system is built with RESTful APIs and iOS frameworks to access the service anywhere with a mobile application for the usability. We hope our system will be further developed in the future to generate sentences with grammatical structures such as adjectives and grammatical numbers.

Tuesday, August 13

Andrew Watson

Chair: Dr. Erika Parsons
Candidate: Master of Science in Computer Science & Software Engineering

1:15 P.M.; DISC 464
Improving Autonomy on the TrickFire Mining Robot

Computers, microcontrollers, and robotic systems have become ubiquitous in recent years. This has enabled the development of robotics applications in different areas, from autonomous vehicles [1] to hobby electronics [2]. In addition, with an increased interest in space exploration, institutions like NASA are trying to encourage the research and implementation of robotics systems in universities through competitions such as the NASA Robotics Mining Competition [3]. Robotic systems like a Mars rover require complex design and implementation, consisting of both hardware and software systems working in constant interaction with each other. The UWB TrickFire Robotics Team [4][5] has recently participated in this competition with their robot. However, the robot needs various improvements at different levels and functionalities. The work done in this Capstone project, has focused on researching and developing strategies to improve the performance of the autonomous navigation functionality, as well as refining the process used in the continued development and testing of the robot. To accomplish the former, this project involved the integration of a new Intel RealSense D435 depth camera [6] and adding 4K cameras, as well as collaborating on and coordinating the implementation of the CAN bus and new motor controllers. One of the relevant tools used in this work was the RealSense [6], which was found to work well for detecting obstacles both indoors and outdoors. The robustness of the RealSense allows for the correction of an issue with the previous sensor, which was both not tolerant of interference from natural light, as well as failing to detect an obstacle during critical operation. One of the biggest challenges related to the development and implementation of this project was that the robot’s software architecture cannot handle ultra-high-resolution images from the new camera hardware without suffering reduced throughput and high latency. Profiling of different settings revealed that the system operates best at high-definition with a moderate frame rate which sufficiently bounds error for safe operation. The other aspect of this project has been related to integration work, which has been substantial because very much like software development, working with physical hardware requires conscientious methodology, attention to detail, and methodical troubleshooting procedures.

[1] “Cat | Autonomous Mining - Expanding Options for Hauling | Caterpillar.” [Online]. Available: https://www.cat.com/en_US/by-industry/mining/articles/catcommandforhauling-expanding.html. [Accessed: 18-Jul-2019].
[2] “Build a robot buggy - Introduction | Raspberry Pi Projects.” [Online]. Available: https://projects.raspberrypi.org/en/projects/build-a-buggy. [Accessed: 30-Jul-2019].
[3] “RMC 2019 Registration, Rules and Rubrics - 09.06.2018.” .
[4] “(4) University of Washington Bothell TrickFire Robotics: Overview | LinkedIn.” [Online]. Available: https://www.linkedin.com/company/trickfire-robotics/. [Accessed: 30-Jul-2019].
[5] “TrickFire Robotics,” TrickFire Robotics. [Online]. Available: https://trickfirerobotics.com. [Accessed: 30-Jul-2019].
[6] “Depth Camera D435,” Intel® RealSenseTM Depth and Tracking Cameras. [Online]. Available: https://www.intelrealsense.com/depth-camera-d435/. [Accessed: 28-Jul-2019].

Back to top

Questions: Please email cssgrad@uw.edu