Honors Theses

Date of Award

Fall 12-8-2023

Document Type

Undergraduate Thesis

Department

Computer and Information Science

First Advisor

Charles Walter

Second Advisor

Jeff Lucas

Third Advisor

Nathan Oakes

Relational Format

Dissertation/Thesis

Abstract

In recent years, cameras have become ubiquitous in daily life, constantly surveilling, and taking in information. This leads to a potential security risk of the invasion in one’s privacy without their knowledge or any ability to prevent the privacy threat. While cameras alone are an issue, they are often only in locations where a user has some expectation of a loss of privacy, such as public locations with security systems. However, systems that rely on cameras to operate correctly, including autonomous vehicles, are becoming a more prominently used technology while often appearing in places where an average person has some expectation of privacy. With this technology, comes the use of many different sensors that collect numerous amounts of data, which can be used for malicious intent. As these technologies advance, security systems must advance to keep pace, protecting the security and privacy of users. In this thesis, I show that it is possible to increase security and, more specifically privacy, within autonomous vehicles. I used the Webots simulation platform, a well-known tool for simulation of autonomous robotics and autonomous vehicle systems. Webots allows a simulation that utilizes the same sensors autonomous vehicles currently use while allowing flexibility in the development of security and privacy methods. Using Webots, I created a system that uses a man-in-the-middle attack as a defense mechanism, intercepting data streaming from the camera, doing initial processing, and returning only textual information about what the camera sees, preventing any information that could cause a privacy violation from making it off the vehicle. I utilize an e-puck robot with controller code that represents the on-board computer used to make decisions within an autonomous vehicle and my solution of a “privacy preserver” used to privatize information received from the camera sensor to help navigate through a maze with colors simulating objects to be categorized. I test this solution to ensure that all information utilized by the on-board computer has been sufficiently anonymized. I show that, through simulation, the “privacy preserver” concept is viable through simulation.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.