PERCEPOLIS
Our Goal
The PERCEPOLIS project aims to design and create a web application that employs a set of algorithms to personalize and optimize a student’s path to graduation. The algorithms automate and streamline the academic advising process, an area that has largely gone untouched through the rise of technology over the past two decades. The algorithms are designed to help the student tailor their college career and degree to their interests while also ensuring that they complete their degree quickly, but at a pace right for them. The application consists of four algorithms, each enumerated below. The repositories for the project can be found below, however, access is required for viewing.
Algorithms
- Student Performance Evaluation
- Evaluate a student’s past performance and calculate a suggested credit hour limit
- Course Selection and Optimization
- Select courses based on the student’s degree requirements, career interests, and course availability. Once all courses on the path to graduation have been selected, the algorithm puts the courses into semesters in an optimized ordering that represents the fastest path to graduation.
- Course Scheduling
- Student Performance Evaluation
- Finds all possible schedules for a student for the upcoming semester. Once a schedule has been selected, the student is auto-enrolled, eliminating the need for the student to stress about enrollment times.
- Course Difficulty
- Evaluate a student’s past performance and calculate a suggested credit for a student’s past performance and that of similar students to determine the probability that the student will pass a given course.
Scholars
Current
- Mitchell Skaggs – Accelerated Master’s Student/Server-side Developer
- Arianna Chaves – Accelerated Master’s Student/Server-side Developer
- Colton Walker – Accelerated Master’s Student/Server-side Developer
- Austin Beckerdite
Previous
- Nic Dobbins – Master’s Student/Server-side Lead
- Sanfan Liu – PhD Student/Server-side Developer
- Hans Hernandez – Server-side Developer
- Hayden Dixon – Client-side Developer
- Thomas Gleiforst – Client-side Developer
- Antonio Kotoni – Undergraduate Scholar
Repositories
Documentation
This repository houses all documentation related to the project.
Client
This repository houses the code for the application web pages. The application has not been published yet.
Server
This repository houses the code for the application algorithm and data operations. The application has not been published yet.
Survivability of Autonomous Vehicles
Our Goal
The Survivability project aims to create qualitative and quantitative models for the survivability of autonomous vehicles under faults and well-correlated failures. This involves the investigation of autonomous vehicle vulnerabilities, which provides the prioritization of attacks based on their probability of success. The insights gained from this experimentation will help fortify the platform against future attacks through criticality analysis, with the goal of increasing survivability.
Project Overview
Perform qualitative survivability analysis of autonomous vehicles communicating with other autonomous vehicles and the intelligent transportation system. This is done by performing two analyses:
- AV Vulnerability Analysis
- Attacks are prioritized by their probability of success and inform quantitative survivability models
- Component Importance Analysis
- Increases the survivability of the AV system through individual component security hardening and defense in depth.
Scholars
- Justin King – CISSP, PhD Student
- Curtis Brinker – Accelerated Master’s Student
- Anna Muecke – Undergraduate Scholar
- Ellie Jackson – Undergraduate Scholar
- Ryan Brothers – Undergraduate Scholar
Repositories
Survivability
This repository is the main repository for the project. This repository is not open to the public.
Traffic Prediction
Our Goal
The Traffic Prediction group is working to create an application that predicts dense road segments by using information that can be gained from a cell phone or other routing device, such as heading, destination, and location. Our prediction method attempts to be as accurate as possible by accounting for predicted dense areas when predicting dense areas further in the future. We also attempt to be as computationally efficient as possible by reducing the areas that are searched for dense roads by using dynamically sized cells. Having the knowledge of where roads are going to be busy and how busy they will be will allow drivers to plan their routes accordingly and avoid traffic saving both time and money. The phases of the traffic prediction algorithm can be found below.
Scholars
Current
- Bikis Muhammed – Accelerated Master’s Student
- Jeremy Thomas – Undergraduate Scholar
- Owen Miller-Fast – Undergraduate Scholar
Previous
- Karl Frank – Undergraduate Scholar
- Jake Mason – Undergraduate Scholar
- Jeremy Thomas – Undergraduate Scholar
Repositories
Traffic Prediction
This repository is the main repository for the project. This repository is not open to the public.
IEMI
Our Goal
The Electromagnetic Interference (EMI) Detection group aims to develop a software-based methodology to detect the presence of EMI and study the effects it has on software. Modern electronics are becoming more susceptible to EMI due to their rising clock speeds and smaller overall size which makes it ever so important for critical systems to understand how EMI affects them. Having the ability to understand when and how a device is being affected by EMI will allow critical systems to further harden against the effects of EMI and provide more reliable services. Furthermore, a software-based analysis method allows for a better understanding of hardware that is infeasible to equip with laboratory sensors such as large-scale systems or consumer hardware. Our technique uses a lightweight watchdog to log changes in key registers and uses these register values to construct states. When the device is active these states are recorded and used to give insight via categorical time series statistics and classification algorithms. Below you can find information about the device we use and specific detection methods.
Device
For our experiments, we examine the operation of a USB 2.0 host controller on a Rock Pro 64. We chose the Rock Pro 64 as it is a well-documented and affordable system that has a built-in USB 2.0 controller. For further information on the Rock Pro 64 system please visit the link below.
Scholars
- Joel Schott – Doctorate Student
- Evan Hite – Undergraduate Scholar
- Connor Jones – Accelerated Master’s Student
- Austin Potter – Undergraduate Scholar
Methods
- Time Series Statistics
- Dispersion
- Serial Dependence
- Classification
- Hidden Markov Model
- Recurrent Neural Network Long/Short-Term Memory Classifier
- Artificial Neural Network
- Support Vector Machine
- Random Forest Classifier
- Gradient Boosted Classifier
Additional Information
Device Info
This website contains additional information about the Rock Pro 64 system.