Honors College Theses

Publication Date

5-3-2021

Major

Electrical and Computer Engineering (B.S.)

Release Option

Restricted to Georgia Southern

Faculty Mentor

Rami Haddad

Abstract

This project aimed to compare the performance of multiple different artificial intelligence training methods using their performance controlling a vehicle on a testing track. The specific methods used in this project were a Dense Convolutional Network (DCN) and Residual Neural Network (RNN). Using each of these methods, the vehicle was able to go around the track successfully, though each had different levels of error and training times. Based on the results of speed vs errors made, the best model found in this project was that of the DCN with an average accuracy of 99.61%. However, the DCN needed over 28 minutes to train, while the RNN only needed 13 minutes to get 96.36% on the same dataset. This made the RNN the best model for speed and quick training requirements.

Copyright

This work is archived and distributed under the repository's Standard Copyright and Reuse License (opens in new tab). End users may copy, store, and distribute this work without restriction. For all other uses, permission must be obtained from the copyright owners or their authorized agents.

Download

Share

COinS