Committee Chair

Wu, Weidong

Committee Member

Sartipi, Mina; Osman, Osama A.; Fomunung, Ignatius

Department

Dept. of Civil and Chemical Engineering

College

College of Engineering and Computer Science

Publisher

University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)

Abstract

The exponential traffic growth hasn't been well-handled by traditional control systems. Adaptive controls are necessary at signalized intersections since they foresee traffic demand based on AI approaches and make decisions ahead of time. These approaches also boost traffic safety by predicting near-crash events leveraging cutting-edge datasets like LiDAR. This thesis addresses such applications of machine learning and deep learning approaches using emerging traffic datasets. A novel deep learning model, MGCNN is suggested for short-term turning volume prediction using GRIDSMART data from the MLK corridor in Chattanooga, Tennessee. During assessments for 1-to-5-minute future prediction, MGCNN surpasses contemporary models with 0.9 MSE. Traditional machine learning models are applied efficiently for forecasting speed and arrival at green with 0.04 and 0.05 MSE. Convolutional Gated Recurrent Neural Network model is proposed for near-crashes prediction that shows 100% recall, precision, and F1-score: accurately predicting all near-crashes based on LiDAR data from Georgia intersection, MLK.

Acknowledgments

I give thanks to God for the success of this research, and to everyone who contributed directly and indirectly to it. I want to especially thank the faculty and staff members of the Civil engineering Department and the College of Engineering and Computer Science. I wish to express sincere gratitude to my first supervisor, Dr. Osama A Osman who saw potential in me to work with him and guided me through my mistakes and success. I give thanks to my co-advisors, Dr. Mina Sartipi and Dr. Weidong Wu who took care of me and supported heavily when I needed them most. I convey my sincere thanks to Dr. Ignatius Fomunung, Dr. Joseph Owino, and Dr. Mbakisya Onyango for watching over me and supporting me throughout the years of graduate study. I thank Ms. Karen Lomen for all the support she provided whenever and however I needed it academically. I thank the Department of Energy (DOE) for providing funds for my research. I want to express my deepest gratitude to my family for being a genuine and encouraging inspiration which helped me to come this far. I also thank Murad Al Qurishee and his family for their unwavering support, inspiration, and comfort whenever I needed it. I extend my thanks to Syed Md. Tareq for watching over me and guiding me in my stay beyond the border. I am grateful to 862 Oak St, Dr. Willson & Clarence. My colleagues and friends: Maxwell, Amani, Evelyne, Grace, Kelvin, Dumbiri, Aman, Vijayalakshmi, Mahadi, Sabbir, Faiza, and Jibril collectively I thank you all. Your company has been very supportive in my stay in Chattanooga, away from family. Last but not the least, I specially thank myself for all sleepless nights, all hard works that no one notices, being there for myself and not giving up in front of ‘impossible’.

Degree

M. S.; A thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Master of Science.

Date

12-2022

Subject

Deep learning (Machine learning); Traffic safety

Keyword

Machine Learning; Deep Learning; Intersection; Traffic Prediction; Graph Neural Network (GNN); LiDAR

Document Type

Masters theses

DCMI Type

Text

Extent

xiii, 84 leaves.

Language

English

Rights

http://rightsstatements.org/vocab/InC/1.0/

License

http://creativecommons.org/licenses/by/4.0/

Date Available

7-1-2024

Download

Available for download on Monday, July 01, 2024

Share

COinS