Chavan, Kshitija Babaji (2024) AI-driven Autonomous Vehicle using Yolov8 and Deep learning. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (939kB) | Preview |
Preview |
PDF (Configuration manual)
Download (904kB) | Preview |
Abstract
By automating vehicle tracking and control, advanced driver assistance systems, or ADAS, have emerged as an essential component of technology for road safety. The inherent difficulties in developing ADAS are addressed in this study, which offers a thorough remedy that combines deep learning and computer vision techniques. With the help of real-time YOLOv8 object detection, OpenCV-based lane detection, and depth-based distance estimate, the system guarantees safe vehicle following, automated steering assistance, and accurate identification of cars, pedestrians, and barriers. Using eye blink and yawning tracking, a convolutional neural network detects driver fatigue and triggers appropriate alerts. Extensive testing in diverse traffic settings has confirmed the system's effectiveness in detecting drowsiness and preventing collisions.
The suggested ADAS not only reduces accident risks but also marks a substantial development towards accessible, sensor-independent vision-based ADAS solutions, advancing the field of road safety technologies by seamlessly fusing computer visionbased automation with driver monitoring.
Actions (login required)
![]() |
View Item |