Real Time Object Detection in Autonomous Vehicle Using Yolo V8

Abstract Autonomous vehicles rely heavily on real-time object detection to ensure safe and efficient navigation in dynamic environments. This paper explores the implementation of YOLOv8 (You Only Look Once, version 8), a state-of-the-art deep learning model for object detection, within autonomous driving systems. YOLOv8 offers enhanced speed, accuracy, and lightweight deployment capabilities compared to its predecessors, making it highly suitable for real-time applications. The model is trained and evaluated on datasets such as KITTI and COCO to detect and classify various objects including pedestrians, vehicles, traffic signs, and lane markings. The integration of YOLOv8 with on-board vehicle sensors and edge computing units enables rapid inference and low-latency decision-making. Experimental results demonstrate that YOLOv8 achieves high mean average precision (mAP) with low computational overhead, affirming its potential for deployment in real-world autonomous driving scenarios. This work highlights the advantages of YOLOv8 in improving the perception module of self-driving cars and addresses challenges related to detection in complex, real-time traffic conditions. Keywords Real-Time Object Detection Autonomous Vehicles YOLOv8 Deep Learning Computer Vision Convolutional Neural Networks (CNNs) Traffic Scene Understanding Edge Computing Mean Average Precision (mAP) Self-Driving Cars Sensor Fusion Road Safety