REAL-TIME OBJECT DETECTION IN AUTONOMOUS VEHICLES USING DEEP LEARNING

Object detection is a crucial component of autonomous driving technology. Accurate and real-time detection of every object on the road is required to ensure the safe operation of vehicles at high speeds. In recent years, there has been a lot of research into how to balance detection speed with accuracy. Real-time object detection is one of the important technologies applied to autonomous vehicles that allow vehicles to move safely through traffic. This paper focuses on the use of deep learning, the YOLOv8 algorithm in object detection of self-driving cars. The real-world data set of real driving scenarios involved includes streets, roads, and intersections/squares. The powerful interaction of the model with the deep learning algorithms defines the objects and allows for a fast decision-making process applied in autonomous systems. The metrics used to assess the models include detection rates, accuracy of the bounding die placement, and accuracy of the objects’ detection. The outcome is beneficial in refining the object detection methods and advancing the perception capability for self-driven vehicles as well as making driving automation safer.