Proactive Highway Collision Avoidance using YOLOv8 and Ultrasonic Sensors

The Internet of Things (IoT) has become a key component of vehicle-to-vehicle communication with collision avoidance systems. However, certain vehicle models lack the required interaction capabilities, enabling other design techniques. This paper provides an innovative method for a real-time identification system with a vehicle position prediction based on YOLOv8. The machine learning framework enhances stability and prediction accuracy by integrating vehicle dynamics. This method makes use of YOLOv8's capabilities to identify vehicles in complex environments and dynamically alter the prediction based on current patterns and courses of action. The performance of these algorithms has been validated by multiple studies, which demonstrate that they provide better prediction accuracy than current techniques. This development helps to create intelligent modes of transport that can function well in various types of traffic scenarios and car capacities, which in turn improves road safety by providing a strong foundation for active, realtime accident avoidance.