Luggage retrieval systems are essential to the passenger experience at airports, and they're also one of the most intractable problems, given the sheer number of passengers and the near uniformity of bags. Traditional methods, such as manual collection and limited tagging, are inadequate; they often result in delays and dissatisfaction among travelers. This paper presents a design for a YOLO (You Only Look Once) object detection system, an artificial intelligence-based luggage retrieval system for modern airports, which aims to retrieve luggage accurately and efficiently from moving conveyor belts. YOLO is popular for its real-time detection speed, which achieves a compromise between speed and accuracy in fast-moving environments where the flow of baggage is constant and heavy. The system helps in identifying the luggage bags as they move down the conveyor belt. In a single shot, YOLO's detection framework works more efficiently than two-stage detection models. The system can cope with changing light and identify several objects in one frame. Preliminary findings show that the YOLO model performs with high accuracy in detecting objects and that the model runs fast enough to be used in real-time at an airport. This research helps in automating the luggage tracking, makes the baggage claims process more efficient, and improves the overall passenger experience. The results lay the groundwork for the eventual use of AI-powered object-detection systems at airports, and the general application of computer vision to solve logistics problems in dense areas.