This study presents a case of developing and operating a Turtlebot3-based robot in a realworld environment to identify prohibited items such as beverages within a library. The robot moves to designated waypoints and utilizes a camera module to collect...
This study presents a case of developing and operating a Turtlebot3-based robot in a realworld environment to identify prohibited items such as beverages within a library. The robot moves to designated waypoints and utilizes a camera module to collect real-time images, which are then analyzed using a Yolov5-based deep learning model to detect prohibited beverages. Roboflow was used for image labeling and dataset construction to train this model. Additionally, the robots waypoints were efficiently and randomly set, allowing the robot to recognize its location and receive information about the entry status of identified items. The performance of this system was validated through its actual operation.