Real Time Robotic Label Affixing for Irregular Items Using Low Cost Depth Sensing and Computational Algorithms

In many applications, items are transported on conveyor belts, and it is desirable to attach RFID tags for efficient tracking and tracing. A notable example is luggage handling at airports, where workers manually attach RFID tags to luggage upon unloading from planes to ensure proper tracking for transfers. This manual process is both tedious and time-consuming, often leading to repetitive strain injuries for workers. This paper presents a developed system that automates this process using a robotic arm equipped with a vision system. The system leverages a low-cost depth sensor and operates on a non-GPU machine, making it feasible for edge computing. It employs traditional algorithm designs, including region growing and template matching, providing a deterministic approach to label affixing. By deploying service robots for repetitive tasks such as label affixing, we can reduce human fatigue, mitigate repetitive strain injuries, and save time. This service robot effectively replicates manual label affixing tasks, enhancing efficiency and improving overall workflow in luggage handling.