A Comparison of Human Tracking Systems on a Mobile Robotic Platform

Abstract

The field of IoT has a growing interest in robotics, and one of the big fields of this interest is human-robot collaboration. A sub-category of the human-robot collaboration is how a robot can follow a human. There are a few methods of following a person; one would be using image recognition, and another way could be using a rope attached to that person. Image recognition is compared to following a person with a rope to find out the benefits and weaknesses of both methods, comparing smoothness, responsiveness, and accuracy. With the comparison of these two systems, the findings presented in this study indicate that the overall responsiveness of the spool-based system is higher than the camera-based system, the directional responsiveness of the camera-based system is higher than the spool-based system, and the distance estimation of the camera-based system is noisier and less reliable than the spool-based system.