A mobile robot system following smartphone users using global following based on Wi-Fi and local following based on walking matching

Abstract

This paper introduces a mobile robot system that follows a human target, in particular, a robot system designed to follow a smartphone user. The proposed system integrates a global following method based on Wi-Fi radio intensity maps and a local following method based on matching walking states.“ Global following ”means that the robot follows the global trajectory of the human target. Because it is important to keep following when the target is lost in a crowded environment, we propose a global following system that uses Wi-Fi radio intensity and occupancy grid maps. This paper presents some experimental results for a robot being called to follow a human target using a Wi-Fi radio intensity map. “ Local following ”means that the robot always keeps close to the human target and keeps tracking the target without losing it. The human target is detected by cross-correlating the signals indicating the target ’s walking motion, including vertical acceleration data measured by the target user’s smartphone and the horizontal velocity of the legs measured by an LRF installed in the robot. The results show that the system produces a higher cross-correlation coefficient for the human target than for other pedestrians in the vicinity. The paper introduces a method for calculating the cross-correlation coefficients online and presents experimental results which show that the target can be identified from an online calculation of the cross-correlation coefficients.