Development of human tracking method using a LRF and vision when the target is passing through a crowd

Abstract

We propose a robot that supports people's daily life by tracking people using LRF and carrying their baggage. However, when tracking a target person in an environment where many other people exist, it is difficult to keep track of the target person using only an LRF. This is because the data acquired by the LRF are of a two-dimensional planar shape and the other information of a person is not available. In this paper, we propose a process to detect markers by using the vision system and acquire the position of the tracked person under the circumstance when the tracking target cannot be observed from the LRF. Using the proposed method, we can rediscover the target person after losing the position of the person.