Pedestrian identification by associating wearable and environmental sensors based on phase dependent correlation of human walking

Abstract

Providing personal and location-dependent services is one of the promising services in public spaces like a shopping mall. So far, sensors in the environment have reliably detected the current positions of humans, but it is difficult to identify people using these sensors. On the other hand, wearable devices can send their personal identity information, but precise position estimation remains problematic. In this paper, we propose a method of associating wearable accelerometers and foot tracking results using laser range finders in the environment. First we propose an association method based on signal correlation between biped foot and accelerometer. However, in crowded situations, sometimes only one foot of a pedestrian is observed because of occlusion. To cope with the problem, we propose a new evaluation function that focuses on the phase dependent correlation of cyclic walking behavior. Example results of tracking individuals in the environment confirm the effectiveness of this method.