Implementation and performance analysis of smartphone-based 3D PDR system with hybrid motion and heading classifier

In this paper, we present a motion recognition based pedestrian dead reckoning (PDR) system. The difference of smartphone-based PDR system and inertial measurement unit (IMU)-based PDR is that axes of smartphone sensor are changed dependently by pedestrian motions, but IMU-based PDR does not that. To solve this issue, we firstly detect a performed motion and operate a proper PDR algorithm according to the detected motion. We apply this system to many persons to identify its applicability. Especially, a heading classifier having 45 degree resolution is applied. We performed a real field test and analyzed a performance of this system.