Indoor 3D pedestrian tracking algorithm based on PDR using smarthphone

In this paper, we develop the indoor navigation system based on PDR (Pedestrian Dead Reckoning) using various sensors in smartphone. Usually PDR is consisted of step detection, step length estimation and heading estimation. The issue of PDR is step length estimation and to enhance the accuracy of step length, we apply the walking status recognition algorithm using ANN (Artificial Neuron Network). The features used in ANN are extracted through sensor signals of accelerometer and gyroscope. After recognizing the walking status, it is applied to estimate the step length. And when the status is recognized as stop, even if sensor signal is generated by redundant motion or movement of pedestrian, the moved distance is not calculated additionally and distance error is not increased. We use the barometric pressure sensor to extend the positioning area to whole building. To verify the proposed indoor navigation system, we implemented the application for android and conducted the experiment. Through the results, we demonstrated the accuracy of our system.