Short-term motion tracking using inexpensive sensors

Abstract

Current consumer electronics is equipped with various sensors, among which accelerometer, gyroscope, and magnetometer represent typical examples. In this paper, we study the possibility of using these low-cost sensors for 3D motion and orientation tracking. In particular, we thoroughly describe a simple dead-reckoning algorithm for sensor data fusion which produces a 3D path of the device in real time. More importantly, we propose a method of automated stabilization every time the device stands still, which corrects the bias caused by sensor inaccuracies. This method extends the time when motion tracking is reliable. We evaluate the proposed pipeline in a variety of experiments using two common smartphones.