Android smartphone usage as a pedestrian tracking system

Location-based applications need to determine one's position by means of his smartphone. Satellite navigation is not able to work properly in indoor environment and alternatives are required. In this paper, an Android smartphone is used as an Inertial Measurement Unit (IMU) to track a person in an indoor environment. Two different sensors fusion techniques are evaluated and compared through an experimental study; the Un-scented Kalman filter sensor fusion algorithm and the Madgwick fusion algorithm. The evaluation shows an accumulated error less than 0.8% of traveled distance for both sensor fusion techniques.