Android application for indoor location using sensor fusion: ultrasounds and inertial devices

This paper presents an Android application that allows estimating the position of the device in a robust and flexible way combining ultrasound signals emitted by a set of Ultrasonic Local Positioning Systems (U-LPSs), with inertial information obtained by an Inertial Measurement Unit (IMU) carried by the user and communicated via Bluetooth with the portable device. The U-LPSs are located in areas that require a high accuracy (such as entrances and waypoints) and the trajectory between U-LPSs where there is not ultrasound (US) coverage is estimated by the inertial sensors. When the user is inside an U-LPS area again, the trajectory and cumulative error of the inertial sensor are corrected. Therefore, this proposal provides a flexible and robust application, which is capable of adjusting to the demands of a large environment, reducing the number of U-LPSs to place. Each U-LPS consists of five transducers or beacons whose emissions, at a frequency of 41.67kHz, are multiplexed in time and encoded to achieve authentication and reduce interference by multiple access. At reception, an unlimited number of users with their mobile devices could calculate their position autonomously. The ultrasonic signals are captured by an external module that digitizes the signals and sends them to the portable device for processing. At the same time, IMU data is captured and, using an Extended Kalman Filter (EKF), both measures are fused and the possible errors accumulated by the IMU are corrected.