Integrated navigation for pedestrian with building heading algorithm and inertial measurement unit

To improve the positioning accuracy in indoor environments, we use the Inertial Navigation System (INS) algorithm to perform the pedestrian tracking, and the measurements consist of two parts: the velocity error and the heading error. The velocity error can be gotten from the navigation result. Meanwhile, we get the heading error in two ways. First, taking the magnetometer as the heading resource, the magnetic perturbations can be eliminated through a new ellipsoid fitting based calibration algorithm. Furthermore, the building heading algorithm(BHA) is also adopted to aid the Inertial Measurement Unit (IMU), the corresponding building heading can be derived based on the motion path. Finally, the Kalman filter (KF) is utilized to fuse the data in order to compensate the sensor error and navigation solution through a 15-dimentional state vector. As a result, some field trials were taken to prove the validity of the proposed algorithm. By contrast the result of Magnetometer (MAG)/BHA/INS and the Zero Velocity update (ZUPT) algorithms, we proved that the algorithm of MAG/BHA aiding INS can effectively improve the indoor positioning accuracy which shows better performance.