Improving magneto-inertial attitude and position estimation by means of a magnetic heading observer

摘要

This paper studies heading estimation jointly with the attitude and position estimation of a rigid body equipped with inertial and magnetic sensors in indoor environment. In contrast with other indoor dead-reckoning approaches, no assumption is made about the nature of the movement or environment layout. Based on a previous paper, an Extended Kalman Filter is designed, which includes inertial sensor biases and magnetic disturbances. A heuristic model of the dynamic of magnetic heading disturbances is then described and added to the observer. The latter is then evaluated in terms of position and heading error on experimental data, showing that in spite of high levels of disturbances, the magnetic field alone can be used to compute heading.