Orientation Estimation by Means of Extended Kalman Filter, Quaternions, and Charts

An orientation estimation algorithm is presented. This algorithm is based on the Extended Kalman Filter, and uses quaternions as the orientation descriptor. For the filter update, we use measurements from an Inertial Measurement Unit (IMU). The IMU consists in a triaxial angular rate sensor, and an also triaxial accelerometer. Quaternions describing orientations live in the unit sphere of R4. Knowing that this space is a manifold, we can apply some basic concepts regarding these mathematical objects, and an algorithm that reminds the also called “Multiplicative Extended Kalman Filter” arises in a natural way. The algorithm is tested in a simulated experiment, and in a real one