Quaternion Unscented Kalman Filtering for integrated Inertial Navigation and GPS

In this paper, a scaled unscented Kalman filter (SUKF) based on the quaternion concept is designed for integrating inertial navigation system (INS) aided by GPS measurements under large attitude error conditions. In this feedback filter, only the bias effects are considered to be independent states and are used to compensate for navigation errors. To preserve the nonlinear nature of the unit quaternion, the weighted mean computation for quaternions is derived in rotational space as a barycentric mean with renormalization and a multiplicative quaternion-error is used for predicted covariance computation of the quaternion because it represents the distance from the predicted mean quaternion. The updates are performed using quaternion multiplication which guarantees that quaternion normalization is maintained in the filter. Since the quaternion process noise increases the uncertainty in attitude orientation, modeling it as a rotation vector is considered. Simulation and experimental results indicate a satisfactory performance of the newly developed model.