Conceptual

Indoor Position and Attitude Tracking on the SO(3) Manifold with Riemannian Kalman Filters

EKFRie and UKFRie algorithms that exploit the SO(3) Riemannian manifold structure of the rotation matrix (via Riemannian optimization and tangent-space updates) to jointly track the 3D position and orientation of a moving indoor target, achieving substantially lower position/orientation RMSE than conventional EKF and UKF.