Observability analysis and filter design for a vision inertial absolute navigation system for UAV using landmarks

This paper focuses mainly on the research of the vision inertial absolute navigation system (VIANS) for Unmanned Aerial Vehicle (UAV) using landmarks. Firstly, the model of the VIANS is established. Secondly, Extend Kalman Filter (EKF) and Unscented Kalman Filter (UKF) are designed, respectively, to get the estimations of absolute position, absolute velocity and attitude of the UAV. And the nonlinear degree of the VIANS is analyzed to design a more suitable navigation filter for the VIANS. Thirdly, the observability matrix of the VIANS is derived based on the Differential Geometric theory, and the influence of the landmark distribution on the degree of observability of the VIANS is analyzed based on the condition number of the observability matrixes. Finally, simulations are carried out to validate the correctness of observability analysis and the effectiveness of the designed navigation filter.