Inertial-Based Integration With Transformed INS Mechanization in Earth Frame

This article proposes to use a newly derived transformed inertial navigation system (INS) mechanization fuse INS with other complementary systems. Through formulating the attitude, velocity, and position as one group state of double direct spatial isometries $\mathbb{S}{\mathbb{E}_2}(3)$ , has proven be affine, which means that corresponding vector error model will trajectory-independent. In order make in inertial-based integration, both right left models are derived. The INS/GPS INS/Odometer integration investigated two representatives integration. Some application aspects applications presented, include how select model, initialization based covariance, feedback correction definitions. Extensive Monte Carlo simulations land vehicle experiments conducted evaluate performance models. It is shown most striking superiority using their ability handle large initial attitude misalignments, just result log-linearity property Therefore, can used so-called alignment for applications. Moreover, state-space also very preferred long-endurance due filtering consistency caused by its less dependence on global estimate.