Integração de sistema de navegação inercial com restrições de movimento e atualizações estacionárias

Abstract

Inertial Navigation Systems (INSs) are navigation systems composed of inertial sensors (accelerometers and gyroscopes), which are of utmost importance for terrestrial, aerial, space, maritime, underwater, and other vehicle-related applications. Inertial Navigation Systems provide a complete navigation solution (position, velocity and orientation) for the vehicle at a high sampling rate, high bandwidth, being also independent of external signals. Nevertheless, INSs have a disadvantage of accumulating errors in the long term, due to the numerical integrations carried out internally. One solution to mitigate this error acumulation is to integrate the INS with auxiliary sensors, usually via Kalman filters. An alternative approach is to use vehicles’s motion constraints (MCs) and/or stationary conditions (ZUPTs), as virtual sensors in order to aid the INS. Informations, such as zero linear and/or angular velocity, zero velocity along the axis of the rear axis, and in the direction that is perpendicular to it and to the terrain can be inserted into the Kalman filter as if they were provide by true sensors, collaborating, in this way for the reduction in INS error accumulation. This work, therefore, revisits the aforementioned problem, and investigates the detection of stationarity techniques that allow sensory fusion, to improve the navigation solution. Through the conduct of computational simulations and practical experiments, in which full-scale vehicles were equipped with Inertial Navigation System (INS) modules, it was possible to conduct a comparison among different scenarios and assess the performance of various integration techniques concerning the north, east, horizontal, and vertical axes.