Low-cost object tracking with MEMS sensors, Kalman filtering and simplified two-filter-smoothing

摘要

The article focuses on using low-cost inertial navigation systems (INS) for long-term object tracking and makes use of a loose coupling integration method based on Kalman filtering in order to realize a sensor fusion between an INS and GPS. This article shows the performance of two filter smoothing to reduce the growth of errors during GPS outages. A simplification technique is applied to avoid the calculation of inverse covariance matrices for the smoothing, which reduces the possibility of numerical instabilities while increasing the computational efficiency. The research is supported with a series of experiments carried out on campus in order to verify reliability and stability of the overall system. The final solution is low-cost, miniature-size and low-weight, while having an increased accuracy compared to ordinary loosely-coupled systems and is capable of handling multiple GPS outages.