A self-resetting method for reducing error accumulation in INS-based tracking

The error accumulation in the INS systems is one of the major drawbacks which limits their applicability. Several efforts have been made to improve the performance of the INS systems in the tracking applications with limited success in terms of efficiency, complexity and cost. This paper proposes a new method for modelling the actual INS error accumulation and use of automatic self-resetting for reducing error accumulation within the INS system without the need of any external sensors or devices. The new method has been verified and tested for the case of a single-axis accelerometer. Furthermore, the method has been used to recover the trajectory of moving objects in outdoor environments when a full INS system is used. The testing results show that the new method considerably reduces the error accumulation in the INS systems. Moreover, the new method increases the operation time of the INS-based systems from few second to a few minutes.