V3I7P2

Application of the Adaptive Kalman Filter in Sensor Drift Compensation and Accuracy Enhancement of the Inertial Navigation System 

Khuat Quang Tien^1* , Vu Van Son², Nguyen Ha Giang²

Abstract

The Inertial Navigation System (INS) has the advantage of operating independently without relying on external signals; however, it is significantly affected by the accumulated errors from inertial sensors, particularly the drift of gyroscopes and accelerometers. This paper presents a simulation and performance evaluation method for the INS with a model including the states of position, velocity, attitude angles (roll, pitch, yaw), and sensor errors. The Adaptive Kalman Filter (AKF) is applied to simultaneously estimate the system states and errors, allowing dynamic updating of measurement noise characteristics and time-varying sensor drift compensation. Simulation results show that the filter significantly improves trajectory and attitude estimation accuracy while minimizing long-term accumulated errors. The proposed method demonstrates high applicability in aerospace and autonomous vehicle systems.

Keywords:

Inertial Navigation System; Kalman Filter, Adaptive Kalman Filter, Attitude angles, Error estimation, Sensor drift compensation.