Self-Alignment Techniques for Inertial Measurement Units

This paper presents the result of a study on self-alignment algorithms for inertial measurement units (IMU). The primary concern is the fine alignment of an inertial platform whose base is subject to vibration and whose sensors are subject to noise. The main contribution of the paper is a new self-alignment algorithm. The algorithm incorporates together the special property of platform kinematics and the concept of least square regression. Compared to the usual self-alignment algorithm, this algorithm gives a self-alignment which can be ten times more accurate and requires less computer time and memory. In addition, the new algorithm is less sensitive to erratic measurements and computational errors.