A quaternion-based tilt angle correction method for a hand-held device using an inertial measurement unit

Abstract

A gyro-based orientation sensor is prone to orient drift due to an integration step. However, a triaxial accelerometer does not require any integration step to calculate the tilt angles, and the calculated tilt angles do not drift over time. In order to find the tilt angles from a triaxial accelerometer, the sensor should be in an acceleration-free condition. In this paper, an expert system is proposed to identify the stationary state of an inertial measurement unit (IMU). A Kalman filter is designed to reduce the noises of the sensors to make the expert system more reliable. To validate the tilt angle correction method, two different tests are conducted: static and dynamic. When an IMU remains stationary for 30 seconds, almost no angular error is observed: The yaw angle stayed at almost 0deg for 30 seconds, and the roll and pitch angles are derived from the accelerations measured by accelerometers. For the dynamic test, the IMU is moved and then returned to the original orientation. The roll and pitch angles are almost perfectly corrected but the yaw angle exhibits no significant improvement.