Indoor pedestrian navigation using miniaturized low-cost MEMS inertial measurement units

The paper presents an indoor pedestrian navigation system using miniaturized low-cost MEMS inertial measurement units (IMU). The low-cost MEMS IMU comprises of a tri-axial accelerometer, a tri-axial gyroscope, a tri-axial magnetometer, which are widely available in most smartphones, tablets, and other handheld devices. Through comparison tests of the IMU fixed on chest and foot, we found that the foot-mounted location is the proximate position, reflecting human's waking kinetic. The tri-axial accelerometer is used to measure step events and to estimate step stride. Due to magnetic disturbance indoor and gyroscope drift error, we propose a quaternion-based extended Kalman filter (EKF) for heading estimation by combining the magnetometer data and gyroscope data. Because of their complementary characteristics, a better accuracy in the heading solution is got.