Heading Angle Correction With Building Map Model Constraint for Single IMU-Based Pedestrian Wearable Localization

The microelectromechanical system inertial measurement units (MIMU)-based foot wearable inertial localization technique has become a competitive choice for indoor pedestrian localization due to the features of low cost, miniature size, independency, and environmental adaptability. The drift of gyroscope data and the integral calculation will induce cumulative error for attitude angles, resulting in heading angle and velocity errors that seriously affects the localization accuracy. This investigation proposes a heading angle correction method with building map model constraint to suppress the cumulative error in continuous motions. The key contributions include the following. A proportion–integration model and zero angular rate update integrated algorithm to correct the gyroscope data that are integrated to extended Kalman filter (EKF) for heading angle correction. Introduction of the building map model as a heading angle constraint to integrate with the motion parameters for motion trajectory correction especially for large indoor scenes. Design of a foot-wearable device and experimental studies to demonstrate the feasibility and effectiveness of the proposed methods. The presented techniques provide a good reference for improving the accuracy of MIMU-based low-cost and miniaturized wearable devices for pedestrian localization.