EWHT-AIB : enhanced waist-mounted human tracking framework based on array IMU and barometer

Abstract

With the development of the Internet of Things (IoT) and artificial intelligence (AI), indoor location-based services have become an indispensable part of public daily life. The performance of 3-D indoor positioning is constrained by the low performance of consumer-grade micro-electromechanical systems (MEMS) inertial measurement unit (IMU), the lack of effective calibration for the barometer, and the poor adaptability to complex human motion modes. To address the above challenges, this article proposes an enhanced waist-mounted human tracking framework based on array IMU and barometer (EWHT-AIB) that combines array IMU data fusion, precise barometer calibration, and a motion-constrained position-attitude update algorithm to achieve robust and accurate indoor positioning. To enhance array IMU data fusion performance, a weighted data fusion algorithm for array IMU based on the bias instability coefficients is proposed to achieve effective weighted fusion of array IMU data. Subsequently, a barometer calibration algorithm based on nonlinear fitting is proposed to achieve accurate compensation for bias error and scale factor error of the barometer. Finally, a position-attitude update algorithm under motion constraints is designed to achieve accurate pedestrian 3-D indoor positioning using compensated array IMU and barometer data. Comprehensive experiments demonstrate that the proposed EWHT-AIB framework can achieve meter level positioning accuracy under typical indoor environments.