Indoor Positioning Based on the Integration of IMU and UWB

Accurate indoor location information is of utmost importance for the navigation of modern indoor mobile robots. However, achieving accurate positioning solely with ultra-wideband (UWB) technology in indoor environments is challenging due to the susceptibility of UWB measurements to non-line-of-sight (NLOS) effects. To address this issue, this paper proposes an indoor positioning system that integrates an inertial measurement unit (IMU) with UWB. Firstly, an analysis of the errors present in UWB measurement data leads to the development of a two-step calibration method based on least squares (LS) to estimate error parameters. Subsequently, an Extended Kalman Filter (EKF)-based integration localization algorithm for IMU and UWB is proposed to mitigate the impact of NLOS. Through comprehensive simulations, the suggested algorithm's performance is assessed and contrasted with the weighted least squares (WLS) approach. Experimental findings show that the two-step LS-based calibration method estimates error parameters with accuracy. Moreover, the EKF-based IMU and UWB integration algorithm achieves approximately half improvement in localization accuracy compared to WLS, it also demonstrates excellent robustness. Consequently, the suggested indoor positioning system, which integrates IMU and UWB, satisfies the requirements for accuracy for indoor mobile robot navigation.