Research on the application of a UKF-based UWB/IMU fusion localization algorithm in mobile robots

Abstract

Aiming at the problem of decreased positioning accuracy of mobile robots due to occlusion in industrial environments, this paper proposes an ultra-wideband (UWB) and inertial measurement unit (IMU) data fusion method based on untraceable Kalman filter (UKF). The robot position information is obtained through the ranging localization of UWB and the trajectory projection of IMU respectively, and the multi-source information of UWB and IMU is fused in real time based on the UKF algorithm, and the final output is the high-precision localization result. The experimental results show that in the region where the NLOS (NLOS) path loss exceeds the threshold, the fluctuation range of the individual UWB localization error is up to 0.7m, while the localization error of the fusion algorithm can be stabilized within 0.15m, and the localization error can be suppressed to the centimeter level by the fusion algorithm, so as to satisfy the requirements of the high-precision operation in the dynamic industrial environment.