IMU/UWB Sensor Fusion Using Moving Average Filter for Indoor Positioning of Mobile Robot

Abstract

Ultra-wideband(UWB) is a short-range radio technology that can be used for accurate indoor positioning of mobile robots. Although UWB provides superior localization performance compared with conventional wireless-communication-based localization methods, there are still problems with applying UWB to the localization of mobile robots. In this paper, we propose an IMU/UWB sensor fusion algorithm that uses a moving average filter to improve the positioning accuracy of UWB-based localization. The proposed method utilizes the law of large numbers, which states that the average of the samples should be close to the expected value as the size of the samples increases. We produce virtual samples of a current position using previous UWB and IMU data and find the optimal position estimation by averaging the value fo the virtual samples. We also build a simulation environment to evaluate the performance of the proposed method using MATLAB. The simulation results show that the proposed method improves position accuracy by up to 23% compared with a previous method that uses the Kalman Filter.