Key points are not available for this paper at this time.
The increasing demand for indoor positioning information has led to a growing emphasis on indoor localization. Non-Line-of-Sight (NLOS) conditions diminish the accuracy of Ultra-Wide Band (UWB) system positioning, while over time, Inertial Navigation Systems (INS) suffer from accumulating positioning errors. To address these issues, this paper proposes a method that combines UWB and INS sensors. Compared to individual system positioning methods, this approach effectively enhances localization precision, leveraging the complementary strengths of both systems. The paper utilizes Extended Kalman Filtering (EKF) to fuse residual positioning information, and the obtained residual position results are processed using the Multiple Fading Factor Square Root Kalman Filter technique (MSTSCKF). Moreover, during temporal asynchrony, it updates INS positioning and yaw angle information using EKF output for subsequent INS positioning until the next data correction. To further mitigate NLOS effects, a k-means preprocessing method is applied to UWB data. Root Mean Square Error (RMSE) is used as an evaluation metric. Simulation and experimental results demonstrate that the proposed method effectively accounts for NLOS error influences, thereby enhancing navigation and positioning accuracy.
Wang et al. (Mon,) studied this question.