UWB/IMU Integrated Positioning Technology With NLOS Mitigation: An Improved LSSVM and AEKF Fusion Strategy

Ultra-Wideband (UWB) and Inertial Measurement Unit (IMU) integration is a key solution for enhancing positioning accuracy and stability of Automated Guided Vehicles (AGVs) in warehouse environments. However, UWB positioning suffers from degraded accuracy due to environmental noise and Non- Line-of-Sight (NLOS) effects, while standalone IMU positioning accumulates errors over time. To address these issues, this paper proposes a novel UWB/IMU integrated positioning method for AGVs. In this method, a UWB/IMU navigation board is installed on the AGV. A robust adaptive Kalman filter (RAKF) is employed to construct a robust matrix, and the Sage-Husa adaptive factor is introduced to preprocess UWB ranging values and compute the tag position. Subsequently, the Least Squares Support Vector Machine (LSSVM) model is trained using the AGV’s preset simulation positions. The input for the training set is the UWB ranging values based on the simulation positions, and the output is the corresponding real coordinates. The Whale Optimization Algorithm (WOA) is employed to automatically optimize the parameters of the LSSVM. The noise-containing UWB positioning results processed by the RAKF are used as input features for the model to compensate for systematic errors in the RAKF solution, and the model outputs the corrected coordinates. Finally, through an innovation-based adaptive extended Kalman filter, the corrected position information is fused with Inertial Measurement Unit (IMU) data to obtain more accurate positioning results, achieving high-precision localization for warehouse AGVs. Experimental results in both Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) environments demonstrate that this method significantly reduces positioning errors. Particularly in NLOS scenarios, both the root mean square error and the maximum error are notably reduced compared to traditional algorithms. This study effectively enhances the robustness and accuracy of the AGV positioning system, making it suitable for high-precision indoor warehouse applications.