UWB and IMU Fusion Localization System Based on Unmanned Vehicles in Complex Factory Areas

In the face of serious satellite signal occlusion in the factory environment, using GPS and IMU combination for positioning still results in positioning jumps of about 1 meter, which cannot meet the positioning requirements of heavy trucks. Moreover, there is a large amount of metal interference in the factory area, which seriously affects the positioning effect of UWB. To address this problem, this paper proposes an optimized IMU and UWB fusion positioning method. Based on the regional roaming algorithm, the current reliable UWB measurement source is searched, and a pruning mean loss function based on TOA is designed to process the UWB pseudo-range measurement values. The inertial measurement unit (IMU) is introduced, and on the basis of IMU error state update in VINS, residual factors of UWB are added. Through optimization, UWB and IMU are loosely coupled for fusion positioning. Multiple experiments have demonstrated that this method offers a reliable positioning guarantee for heavy trucks operating within the challenging factory environment.