EULF-IUB: An Enhanced UGV Localization Framework Using Tightly Coupled IMU/UWB/Barometer Integration

Accurate 3D localization in Global Navigation Satellite System (GNSS)-denied environments is a critical prerequisite for the autonomous navigation of unmanned ground vehicles (UGVs). However, robust localization remains severely challenged by ultra-wideband (UWB) ranging degradation under multipath and blockage, accumulated inertial errors, and barometer altitude drift. To address these issues, this paper proposes EULF-IUB, an enhanced fusion framework that integrates robust UWB range smoothing, constraint- and geometry-enhanced tightly coupled IMU/UWB localization, and barometric-field-enhanced height correction. Specifically, a Huber-IRLS-based regression smoothing strategy is employed to suppress outliers and abrupt errors in UWB ranging. The smoothed ranges are then fused with inertial propagation in an extended Kalman filter (EKF), where motion constraints and geometry-aided strategies are introduced to mitigate horizontal bias and trajectory jumps. Furthermore, a spatial barometric height field is constructed through anchor-referenced cross-calibration to compensate for altitude drift. Experimental results in outdoor and indoor GNSS-denied scenarios show that EULF-IUB achieves mean 3D positioning errors of 0.20 m and 0.21 m, respectively, reducing the 3D positioning error by 52.3%–61.5% compared with representative methods.