Search papers, labs, and topics across Lattice.
This paper introduces a new geodetically-referenced dataset and evaluation methodology for RTK-SLAM systems that addresses the limitations of SE(3) alignment in standard ATE evaluation, which masks global drift. The dataset features RTK data as a system input and independently-derived ground truth from a geodetic total station, unlike existing datasets that use GNSS as part of the ground truth. Experiments using LiDAR-inertial, visual-inertial, and LiDAR-visual-inertial RTK-SLAM systems demonstrate that SE(3) alignment can underestimate absolute positioning error by up to 76%, while RTK-SLAM achieves centimeter-level accuracy in open-sky and decimeter-level accuracy indoors.
SE(3) alignment hides up to 76% of the absolute positioning error in RTK-SLAM, and this new dataset exposes the problem.
RTK-SLAM systems integrate simultaneous localization and mapping (SLAM) with real-time kinematic (RTK) GNSS positioning, promising both relative consistency and globally referenced coordinates for efficient georeferenced surveying. A critical and underappreciated issue is that the standard evaluation metric, Absolute Trajectory Error (ATE), first fits an optimal rigid-body transformation between the estimated trajectory and reference before computing errors. This so-called SE(3) alignment absorbs global drift and systematic errors, making trajectories appear more accurate than they are in practice, and is unsuitable for evaluating the global accuracy of RTK-SLAM. We present a geodetically referenced dataset and evaluation methodology that expose this gap. A key design principle is that the RTK receiver is used solely as a system input, while ground truth is established independently via a geodetic total station. This separation is absent from all existing datasets, where GNSS typically serves as (part of) the ground truth. The dataset is collected with a handheld RTK-SLAM device, comprising two scenes. We evaluate LiDAR-inertial, visual-inertial, and LiDAR-visual-inertial RTK-SLAM systems alongside standalone RTK, reporting direct global accuracy and SE(3)-aligned relative accuracy to make the gap explicit. Results show that SE(3) alignment can underestimate absolute positioning error by up to 76\%. RTK-SLAM achieves centimeter-level absolute accuracy in open-sky conditions and maintains decimeter-level global accuracy indoors, where standalone RTK degrades to tens of meters. The dataset, calibration files, and evaluation scripts are publicly available at https://rtk-slam-dataset.github.io/.
