Validation of AIS Position Accuracy Using Lidar

Abstract

Automatic Identification System (AIS) data are widely used for vessel traffic monitoring, navigation support, and maritime research, but AIS-reported positions do not necessarily represent the physical vessel hull accurately. The geometric consistency between AIS-derived vessel representations and independently observed hull contours remains insufficiently quantified. This study evaluates AIS-derived vessel geometry using a shore-based Light Detection and Ranging (Lidar) system as an independent geometric reference. A synchronized field campaign on the River Hunte, Germany, combines shore-based Lidar, AIS, and multiple GNSS receivers. The Lidar system is calibrated using RTK-GNSS reference targets. AIS vessel geometries are reconstructed from reported positions, headings, antenna-reference points, and static vessel dimensions and are matched with Lidar-derived hull contours. Across 69 vessel passages, the lateral deviation has a median of 0.62 m and a 95th percentile absolute deviation of 4.68 m, while the longitudinal deviation has a median of −1.81 m and a 95th percentile absolute deviation of 12.41 m. Receiver-side AIS timing differences contribute less than 0.5 m under the investigated conditions and therefore do not explain the dominant deviations. The results indicate that the observed discrepancies are predominantly longitudinal and are influenced by dynamic positioning effects and inconsistencies in the geometric representation of vessels in AIS, including reference-point definitions and static vessel data. Two rare cases with deviations of approximately 50 m are identified as vessel-specific anomalies associated with erroneous static AIS information. The study demonstrates that calibrated shore-based Lidar can provide an independent geometric reference for assessing the spatial consistency of AIS-derived vessel representations in inland waterways.