Seabed-to-Sky Mapping of Maritime Environments with a Dual Orthogonal SONAR and LiDAR Sensor Suite

This paper presents a GNSS-independent mapping framework that generates unified seabed-to-sky maps using LiDAR-IMU odometry and a dual orthogonal Forward-Looking Sonar (FLS) configuration. Localization is provided by the LiDAR-IMU subsystem, while acoustic measurements enrich the map with underwater structure. We extend orthogonal wide-aperture sonar fusion to support arbitrary inter-sonar translations, enabling the use of heterogeneous, non-co-located sensors. In addition, leading-edge extraction is applied to each FLS image to generate line-scan measurements that complement stereo-derived 3D sonar points. To integrate acoustic and LiDAR data, LIO-SAM is modified to incorporate both sonar point clouds and line scans at and between keyframes through motion-interpolated poses, allowing asynchronous sonar observations to contribute continuously to the map. The framework is validated on real-world data collected in Copenhagen harbour, demonstrating real-time operation with map and odometry update rates of approximately 2.65 Hz and 2.85 Hz, respectively. Experimental results show that the proposed system produces consistent, unified 3D representations spanning both underwater and above-surface environments, enabling robust maritime mapping in GNSS-denied conditions.