Autonomous landing of underwater vehicles using high-resolution bathymetry

Abstract

The ability to land on the seafloor expands the envelope of tasks that underwater vehicles can carry out during survey and inspection. However, even though remotely operated vehicles routinely land during their operations, autonomous underwater vehicles (AUVs) lack the sensing and data processing capabilities needed to identify safe, stable landing sites. Here, an algorithm is developed that uses mm resolution bathymetry to detect regions where an AUV of known geometry can safely and stably land on the seafloor. The algorithm uses physical models that consider vehicle geometry, seafloor slope, roughness, friction and currents. It can identify the most suitable of multiple candidate sites based on a landing cost function. The performance of the algorithm is evaluated using seafloor bathymetry data that was obtained using an AUV equipped with a high resolution laser mapping system on the slopes of the Takuyo Daigo seamount in the Northwest Pacific. The algorithm successfully identified multiple landing sites along a 500 m transect on the slopes of the surveyed seamount. The study demonstrates that safe, reliable AUV landing operation is feasible in actual seafloor environments.

More information

Identifiers

ORCID for Adrian Bodenmann: orcid.org/0000-0002-3195-0602

Catalogue record

Export record