RGB-D Camera-based Navigation for Autonomous Underwater Inspection using Low-cost Micro AUVs

Abstract

Commercial RGB-D cameras enable even small scale mobile robots with sophisticated and cost-efficient 3D-scanning capabilities. We transfer this sensory concept to the domain of underwater structure inspections. Hereby, we address the problem of autonomous robot navigation at close distance to the inspected structure. Therefore, we propose a depth-camera based lightweight online planning and navigation methodology which enables micro autonomous underwater vehicles (μAUVs) to autonomously inspect structures with unknown geometry. We demonstrate the effectiveness of our approach in simulation and experiments and discuss the results in the context of prior work. All software and hardware designs used in this work will be made available online.