Planning inspection tasks for AUVs

Underwater installations require regular inspection and maintenance. In the EU FP7 project 288273, PANDORA, we seek to perform these tasks using autonomous underwater vehicles, achieving persistent autonomous behaviour in order to avoid the need for frequent human intervention. In this paper we consider one aspect of this problem, which is the construction of a suitable plan for a single inspection tour. Although we assume that some knowledge of the environment is given (a blueprint of the expected structures), we consider the problems of adapting to discoveries of modified structures (which could be a consequence of unrecorded engineering work or of shifting environmental conditions) and both planning and replanning suitable inspection missions. We demonstrate that these missions can be performed in a simulation and that performance is suitable for deployment in the proposed setting.