Uncertainty management during hybrid autonomous underwater vehicle missions

Abstract

Effective autonomous underwater vehicle (AUV) mission uncertainty management requires identifying optimal mission profile based on multiple uncertain attributes. The new deep underwater glider being developed by the BRIDGES consortium will enable the use of low power AUV to sample three different phenomena: living things, sea mining and oil and gas exploration. In order to have the versatility to address these three requirements, the underwater glider will operate in a hybrid mode, using a buoyancy change and propeller based propulsion system. When we developed a failure model for this AUV we realized that different risk profiles are obtained if we are using different type of propulsion system. Therefore, we propose a multiple attribute utility technique to inform the optimal mission profile in light of mission risks and opportunities.