Learning multiple strategies to perform a valve turning with underwater currents using an I-AUV

Recent efforts in the field of intervention-autonomous underwater vehicles (I-AUVs) have started to show promising results in simple manipulation tasks. However, there is still a long way to go to reach the complexity of the tasks carried out by ROV pilots. This paper proposes an intervention framework based on parametric Learning by Demonstration (p-LbD) techniques in order to acquire multiple strategies to perform an autonomous intervention task adapted to different environment conditions. The manipulation skills of a pilot are acquired thought a set of demonstrations done under different environment circumstances, in our case different levels of water current. The proposed algorithm is able to learn these different strategies and depending on the estimated water current, autonomously reproduce a combined strategy to perform the task. The p-LbD algorithm as well as its interplay with the rest of the modules that take part in the proposed framework are described in this paper. We also present results on a free-floating valve turning task, using the Girona 500 I-AUV equipped with a manipulator and a customized end-effector. The obtained results show the feasibility of the p-LbD algorithm to perform autonomous intervention tasks combining the learned strategies depending on the environment conditions.