Sequential Action Control for models of underactuated underwater vehicles in a planar ideal fluid

This paper uses Sequential Action Control (SAC), a model-based method for control of non-linear systems, for fast, optimal trajectory-tracking tasks in the presence of fluid drift. Through the benchmark example of the kinematic car, it is shown that SAC outperforms a traditional offline projection - based optimization technique in terms of control effort and objective cost. Motivated by recent work on effort-efficient, sight-independent weakly electric fish, this papers also shows that SAC successfully provides control-optimal dynamics to perform short-range underwater maneuvers. Simulation results highlight SAC's robustness to different drift intensities and added mass properties, even when the effective fluid drift is not included in the controller's model.