Abstract—Turbulent flows are often treated as a noisy environment by control algorithms of underwater robots. However, aquatic animals such as fish have learned to take advantage of certain unsteady flow. Periodic complex flow, such as that found in the wake of cylinders has been shown to offer energy saving opportunities to fish. We built a fish-like robot with an integrated pressure sensor array housed in the head. The robot can control its tail beat synchronization with respect to the periodic oscillations in the flow behind a cylinder. We show that vortices, represented here by pressure maxima, can be detected and exploited to increase the swimming efficiency of the robot fish while it remains rigidly mounted to a force plate. Force measurements show an efficiency gain of 23% when the tail beat of the robotic fish is synchronized at a particular phase lag. I. INT RODUCT ION The swimming efficiencies of fish and sea mammals are investigated by many researchers in various disciplines as understanding the principles of their control and locomotion is interesting both for biologists and engineers. Fin locomotion is mostly investigated in steady flows. But as in a real world, flow is seldom steady, it is therefore interesting to investigate how fish negotiate turbulence to minimize their energy consumption. Studies show that fish prefer swimming in predictably oscillating flows [1-4], like the Kármán vortex street generated in the wake of a cylinder at particular flow speeds. In the Kármán vortex street , fish can adapt their tail beat to the vortex shedding frequency and slalom between vortices. The tail beat frequency of fish drops considerably compared to swimming in a steady flow [5], also the muscle activity of fish swimming in periodic turbulence is much lower than it would be suggested by the reduced flow behind a bluff body [6]. This suggests that fish can potentially utilize energy from the flow. Energy harvesting is an interesting topic for underwater robotics because it permits the design and build of vehicles that can survive longer missions. The energetics of finned propulsion has been previously investigated in periodic turbulence and it was found that the efficiency depends on Jaas Ježov and Maarja Kruusmaa are with the Centre for Biorobotics, Tallinn University of Technology, Tallinn, Estonia. {jaas.jezov, maarja.kruusmaa}@ttu.ee Otar Akanyeti is with the Department of Computer Science, University of Verona, Verona, Italy. otarakanyeti@yahoo.com Lily D. Chambers is with the Department of Mechanical Engineering, University of Bath, Bath BA2 7AY, UK. L.D.Chambers@bath.ac.uk the timing of the tail beat. Gopalkrishnan et al. [7] found that the efficiency of an actuated hydrofoil in a controlled turbulence depends on the phase of the hydrofoil interaction with the vortices shed from behind a cylinder. They showed that there are 3 different interaction modes that leave visually distinctive vortex patterns in the wake of the hydrofoil. The mode of interaction depends on the phase of the foil flapping with respect to the vortex location. Triantafyllou et al. [8] highlights that the most thrust is produced with a destructive vortex merging mode. In this mode a vortex produced by the hydrofoil is merging with a vortex from the cylinder rotating in the opposite direction. These vortices destroy each other producing a pattern of weaker vortices in the wake of the hydrofoil. This mode is less efficient than a vortex pairing mode as it requires more power. In the vortex pairing mode a pattern of mushroom like double vortices is produced. The most inefficient is a constructive vortex merging mode where vortices from the cylinder and the hydrofoil are rotating in the same direction at the time of merging. In the vortex pairing mode hydrofoils and fishlike three- dimensional bodies can be actuated even with efficiency over 100% [7-8]. In their studies the controlled turbulence was achieved by oscillating the cylinder laterally in the flow. The angle of the hydrofoil was changed periodically while the pivot point of the hydrofoil was oscillated laterally. The phase difference was measured between the oscillations of the cylinder and the hydrofoil. In the above mentioned studies the body interaction with the turbulence was controlled experimentally, which, from robotics applications point of view, is not a realistic precondition. In robotics, the control problem is the reverse, instead of controlling the turbulence, the tail beat of the fin should be controlled to gain better efficiency. The hydrofoils in those experiments were also moving in a lateral direction, which is similar to the biological fish kinematics in the wake of a cylinder [5]. Fish-like robots are usually not directly controlled in the lateral direction. Since we do not have the control over the hydrodynamic environment and the lateral translation, two questions can be asked when investigating swimming efficiency in the Sensing oscillations in unsteady flow for better robotic swimming efficiency Jaas Ježov, Otar Akanyeti, Lily D. Chambers, Maarja Kruusmaa