Role of three-dimensionality in navigation behind bluff-body wakes*

Small-scale autonomous vehicles encounter unsteady flow conditions in rescue, surveillance, and sensing operations. Since they are equipped with finite actuation and sensory capabilities, trajectory planning in unsteady wakes is crucial. In this work, we study the role of three-dimensionality in navigation behind bluff body wakes. We use finite-horizon model-predictive control for trajectory planning in a three-dimensional cylinder wake at a Reynolds number of 300. We compare the 3D wake navigation performance with the 2D wake for several wake-crossing scenarios. Successful navigation is possible with a time horizon as low as one-tenth of the wake-shedding period. Further, the 3D wake navigation can be faster than the 2D wake navigation by taking advantage of the secondary vortices and lower spanwise coherence in 3D wakes. We find that the trajectory planners can leverage the secondary transverse vortices to effectively redirect toward the target. This study has the potential to develop sensor-friendly navigation strategies for autonomous vehicles in unsteady flows.