Biomimetic sensing of incident flow of underwater vehicles at small to large angles of attack using a physics-based estimation model

Sensing of the incident flow plays a vital role in precise maneuvering and control of underwater vehicles. Inspired by the lateral line of fish, this study develops a biomimetic flow sensing approach for identifying incident flow speed and angle of attack (AoA). A model-based estimation framework is established by integrating the boundary element method with the unscented Kalman filter. The model uses the distributed pressure data to estimate the incident flow speed and AoA. Computational fluid dynamics (CFD) and potential flow theory are employed to investigate the characteristics of flow around the vehicle; the pressure-sensing points that are sensitive to the flow are selected to install pressure sensors. CFD simulation and towing tank experiment demonstrate that the proposed model can estimate the AoA and incident flow speed accurately across a wide range of angles of attack. The biomimetic flow sensing model has wide application values in the hydrodynamic estimation and motion control of underwater vehicles in complex flow conditions.