Abstract

Drag and added inertia coefficients for an underwater manipulator of revolute joints are time and shape dependent. This study proposes an extension of the method to determine transient hydrodynamic coefficients, which was originally developed for a single degree-of-freedom (SDOF) underwater manipulator of a cylindrical arm, for the arm of a square cross-section. This method is based on the normalization of coefficients using the dimensionless distance traveled by the centroid of water displaced by the arm. This results in simple approximation functions independent of kinematic conditions and dependent only on the dimensionless distance traveled. Surrogate model creation via machine learning methods requires many data sets obtained as a result of experiments or numerical simulations. The proposed normalization simplifies and speeds up the process of creating metamodels for drag and added inertia torques for SDOF manipulators because only a few time histories of the hydrodynamic load are required for regression. The results of this study indicate that the method based on such a normalization can be applied to a SDOF underwater robotic arm of a square cross-section with good accuracy, thus widening the scope of the original method.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call