Automated Drag Optimization of a Torpedo Shaped AUV Using B-spline and Genetic Algorithm

Torpedo-shaped autonomous underwater vehicles (AUV) offer versatile capabilities such as surveying underwater environment, collecting data, performing inspections and even assisting in search and rescue operation. However, their design still attract the attention of manufacture due to the various parameters that affect the final shape. This article aims to optimize the shape of a torpedo-shaped (AUV) through the utilization of B-splines for shape representation and the application of genetic algorithms for form optimization, in which the main objective is to minimize the drag. The automated optimization procedure uses minimum control points to form the nose and the tail of the AUV whereas the cylindrical part remain unchanged. To evaluate the performance, the computational fluid dynamics (CFD) model as high fidelity model is used and it is calibrated for a DREA type UV. The overall strategy presents a drag reduction about 19 %.