This work describes the development of a computational model created to autonomously navigate a fishlike underwater vehicle with a multivertebra spine and a flexible tail. First, a two-dimensional numerical simulation, developed previously, and based on an unsteady panel method, was coupled with the vehicle, spine, and tail dynamics. Numerous test cases spanning a wide range of swimming conditions were processed with the simulation, and these data were subsequently used to train an artificial neural network. The navigation procedureeural network uses as input a set of vehicle geometric and state variables, and from that, it predicts the needed vehicle body deformations, i.e., necessary spine vertebrae rotations, so that the vehicle can navigate through a set of given way points. Validation results are presented that show the accuracy of the present flow-structure interaction solver, and several test cases of autonomous navigation are presented to show the capabilities of the current method.
展开▼
