Cooperative Estimation of a Three-Dimensional Flow Field Using Multiple AUVs and Local Measurements

Abstract

This brief proposes a three-dimensional flow field estimation algorithm using multiple autonomous underwater vehicles (AUVs) and local measurements. Taking into account the fact of negligible vertical velocity, the layered flow field model is established in a three-dimensional scene, which incorporates the property of incompressibility and increases the span of model parameters. The flow field affects the movement of each vehicle and it may result in motion-integration errors when comparing its predicted movement. Meanwhile, local measurements of the unknown flow field are stored as time-dependent variables due to the lack of position while vehicles are submerged. Then, we formulate the nonlinear constraints based on the motion-integration errors and local measurements. The cooperative algorithm is proposed to solve the flow field estimation problem composed of these constraints and the effectiveness of the proposed estimation algorithm is validated with simulations.