Distributed Flow Estimation for Autonomous Underwater Robots Using Proper Orthogonal Decomposition-Based Model Reduction

Abstract

Flow estimation plays an important role in the control and navigation of autonomous underwater robots. This paper presents a novel flow estimation approach that assimilates distributed pressure measurements through coalescing recursive Bayesian estimation and flow model reduction using proper orthogonal decomposition (POD). The proposed flow estimation approach does not rely on any analytical flow model and is thus applicable to many and various complicated flow fields for arbitrarily shaped underwater robots, while most of the existing flow estimation methods apply only to those well-structured flow fields with simple robot geometry. This paper also analyzes and discusses the flow estimation design in terms of reduced-order model accuracy, relationship with conventional flow parameters, and distributed senor placement. To demonstrate the effectiveness of the proposed distributed flow estimation approach, two simulation studies, one with a circular-shaped robot and one with a Joukowski-foil-shaped robot, are presented. The application of flow estimation in closed-loop angle-of-attack regulation is also investigated through simulation.

Laplacian regularized motion tomography for underwater vehicle flow mapping with sporadic localization measurements

“…A sensor array that measures pressure gradients and local flow velocities has been used in [6]. Orthogonal decomposition technique has been applied in [7] to compute flow model parameters, with the help from dynamic mode decomposition and the Koopman operator representation [8,9]. [10,11] develops a data assimilation scheme that adaptively constructs the basis functions from the Eulerian flow field prediction model.…”

Laplacian Regularized Motion Tomography for Underwater Vehicle Flow Mapping with Sporadic Localization Measurements

“…A sensor array that measures pressure gradients and local flow velocities has been used in [6]. Orthogonal decomposition technique has been applied in [7] to compute flow model parameters, with the help from dynamic mode decomposition and the Koopman operator representation [8,9]. [10,11] develops a data assimilation scheme that adaptively constructs the basis functions from the Eulerian flow field prediction model.…”