Structured sparse methods for active ocean observation systems with communication constraints

Actuated sensor networks enabled by underwater acoustic communications can be efficiently used to sense over large marine expanses that are typically challenged by a paucity of resources (energy, communication bandwidth, number of sensor nodes). Many marine phenomena of interest admit sparse representations, which, coupled with actuation and cooperation, can compensate for being data starved. Herein, new methods of field reconstruction, target tracking, and exploration-exploitation are provided, which adopt sparse approximation, compressed sensing, and matrix completion algorithms. The needed underlying structure (sparsity/low-rank) is quite general. The unique constraints posed by underwater acoustic communications and vehicle kinematics are explicitly considered. Results show that solutions can be practically implemented, even over large ocean spaces.

Introduction

Future advances in ocean monitoring, offshore industry, and basic marine sciences will rely heavily on our ability to jointly consider communication, actuation, and sensing in a unified system that includes remote instruments, underwater vehicles, human operators, and sensors of all types. These tasks will require methods to detect and track large-scale ocean phenomena such as algal blooms, oil spills, ocean currents, and hydrothermal vents, as well as man-made signals such as those emanating from an airplane's black box. We envision a scenario as depicted in Fig. 1, where multiple autonomous underwater vehicles (AUVs) interact and coordinate via acoustic communications with a network of sensors to detect and track a phenomenon of interest. The underlying network architecture includes both static, communication-enabled sensor nodes, as well as actuated nodes in the form of AUVs. Thus, our system needs to control and move some of the nodes to achieve its sensing and communication goals. Moreover, the choices made regarding communication, control, and sensing are interdependent [1]. Figure 1.

The envisioned network architecture for future ocean observing systems, comprising both fixed and moving nodes capable of advanced sensing and wireless communications.