Session: 04-06-01 Underwater Vehicles & Subsea Communications I

Submission Number: 180274

Development and Installation of the Kilo Nalu Coastal Robotic Testing Range 

Marine operations are increasingly performed using uncrewed underwater vehicles (UUVs). UUVs are used for ship hull inspections, laying oil pipelines and telecommunication cables and other seafloor construction operations, environmental measurement and sampling, coastal surveillance, search and rescue missions, and bathymetric mapping. Sea mine countermeasure operations are performed with a combination of towed sonar arrays and piloted UUVs. In general, UUVs offer several advantages to submersibles and other manned vehicles, including lower costs, reduced risk to human operators, and smaller vehicle size.

However, there are several unique challenges to operating UUVs that prevents a more ubiquitous robotic presence in the ocean. Communications are severely limited, as radio signals attenuate rapidly underwater, acoustic communication is very low bandwidth, and optical communication is either very short range or requires often unattainable alignment accuracy. The deeper regions of the ocean have little to no natural light, and are subject to crushing pressures. On top of this, ocean environments are cluttered with water currents and fluctuations which impart significant forces onto marine. This last issue makes operating marine robots especially challenging in coastal and littoral regions where wave oscillations create very chaotic flows in shallow depths with abundance of obstacles.

Autonomous marine robots and their control algorithms are often developed in a controlled laboratory testing tank, such that background flow is either stationary or a uniform velocity. This is useful for initial development, but it means there is a significant gap between the conditions used for development and the challenges that exist in the field. This paper describes the development of a robotic testing range in a natural coastal environment off the south shore of O`ahu that will provide communication and real-time positioning data to vehicles. By eliminating navigation error/uncertainty, platforms, subsystems, and control algorithms can be developed and optimized for realistic coastal flows independent of that significant complication. In later stages of development, once a given vehicle platform has been optimized, the positioning data can be used as an independent measurement to validate and optimize navigation sensing and mapping algorithms. Allowing advancement of underwater robot technologies from academic research to robust ocean-ready platforms.

Positioning and communication equipment was integrated into existing seafloor infrastructure, the Kilo Nalu Observatory (KNO), creating a robotic testing facility in a natural coastal environment with acoustic communication and high-accuracy, real-time tracking capabilities. A custom-built acoustic long-baseline (LBL) array was added to provide this simultaneous localization and communication. The LBL array consists of 6 nodes, each of which is comprised of an S2C M 18/34 acoustic mini-modem mounted to an aluminum tripod, providing a total measurement volume of 100m x 100m x 12m.

A custom Science Instrument Interface Module (SIIM) was developed to interface the LBL system with the existing KNO infrastructure. Localization/triangulation requires that messages sent back and forth between vehicles and the LBL are timestamped with a consistent timebase to calculate message time of flight (TOF), and thus distance. This timing synchronization is accomplished using the Global Positioning System (GPS) pulse-per-second (PPS) timing signal. The PPS signal coming from the KNO primary node is at TTL levels consistent with standard GPS receiver outputs. The SIIM regenerates this signal on 6 separate lines, but also converts from TTL to RS-232 voltage levels to improve signal integrity as it is transmitted across 50m cables to get to the LBL nodes positioned radially around the testing range. Seafloor sensors are integrated to input live water sound velocity measurement into the system. The paper describes a procedure used to synchronize the timebase between UUV and LBL modems to within 1 microsecond, as well as a procedure to georeference the nodes on the seafloor to extremely accurate geodetic positions. After georeferencing, the acoustic long-baseline (LBL) array positions are validated by independent acoustic TOF measurements showing strong agreement with an average error of 0.14%.

Presenting Author: Michael Krieg University of Hawaii

Presenting Author Biography: Dr. Krieg received his BS and PhD degrees in Aerospace Engineering Sciences from the University of Colorado at Boulder in 2006 and 2012, respectively. His research interests include marine robotics, bioinspired propulsion and sensory systems for uncrewed vehicles, automation and control theory, soft robotics, and resident underwater vehicles. He is currently an Associate Professor in the Ocean and Resources Engineering Department within the School of Ocean and Earth Science and Technology at the University of Hawai'i at Manoa.

Authors:

Michael Krieg University of Hawaii