Search

To realize the potential of autonomous underwater robots that scale up our observational capacity in the ocean, new techniques are needed. Fleets of autonomous robots could be used to study complex marine systems and animals with either new imaging configurations or by tracking tagged animals to study their behavior. These activities can then inform and create new policies for community conservation. The role of animal connectivity via active movement of animals represents a major knowledge gap related to the distribution of deep ocean populations. Tracking underwater targets represents a major challenge for observing biological processes in situ, and methods to robustly respond to a changing environment during monitoring missions are needed. Analytical techniques for optimal sensor placement and path planning to locate underwater targets are not straightforward in such cases. The aim of this study was to investigate the use of reinforcement learning as a tool for range-only underwater target-tracking optimization, whose promising capabilities have been demonstrated in terrestrial scenarios. To evaluate its usefulness, a reinforcement learning method was implemented as a path planning system for an autonomous surface vehicle while tracking an underwater mobile target. A complete description of an open-source model, performance metrics in simulated environments, and evaluated algorithms based on more than 15 hours of at-sea field experiments are presented. These efforts demonstrate that deep reinforcement learning is a powerful approach that enhances the abilities of autonomous robots in the ocean and encourages the deployment of algorithms like these for monitoring marine biological systems in the future., Funding: The European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie, grant agreement No 893089 (IM). The Spanish Ministerio de Economia y Competitividad under the project SASES, grant agreement No RTI2018-095112-B-I00 (IM). The Spanish Ministerio de Economia y Competitividad under the project BITER-ECO, grant agreement No PID2020-114732RB-C31 (IM, JN). The Spanish Ministerio de Economia y Competitividad under the project BITER-AUV, grant agreement No PID2020-114732RB-C33 (NP)., Peer Reviewed, Postprint (author's final draft)

To realize the potential of autonomous underwater robots that scale up our observational capacity in the ocean, new techniques are needed. Fleets of autonomous robots could be used to study complex marine systems and animals with either new imaging configurations or by tracking tagged animals to study their behavior. These activities can then inform and create new policies for community conservation. The role of animal connectivity via active movement of animals represents a major knowledge gap related to the distribution of deep ocean populations. Tracking underwater targets represents a major challenge for observing biological processes in situ, and methods to robustly respond to a changing environment during monitoring missions are needed. Analytical techniques for optimal sensor placement and path planning to locate underwater targets are not straightforward in such cases. The aim of this study was to investigate the use of reinforcement learning as a tool for range-only underwater target-tracking optimization, whose promising capabilities have been demonstrated in terrestrial scenarios. To evaluate its usefulness, a reinforcement learning method was implemented as a path planning system for an autonomous surface vehicle while tracking an underwater mobile target. A complete description of an open-source model, performance metrics in simulated environments, and evaluated algorithms based on more than 15 hours of at-sea field experiments are presented. These efforts demonstrate that deep reinforcement learning is a powerful approach that enhances the abilities of autonomous robots in the ocean and encourages the deployment of algorithms like these for monitoring marine biological systems in the future

© The Author(s), 2022. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Zhang, Y., Yoder, N., Kieft, B., Kukulya, A., Hobson, B. W., Ryan, S., & Gawarkiewicz, G. G. Autonomous tracking of salinity-intrusion fronts by a long-range autonomous underwater vehicle. Ieee Journal of Oceanic Engineering, (2022): 1-9, https://doi.org/10.1109/JOE.2022.3146584., Shoreward intrusions of anomalously salty water along the continental shelf of the Middle Atlantic Bight are often observed in spring and summer. Exchange of heat, nutrients, and carbon across the salinity-intrusion front has a significant impact on the marine ecosystem and fisheries. In this article, we developed a method of using an autonomous underwater vehicle (AUV) to detect a salinity-intrusion front and track the front's movement. Autonomous front detection is based on the different vertical structures of salinity in the two distinct water types: the vertical difference of salinity is large in the intruding saltier water because of the salinity “tongue” at mid-depth, but is small in the nearshore fresher water due to absence of the salinity anomaly. Every time the AUV crosses and detects the front, the vehicle makes a turn at an oblique angle to cross the front, thus zigzagging through the front to map the frontal zone. The AUV's zigzags sweep back and forth to track the front as it moves over time. From June 25 to 30, 2021, a Tethys-class long-range AUV mapped and tracked a salinity-intrusion front on the southern New England shelf. The frontal tracking revealed the salinity intrusion's 3-D structure and temporal evolution with unprecedented detail., This work was supported in part by the National Science Foundation under Grant OCE-1851261 and in part by the David and Lucile Packard Foundation.

Our vast and dynamic ocean is largely undersampled and its inhabitants poorly understood. Imaging systems, supported by machine learning-based automated analysis, promise to help us establish a more consistent presence in this difficult to access habitat. But on-going challenges with ML-systems keep the dream of an ocean teeming with robots capable of autonomously reporting observations or tracking rare individual organisms just out of reach. Indeed, the very signals we wish to measure – rapidly fluctuating, spatially and temporally heterogeneous biological distributions – confound supervised machine learning systems that are encoded with a static view of the world. Bridging the gap between these imperfect systems and deployable instrumentation requires supervised autonomy: keeping a human-in-the-loop to ensure the machine is doing the right thing. Here we describe two supervised autonomy experiments that push the limits of allowing in situ platforms to make decisions base on imagery. In the first, we couple a multi-object detector and a 3D stereo tracker on a Remotely Operated Vehicle to find and track midwater organisms while a human observer confirms the chosen target. In the second, we demonstrate transmitting selected Regions of Interest, chosen by a real time binary detector, from an Autonomous Underwater Vehicle at depth via an acoustic modem. Together, these experiments represent steps toward allowing a computer to trigger a behaviour change based on visual data to track a salient feature. Both fundamentally rely on a human operator to confirm the machine’s choice. Enabling this type of constructive interaction between potentially inaccurate machines and human supervisors is critical to ensure that our in situ systems operate in an efficient and unbiased manner.

© The Author(s), 2020. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Zhang, Y., Kieft, B., Hobson, B. W., Ryan, J. P., Barone, B., Preston, C. M., Roman, B., Raanan, B., Marin,Roman,,III, O'Reilly, T. C., Rueda, C. A., Pargett, D., Yamahara, K. M., Poulos, S., Romano, A., Foreman, G., Ramm, H., Wilson, S. T., DeLong, E. F., Karl, D. M., Birch, J. M., Bellingham, J. G., & Scholin, C. A. Autonomous tracking and sampling of the deep chlorophyll maximum layer in an open-ocean eddy by a long-range autonomous underwater vehicle. IEEE Journal of Oceanic Engineering, 45(4), (2020): 1308-1321, doi:10.1109/JOE.2019.2920217., Phytoplankton communities residing in the open ocean, the largest habitat on Earth, play a key role in global primary production. Through their influence on nutrient supply to the euphotic zone, open-ocean eddies impact the magnitude of primary production and its spatial and temporal distributions. It is important to gain a deeper understanding of the microbial ecology of marine ecosystems under the influence of eddy physics with the aid of advanced technologies. In March and April 2018, we deployed autonomous underwater and surface vehicles in a cyclonic eddy in the North Pacific Subtropical Gyre to investigate the variability of the microbial community in the deep chlorophyll maximum (DCM) layer. One long-range autonomous underwater vehicle (LRAUV) carrying a third-generation Environmental Sample Processor (3G-ESP) autonomously tracked and sampled the DCM layer for four days without surfacing. The sampling LRAUV's vertical position in the DCM layer was maintained by locking onto the isotherm corresponding to the chlorophyll peak. The vehicle ran on tight circles while drifting with the eddy current. This mode of operation enabled a quasi-Lagrangian time series focused on sampling the temporal variation of the DCM population. A companion LRAUV surveyed a cylindrical volume around the sampling LRAUV to monitor spatial and temporal variation in contextual water column properties. The simultaneous sampling and mapping enabled observation of DCM microbial community in its natural frame of reference., 10.13039/501100008982 - National Science Foundation 10.13039/100000936 - Gordon and Betty Moore Foundation 10.13039/100000008 - David and Lucile Packard Foundation 10.13039/100016377 - Schmidt Ocean Institute 10.13039/100000893 - Simons Foundation

Knowing the displacement capacity and mobility patterns of industrially exploited (i.e., fished) marine resources is key to establishing effective conservation management strategies in human-impacted marine ecosystems. Acquiring accurate behavioral information of deep-sea fished ecosystems is necessary to establish the sizes of marine protected areas within the framework of large international societal programs (e.g., European Community H2020, as part of the Blue Growth economic strategy). However, such information is currently scarce, and high-frequency and prolonged data collection is rarely available. Here, we report the implementation of autonomous underwater vehicles and remotely operated vehicles as an aid for acoustic long-baseline localization systems for autonomous tracking of Norway lobster (Nephrops norvegicus), one of the key living resources exploited in European waters. In combination with seafloor moored acoustic receivers, we detected and tracked the movements of 33 tagged lobsters at 400-m depth for more than 3 months. We also identified the best procedures to localize both the acoustic receivers and the tagged lobsters, based on algorithms designed for off-the-shelf acoustic tags identification. Autonomous mobile platforms that deliver data on animal behavior beyond traditional fixed platform capabilities represent an advance for prolonged, in situ monitoring of deep-sea benthic animal behavior at meter spatial scales

© The Author(s), 2019. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Zhang, Y., Ryan, J. P., Kieft, B., Hobson, B. W., McEwen, R. S., Godin, M. A., Harvey, J. B., Barone, B., Bellingham, J. G., Birch, J. M., Scholin, C. A., & Chavez, F. P. Targeted sampling by autonomous underwater vehicles. Frontiers in Marine Science, 6 (2019): 415, doi:10.3389/fmars.2019.00415., In the vast ocean, many ecologically important phenomena are temporally episodic, localized in space, and move according to local currents. To effectively study these complex and evolving phenomena, methods that enable autonomous platforms to detect and respond to targeted phenomena are required. Such capabilities allow for directed sensing and water sample acquisition in the most relevant and informative locations, as compared against static grid surveys. To meet this need, we have designed algorithms for autonomous underwater vehicles that detect oceanic features in real time and direct vehicle and sampling behaviors as dictated by research objectives. These methods have successfully been applied in a series of field programs to study a range of phenomena such as harmful algal blooms, coastal upwelling fronts, and microbial processes in open-ocean eddies. In this review we highlight these applications and discuss future directions., This work was supported by the David and Lucile Packard Foundation. The 2015 experiment in Monterey Bay was partially supported by NOAA Ecology and Oceanography of Harmful Algal Blooms (ECOHAB) Grant NA11NOS4780030. The 2018 SCOPE Hawaiian Eddy Experiment was partially supported by the National Science Foundation (OCE-0962032 and OCE-1337601), Simons Foundation Grant #329108, the Gordon and Betty Moore Foundation (Grant #3777, #3794, and #2728), and the Schmidt Ocean Institute for R/V Falkor Cruise FK180310. Publication of this paper was funded by the Schmidt Ocean Institute.

The use of autonomous underwater vehicles for ocean research has increased as they have a better cost/performance ratio than crewed oceanographic vessels. For example, autonomous vehicles (e.g. a Wave Glider) can be used to localise and track underwater targets. Whereas other researchers have been focused on target tracking using acoustic modems, here we present a novelty method called area-only target tracking. This method works with commercially available acoustic tags, thereby reducing the costs and complexity over other tracking systems. Moreover, this method can be used to track small targets such as jellyfishes due to the tag's size. The methodology behind the area-only technique is shown, and results from the first field tests conducted in Monterey Bay area are also presented.

Underwater localization is one of the main problems that must be addressed in subsea exploration, where no global positioning system (GPS) is available. In addition to the traditional underwater localization systems, such as long base line (LBL), new methods have been developed to increase the navigation performance and flexibility and to reduce the deployment costs. For example, range-only and single-beacon (ROSB) is based on an autonomous vehicle that localizes and tracks different underwater targets using slant range measurements carried out with acoustic modems. This paper presents different strategies to improve ROSB tracking methods. The ROSB target tracking method can be seen as a hidden Markov model (HMM) problem. Using Bayes' rule, the probability distribution function of the HMM states can be solved by using different filtering methods. Here, we present and compare different methods under different scenarios, both evaluated in simulations and field tests. The main mathematical notation and performance of each algorithm are presented, where best practice has been derived. From a methodological point of view, this paper advanced the understanding of accuracy that can be achieved by using the ROSB target tracking methods with autonomous underwater vehicles

© The Author(s), 2017. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Journal of Field Robotics 35 (2018): 705-716, doi:10.1002/rob.21771., For robots to succeed in complex missions, they must be reliable in the face of subsystem failures and environmental challenges. In this paper, we focus on autonomous underwater vehicle (AUV) autonomy as it pertains to self‐perception and health monitoring, and we argue that automatic classification of state‐sensor data represents an important enabling capability. We apply an online Bayesian nonparametric topic modeling technique to AUV sensor data in order to automatically characterize its performance patterns, then demonstrate how in combination with operator‐supplied semantic labels these patterns can be used for fault detection and diagnosis by means of a nearest‐neighbor classifier. The method is evaluated using data collected by the Monterey Bay Aquarium Research Institute's Tethys long‐range AUV in three separate field deployments. Our results show that the proposed method is able to accurately identify and characterize patterns that correspond to various states of the AUV, and classify faults at a high rate of correct detection with a very low false detection rate., Office of Naval Research Grant Number: N00014‐14‐1‐0199; David and Lucile Packard Foundation

Underwater localization using acoustic signals is one of the main components in a navigation system for an autonomous underwater vehicle (AUV) as a more accurate alternative to dead-reckoning techniques. Although different methods based on the idea of multiple beacons have been studied, other approaches use only one beacon, which reduces the system’s costs and deployment complexity. The inverse approach for single-beacon navigation is to use this method for target localization by an underwater or surface vehicle. In this paper, a method of range-only target localization using a Wave Glider is presented, for which simulations and sea tests have been conducted to determine optimal parameters to minimize acoustic energy use and search time, and to maximize location accuracy and precision. Finally, a field mission is presented, where a Benthic Rover (an autonomous seafloor vehicle) is localized and tracked using minimal human intervention. This mission shows, as an example, the power of using autonomous vehicles in collaboration for oceanographic research

© The Author(s), 2015. This article is distributed under the terms of the Creative Commons Attribution License. The definitive version was published in Journal of Field Robotics 33 (2016): 67-81, doi:10.1002/rob.21617, Coastal upwelling is a wind-driven ocean process that brings cooler, saltier, and nutrient-rich deep water upward to the surface. The boundary between the upwelling water and the normally stratified water is called the “upwelling front.” Upwelling fronts support enriched phytoplankton and zooplankton populations, thus they have great influences on ocean ecosystems. Traditional ship-based methods for detecting and sampling ocean fronts are often laborious and very difficult, and long-term tracking of such dynamic features is practically impossible. In our prior work, we developed a method of using an autonomous underwater vehicle (AUV) to autonomously detect an upwelling front and track the front's movement on a fixed latitude, and we applied the method in scientific experiments. In this paper, we present an extension of the method. Each time the AUV crosses and detects the front, the vehicle makes a turn at an oblique angle to recross the front, thus zigzagging through the front to map the frontal zone. The AUV's zigzag tracks alternate in northward and southward sweeps, so as to track the front as it moves over time. This way, the AUV maps and tracks the front in four dimensions—vertical, cross-front, along-front, and time. From May 29 to June 4, 2013, the Tethys long-range AUV ran the algorithm to map and track an upwelling front in Monterey Bay, CA, over five and one-half days. The tracking revealed spatial and temporal variabilities of the upwelling front., This work was supported by the David and Lucile Packard Foundation.

Nowadays, the use of autonomous vehicles for ocean research has increased, since these vehicles have a better cost/performance ratio than crewed vessels or oceanographic ships. For example, autonomous surface vehicles can be used to localize underwater targets. This paper describes a mission to find a crawling robot - Benthic Rover - on the abyssal plain in the north eastern Pacific, using single-beacon localization from onboard a Wave Glider autonomous surface vehicle. While the Wave Glider is moving around the surface in the target zone, it takes ranges between the target and itself using acoustic modems. With these ranges it can compute the target location, as a Long Baseline (LBL) system. The benefit of this approach is the reduction of cost and complexity relative to deployment of a traditional shipboard LBL system. Additionally, this is a mobile system, and can cover long distances, and can geolocate multiple targets over a large area., Peer Reviewed