Source: UNIVERSITY OF VIRGINIA submitted to NRP
COLLABORATIVE RESEARCH: NRI: OCEAN-POWERED ROBOTS FOR AUTONOMOUS OFFSHORE AQUACULTURE
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1027622
Grant No.
2021-67021-35975
Cumulative Award Amt.
$250,000.00
Proposal No.
2021-10978
Multistate No.
(N/A)
Project Start Date
Sep 1, 2021
Project End Date
Aug 31, 2025
Grant Year
2021
Program Code
[A7301]- National Robotics Initiative
Recipient Organization
UNIVERSITY OF VIRGINIA
(N/A)
CHARLOTTESVILLE,VA 22901
Performing Department
(N/A)
Non Technical Summary
According to the United Nations Food and Agriculture Organization (FAO), 76% of the 600 fish stocks they track are either fully exploited, overexploited, or depleted. The fishing industry has been forced to turn to aquaculture, an alternative means ofproduction. Aquaculture has been the fastest-growing source of animal protein since 1990. In 2018, global fish productionreached 179 million tons, 45% of which, valued at $250 billion, came from aquaculture production. The FAO has predicted that by 2030 the global aquaculture market is projected to be more than $529 billion. Aquaculture production has increaseddramatically to five times the level of production in the 1990s, while wild fish production has decreased. Compared with coastalinshore aquaculture, offshore fish farming may yield 10-100 times the fish production, and is an effective strategy to decreasepressure on wild natural resources, with lower environmental impact.However, the U.S. globally remains a relatively minor aquaculture producer, ranked 16th in 2018 on a global scale, although it isthe leading global consumer of aquaculture products, importing 90% of its seafood from abroad. Blue Ocean Mariculture Inc.(HI) remains the only ffshore aquaculture farm in operation in the USA. In 2020, Presidential Executive Order 13921 onseafood was signed to boost the infant US aquaculture industry. This Executive Order prioritized aquaculture development inocean waters, especially in the U.S. exclusive economic zone. It is pivotal to develop and implement advanced technologies toaccelerate domestic aquacultural production.Huge challenges exist for offshore aquaculture operations and maintenance. Preliminary research was performed by the projectteam by interviewing offshore fish farm industry leaders and reviewing the available literature on offshore fish farms. Among themost labor-intensive and high-risk tasks are cleaning and dead fish removal. According to Open Blue Sea Farms, Inc., humandivers are sent to fish pens to remove dead fish every day, since dead fish will attract sharks and marine mammals, anddamage the fish pens. They also conduct monthly cleaning work to mitigate biofouling of the metal fencing comprising the pens.Needless to say, the offshore environment is subject to high waves and more aggressive sea creatures, which increase both theworkload and danger of human divers.These challenges motivate us to develop a sustainably powered autonomous robotic system, including both an autonomoussurface vehicle (ASV) and an autonomous variant of a tethered remotely operated vehicle (which we term AROV), to improvethe operation, maintenance, and monitoring processes and increase overall fish production at offshore fish farms.
Animal Health Component
50%
Research Effort Categories
Basic
0%
Applied
50%
Developmental
50%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
6017210202060%
6017210208040%
Goals / Objectives
Broad Goals of the Collaborative ProjectThe United States is a minor aquaculture producer, while it is the leading global importer of fish and fishery products. The aim ofthis project is to develop an autonomous, ocean-powered robotic system to support offshore fish farming, which opens newpossibilities for advanced technology to aid the proliferation of domestic aquaculture. To operate sustainably far offshore, thesystem will power itself by harvesting energy from ocean waves. An autonomous surface vehicle (ASV) will serve a dualpurpose as both a mission management system and a wave energy converter. The ASV will deploy an autonomousunderwater robot embodied in a tethered, omnidirectional remotely operated vehicle (ROV), which will be termed anautonomous remotely operated vehicle (AROV). The AROV will address two of the most labor-intensive and time-consumingtasks associated with the maintenance and upkeep of an offshore fish farm: persistently cleaning fish pen fencing to avoid theaccumulation of biofouling, and frequently removing dead fish from the mort traps at the bottom of their pens. The ASV's noveldesign and energy-efficient control will permit it to harvest energy while supporting the AROV's subsea tasking. The AROV,equipped with continuum manipulators and novel sonar-based perceptual capabilities, will be uniquely capable of conservingenergy while cleaning by climbing along a fish pen's fencing, minimizing its use of thrusters. The project will be executed by anexperienced team with diverse expertise spanning wave energy harvesting, target tracking and control, grasping andmanipulation, underwater perception and navigation, and fisheries ecology and management.Goals of the UVA TeamFurukawa will lead the project as the Lead PD. Furukawa's first goal is toadminister the project as the Lead PD and lead the project successfully to produce the expected outcomes. Technically, the UVA team will be in charge of the development and navigation of the ASV. The ASV will serve as a wave energy converter (WEC) to harvest ocean energy. Since the WEC will be designed at Virginia Tech (VT), the UVA team will collaborate with the VT team and complete the development of the ASV. The mission of the ASV is to allow the AROV to localize and navigate itself autonomously to perform both dead fish removal and cleaning path following. Since the Stevens Institute of Technology (Stevens) team will develop the AROV and its autonomous navigation scheme, the UVA team will collaborate with the Stevens team and develop an autonomous navigation scheme for the ASV.
Project Methods
Overall MethodThe entire system will consist of an ASV and an AROV. The ASV will serve as a WECto harvest oceanenergy, supplying power to itself and to the AROV for its mission. The AROV will localize and navigate itself autonomously to performboth dead fish removal, and to execute cleaning paths along the boundary of the fish pens (typically30-70 meters across). The AROV will perform frequent, light-duty cleaning of the fish pen fencingand dead fish collection with unprecedented energy efficiency, using low-cost modular continuummanipulators to grip the surrounding fencing. As the AROV may travel beyond the length of itstether cable, the ASV performs "mild" autonomous following.Each of the proposed technologies isstrongly coupled with the expertise of a PD. For ASV development, PD Zuo will work on power generation with his ocean energy harvesting expertise. PD Englot will work on AROV navigation with his underwater vehicle autonomy expertise. Co-PDWang will work on AROV manipulation with his robotic armdesign expertise. PD Furukawa will work on ASV navigation with his multi-robot cooperation expertise. Co-PD Jiaowill work on technology assessment with her fishery quantitative assessment expertise. Senior Personnel (SP) Cordero willserve as Chair of an Industry Advisory Board (IAB) formedwith representatives of leading fish farm companies. Furukawa will serve as the Lead PD for thisproject due to his robotics research experience over two decades and extensive experience in leadingindustry-funded projects.ASV Integration and Navigation (UVA Team)The proposed WEC simultaneously serves as an ASV, harvesting energy both to operate the AROV, and to control the ASV as it follows theAROV with minimal energy consumption. The state constraint to satisfyis that the distance between the ASV and the AROV must be less thana threshold distance, which is well less than the maximum cable length.The proposed control scheme is as follows. If the AROV is on afence-cleaning or dead fish collection mission, the ASV will be connectedto a rail installed along the circumference of the fish pen while keepingoriented tangentially to the circumference. The ASV is stationary unlessthe ASV-AROV distance exceeds the threshold, in which case it moves further along the circumference. This kinematic constraint simplifies theASV control by making it a 1-D problem.The ASV control becomes fully two-dimensional when the AROVmoves from one fish pen to another. Given the state of the AROV at the current step, the state of the AROV at future steps can be probabilistically predicted using the Chapman-Kolmogorov equation. In order tominimize the energy consumption, the simple solution proposed for ASV control is linear motionto the peak of the belief at the predefined multi-step lookahead, based on receding horizon control (RHC). This simple linear servoing can be revisited as a non-trivial control solution when energyefficiency is concerned. It is to be noted that the control scheme can also be used for engaging witha target, such as a fish pen's rail, if the target can be localized by the ASV.Lastly, the localization of the ASV as well as its target should be discussed. Simultaneouslocalization of the ASV and any target is classified as a tracking-and-localization (TAL) problem andhas been solved previously using an Extended Kalman Filter (EKF) based multi-stage localizationtechnique by PD Furukawa. Since it is on the surface, the ASV will be equipped withstandard localization sensors including a GPS, an IMU, a LiDAR and a stereo camera, which areneeded for the proposed multi-stage localization technique. The proposed technique will be usedto estimate the ASV pose and the target pose.

Progress 09/01/23 to 08/31/24

Outputs
Target Audience:The primary audience targeted in the first year was the project's Industry Advisory Board (IAB) that consists of industry partners including InnovaSea, BlueOcean Mariculture, OpenBlue, Ocean Era, E-Wave Technologies, and NEC Corporation. These are the companies that are most known and working most intensively in the field of automated aquaculture. Therefore, it is most important that these companies remain as the IAB members, and wehost annual meetings with them. The second audience targeted in the third year was the other aquaculture community and the robotics community. Our technologies are being developed for the aquaculture industry, so we should have the IAB open to the other aquaculture-related companies and explore more opportunities to advance automated aquaculture technology. For the robotics community, aquaculture is relatively an untouched area. Making the community aware of the importance of automated aquaculture is one of our missions, so the robotics community is included in our targeted audience. Changes/Problems:There are no major changes/problems in approach. The only minor change made for the UVA team is that the UVA team is that the UVA team is conducting experimentation independently after developing an integrated ASV-AROV system. However, the team will integrate its estimation and control system soon for the AROV at SIT and the WEC system at UM when all the systems are ready for integration. What opportunities for training and professional development has the project provided?The UVA team consists of not only graduate students but also undergraduate research students and senior students forming a capstone project. For all the undergraduate students, an opportunity to learn Robot Operating System and its packages, which allow the robot software development with minimum efforts, was provided. In addition, I provided MAE4260 Robotic Autonomy and MAE4270 Experimental Robotics courses to undergraduate research students who are participating in the project to develop autonomy for the ASV-AROV system. Senior students in the capstone team have opportunities to communicate with the project team to identify customer needs. Graduate students and undergraduate research students have opportunities to participate in the monthly meetings of the project team and the bi-annual meetings where the Industry Advisory Board members participate. How have the results been disseminated to communities of interest?We have an Industry Advisory Board (IAB) that consists of industry partners including InnovaSea, BlueOcean Mariculture, OpenBlue, Ocean Era, E-Wave Technologies, and NEC Corporation. Since we have annual meetings with them, we first disseminate our results to these companies. IAB is open to other companies. We are continuously finding opportunities to communicate with other companies, and on November 25, 2024, Mitsubishi Corporation visited UVA with strong interest in our project. As a team, we made a few plans to disseminate our project externally. One is Aquaculture America, which is to be held on March 6-10, 2025 in New Orleans. The other was IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) where we are planning to propose a workshop on Marine Robotics. What do you plan to do during the next reporting period to accomplish the goals?After receiving the one-year no-cost extension, the team is planning to integrate their systems so that the planned goal can be achieved. The base system will be the University of Virginia (UVA)'s ASV and AROV. University of Michigan's Wave Energy Converter (WEC) will be implemented into the ASV for energy harvesting. Stevens Institute of Technology's manipulator and localization and navigation algorithms will be implemented into the AROV. UVA will develop autonomous navigation to ASV and AROV for cooperative navigation. The integrated system will be finally demonstrated in an open water environment.

Impacts
What was accomplished under these goals? Furukawa has administered the project as the Lead PD and led the project successfully to complete the third year following the third year. Technically, the UVA team is in charge of the autonomous cooperation of an ASV and an AROV. In the third yea, the team enhanced the two hardware systems: An ASV with the maximum dimension less than 3 feet, which was fully custom-designed and developed by the UVA team and can be controlled wirelessly. The ASV is now equipped with a sonar and GPS such that it can be autonomously driven while localizing itself and the AROV An AROV with a BlueROV2 as a base platform, which has an end-effector for net cleaning. The AROV is now equipped with a camera so that it can localize itself with respect to the ASV. The team also developed a new ASV for general use, incorporates the architecture of the original ASV. In addition, the team worked on particularly on the cooperative localization of the ASV and AROV: The ASV and the AROV are each equipped with a sonar and a camera and cooperatively localizes the AROV. A simulator incorporates dynamics models of the ASV and AROV so that the localization can be achieved accurately.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2024 Citation: Gilchrist Johnson and Tomonari Furukawa, Autonomous Platooning of General Connected Vehicles Using Bayesian Receding Horizon Control, Vehicular 2024, March 10-14, 2024, Athens, Greece, 2024
  • Type: Conference Papers and Presentations Status: Under Review Year Published: 2025 Citation: Malori Oxford, Nathan Vu, Brendan Englot and Tomonari Furukawa, Cooperative Localization of Heterogeneous System of Marine Robots Using Hybrid EKF and Grid-Based Method, 2025 IEEE International Conference on Robotics and Automation, May 19-23, 2025, Atlanta, 2025
  • Type: Conference Papers and Presentations Status: Awaiting Publication Year Published: 2025 Citation: Tomonari Furukawa, Long Wang, Brendan Englot and Lei Zuo, Light-duty Autonomous Underwater Robot for Offshore Aquaculture, Aquaculture America, March 6-10, 2025, New Orleans, 2025
  • Type: Conference Papers and Presentations Status: Awaiting Publication Year Published: 2025 Citation: Tomonari Furukawa, Malori Oxford and Nathan Vu, Cooperative Localization and Navigation of Unmanned Underwater and Surface Vehicles for Open Ocean Aquaculture, Aquaculture America, March 6-10, 2025, New Orleans, 2025


Progress 09/01/22 to 08/31/23

Outputs
Target Audience:The primary audience targeted in the first year was the project's Industry Advisory Board (IAB) that consists of industry partners including InnovaSea, BlueOcean Mariculture, OpenBlue, Ocean Era, E-Wave Technologies, and NEC Corporation. These are the companies that are most known and working most intensively in the field of automated aquaculture. Therefore, it is most important that these companies remain as the IAB members. For this reason, we began to host bi-annual meetings with them. The second audience targeted in the first year was the other aquaculture community and the robotics community. Our technologies are being developed for the aquaculture industry, so we should have the IAB open to the other aquaculture-related companies and explore more opportunities to advance automated aquaculture technology. For the robotics community, aquaculture is relatively an untouched area. Making the community aware of the importance of automated aquaculture is one of our missions, so the robotics community is included in our targeted audience. Changes/Problems:There are no major changes/problems in approach. The only minor change made for the UVA team is that the UVA team is that the UVA team is conducting experimentation independently after developing an integrated ASV-AROV system. However, the team will integrate its estimation and control system soon for the AROV at SIT and the WEC system at UM when all the systems are ready for integration. What opportunities for training and professional development has the project provided?The UVA team consists of not only graduate students but also undergraduate research students and senior students forming a capstone project. For all the undergraduate students, an opportunity to learn Robot Operating System and its packages, which allow the robot software development with minimum efforts, was provided. In addition, I provided MAE4260 Robotic Autonomy and MAE4270 Experimental Robotics courses to undergraduate research students who are participating in the project to develop autonomy for the ASV-AROV system. Senior students in the capstone team have opportunities to communicate with the project team to identify customer needs. Graduate students and undergraduate research students have opportunities to participate in the monthly meetings of the project team and the bi-annual meetings where the Industry Advisory Board members participate. How have the results been disseminated to communities of interest?We have an Industry Advisory Board (IAB) that consists of industry partners including InnovaSea, BlueOcean Mariculture, OpenBlue, Ocean Era, E-Wave Technologies, and NEC Corporation. Since we have bi-annual meetings with them, we first disseminate our results to these companies. IAB is open to other companies. We are continuously finding opportunities to communicate with other companies. As a team, we made a few plans to disseminate our project externally. One is Aquaculture America, which is to be held on February 23-26, 2023 in New Orleans. I gave an invited talk at the conference as the Lead PD and try to interact with the audience. The other was IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) where we proposed a workshop on Marine Robotics. Unfortunately our proposal was rejected, but we are finding another opportunity to disseminate our results to both the aquaculture and the robotics communities. What do you plan to do during the next reporting period to accomplish the goals?As planned in the proposal, we plan to develop technologies that allow the fish net cleaning and the dead fish removal processes by the ASV-AROV system. Since a prototype ASV-AROV system and a simulator were developed in the second year, the UVA team will focus on completing and demonstrating an autonomous navigation strategy to move from one fish pen to another. In addition, the team will design and develop other essentials for the ASV that have not been completed in the pastyears. These include the development of a gripper which locks the ASV to the fish pen since the aim of the ASV is to harvest energy through the Wave Energy Converter (WEC) and minimize energy usage for itself. Close collaboration with Stevens Institute of Technology (SIT) for the ASV-AROV system and that with University of Michigan (UM) for the integration of ASV and WEC will be performed continuously. SIT works on the AROV whereas UM works on WEC. Collaboration with them is essential for the completion of the project.

Impacts
What was accomplished under these goals? Furukawa has administered the project as the Lead PD and led the project successfully to complete the second year following the first year. Technically, the UVA team is in charge of the autonomous cooperation of an ASV and an AROV. The team developed the following two hardware systems: A prototype ASV with the maximum dimension less than 3 feet, which was fully custom-designed and developed by the UVA team and can be controlled wirelessly A prototype AROV with a BlueROV2 as a base platform, which has an end-effector for net cleaning In addition, the team worked on estimation, control and software: A set of possible estimation and control approaches that allow an ASV and an AROV to move from one fish pen to another A simulator that validates the autonomous behavior of an ASV and an AROV in an offshore fish farm environment including fish pens and fish

Publications

  • Type: Theses/Dissertations Status: Published Year Published: 2023 Citation: Kristen Babel, Alvaro Crisanto, Brian Richard, Peter Stauffer and Charlie Tilney-Volk, Design of a Co-Navigational Aquaculture Vehicle System, University of Virginia, 2023.


Progress 09/01/21 to 08/31/22

Outputs
Target Audience:The primary audience targeted in the first year was the project's Industry Advisory Board (IAB) that consists of industry partners including InnovaSea, BlueOcean Mariculture, OpenBlue, Ocean Era, E-Wave Technologies, and NEC Corporation. These are the companies that are most known and working most intensively in the field of automated aquaculture. Therefore, it is most important that these companies remain as the IAB members. For this reason, we began to host bi-annual meetings with them. The second audience targeted in the first year was the other aquaculture community and the robotics community. Our technologies are being developed for the aquaculture industry, so we should have the IAB open to the other aquaculture-related companies and explore more opportunities to advance automated aquaculture technology. For the robotics community, aquaculture is relatively an untouched area. Making the community aware of the importance of automated aquaculture is one of our missions, so the robotics community is included in our targeted audience. Changes/Problems:There are no major changes/problems in approach. The only minor change made for the UVA team is that the UVA team is now developing its own prototype ASV-AROV system. The WEC system is developed at UM whereas the AROV is developed at SIT. In order that the UVA team can make a progress on ASV-AROV collaboration, the team has decided to develop a prototype system in addition to creating simulation environments. What opportunities for training and professional development has the project provided?The UVA team consists of not only graduate students but also undergraduate research students and senior students forming a capstone project. For all the undergraduate students, an opportunity to learn Robot Operating System and its packages, which allow the robot software development with minimum efforts, was provided. In addition, I provided MAE4260 Robotic Autonomy and MAE4270 Experimental Robotics course to undergraduate research students who are participating in the project to develop autonomy for the ASV-AROV system. Senior students in the capstone team have opportunities to communicate with the project team to identify customer needs. Graduate students and undergraduate research students have opportunities to participate in the monthly meetings of the project team and the bi-annual meetings where the Industry Advisory Board members participate. How have the results been disseminated to communities of interest?We have an Industry Advisory Board (IAB) that consists of industry partners including InnovaSea, BlueOcean Mariculture, OpenBlue, Ocean Era, E-Wave Technologies, and NEC Corporation. Since we have bi-annual meetings with them, we first disseminate our results to these companies. IAB is open to other companies. We are continuously finding opportunities to communicate with other companies. As a team, we have made a few plans to disseminate our project externally. One is Aquaculture America, which is to be held on February 23-26, 2023 in New Orleans. I will give an invited talk at the conference as the Lead PD and try to interact with the audience. The other is IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). We plan to host a workshop on Marine Robotics in this robotics conference and present a few papers from our team. We plan to disseminate our results to both the aquaculture and the robotics communities. What do you plan to do during the next reporting period to accomplish the goals?As planned in the proposal, we plan to develop technologies that allow the fish net cleaning and the dead fish removal processes by the ASV-AROV system. For the purpose, the UVA team will focus on completing the prototype ASV-AROV system and its autonomous navigation strategy to move from one fish pen to another. In addition, the team will design and develop any other essentials for the ASV. These include the development of a gripper which locks the ASV to the fish pen since the aim of the ASV is to harvest energy through the Wave Energy Converter (WEC) and minimize energy usage for itself. Collaboration with Stevens Institute of Technology (SIT) for the ASV-AROV system and that with University of Michigan (UM) for the integration of ASV and WEC will be performed more intensively. SIT works on the AROV whereas UM works on WEC. Collaboration with them is essential for the completion of the project.

Impacts
What was accomplished under these goals? Furukawa has administered the project as the Lead PD and led the project successfully to complete the first year. Technically, the UVA team came up with the design of an ASV that has a similar size as the BlueROV2. By developing it and integrated with the BlueROV2,a prototype ASV-AROV system, which can be used for experimenting the autonomous navigation of the ASV-AROV system from one fish pen to another. In addition, the UVA team has developed a simulation environment with the components of ASV, AROV and fish pen. This allows the navigation capability of the ASV-AROV system to be tested without the actual experiments. Techniques for the ASV to track the AROV have also been explored.

Publications

  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2023 Citation: Tomonari Furukawa, Ocean-Powered Robots for Autonomous Offshore Aquaculture, Aquaculture America 2023, February 23-26, 2023, New Orleans, Louisiana USA